A DISSERTATION SUBMITTED TO THE FACULTY OF THE GRADUATE SCHOOL OF THE UNIVERSITY OF MINNESOTA BY Paul Lesperance IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY Professor Andrea Berlin August 2010 © Paul Lesperance, 2010 i Acknowledgements I have benefitted greatly from the aid and support of many people and organizations during the writing of this dissertation. I would especially like to thank my advisor, Professor Andrea Berlin, for all her help and advice at all stages of the production process as well as for suggesting the topic to me in the first place. I would also like to thank all the members of my dissertation committee (Professor Susan Herbert of the University of Michigan, as well as Professors Philip Sellew and Nita Krevans of the University of Minnesota) for all their help and support. During the writing process, I benefitted greatly from a George A. Barton fellowship to the W. F. Albright Institute of Archaeological Research in Jerusalem in the fall of 2009. I would like to thank the fellowship committee for giving me such a wonderful and productive opportunity that helped me greatly in this endeavour as well as the staff of the Albright for their aid and support. I would also like to thank both Dr. Donald Ariel of the Israel Antiquities Authority for his aid in getting access to the material and his valuable advice in ways of looking at it and Peter Stone of the University of Cincinnati whose discussions on his work on the pottery from Kedesh helped to illuminate various curious aspects of my own. Any errors are, of course, my own. ii Table of Contents List of Illustrations (Maps, Charts & Figures) iii Chapter 1: Introduction 1 Chapter 2: Other Archives 30 Chapter 3: The Catalogue 66 Anchors 76 Bucrania 91 Caduceus 117 Columns 127 Composites 135 Cornucopias 152 Ears 164 Floral Motifs 174 A Ligature 196 Lyres 199 Masks 212 Pilei of the Dioscuri 245 Shields 271 Stars 284 Thunderbolts 295 Vessels 320 Unknowns 345 Chapter 4: The Geographic Distribution of the Kedesh Symbol Motifs 351 Chapter 5: Observations on the Nature and Possible Meanings of the Symbols of the Kedesh Sealings 388 Chapter 6: Conclusion 414 Bibliography 418 Appendix The Kedesh Symbol Artist 445 iii List of Illustrations (Maps, Charts & Figures)* Maps: Number Description Page 2.1 The archives in the Ancient World 60 3.1 A map of the regions in the catalogue 73 3.2 The distribution of the plain anchor in the Persian period 87 3.3 The distribution of the plain anchor in the Hellenistic period 88 3.4 The distribution of anchors with horsehead protomes in the Hellenistic period 89 3.5 The distribution of anchors with dolphins in the Hellenistic period 90 3.6 The distribution of plain bucrania in the Persian period 105 3.6a Detail map showing the Aegean region 106 3.7 The distribution of bucrania with objects in the Persian period 107 3.7a Detail map showing the Aegean region 108 3.8 The distribution of plain bucrania in the Hellenistic period 110 3.8a Detail map showing the Aegean region 111 3.8b Detail map showing Italy 112 3.9 The distribution of bucrania with objects in the Hellenistic period 114 3.9a Detail map showing the Aegean region 115 3.9b Detail map showing Italy 116 3.10 The distribution of caducei in the Persian period 125 3.11 The distribution of caducei in the Hellenistic period 126 3.12 The distribution of columns in the Persian period 133 3.13 The distribution of columns in the Hellenistic period 134 3.14 The distribution of janiform heads in the Persian period 148 3.15 The distribution of janiform heads in the Hellenistic period 149 * Please note that this list does not include the photographs and line drawings of each individual seal that are integrated into that seal’s catalogue entry. iv 3.16 The distribution of composite eagles in the Hellenistic period 150 3.17 The distribution of composite hippocamps in the Hellenistic period 151 3.18 The distribution of possible cornucopias in the Persian period 162 3.19 The distribution of cornucopias in the Hellenistic period 163 3.20 The distribution of ears in the Persian period 171 3.21 The distribution of ears in the Hellenistic period 172 3.22 The distribution of ears with hands in the Hellenistic period 173 3.23 The distribution of flowers in the Persian period 190 3.24 The distribution of flowers in the Hellenistic period 191 3.25 The distribution of (eight-petal) rosettes in the Persian period 192 3.25a Detail map showing the Aegean region 193 3.26 The distribution of (eight-petal) rosettes in the Hellenistic period 194 3.26a Detail map showing the Aegean region 195 3.27 The distribution of ligatures in the Hellenistic period 198 3.28 The distribution of lyres in the Persian period 207 3.28a Detail map showing the Aegean region 208 3.29 The distribution of lyres in the Hellenistic period 209 3.29a Detail map showing the Aegean region 210 3.29b Detail map showing Italy 211 3.30 The distribution of New Comedy slave masks 237 3.30a Detail map showing the Aegean region 238 3.30b Detail map showing Italy 239 3.31 The distribution of satyr masks in the Persian period 240 3.32 The distribution of satyr masks in the Hellenistic period 241 3.32a Detail map showing the Aegean region 242 3.33 The distribution of Phoenician masks in the Persian period 243 v 3.34 The distribution of Phoenician masks in the Hellenistic period 244 3.35 The distribution of pilei in the Persian period 265 3.36 The distribution of pilei flanking an item in the Persian period 266 3.37 The distribution of pilei in the Hellenistic period 267 3.37a Detail map showing the Aegean region 268 3.38 The distribution of pilei flanking an item in the Hellenistic period 269 3.38a Detail map showing the Aegean region 270 3.39 The distribution of oval shields in the Persian period 280 3.39a Detail map showing the Aegean region 281 3.40 The distribution of oval shields in the Hellenistic period 282 3.40 Detail map showing the Aegean region 283 3.41 The distribution of single stars in the Persian period 291 3.41a Detail map showing the Aegean region 292 3.42 The distribution of single stars in the Hellenistic period 293 3.42a Detail map showing the Aegean region 294 3.43 The distribution of plain thunderbolts in the Persian period 308 3.43a Detail map showing the Aegean region 309 3.43b Detail map showing Italy 310 3.44 The distribution of winged thunderbolts in the Persian period 311 3.44a Detail map showing the Aegean region 312 3.44b Detail map showing Italy 313 3.45 The distribution of plain thunderbolts in the Hellenistic period 314 3.45a Detail map showing the Aegean region 315 3.45b Detail map showing Italy 316 3.46 The distribution of winged thunderbolts in the Hellenistic period 317 3.46a Detail map showing the Aegean region 318 3.46b Detail map showing Italy 319 3.47 The distribution of kantharoi in the Persian period 333 3.47a Detail map showing the Aegean region 334 3.48 The distribution of kantharoi in the Hellenistic period 335 3.48a Detail map showing the Aegean region 336 vi 3.49 The distribution of Panathenaic amphora with Palm branch in the Persian period 338 3.49a Detail map showing the Aegean region 339 3.50 The distribution of Panathenaic amphora with Palm branch in the Persian period 341 3.50a Detail map showing the Aegean region 342 3.50b Detail map showing Italy 343 3.51 The distribution of lagynoi in the Hellenistic period 344 3.52 The distribution of UNK2 (as scene of slave on altar) in the Hellenistic period 349 3.53 The distribution of UNK1 (as pyramid capped baetyl) in the Persian period 350 4.1 Sites with multiple occurrences in significant regions (more than 10 total occurrences) in the Persian period 358 4.2 Sites with multiple occurrences in significant regions (more than 40 total occurrences) in the Persian period 372 Charts: Number Description Page 3.1 The distribution of the plain anchor in the Persian period 86 3.2 The distribution of the plain anchor in the Hellenistic period 87 3.3 The distribution of anchors with horsehead protomes in the Hellenistic period 88 3.4 The distribution of anchors with dolphins in the Hellenistic period 91 3.5 The distribution of plain bucrania in the Persian period 105 3.6 The distribution of bucrania with objects in the Persian period 107 3.7 The distribution of plain bucrania in the Hellenistic period 110 3.8 The distribution of bucrania with objects in the Hellenistic period 113 3.9 The distribution of caducei in the Persian period 125 vii 3.10 The distribution of caducei in the Hellenistic period 126 3.11 The distribution of columns in the Persian period 132 3.12 The distribution of columns in the Hellenistic period 133 3.13 The distribution of janiform heads in the Persian period 148 3.14 The distribution of janiform heads in the Hellenistic period 149 3.15 The distribution of composite hippocamps in the Hellenistic period 150 3.16 The distribution of possible cornucopias in the Persian period 161 3.17 The distribution of cornucopias in the Hellenistic period 162 3.18 The distribution of ears in the Persian period 170 3.19 The distribution of ears in the Hellenistic period 171 3.20 The distribution of ears with hands in the Hellenistic period 172 3.21 The distribution of flowers in the Persian period 190 3.22 The distribution of flowers in the Hellenistic period 191 3.23 The distribution of (eight-petal) rosettes in the Persian period 192 3.24 The distribution of (eight-petal) rosettes in the Hellenistic period 194 3.25 The distribution of ligatures in the Hellenistic period 197 3.26 The distribution of lyres in the Persian period 206 3.27 The distribution of lyres in the Hellenistic period 208 3.28 The distribution of New Comedy slave masks 237 3.29 The distribution of satyr masks in the Persian period 240 3.30 The distribution of satyr masks in the Hellenistic period 241 3.31 The distribution of Phoenician masks in the Persian period 243 3.32 The distribution of Phoenician masks in the Hellenistic period 244 viii 3.33 The distribution of pilei in the Persian period 264 3.34 The distribution of pilei flanking an item in the Persian period 265 3.35 The distribution of pilei in the Hellenistic period 266 3.36 The distribution of pilei flanking an item in the Hellenistic period 269 3.37 The distribution of oval shields in the Hellenistic period 282 3.38 The distribution of single stars in the Persian period 291 3.39 The distribution of single stars in the Hellenistic period 293 3.40 The distribution of plain thunderbolts in the Persian period 307 3.41 The distribution of winged thunderbolts in the Persian period 311 3.42 The distribution of plain thunderbolts in the Hellenistic period 314 3.43 The distribution of winged thunderbolts in the Hellenistic period 317 3.44 The distribution of kantharoi in the Persian period 333 3.45 The distribution of kantharoi in the Hellenistic period 335 3.46 The distribution of Panathenaic amphora with Palm branch in the Persian period 337 3.47 The distribution of Panathenaic amphora with Palm branch in the Hellenistic period 340 3.48 The distribution of lagynoi in the Hellenistic period 344 3.49 The distribution of UNK2 (as scene of slave on altar) in the Hellenistic period 349 3.50 The distribution of UNK1 (as pyramid capped baetyl) in the Persian period 350 4.1 The total distribution of the symbol motifs in the Persian period 357 4.2 The total distribution of the symbol motifs in the Hellenistic period 371 ix 4.3 The total number of coins in motif categories by century at Athens 381 4.4 The proportion of coin motif categories by century at Athens 381 Figures: Number Description Page 1.1 A plan of the PHAB with rooms highlighted to indicate the sealings findspots 22 3.1 Harpocrates seated on a flower (K99 0015) 93 3.2 A bird sitting atop a column (K99 0136) 127 3.3 Two birds sitting on a basin atop a column (K00 0272) 128 3.4 An Eros riding a hippocamp from Delos 141 3.5 A bird sitting on an ear from Seleucia 165 3.6 A hand holding an ear from Seleucia 166 3.7 Pilei of the Dioscuri flanking a Uraeus from Paphos 250 3.8 A lararium painting from Pompeii showing two lares flanking a bearded genius loci on a columnar altar 253 5.1 The abduction of Ganymede by Zeus in the form of an eagle (K99 0086) 411 App.1 The seals of the Kedesh symbol artist 448 1 Chapter 1: Introduction This dissertation will examine a subset of 125 of the 2043 baked clay sealings or bullae from the site of Kedesh, consisting of images of symbols and inanimate objects. The 2043 bullae were recovered from the site of Kedesh in northern Israel during the summers of 1999 and 2000. The total 2043 bullae represent the remains of what was once a large archive dating to the Hellenistic Period. Originally, the bullae had been fixed to string which bound rolls of papyrus, probably records of a legal, commercial and/or administrative nature. The conflagration that destroyed the papyrus documents of the archive at Kedesh was also responsible for baking the clay of the bullae and thereby preserving them. I will take a moment here to clarify some of my terminology. In the following work, I will be making numerous references to a set of interrelated terms, specifically sealing, bulla, seal, and seal impression, which need to be defined in order to avoid confusion. To begin, a seal is at its basis an instrument that a party uses to create a mark identifying that party for the purposes of indicating the party’s ownership of the marked object or assent to the contents of a document. The mark usually consists of an image and/or inscription that have been carved into the surface of the seal in a technique known as intaglio so that, when impressed into a soft surface, the seal leaves behind a positive or raised image. For my purposes, a seal can include a wide variety of actual instruments such as purpose designed stone seals to metal rings with intaglio designs as long as one of their purposes was to make such an identifying mark. The raised image left by a seal is a seal impression. It represents the actual act of 2 presenting the party’s identity for the purposes of claiming ownership or assent to a documented transaction. The (originally) soft plastic material, usually clay, which bears the impression, is a sealing or bulla. A sealing may have multiple impressions or only one, in which case the sealing and the seal impression become in effect synonymous. At Kedesh, each sealing (or bulla) is fairly small, with the largest in the archive being 23 millimeters in length. Originally, the sealings had been soft clay, but they were baked in a fire turning them into terracotta. In some cases, there are traces on the back of the sealings of the papyrus fibers of the document or of the string that tied the document and to which the sealing was originally attached. Each of the sealings bears a single seal impression made by a seal that carried an image. The impressed images can reach a preserved size of 20 millimeters, though usually they range between 8 and 15 millimeters. There is a wide variety of subjects and motifs to be found on the seal impressions of the Kedesh bullae, including amongst others images of deities, royal portraiture and animals. This work will examine the category of these motives consisting of symbols and inanimate objects. Specifically this work will deal with those impressions of symbols and objects which do not bear inscriptions. Those sealings which have symbols and bear inscriptions have already been published by Ariel and Naveh.1 They will, therefore, not form part of the primary corpus for examination, but will be referred to when appropriate. Excluding these inscribed impressions, there are 125 sealings that depict symbols and objects. 1 Ariel & Naveh, 2003, pp. 61-80. 3 The Definition of a Symbol: Generally, symbols have an extremely important role both in art and in human interaction as a whole. Cassirer, for instance, sees the concept of symbols as being fundamental to human consciousness and interaction.2 Furthermore, for her part, Mary Douglas has seen in the symbol the basis for all human communication.3 Considering this importance, we must take special care special care to delimit and specify what exactly a symbol is. As such, when putting together the corpus of sealings that forms the basis of this work, I have had to consider and define what being a symbol actually means in regards to the images that I am examining. In general, I have steered away from the more theoretical discussions of symbols found in the philosophy of Cassirer and the study of semiotics as being overly broad or inappropriate for the material at hand. After all, for its part, the discipline of semiotics is not so much interested in the sign itself as with examining and defining the process of sign-making.4 We can note that the images of the symbols are all semiotically icons, at least in part, in that they are mostly images of specific objects.5 At the same time, however, the same can be said of all the identifiable images on the Kedesh sealings, not just the symbols. Firth, in his 1973 book on public and private symbols, gives a more anthropological definition of symbols and symbolism, stating: 2 Cassirer, 1979, pp. 190-195. 3 Douglas, 1973, p. 29. 4 Bal, & Bryson, 1991, p. 174. 5 Bal & Bryson, 1991, pp. 189-90. 4 The essence of symbolism lies in the recognition of one thing as standing in for (re-presenting) another, the relation between them normally being that of concrete to abstract, particular to general. The relation is such that the symbol by itself appears capable of generating and receiving effects otherwise reserved for the objects to which it refers.6 Again, the problem with this definition, as with the other theoretical concepts of symbols in semiotics and elsewhere, is that it is too broad, and encompasses too great a scope of possible ideas and motifs for our purposes. Theoretically, under Firth’s definition, we could examine the majority of images from the Kedesh archive, including creatures, deities, and even mythological scenes with symbolic meanings attached to them within this work under the guise of “symbol.” That would, however, be much too elaborate and cumbersome an endeavor for our purposes here. It is necessary, rather, to restrict our definition of symbol and thereby limit the scope of material to examine. The purpose of this dissertation is to examine the images of inanimate objects and symbols that appear on the Kedesh sealings. The material that I am examining is thus roughly equivalent to the modern warning emblems for hazardous materials, like biohazard, explosive, toxic, and corrosive. As such, we need a definition of the term “symbol” in which the symbol is by its nature a symbol or, rather, a symbol qua symbol. I have therefore chosen to define a symbol on a more intuitive and less theoretical level. As such, I believe that most people would recognize the images that I will discuss as symbols without having to have studied semiotic theory. I have based my definition on a number of characteristics, both positive (what a symbol is) and negative (what a 6 Firth, 1973, p. 15. 5 symbol is not). First and most importantly, a symbol is a conceptually complete visual object. The image may be made up of a single element or many elements combined, but it is complete in itself. The depiction may range from the naturalistic representation to the abstract. The symbol does not require external modifiers, such as additional visual elements or information to be complete. For instance, an image of a lyre is complete in and of itself. It is the image of a musical instrument. In certain cases, the object depicted may be part of a larger whole, such as in the case of a human body part. Yet again, even here the human body part, such as an ear, is conceptually complete in itself and separate from the larger entity, referring specifically to the ear and not necessarily the larger person to which it would be attached. This fact raises another characteristic. For our purposes, a symbol is not a creature. It lacks animation. And so the body part of a creature is a symbol, such as a human ear or the head of a bull since they are conceptually complete in themselves, being an ear or a head, but lack the animate force depicted in an image of an entire live creature, like a portrait or the image of a mythological creature. In certain cases, animate force may be present, but only in a problematic fashion. This is the case with variants in the category of composites which do show a creature. Here, however, the creature is constructed in such a way so as to be nonsensical: an eagle with the body formed from a human head or a hippocamp made up of the head of a horse, the body composed of a satyr mask, and the tail in the form of a cornucopia. Such creatures are conceived in such a way so that they do not represent a true creature, even a mythological one, but rather an abstraction in the form of one. They are thus symbols 6 and not intrinsically creatures. The conceptual completeness of a symbol means that it can carry allegorical or analogical meaning but not narrative. A symbol can represent things other than its concrete self. Indeed, this is, according to Firth, the defining concept of the term “symbol” in general. The attribute of a god, for instance, can stand for that god, without requiring the presence of the same divinity to be conceptually complete. On the other hand, a scene from a narrative usually requires both animate beings and concepts that exist outside the scene itself, such as the other portions of the narrative, in order to be complete. In essence, a symbol can refer to something other than itself but does not require that thing for completion. The Symbols on the Kedesh Sealings: Altogether, 125 sealings from Kedesh fall within the parameters of a symbol that I have set. These sealings fall into a one of seventeen motif categories, consisting of one or more variants on a symbol motif. These categories consist of anchors (with 9 sealings), bucrania (8 sealings), caducei (1 sealing), columns (2 sealings), composites (nine sealings), cornucopias (7 sealings), ears (3 sealings), floral motifs (7 sealings), ligatures (1 sealing), masks (31 sealings), pilei of the Dioscuri (23 sealings), shields (4 sealings), thunderbolts (8 sealings), and vessels (5 sealings) as well as a final category of two sealings whose identity is not quite clear. Each of these categories will form the basis of a separate section in the catalogue, allowing us to focus on each motif on its own. These 125 sealings each represent the one-time application of a seal on clay for 7 legal documentary reasons. We do not have any of the original seals that made these impressions. Considering the period, in all likelihood, the original seals came in the form of rings, either completely metal or with an engraved stone.7 Most of the sealings at Kedesh seem to have been the singular use of any particular seal. In a minority, however, the same seal appears to have been responsible for the images on multiple sealings. As such, my discussions within the catalogue will be based on the seal, not the sealing, thereby grouping all uses of the same seal together. The total number of seals that appear to be represented in this corpus is 91. Symbols in Ancient Art: Symbols, as I have defined them, do not exist solely on the sealings at Kedesh. They have a long history of use in ancient art in Greece and the Near East in various media and usages. A brief examination of these appearances is necessary to help contextualize the use of symbols within the overall world of ancient art. To this end, I will examine how symbols were used in different places, namely Greece and the Near East, focusing on the first millennium BCE, but going further back where appropriate. This process is somewhat complicated by the idiosyncratic and restrictive definition of symbol that I am using, since the scholarship that exists may deal only tangentially with what I have defined as “symbols”, even when discussing “symbols” in general. The ancient Near East was a vast area and, in the first millennium as in other periods, was populated by many separate cultures in different regions like the Babylonians, Assyrians, Egyptians, and Phoenicians just to name a few. Each of these 7 Boardman, 1997, p. 78. 8 cultures had its own art and ways of dealing with symbols. As such, scholarship on the subject tends to focus on specific symbols on a culture by culture, motif by motif basis. In Mesopotamia, for instance, there was a longstanding tradition, dating back to the second millennium BCE of using symbols of greater or lesser abstraction to stand in for specific deities. Examples of this phenomenon can be seen on two 12th century kudurrus, or boundary stones, that were found at Susa and that originally came from Babylonia.8 Here, the various gods that oversaw the legal force of the kudurru and its text were represented by means of items or symbols specific to each god that were themselves placed on bases and/or the backs of various creatures also associated with the deities. The practice continued well into the first millennium so that the gods continued to be represented by various symbols on the Esarhaddon stele, on an 8th century boundary stone of the Babylonian king Marduk-apla-iddina II and on Neo-Babylonian stamp seals that show a worshipper before a divine symbol.9 As such, these divine symbols were used both as symbolic motifs within a scene, as when they were confronted by worshippers in Babylonian glyptic, and as completed symbols in themselves as when they appear outside any scene on the kudurru’s. The use of these divine symbols was especially popular in the 6th-5th centuries BCE, after the fall of Assyria and during the subsequent rise of Babylonia, where they appear, for instance, above the figure of the king on a stele of Nabonidus.10 The most popular symbols were the marru, or the schematized spade and stylus of the important Babylonian gods 8 Harper & Amiet, 1992, #115 & #116. 9 Farbridge, 1970, pp. 160 & 162; Ornan, 1995, fig.1. 10 Ornan, 1995, p. 39, fig.8. 9 Marduk and Nabu.11 Other symbols that appeared in seals and on monuments in the first millennium included the stylized sun disk, or nipḫu, that was the symbol of the god Shamash.12 Ornan notes that this practice of representing gods by symbols was not alien to Mesopotamia but was rather a concept running parallel to the more dominant practice of anthropomorphic divine representation throughout Mesopotamian history.13 Indeed, the practice of representing the gods by symbols continued longer than the native empires of Mesopotamia, with some schematized examples appearing on 16 sealings from a Hellenistic archive at the Greek foundation of Seleucia-on-the-Tigris.14 Ancient Egypt, of course, is well known for its use of symbols, mostly in the form of its hieroglyphic writing system, starting from the Old Kingdom and lasting through the Ptolemaic period. Several scholarly works discuss the use of symbols in Egypt. For instance, Clark in his book discusses many symbols in Egyptian art, including the symbols of the djed pillar, the eye, and the representation of the ka, as well as the overall use of symbols.15 According to Clark, the very essence of Egyptian art is a symbolism that is heavily interwoven into Egyptian religion and thereby manages to maintain a deep continuity of meaning over a very considerable amount of time.16 Other scholarship considers specific symbols within the overall Egyptian repertoire. For instance, in a 1982 article, Schwabe, Adams and Hodge examined the ancient Egyptian belief in the spine being the seat of life and the vital force, and then used this to argue that the ankh symbol, which means life in hieroglyphics, originated as a representation of 11 Ornan, 1995, p. 40. 12 Ornan, 1995, p. 45. 13 Ornan, 1995, pp. 43 & 49. 14 Invernizzi, Bollati, & Messino, 2004, #252-266. 15 Clark, 1959. 16 Clark, 1959, pp. 25-7 & 29. 10 a bull’s vertebra.17 For their part, the Phoenicians had an extensive corpus of symbol motifs that appeared in their art. Several of these motifs, such as the eye of Horus, were borrowed from other cultures, including the Egyptians.18 Scholarly treatment of Phoenician symbols tends to be on a case by case basis. One good example of this practice concerns Phoenician style masks which generally appear in funerary or cultic contexts, and on the subject of which several scholars have commented.19 This motif is especially relevant to this work as Phoenician style masks appear on a couple of sealing from Kedesh. As such, the motif will be discussed fully below in the appropriate portion of the catalogue. Another case of the Near-Eastern use of symbols consists of a series of signs, called linear devices, which appear on pyramidal stamp seals from Anatolia in the Persian period, which Boardman discusses in two articles.20 He localizes them in and around Persian period Lydia, noting the presence of Lydian inscriptions on many of the seals.21 These linear devices are highly abstract. They appear to bear some inspiration for their forms from letters, especially Aramaic, but they do not generally resemble any specific letter or monogram of letters. Other forms resemble objects such as an Egyptian cartouche or even a flower. The function of the linear devices on the seals seems to have 17 Schwabe, Adams & Hodge, 1982, pp. 445-479. 18 For examples of Phoenician reuse of the Egyptian motifs like the eye of Horus and the djed pillar see Acquaro, 1988, #245-6, #287, #625, #672, #679, #681, #683-4. These are all amulets in different materials, dating from the 7th to the 4th centuries BCE. 19 For instance see Ciasca, 1988, pp. 354-370; Cintas, 1946; Karageorghis, 1996, pp. 813-821; Piccard, 1965, pp. 7-115. 20 Boardman, 1970b, pp. 19-45; Boardman, 1998, pp. 1-13. 21 Boardman, 1970b, p. 20; Boardman, 1998, p. 6. Boardman notes that linear devices appear just prior to the Persian conquest of Lydia and that they were used on seals concentrated in and around Lydia, especially in western Anatolia. 11 been as personal marks or blazons.22 Indeed, in one case at least, a Lydian inscription explicitly identifies a linear device as just that.23 Boardman notes that the linear devices would have been especially useful in individuating seals whose main figural motifs were very close in actual representation, like the popular figural motifs of Persian glyptic.24 The linear devices were not conceptually part of the seal’s scene but were rather a means per se of conveying identity in possibly confusing contexts. They therefore display the traits of visual objectivity, completeness and allegorical meaning that define symbols for our purposes. Boardman also originally put the origin of this practice in the marks and graffiti put on Greek pottery, but later saw a much closer relationship between the linear devices on the seals and mason’s marks that appear in various places in Anatolia prior to the arrival of the Persians in 547 BCE, including Sardis.25 In Greece, the picture is significantly different. Here, the artistic development has long been fit into a great, overarching scheme of stylistic development and periods. The Geometric period possessed an aesthetic, based almost entirely on abstract designs, that was different from that of the preceding Mycenaean age or the subsequent Orientalizing and Archaic periods. The focus on geometric designs, especially in vase painting, renders the identification of symbols and the examination of how they were used problematic. Courbin, for example, saw Argive Geometric art as having very little symbolic content in any sense of the word, being rather decorative and 22 Boardman, 1998, p. 4. 23 Boardman, 1970b, p. 20, 24 Boardman, 1970b, p. 21. 25 Boardman, 1998, pp. 5-7. 12 representational.26 Others have taken a different approach, seeing symbols with specific meanings in the filler motifs on Geometric vases. Roes, for his part, saw the swastika motif as maintaining its original Indic solar/life connotations in its use on Greek pots.27 Both Hurwitt and Boardman have separately taken more nuanced approaches by stating that some motifs in Geometric art might have had meaning, and therefore risen to the level of possible symbols, in certain contexts. Hurwitt, for instance, notes that the motif of a double triangle might represent a double axe and, therefore, might have ritual connotations in some contexts or on some pots.28 Boardman takes a yet different approach by noting that abstract designs like lines of lozenges and zigzag patterns might stand in for specific items within 8th century Geometric figural scenes, such as a horse’s halter rope or water and thus possibly might maintain some of those meanings even when the motifs are outside the scenes and thereby possess the completeness that in part defines a symbol.29 The 8th century figural scenes on Geometric pots also bore within themselves the seeds of a massive artistic shift away from abstract geometric forms towards an interest in figural representation, especially representations of the human form. Indeed, according to Hurwitt, the work of the Dipylon Master with its incorporation of human scenes into Attic vase painting signaled the start of a shift towards the human figure being the central focus of Greek art.30 This artistic shift is also reflected in a shift of scholarly interest in the same direction towards representations of the human figure. 26 Courbin, 1966, p. 495. 27 Roes, 1933. 28 Hurwitt, 1993, p. 14 (note). 29 Boardman, 1983, pp. 16-22. 30 Hurwitt, 1993, p. 33. 13 Potential symbols did not disappear from vase painting immediately in favor of the purely human scene. Filler motifs, for instance, continued to be used. Indeed, Payne in his 1931 work on Corinthian art notes many different kinds, mostly floral in nature, that appear on Corinthian pottery in the 7th century BCE.31 He does not, however, ascribe to these ornaments any kind of meaning or purpose other than ornamental. Rather, we must turn to a different medium, one preserved only indirectly, in order to see the development of symbol use in Greek art after the Geometric period. This medium consists of shield decoration. This decoration appeared most popularly on a specific type of round shield known as the hoplon. This was the eponymous equipment of the citizen hoplite whose own rise along with the favored phalanx tactic in the late 8th century mirrored the development of the polis.32 While no actual hopla have survived with decoration intact, George Chase has compiled and examined a catalogue of the shield decorations mostly from what appears in scenes on Greek pottery, especially the Attic Black and Red-Figured pottery.33 Chase postulates that there was a high degree of congruence between the shield devices that the artists painted on vases and the ones that actually existed because of the tendency of the artists to depict pieces of arms and equipment familiar to them in their lives and because of the similarity between literary attestations of shield devices, such as the shield scene from Aeschylus’ Seven Against Thebes that Froma Zeitlin has studied in depth, and the ones that are depicted.34 It is among the devices of these shields that we truly first see images that fit with 31 Payne, 1931, pp. 36, 144-5, 147-8, & 156-7. 32 Hurwitt,1993, p. 39. 33 Chase, 1902,pp. 61-127. 34 Zeitlin, 2009. 14 concept of symbols that we have defined above. Indeed, Chase has catalogued several of the motifs that appear at Kedesh as shield devices in scenes pottery, including the anchor, bull’s head, flower, kantharos, rosette, star, and thunderbolt.35 The reason for this is interesting in that it illustrates parallels with the use of symbols on seal(ing)s, such as we have at Kedesh. Like a seal, a hoplon provided a constrained visual space that was more or less two dimensional. It could be viewed in its entirety in a single glance, unlike a pot whose three dimensional form requires someone to shift perspective in order to see the whole of the decoration. Certainly a shield was much bigger than a seal in size, but the image upon a shield would have had to be clear over a much greater distance than that of a seal. Any shield device would need to be both relatively simple visually and complete in the circumscribed area of the shield if the device was to be seen and comprehended at a distance. Such factors would tend to favor the use of symbols as I have defined them for the use as shield devices, since they are relatively simple images that are inherently complete. The purpose of shield devices, like that of a seal, was to individuate and identify the user. The panoply of the hoplite, like that of the later medieval knight which foreshadowed the rise of heraldry, could often obscure the identity of the wearer in battle, especially when using the earlier Corinthian helmet that would have hidden the entirety of the face. Shield devices allowed individual hoplites to recognize one another in order to differentiate friend from foe in a mêlée and maintain the high discipline 35 Chase, 1902, #II, #LX, #LXV, #CXII, #CXIX, #CCXLIV, #CCLIV. The appearance of these motifs here does not constitute their use as symbols, since they are part of a scene on painted pottery and, therefore, lack the completeness that is one of our defining characteristics. As such, they will not be counted among the appearances of these motifs as symbols within the catalogue below. Their presence, however, does point to the possibility of their use as symbols as real devices on real shields. 15 necessary for a phalanx in confusing battlefield situations. Thus the hoplon was a token of the identity of the user just as a seal would be in different situations.36 Indeed, Chase classifies the shield devices into 12 categories including decorative, referents to the cults of gods, referents to polis or polity, referents to family or descent, referents to rank, referents to the deeds and fortunes of the user, referents to personal characteristics, devices meant to inspire fear in the enemy, devices symbolic of the user’s name, devices of purely personal choice or caprice, devices composed of many elements, devices which copy works of art.37 Most of these classifications would work equally well with little modification for describing the choice of motifs in general on seals and not just on shields. Altogether, the shield devices, starting at the end of the Geometric period and lasting through the Classical, provided a medium for the first time in Greek art that was especially conducive to the use of symbols as symbols and not just a symbolic motif potentially integrated into larger whole. Interestingly, the symbols used as shield devices, as with the Lydian linear devices of the Persian period and the Kedesh symbol sealings of the Hellenistic, had the function of individuating a person in specific social contexts, be it battle or legal transactions so that it would appear that symbols from a relatively early period in the Aegean are commonly linked to the matter of identity. Alongside shield devices, another medium in which symbols appear is on 36 See Zeitlin, 2009, pp. 36-7, 42-3, 50-1, 56-7, 62, 67-68. Indeed, according to Zeitlin, Aeschylus plays with the idea of the combatants’ shield devices as a means not only of identifying the combatants but also of characterizing them, their personalities and even their place in the world. For instance, the first attacker, Tydeus, has a device showing a primordial night sky, which fits with his position as the first named attacker. Thus, the shield devices become essential to the identity of the character. 37 Chase, 1902, pp. 91-2. 16 coinage, which first began to appear in the late 7th century in western Anatolia.38 Like the hoplon shield, the field of a coin is both constrained and two dimensional, the kind of characteristics that would once again favor visually simple and inherently complete images like symbols. Indeed, Seltman in his book on Greek coinage explicitly links the motifs that appear on early Athenian coinage to the shield devices of the nobles who minted them and who, according to him, served as the inspiration for the shield devices that appeared on Athenian Black and early Red Figure pottery.39 Certainly several of the Athenian coins of the second half of the sixth century displayed symbols as types, including an amphora, a bull’s head and a four-spoke wheel which Chase also notes among his catalogued shield devices.40 Whatever the case, coinage would continue to play a significant role in the use of symbols through the Classical and Hellenistic periods and beyond. For the Hellenistic period, scholarship that treats symbols synthetically does exist. Onians, for instance, in his book on the Hellenistic Greek world view deals tangentially with the use of symbols through his discussion of the importance of allegory in Hellenistic art.41 Onians specifically focuses on images that appear on tombstones. In other cases, works may track and discuss specific motifs or classes of motifs across multiple media, such as Chapouthier’s work on representations of the Dioscuri which includes cases where they are represented by their helmets or pilei.42 Another example of a category of symbol that has benefitted from scholarly attention consists of the 38 Carradice, 1995, p. 21. 39 Seltman, 1955, pp. 47-8 & fig.7. 40 Kroll, J.H. & A. S. Walker, 1993, #1 & #3-5; Chase, 1902, I, LX, CCLIX. 41 Onians, 1979. 42 Chapouthier, 1935. 17 composite or gryllos, of which there are several examples at Kedesh. Several scholars, including Boardman and Roes, have examined this motif group and discussed its origins and its purpose.43 Still, symbols oftentimes do not receive the same level of scholarly treatment as other kinds of motifs. They do not have the same kind of cachet of interest associated with images of divinities, mythological scenes or portraits of rulers because of the ability of these other types of motifs to link to the narratives of mythology and history. As a result, these other scenes often get pride of place in the scholarly treatments of the various other Hellenistic archives. For instance, the large Hellenistic archive at Delos has seen two main volumes published, one on the images of Eros and another encompassing the public seals, as well as the images of Apollo, Artemis, Hecate, and Helios.44 There has also been a publication dealing with the theatrical scenes, including masks, in general from the archive.45 Sealings bearing images of symbols occasionally appear in these publications, but they are not the focus of any one publication, nor are they treated synthetically as a group. Another set of Hellenistic archives at Seleucia on the Tigris fare somewhat better in that they have been published in their entirety, including the symbols.46 Still, the size and organization of the material means that the symbols are often treated separately in different sections without any attempt to examine the symbols overall or discuss them as a group. Generally, the presence of symbols may be noted and the images described, but more is rarely undertaken. 43 See Boardman, 2003, pp. 123-32; Roes, 1933, pp. 232-5. 44 Boussac, 1992; Stampoulidis, 1992. 45 Boussac, 1997, pp. 145-164. 46 Invernizzi, Bollati,. Messino, & Maolo, 2004; McDowell, 1935. 18 Symbols as a Path for Research: Symbols can offer a great deal of information both as individual motifs and as an overarching category to those willing to examine them. In this regard, the 125 sealings from Kedesh represent a unique opportunity. The corpus is both diverse enough to provide a well fleshed out examination and compact enough to allow such an examination. The importance of these sealings derives in part from the use of the symbols, both functionally as images on a seal and conceptually in the very nature of a symbol as image or motif. There are many levels or avenues to study and explore in this matter. First, I will use these sealings to help to place the Kedesh archive within its larger geographical and chronological context. The Kedesh archive did not exist within a cultural vacuum, but rather as part of a larger whole. The appearance of each motif at Kedesh represents a single point defined by its date, location and medium in the use of that motif. If the motif appeared elsewhere, those occurrences represent similarly defined points that make up a global whole of the use of the motif. We can therefore compare the appearance of a motif in the Kedesh symbol sealings to this larger whole by tracking and plotting the use and appearances of examples of the same motives in a variety of media in other places and times in the ancient world. In doing so we will create a series of diachronic maps of the various Kedesh motifs consisting of all the points by media, located geographically. These points will form in essence a use-constellation of each motif. This information in turn will allow me to both contextualize and characterize the use of symbol motifs at Kedesh. I will examine whether the Kedesh symbol motives 19 appear widely with numerous examples in many places, indicating a high level of integration into the larger iconographic culture, or whether the motives at Kedesh are more idiosyncratic and localized. I will also go beyond the general view of the integration of Kedesh into the wider ancient world, by looking to see if there are links to specific regions. If, for example, a motif that is limited to a specific area, or is most popular in a specific area, appears at Kedesh, then it may indicate a link between the two places. In the same vein, I will examine the use of the motives in a temporal context, seeing if there are motifs that were popular earlier, indicating a conservative taste among the users of the Kedesh archive, or maybe that there are motifs that became popular later in the wider ancient world, representing an innovative element. Once the motifs of the Kedesh symbol sealings have been properly contextualized both in time and in place, I will examine the iconographic functionality of the motives that were used. Some symbols may express political allegiance to specific polities among the Hellenistic kingdoms or cities. Others may express or represent ethnic and/or religious affiliation. An examination of the symbol sealings may also shed light on other characteristics of the images. The Context of Kedesh and its Archive: Before beginning it is necessary to situate those sealings showing symbols within the Kedesh archive as a whole. The site of Kedesh is located in the extreme north of modern-day Israel, some 10 kilometres northwest of the site of Hazor in the Upper Galilee. The site is situated in a shelf like valley overlooking the Hula valley from the east. It also lies on an ancient route between Hazor and Tyre. The site itself consists of a 20 large double tell, or mound that is roughly 30 hectares in area.47 The first known reference to Kedesh comes from a group of 19th century BCE execration texts from Egypt.48 It also appears in a list of Canaanite cities that Seti I captured in the 13th century.49 The site is mentioned several times in various books in the bible. In the Book of Joshuah, Kedesh was originally a Canaanite city that the Israelites conquered and turned into a fortified city of the tribe of Naphtali as well as a “city of refuge” for those guilty of unintentional homicide.50 The Second Book of Kings also relates how the city was captured by the Assyrian king Tiglath-Pileser III around 732 BCE.51 Later texts of the Hellenistic and Roman periods also mention the site. It is mentioned by the 3rd century Zenon papyri as a place where Zenon got supplies and had a bath.52 In the Book of Maccabees, Kedesh was the site of the encampment of the forces loyal to Demetrius II prior to his battle with Jonathan’s forces on the plains of Hazor in the mid second century BCE.53 Josephus, for his part, mentions the site several times in both the Antiquities, where he sites it between Tyre and Galilee and mentions the story of it being a “city of refuge,” and the Wars, where he notes the site on multiple occasions as being a settlement of Tyrians.54 The archive was recovered from the northwest corner of a large rectangular building of Hellenistic date, which measures 56 meters on its east-west axis and 40 on its north-south one. This building lay at the southern edge of the tell. The building’s layout 47 Herbert & Berlin, 2003, p. 13. 48 Herbert & Berlin, 2003, pp. 13-15. 49 Herbert & Berlin, 2003, p. 15. 50 Josh. 12:22, 19:37, 20:37, 21:32. 51 2 Kings 15:29. 52 Herbert & Berlin, 2003, p. 15. 53 1 Macc. 11: 63-74 54 Josephus AJ 5.91, 13.154-62; JW 2.459 & 4.104. 21 consisted of a number of rooms focused around a large paved courtyard in the western third of the building. Among its rooms, besides the archive facilities, the building contained decorated facilities for reception, extensive storage rooms and even bathing facilities.55 Rhodian amphora handle stamps show that the building was active during the first half of the second century BCE, during the period of Seleucid control of the region.56 At the same time, this structure is a modification of a similar building that dates back to the Persian period. The excavators have termed this structure the Persian- Hellenistic Administrative Building (or PHAB). The size of the structure and the similarities of layout between the structure and governors’ palaces at other sites like Nippur, Dura, Lachish and Ai Khanoum have led the excavators to identify it as an administrative structure.57 As such, the structure served in the first half of the second century as the seat of an official who was in charge of the local Seleucid administrative unit, possibly the eparch of Galilee or the strategos of Coele-Syria.58 The PHAB went out of use as an administrative building shortly after 145 BCE, when the Maccabees defeated the Seleucids in a battle at nearby Hazor.59 Shortly after the abandonment, somebody set fire to the archive, and only the archive, possibly as a means of destroying the tax records there. Sometime after the fire, squatters set up residence in the building.60 55 Herbert & Berlin, 2003, pp. 18-38. 56 Herbert & Berlin, 2003, p. 23. 57 Herbert & Berlin, 2003, pp. 19 & 21. 58 Herbert & Berlin, 2003, p. 53. 59 Herbert & Berlin, 2003, p. 15; 1 Macc. 11: 63-74. 60 Herbert and Berlin, 2003, p. 54. 22 As previously mentioned, the evidence for the archive consists of some 2043 burnt clay sealings of which 1765 bear legible impressions, with the others being either too blurry or fragmentary for analysis. These sealings were uncovered in the northwest rooms of the compound (trench CB 4.7/4.8). The room showed signs of burning and roof-collapse above which was a layer of early Roman material, while the material with the sealings dated to the early and mid 2nd century BCE.61 A few of the impressions are definitely of an official or public nature as indicated by their inscriptions or the presence of Seleucid symbols, such as the anchor with horse-head protome. Indeed, one group of nine sealings, representing the use of a single seal, even bears the symbol of the 61 Herbert & Berlin, 2003, pp. 22-24. 23 Phoenician goddess Tanit and an inscription in Aramaic giving the title of an official.62 Also some twenty percent of the sealings bear portraits of which the grand majority are of Seleucid monarchs, some or all of which may derive from official correspondence between the royal court to the administrator in charge of Kedesh. For the rest of the sealings, some seventy five percent bear images from Greek mythology or are human figures. The 125 sealings that show symbols represent six percent of the overall total. This is a small but significant fraction. The symbols at Kedesh were not the most popular group of motifs that appeared in the archives, but they still appeared frequently enough that they provide a platform from which to study how the archive behaved within its larger cultural context. Several different pieces of evidence point to the composition of the site’s population in the Hellenistic period. The inscribed sealings use either Aramaic (for instance, the seals with the sign of Tanit and the official’s title), Greek (for example, a seal bearing the name of the site “ΚΥ∆ΑΣΣΑ” and an image of an ear of wheat and a cluster of grapes) or a combination of the two as with a seal bearing a caduceus and references to the common assembly of Tyre in both Greek and Aramaic.63 As previously noted, the ancient historian Josephus repeatedly refers to Kedesh in the context of Tyre, even claiming that it was a village of the Tyrians.64 This evidence suggests that the population of the site was a mixture of Greek and Phoenician, with Hellenized Phoenicians predominating.65 Thus, Kedesh bears cultural links to both the Hellenic and 62 Ariel & Naveh, 2003, pp. 62-64. 63 Ariel & Naveh, 2003, #1, #2, & #5. 64 Josephus AJ 13.154-62; JW 2.459 & 4.104. 65 Herbert & Berlin, 2003, pp. 50-53. 24 the Phoenician traditions, which we should remember in examining the material of the archive in order to see how these large cultural traditions are reflected in the corpus of Kedesh symbol sealings. Kedesh and the outside World: The archive of Kedesh is one of a select group of Persian and Hellenistic archives that have been uncovered. These archives, including the previously mentioned ones at Delos and Seleucia represent one of the prime sources of material for comparison to the Kedesh symbol motifs. Each of these archives presents us with the impressions of seals preserved in a given time and place. They thus allow us a glimpse into the larger context of the exact medium represented by the Kedesh archive. I will discuss these archives in detail separately in the next chapter. Besides the evidence from the various archives in the Mediterranean and the Near East, many other media also demonstrate instances of the images found on the Kedesh symbols sealings, providing valuable points of congruence for the examination of the sealings. These media include instances in gems and rings, coins, vessels, sculpture and mosaics. Engraved Gems and Rings: In a way, the evidence for engraved gems and rings does not constitute a truly separate medium from that of the sealings, since it was by impressing examples of the former into clay that the latter were created. The two media are in fact reflections of one another. Unlike sealings, however, engraved gems and rings are often inherently durable, 25 surviving quite well, independent of the vagaries of nature such as the fire necessary to bake the clay of the sealings. The seals themselves, therefore, often survive much better than the sealings they produced. As such, the seals represent the more complete corpus of motifs, which were being used at a given time. The flip side of this is that, whereas sealings are usually recovered archaeologically with specific findspots and dates noted, the same can not always be said for the seals. The inherent value of the engraved gems and rings means that they could move quite fluidly through many centuries and collections after their initial discovery. As such, their original context is often unclear. Still, the gems and rings can be of use. They provide a good broad basis of motifs, especially certain ones which may be rare otherwise in the archaeological record. Their stylistic aspects have also allowed scholars to provide dates for them even when specific contexts are lacking. They are usually found in the very numerous publications of various collections, both of museums and of private individuals.66 There are also several large scholarly works which present the gems and rings from multiple collections, notably the seminal work of Furtwängler as well as those of Boardman, and Richter.67 In some, unfortunately rare, cases the findspots of the gems or rings are also reported, allowing us to situate them directly into the use-constellation of the motif. Coins: 66 A variety of texts is worth noting here: Boardman & Scarisbrick, 1977; Boardman & Vollenweider, 1978; Bordreuil, 1986; Brandt, 1968; Brandt & Schmidt, 1970; Forbes, 1978; Gerke, Scherf & Zazoff, 1970; Gramatopol, 1974; Speleer, 1912; Spier, 1992; Vollenwieder, 1979; Zwierlein-Diehl, 1969. 67 Furtwängler, 1965 reprint of 1900 text; Boardman, 1970a; Richter, 1968. 26 As noted, coins were a prime medium for displaying symbols, due to their form and early connections the shield blazons. Likewise, coins and sealings also share a special relationship in regards to the motifs they depict. Both the seals that produced the sealings and the dies that produced the coins were made using the intaglio technique. Indeed, artists may have worked in both media over the course of their careers.68 As such, some motifs were copied between the two media. This is especially true of seals which can be termed as “official”, representing the governments of the Hellenistic kingdoms or cities that existed at the time of the Kedesh Archive. For instance, seals bearing the motifs of an anchor with a horse’s head, both of which appear separately on Seleucid coins, also appear at Kedesh. As such, Hellenistic coins provide an additional corpus of motifs with which to compare the Kedesh symbol sealings. The nature of numismatic scholarship also means that, even in the case of large museum collections where the find-spots may not be known, the mint and date range of the coin’s production can still often be deduced. Thus coins are by their nature often easy to locate within the use-constellation of a given motif. Especially helpful have been the many catalogue volumes of the British Museum’s coin collection along with Newell’s treatises on the Seleucid Mints, Mørkholm’s 1991 work on the coinage of the Early Hellenistic Coinage, Kroll and Walker’s volume on the coins from the Athenian Agora, and two works on early Palestinian coinage by Meshorer and Qedar and Gerson respectively.69 The American 68 Vermeule, 1966, #3&4. Vermeule notes that these two gems were probably cut by die engravers for different mints, including Corinth, based on the closeness of their images to types portrayed on coins. 69 Newell, 1977; Newell, 1978; Mørkholm, 1991; Kroll & Walker, 1993; Meshorer & Qedar, 1999; Gerson, 2001. 27 Numismatic Society has also put the catalogue of their collection online, which has proven invaluable in tracking the appearance of different motifs in coins.70 Sculpture: The Kedesh symbol motifs also appear in sculpture. This includes both large scale stone sculpture and terracotta figurines. The occurrences of the motifs in sculpture are often published on a site by site basis. There are, however, some sources that deal with groupings of sculpture across larger areas. Among these sources is Webb’s 1996 work on the architectural sculpture in Asia Minor and the Aegean Islands.71 Also of interest are Acquaro’s catalogue in Moscati’s 1988 book on the Phoenicians, Chapouthier’s book on the Dioscuri and Webster’s 1995 book on the monuments illustrating New Comedy.72 These last three sources provide synthetic overviews of a number of different media within the purview of each text, including many examples of sculpture. Vessels: Vessels of various types, including lamps, also occasionally bore examples of the Kedesh symbol motifs. The vessels themselves are mostly ceramic, but include instances in other materials, like bronze. Here, the several volumes dealing with the Classical and Hellenistic ceramics from the Athenian Agora have been most useful, as have 70 ANS Online Catalogue (http://numismatics.org/collection/accnum/list). 71 Webb, 1996. 72 Acquaro, 1988; Chapouthier, 1935; Webster, 1995. 28 Mlynarczyk’s 1997 work on Alexandrian lamps.73 Works with a broader regional scope that encompass vessels include Webster’s previously mentioned work and Bailey’s work on the Greek lamps of the British Museum.74 Mosaics: One final medium to consider for occurrences of the Kedesh motifs is that of mosaics and paintings. Here again, some of the motifs which appear in the Kedesh symbol sealings are also to be found, especially in the forms of decorative medallions or panels in the mosaics or paintings. One particularly fertile locus for comparison consists of the mosaics from Delos.75 Another group for comparison consists of the Hellenistic tomb paintings from sites such as Alexandria and Marisa.76 73 Howland, 1997; Moore, Philippides & von Bothmer, 1982; Rotroff, 1997; Mlynarczyk, 1997. 74 Bailey, 1975. 75 Bruneau, 1972. 76 Brown, 1957; Venit, 2002. 30 Chapter 2: Other Archives The Kedesh archive is not unique. It was one example among many other archives from around the Ancient Mediterranean and Near Eastern worlds (see Map 2.1). These other archives provide the most valuable sources of comparanda for the sealings in this study, being separate examples of the specific medium. Furthermore, by using these archives, it will be possible to localise the Kedesh archive in general and the impressions of symbols in particular within the larger sphragistic universe of the Hellenistic world, thereby allowing us to situate and compare the corpus within the traditions which were prevalent in that time and place. In order to do so, we can not simply limit ourselves to archives of the Hellenistic period alone. Instead, we need to include archives from earlier, namely in the sixth through fourth centuries. Certain of the archives also continued on into the Roman Period. These archives that exceed the boundaries of dates of the Hellenistic period allow for comparison of sealing images through a long span of time. Thus, it will possible to situate the impressions of this study not only geographically and see if they reflect influence from specific geographical regions but also within the temporal continuum, thereby seeing if the uses of specific symbols reflect any conservative or innovative trends for the mid second century B.C.E. Described below are the various archives which have been uncovered and published from the regions of the Mediterranean and the Near East and which date from the sixth through the first centuries B.C.E. The entries are organized by geographic regions with the archives in a region being set out alphabetically. The regions are laid out 31 more-or-less geographically with consideration given to discrete cultural areas. As such, the order consists of the Levant followed by Egypt, the area of the Aegean, central Anatolia, Sicily, the region of Mesopotamia and areas to the east, and, finally, north Africa. This order, with suitable expansions to include areas with no archives, will continue to be used in the catalogue chapter (see below). The Levant and Cyprus: Alexandria-at-Issus: One possibly Hellenistic archive, in the form of some 880 sealings, has been recovered from underneath the floor of a Roman Period House in Alexandria-at-Issus. These sealings are in generally poor shape due to damage from fire and building collapse. Most appear relatively small in size, though some fifteen are much larger and may represent official seals. Some scenes which have been preserved include the busts of Dionysus and Artemis and the head of Alexander in the guise as Heracles. Other seals show what appear to be generic heads, which may be Roman.77 Jerusalem: A collection of impressed bullae along actual seals has been discovered, purportedly altogether in a ceramic pot somewhere in the region of Jerusalem. Unfortunately, these bullae were uncovered through illicit excavation and directly sold to a Jerusalem antiquities dealer, so nothing else is known of the original context. The collection consisted of more than seventy bullae and two seals.78 Sixty-seven of these 77 Plantzos, 1999, p. 30; Seyrig, 1940, pp. 96-98. 78 Avigad, 1976, p. 1. 32 bullae with the two seals have been published by Avigad. These bullae were impressed by oval stamp seals, each bearing an inscription of varying length in Aramaic script. All of the seals and sealings are aniconic, bearing no visible pictures.79 Avigad has dated these seals and sealings to the Persian period, between 515 and 445 BCE.80 Nea Paphos: In 1970, UNESCO conservators working on the Late Roman House of Dionysus in Nea Paphos, Cyprus, removed one of the mosaic floors to reveal a quarry that had been filled in order to make the substratum for the mosaic. Excavating this fill stratigraphically, the restorers discovered, one meter below the level of the mosaic, a thick layer of greasy ashes which contained a large number of impressed clay sealings. In all, the team recovered around 11000 sealings and fragments from this layer, with some coming from the layers immediately above and below, representing the material from a single archive that had been destroyed under unknown circumstances and reused in a secondary deposit. The sealings had been attached originally to papyrus documents, which had been destroyed in a fire which had also baked the clay sealings, thereby preserving them. Most of the sealings still bear longitudinal piercings made from the string that tied the papyrus rolls. The sealings seem to almost all bear impressions made by finger rings. The other finds, such as lamps, coins and terra cottas, from the same layers as well as subject and stylistic considerations from the sealings themselves date from the second century BCE to the first century CE. The excavators believe the archive to have served some sort of public function, being termed either a Nomophylakeion or an 79 Avigad, 1976, pp. 3 & 16-17. 80 Avigad, 1976, p. 32. 33 Archeion in publication. The sealings, only two percent of which are inscribed, bear a wide variety of images, including images of various deities; portraits of the later Ptolemies and Seleucids as well as portraits of philosophers and of private individuals; and images of animals, birds, masks, and other objects. Many of the images have numismatic antecedents, such as a series which shows the Temple of Aphrodite in Palaipaphos, which appears on city coins from the Greco-Roman Period. The excavators have linked this public archive with the Cypriot Federation (Koίνον Κυpiρίου). Ιnscriptions attest that this body arose in the second century BCE and that it had rights over religious matters and the minting of small coins. It was centered in Paphos itself.81 There is as of yet no final publication of the archive, though several small groups of the sealings have received treatment in publication, including the Ptolemaic portraiture, images of Eastern divinities, and some of the inscribed sealings.82 Among the published examples, certain examples are especially relevant to this study. These include an image of a coiled Uraeus flanked by pilei of the Dioscuri, an image of Hathor done as a bucranium topped by the goddess’ headdress, two images of a bucranium bearing the inscribed name of Kourion in the genitive, another image of a bucranium bearing the name of the city of Ledroi in the genitive, and a seal bearing a ship’s prow with the name of Kition in the genitive.83 Wadi Daliyeh: The Wadi ed-Daliyeh material is very interesting in regards to Kedesh, being the closest archive physically to the site. It consists of a number of papyrus documents and 81 Nicolaou, 1972, p. 315; Nicolaou, 1978, pp. 849-850; Michaelidou-Nicolaou, 1979, p. 413. 82 Kyrieleis, 1996, pp. 315-320; Nicolaou, 1978, pp. 849-853; Michaelidou-Nicolaou, 1979, pp. 413-416. 83 Nicolaou, 1978, #11-12; Michaelidou-Nicolaou, 1979, pp. 414-415. 34 associated seal impressions found in a cave near the site of Samaria and dating to 375 and 334 BCE84, which is the absolute end of the Achaemenid Period, when Alexander was passing through the region. Indeed the reason for the deposition of the sealings and the papyri was the flight and eventual slaughter of some of Samaria’s leading citizens after the murder of Alexander’s prefect.85 There are a total of 128 sealings which were recovered along with scraps of the original papyrus documents through a combination of purchase from the original Bedouin discoverers and controlled excavation. Two gold seal rings were also recovered, though no impressions of them were found within the archive. A total of sixty-two sealings were legible and have been published along with the rings. Each sealing consisted of a single seal impression upon a small bulla of clay which was attached to loops of string binding the scroll. As each bulla possessed only a single impression, a single scroll might possess multiple bullae in order to accommodate the various parties to a transaction. The imagery includes both Greek and Near Eastern styles, with thirty- nine impressions and one gold ring having Greek style imagery and with twenty one impressions and the other ring being styled as Near Eastern. Interestingly two of the impressions bear Paleo-Hebrew inscriptions.86 Of the Greek impressions, some twenty- nine were apparently made by the bezel of a seal-ring while ten were made by a scaraboid or other form of stamp seal. On the other hand, only three Near Eastern impressions were made by rings. Instead, thirteen were impressed apparently by some 84 Winn Leith, 1997, p. 6. 85 Winn Leith, 1997, p. 7. 86 Winn Leith, 1997, pp. 18-21. 35 form of stamp seal, and three may have been made by a cylinder seal.87 In all, the sealings displaying Greek iconography predominate in the archive.88 Images in the Near Eastern style generally either portray some version of the Persian hero, battling one or two animals such as lions or sphinxes, or a scene simply with animals. Those bullae classified as being Greek in style are more variable in their depictions. Categories include standing youths, mature males, Dionysiac imagery, male and female couples and plant or animal designs, such as a sphinx or a dove.89 None of the legible sealings showed images of either objects or symbols by themselves such as those found at Kedesh. Egypt: The situation of Hellenistic archives in Egypt is somewhat problematic. The climate of Egypt has allowed for the preservation of a great deal of papyrus documents dating to the Hellenistic Period and earlier. However, oftentimes these documents and archives are recovered, sometimes through clandestine operations, without their sealings, or they are published without references thereto. Their use as comparanda to the Kedesh archive, therefore, is rather minimal. As a result, there are only a couple of Hellenistic archives which stand as an exception to this trend and preserve their sealings intact alongside the documents. Edfu: One possible Hellenistic Period archive, as represented by its clay sealings, is 87 Winn Leith, 1997, p. 23. 88 Winn Leith, 1997, pp 26 & 30. 89 Winn Leith, 1997, pp. 28-31. 36 now divided between the collections of the Royal Ontario Museum and the Allard Pierson Museum in Amsterdam. This archive was not recovered from a controlled excavation, but rather was found illicitly, supposedly from a single pot in Edfu, and smuggled out of Egypt at the turn of the twentieth century.90 They must thus be treated with a fair amount of caution as their context is now lost. Still, scholars generally treat the sealings as a single archive and there does not appear to be any reason inherent in the sealings themselves to doubt the admittedly nebulous story of their origins. As such, it should be safe to deal with the sealings as a single chronologically and geographically bounded Hellenistic Egyptian archive. In total there are some 647 sealings in the hoard, divided roughly equally between Toronto and Amsterdam. Each sealing bears only one impressed image.91 It seems that the hoard can be dated to the later Ptolemaic Period, with portraits of Ptolemy VIII Euergetes II and Kelopatra VII being recognized among the sealings.92 One seal can be explicitly dated to the reign of Ptolemy IX Soter II (116- 106 BCE). The images on the sealings come from both Egyptian and Greek cultural contexts. For instance, distinctly Egyptian imagery appears on 118 sealings, representing the use of 56 distinct seals. Among these are the cartouches with the names and titles of several individuals including Ptolemy IX Soter II as well as people identified as members of the priesthood of Isis at Philae in hieroglyphics. One of these seals even bears the hieroglyphic cartouche of Senwosret III of the 12th Dynasty. The appearance of types that can be classified as distinct symbols or objects within the Egyptian group is, however, fairly limited. Five of the published impressions bear the image of a crowned 90 Plantzos, 1996, p. 307. 91 Plantzos, 1996, p. 307; Milne, 1916, pl. IV. 92 Plantzos, 1996, pp. 308 & 312. 37 hawk-head pierced by a vertical harpoon, which is a symbol of Edfu. Another sealing bears an image of the hieroglyphic “ankh” symbol paired with the harpoon of Horus. A further three impressions bear images of symbols which, though visibly distinct, are unidentified.93 The remainder of the Edfu sealings have either ‘Greek” or ‘Greco- Egyptian” images. Milne has grouped these into several units, including genre (animals, thunderbolt, and helmet), Greek and Greco-Egyptian mythological scenes including depictions of gods such as Dionysus and Isis, male and female portraiture, especially of members of the Ptolemy clan, and grouped portraiture.94 In regard to the present work, it is the first and numerically smallest category that is of interest to us. Milne includes 14 sealings in his genre category of which eleven display images of animals, including horses, a dog, eagles, an owl, seated griffins, hawks and a bee respectively. Of the three remaining sealings in this category, Milne notes the image of an ear of grain with leaves, an image of a Corinthian helmet and also a sealing with the impression of a winged thunderbolt and a janiform head.95 These last two are the only points of iconographic congruity between the Edfu sealings and the Kedesh symbol sealings. Elephantine: One of the exceptions to the problem of archival loss of context in Egypt comes from the Nilotic island of Elephantine. Here, some thirty-five seal impressions were recovered attached to a total of eight papyrus documents that they sealed. These documents are all private contracts, thus indicating that the archive itself was private 93 Murray, 1908, pp. 62-70, see especially #43-47 and #53-56; Milne, 1916, pp. 87-88; Plantzos, 1999, p. 27. 94 Milne, 1916, p. 88. 95 Milne, 1916, #1-14, #32. 38 though possibly associated with a syggraphophylax, a private agent who kept other peoples’ records and who gradually replaced certain royal institutions with similar functions in Ptolemaic Egypt.96 On these particular documents, the roll was sealed by usually several clay bullae, in the form of small clay blobs, which each bear multiple impressions, unlike Edfu. The date of these seals, from the papyri, seem to date from the late fourth century BCE through the last quarter of the third, and thus mark the very end of the Persian Period through the beginning of the Hellenistic Period.97 Stylistically, most of the seals look to be fourth century workmanship, though some could be earlier or later.98 Only two of the impressions bear non-Greek motifs. One of these bears an image of the Egyptian god Thoth as a seated ape, while the other appears to be an imitation of the scarabs of Thutmosis III. The production of these imitations began in the 26th Dynasty and apparently lasted into the Hellenistic Period. Interestingly, the scarab of Thutmosis III was used by a man named Drimachos, while the Thoth ape seal was used by a woman named Callista from Temnos.99 Two other seal impressions, on documents dated to 223/2 BCE, had Egyptianizing motives. One had an image of Isis and Serapis, while the other showed a female head with the hairstyle common in the “Isis of Lybia”- type.100 The majority of the seal impressions seemed, however, to be Greek in character. There are the heads and full figures of several Greek gods or heroes, such as Eros, Heracles, Dionysus, Artemis and Apollo.101 Other scenes include both male and female 96 Plantzos, 1999, p. 24; Rubensohn, 1907, pp. 6-7. 97 Vandorpe, 1996, pp. 234-235 & 258-259. 98 Rubensohn, 1907, p. 9. 99 Plantzos, 1999, pp. 23-24; Rubensohn, 1907, #10 & 33. 100 Rubensohn, 1907, #32 & 34. 101 Rubensohn, 1907, #1, 3, 6, 11-14 & 30. 39 heads, scenes of humans, animal scenes and a gorgoneion.102 In regards to this work, the sealings of the greatest interest are the images of a shield bearing the image of an Ionic column topped by a helmet crest, a feminine herm, and four images of masks of which one is comic, a second is of a young Dionysus, the third of an old man and the fourth of a young woman.103 Greece, the Aegean and Western Anatolia: Callipolis: Another archive has been uncovered from an upper-class house in the Aetolian city of Callipolis. In all, several hundred clay sealings, which once sealed now missing papyrus or parchment documents, have been recovered. It has been surmised that the archive might have belonged to one Agetas and his son Lochagos, who were both attested as often serving as strategoi in the late third-early second centuries BCE. The archive has been dated between 279 and 168 BCE. The seals in the archive probably represent correspondence between various parties, including other city-states and kingdoms, and the family as well as a variety of different transactions. A significantly large portion of these seem to have represented the state seals of nearby cities and series of royal portraiture as well as portraits of what appear to be Roman officials. Among the other seal impressions are Greek deities like Poseidon and Apollo, human figures and animals.104 Furthermore, a total of 145 sealings, representing the use of 70 distinct seals, display images that can be considered as either symbols or objects, falling into several 102 Vandorpe, 1996, pp. 259-260; Rubensohn, 1907, #5, 7, 15-24, 26-29 & 31. 103 Rubensohn, 1907, #4, 8-9, 25, 30, 35. 104 Pantos, 1996, pp. 188-193; Plantzos, 1999, p. 31. 40 distinct categories. Some of the Callipolis categories also appear in the Kedesh symbol sealings while others do not. Among the categories that appear at Kedesh are anchors, bucrania, composites, decorated shields, lyres, masks, stars, and thunderbolts. Callipolis, however, also has examples of tripods, monograms, tridents, wreaths and other types of weapons, such as helmets and spear points that do not appear in the Kedesh corpus.105 Daskyleion: A Persian Period archive has been discovered in north-western Anatolia at the site of Daskyleion, located on the shore of Lake Manyas near the modern Turkish village of Ergili. Daskyleion served as the satrapal capital of Hellespontine Phrygia under a hereditary line of Persian satraps in the fifth and fourth centuries BCE. Afterwards, the forces of Alexander captured the site, an occasion marked by a burned destruction level.106 The archive itself seems to have had a life of about a century, spanning from the second quarter of the fifth century to the first quarter of the fourth.107 The evidence for the archive consists of 406 complete or fragmentary baked clay bullae bearing seal impressions, which were once attached to now destroyed documents or objects. Most of the bullae are small round or oval clay lumps with seal impressions on the outside and impressions from string and papyrus fibres on the inner side. A few are somewhat larger, plano-convex in shape and similar in form to the “Labels” from the Persepolis Treasury Archive.108 The Daskyleion bullae all bear seal impressions, usually one but occasionally 105 Pantos, 1985, pp. 15-31 & 97-162 (#1-10, 97-142). 106 Kaptan, 2002, pp. 1 & 5-8. 107 Kaptan, 2002, p. 27. 108 Kaptan, 2002, p. 13-14. 41 two, representing a total of 185 discrete seals.109 These seals, which have been published by Deniz Kaptan, display a variety of Persian and Greek stylistic elements.110 The vast majority of the seals represented in the impressions consist of images of Persian court and religious rituals, hunting and heroic combat, animals, and Greek heroes or deities. There are, however, a few seals that display images of symbols or objects, mostly plants.111 These few examples consist of four seals, used in a total of four impressions that depict a rose, a date palm with an alphabetic character, the crown of a date palm, and an ornate decorative cup.112 An additional fragmentary sealing shows an ear of grain that was probably a secondary motif on the original seal, and another sealing that shows an unclear rectangular symbol or object of some sort, perhaps a building.113 Delos: The island of Delos in the southern Aegean has produced the remains of a very large archive of Hellenistic date. This archive was discovered in a house in the Skardhana district of the ancient city. The evidence for the archives consists of small oblong bullae of clay with imprints of the papyrus fibres on the back and a channel for the passage of a string. Upon these bullae, as few as one and as many as thirteen seals could leave their impressions.114 These bullae were stored in a room in the north-western corner of the house, which had apparently been outfitted with a masonry bench as well as shelves and/or chests for the storage of the various documents. Interestingly, bullae showing copies of the same seal impression were often recovered from the same areas 109 Kaptan, 2002, pp. 3 & 24. 110 Kaptan, 2002, pp. 3-5. 111 Kaptan, 2002, pp. 28-105. 112 Kaptan, D. 2002, DS 130, DS 146, DS 156 & DS 180. 113 Kaptan, 2002, DS 159, DS 143. 114 Boussac, 1988, p. 307. 42 during the excavations, thereby indicating some possible organizational principles.115 In regards to the dates of the archive, the end terminus is set by the burning of the house in 69 BCE, which baked the bullae, thereby preserving them. The other terminus is less fixed, One seal impression with Phoenician and Greek inscriptions bears a Seleucid date which gives the year 128/127 BCE. Furthermore, the corpus also possesses sealings with portraits of Seleucids or Ptolemies from the late second century to the early first. Still, there is no hard or fast point to mark the beginning of the archive. If one were to consider the dated seal as marking some sort of beginning, then it would seem that the archive functioned for at least some sixty years in the late-second and early-first centuries BCE.116 A total of around 25000 seal impressions have been recovered from the house, representing some 14000 different seal types.117 The sheer variety of the seals within the archive, as well as the multiplicity of regional influences, means that Delos would probably be a profitable tool of comparison to the Kedesh material, if it were entirely published. Among the published impressions, there are some large official sealings, including the city seals of nearby Cycladic islands, the seals of Ephesus and Colophon, some 17 from Seleucid officials and another 17 from the Epimeletai of the island itself, but these sealings are in the definite minority. Instead, the vast majority of the impressions seem to represent the smaller seals of private individuals.118 The preponderance of private seals as well as their number and variety might indicate that the 115 Boussac, 1988, pp. 310-312. 116 Boussac, 1988, p. 318. 117 Plantzos, 1999, p. 32. 118 Boussac, 1993, p. 682. 43 owner of the house provided the service of storing other people’s documents, like a solicitor.119 Some show mythological scenes as well as Greek or Greco-Egyptian gods, such as the Isiac triad, or Apollo, or Helios. Interestingly, one series, showing the cult statue of Delian Apollo may have actually been made locally. Other impressions show scenes of heroes, though the subject matter is fairly limited with scenes of Herakles and the Trojan cycle being most popular. Portraiture is also common among the impressions. Most of these portraits do not seem to depict anyone recognizable, such as the faces of Hellenistic kings, and therefore probably represent the private owners of the seals themselves. There are also many scenes of animals and of daily life. Furthermore, there are numerous impressions that display either symbols or objects, among which there are many images of composites and theatrical masks. The sheer variety of scenes, with influences from Greece, Egypt, Syria and Italy, demonstrates a certain cosmopolitan quality congruent with the status of the island as a free port in the Late Hellenistic Period.120 As noted, the archive is in the process of final publication, with two volumes having been published so far: one dealing with the small numbers of official seals from various kingdoms and cities as well as the gods Apollo, Artemis, Hecate and Helios, while the other deals with the Erotic cycle.121 Neither of these volumes deals specifically with images of symbols or objects, but examples do appear. For instance, several of the public seals in volume 1 display symbols or objects like the Seleucid anchor with horsehead protome, the ornamental krater of Naxos, the bunch of grapes of Mykonos and 119 Boussac, 1993, pp. 683-684. 120 Boussac, 1988, pp. 333-338. 121 Boussac, 1992; Stampoulidis, 1992. 44 the kithara of Collophon.122 Likewise, the second volume presents some scenes of composites or “grylloi” involving Eros.123 Boussac has also published an article that examines the large number of seal impressions that portray theatrical scenes, especially the images of masks, which make up the majority of the impressions.124 According to Boussac, there are more than 500 images of comic masks, representing mostly a variety of slave and young man roles from New Comedy, as compared to some fifty tragic masks.125 She also mentions the presence of numerous Dionysiac masks, but does not go into detail as they are not part of the theatrical cycle.126 Thesprotia: An archive has also been found in the region of Thesprotia, which lies in the extreme north-west of modern Greece, at an ancient settlement identified as either as Titane or Gitane. This settlement apparently functioned as a regional political centre for Thesprotia and Epirus in the Hellenistic Period. The archive, which consisted of 2500 clay seal impressions for now lost papyrus documents, was uncovered in the remains of a large building, “Building A,” which measured 31 x 41 m. with various dwelling units grouped around a central courtyard, and which was situated next to the theatre. The seal impressions were concentrated in a pot located in a small room to the southeast of the central courtyard, wherein lay the remains of a possible box for archival storage in the form of wooden fragments and sheets of bronze and lead. Most of the sealings mimic the forms found on Epirote coins as well as the portraits of kings and possibly local 122 Boussac, 1992, SP 3-4, Sp 6, SP 8. 123 Stampoulidis, 1992, #605, & #741-764. 124 Boussac, 1997, pp. 145-164. 125 Boussac, 1997, p. 148. 126 Boussac, 1997, pp. 152-156. 45 dignitaries. There are also depictions of animals, birds, weapons, various objects, and male and female deities, eleven of which also bear an inscription with the name of the city. Some of the sealings were found in the same deposits as Epirote coins which date between 238 and 169 BCE. It is a possibility that the archive went out of use, therefore, after the Battle of Pydna.127 Central and Eastern Anatolia: Artaxata: The site of Artaxata, the capital of ancient Armenia, has produced the evidence, in the form of clay sealings, of two archives from the Hellenistic and Early Roman Periods. The sealings were found in a single layer which dates between the founding of the city in 180 BCE by Artashes I and the rebuilding of the city in 60 CE after a destruction event.128 Both archives were also found in habitation areas within the bounds of the city.129 Neither archive has received final publication yet, though there are a couple of brief preliminary publications of the sealings with pictures.130 The first archive consists of some 8000 sealings of which 6000 are in good condition. The second archive is much smaller with only twenty sealings of which fifteen are in good condition. A total of 1035 distinct seals have been discerned. The sealings generally have only one seal impression each, though a minority have up to four impressions on one sealing. Each sealing has from one to six holes from the thread originally passing through it. The reverse of the sealings indicates that they were affixed to documents, often parchment, or 127 Preka-Alexandri, 1996, pp. 196-198; Plantzos, 1999, p. 32. 128 Kachatrian, 1996, pp. 365-366. 129 Khachatrian, 1996, p. 366; Manoukian, 1996, p. 371. 130 Kachatrian, 1996, pp. 365-370; Manoukian, 1996, pp. 371-373. 46 pieces of leather. Metallic rings or gems cut into a variety of shapes appear to have made the majority of the impressions, though one imprint was apparently made by a cylinder seal. About thirty impressions bear inscriptions in Greek or Aramaic.131 The iconography on the sealings is quite varied and illustrates links between Artaxata and many areas of the ancient world, including Persia, Greece, Rome and Egypt. Among the various types of images, scenes showing a cavalryman riding down an infantryman and scenes of hunting are very popular.132 Also popular are portraits, both of private individuals and of various kings or important Romans like Marcus Antonius, indicating their possible function as official seals.133 Also present are images of Greek gods, various animals, and mythological figures.134 Finally, symbols and objects are also present, including images of amphorae, vegetal motifs, Janiform composites, Isiac symbols, astrological symbols, bucrania, a hand holding a palm branch, a dolphin with an amphora inside a wreath border, and a ligature, which is very close to an example in the Kedesh symbols sealings (LIG1).135 Doliche: The case of the archive of Doliche is perhaps the most nebulous of all the archives of the ancient Mediterranean. Indeed, it may not even be a single archive with a single date range. Like the sealings from Edfu above, the original context has been lost due to illicit excavations. Furthermore, unlike Edfu, there is not even a possibly suspect tale of their discovery to point to. Instead, the ‘archive’ consists of some 326 impressed 131 Khachatrian, 1996, pp. 366-367; Manoukian, 1996, p. 372. 132 Manoukian, 1996, fig.3-4. 133 Khachatrian, 1996, pp. 367-368. 134 Kachatrian, 1996, fig.1-4, 11-14, 16, 21-31& 33-34. 135 Kachatrian, fig. 17-20, 32 & 35-42. 47 clay sealings that were purchased individually and in several discrete groups from dealers in Aleppo, Istanbul, Izmir, Gaziantepe, and Beirut and that are now housed in several collections including those of the Bibliothèque National in Paris, the Staatliche Munzsammlung in Munich and the Cabinet Royal des Médailles in the Hague as well as several private collections. The impressions on the sealings consist of both the large, and often inscribed, official seals of cities as well as the smaller oval seals of individuals. There are several reasons to see these disparate groups of sealings as originating from a single source or archive. First, the various groupings all share similar types of official seals in common, though no single grouping has the entire set. Among these types are images of the goddess Tyche with an inscription of the city name in the Greek genitive case, and scenes of Zeus Dolichenus grasping the hand of a Roman emperor which also has the same inscription. Moreover, many of the sealings with smaller impressions from the different groups were apparently made by the same seals. The arguments for localising the find spot at Doliche include the repeated appearance of the city name on several of the public seals, the appearance of several figures associated with the cult of the city’s tutelary god, including Zeus Dolichenus and the Dioscuri as Castores Dolicheni, and the fact that many of the sealings were purchased in Gaziantepe, the nearest city to Duluk (ancient Doliche). However, some eleven sealings bear images of the goddess Tyche and an inscription naming the city of Nicopolis, which lies some 90 kilometres from Doliche, while another bears the name of Cyrrhus another neighbouring city. All of this may point to the presence of several separate archives in the area, as opposed to a single one. At the same time, however, we should note that Kedesh itself has sealings bearing the impressions of seals from Tyre and Sidon, and Delos, as 48 previously mentioned, has impressions of seals from Colophon, Mykonos and Naxos.136 The date of the ‘archive’ is also hard to pin down, being derived largely from the style and the subjects of the sealings with a few of the official seals bearing dates in unspecified eras.137 Among the recognizable portraits, for instance, are examples of Seleucus II and Septimius Severus, indicating a wide span of time that spans from the third century BCE to the late second, early third century CE.138 Only a few of the sealings bear images of symbols. These include a shell, an ear of wheat with a bunch of grapes, an owl on a helmet, a bird on a rose, and something which could either be a vase or a helmet.139 Sicily: Selinus: One archive comes from Magna Graeca, specifically the ancient city of Selinus in the south-east Sicily. The archive was found in the ruins of ‘Temple C’ on the acropolis, and thus may be classified as being public in nature.140 The archive dates between two destructions of the temple by fire, one in 409 BCE by the Carthaginians and one in 250 by the Romans. The evidence of the archive consists of some 688 burnt clay bullae for papyrus documents which were burned in the latter fire. Some 589 of the bullae bear legible seal impressions, representing the use of 431 distinct seals.141 Each bulla could 136 Ariel & Naveh, 2003, #2, 4, & 7; Boussac, 1992, SP 3-4, Sp 6, SP 8. 137 Klose, 1984, pp. 63-76; Maaskant-Kleibrink, 1971, pp. 23-63; Weiss, 1992, pp. 171-193; Seyrig, 1940, pp. 85-107. 138 Seyrig, 1940, #14; Maaskant-Kleibrink, 1971, #9. 139 Klose, 1984, #50; Maaskant-kleibrink, 1971, #77, #79, #83-84, & #122. 140 Salinas, 1883, pp. 288. 141 Zoppi, 1996, pp. 327-329; Salinas 1883, p. 298. 49 bear between one and three seal impressions.142 Standing anthropomorphic figures and the heads of mortals and gods, especially Heracles, were quite popular and represented by the use of 141 and 107 different seals respectively, representing the majority of the seal types present. In addition, images of animals were also present with 97 distinct seals. One scene in this category, showing a dolphin over a club in a square field, appeared on more than 300 bullae. Another category consisted of varied objects with 64 separate types. A small group or 22 different types has been classified as ‘Oriental’, bearing the images of deities and Punic inscriptions. It may even be that the popular depictions of Heracles represent a syncretism with the Punic Melqart, thereby representing Carthaginian influence in the area during the fourth and third centuries. Some scenes of Heracles, especially those depicting Heracles with a sigma beneath him, may represent the official seals of the city.143 Many of the iconographic categories present in the Kedesh symbol sealings also possess examples of images among the Selinus bullae, including bucrania, caducei, cornucopias, columns, composites (mostly in the form of Janiform heads), decorated shields, flowers, masks, thunderbolts and vessels.144 The East: Nippur: One textually well-published Persian Period archive from Mesopotamia consists of the archive of the Murašû family from Nippur. This archive consists of close to nine 142 Salinas, 1883, p. 291. 143 Zoppi, 1996, pp. 330 & 335; Plantzos, 1999, p. 23; Salinas, 1883, p. 298. 144 Salinas, 1883, pp. 307-313. 50 hundred clay tablets and fragments uncovered from Nippur by the University of Pennsylvania in 1893 and now residing primarily in The University Museum of the University of Pennsylvania and in Istanbul. These tablets all date between 454 and 404 BCE.145 Of the tablets with preserved edges, some 82% bear at least one seal impression, for a total of 657 different impressions, representing 586 separate owners, whose names were noted in the text of the documents. Of the seals that are present, there are roughly equal numbers of impressions made by cylinder seals, stamp seals and finger rings, with these last becoming much more frequent in the documents of the archive postdating 425 BCE.146 The motifs used on these seals are almost all figurative with humans and/or animals. Many of these scenes are Mesopotamian or Near Eastern in style, including Contest Scenes, and Human Headed Monsters. However, 10% of the total represents western or Hellenic motifs, such as nude warriors and composite busts of humans and animals.147 Persepolis: Two distinct yet related archives were uncovered at the site of Persepolis, one of the capital cities of the Achaemenid Empire. Both have been named for their findspots within the site. The earliest was the Persepolis Fortification Archive found in several rooms making up a portion of the fortification system on the northern end of the site. This archive consists of 20,000 to 30,000 inscribed and anepigraphic clay tablets and fragments of which Richard Hallock published texts of a representative sample of 2087, 145 Bregstein, 1996, pp. 53-54. 146 Bregstein, 1996, pp. 54-55. 147 Bregstein, 1996, pp. 58-59. 51 the majority of which bear seal impressions.148 These tablets date from 509 to 494 BCE, during the reign of Darius I. They consist of several different types of administrative texts, with each type having its own sealing protocol.149 Recently, the published tablets, which are mostly in Elamite, though there are examples in Babylonian, Aramaic, Greek and Phrygian, have formed the basis for an extensive seal study by Mark Garrison and Margaret Cool Root. From the 2087 texts published by Richard Hallock, Garrison and Root have identified 1162 distinct seals, of which they have thus far published 312, those dealing with heroic encounters, in the first volume of a three volume set. The other two volumes have yet to be published. The second volume will consist of seals, representing 30% of the total Fortification Corpus, that display images of human activity. The largest group of such scenes display images of archers. Other scenes in the volume will include images of hunting, chariots, processions of figures, worship and cult scenes, seated figures, genre scenes and studies of isolated figures. The third volume will include images of animals, creatures, plants and geometric devices and will represent 43% of the total fortification corpus. Here, images of animals, especially animal on animal contests predominate, but there are a few examples that display floral or geometric motives.150 The second Persepolis archive is the much smaller and somewhat later Treasury Archive, named once again for its findspot. This archive consists of some 198 (near) intact inscribed clay tablets with some 548 smaller fragments and 198 sealed clay labels. This last category includes both the category of the sealed anepigraphic tablets of the first archive and hand moulded impressed clay bullae which have survived. The texts of 148 Hallock, 1969. 149 Garrison, & Cool Root, 2001, pp. 1-3 & 10. 150 Garrison & Cool Root, 2001, pp. 1 & 42-44. 52 the tablets date to between 492 and 459 BCE. Almost all of them are in Elamite, though one is in Babylonian. The archive is not only smaller but less varied than the Fortification archive. All of the texts deal with payments made from the treasury and authorized by Treasury Agents. Almost every tablet was sealed, but only by one seal. The ‘labels’ may each bear multiple seals. The entire archive is represented by only 77 distinct seals, consisting of 43 cylinder seals and 34 stamps or rings, which only appear on the labels. These seals have not been the focus of an extensive study like the Fortification Archive, though there is a good deal of overlap between the two archives as several seals appear in both.151 The texts of the tablets, however, have been published by George Cameron.152 Seleucia-on-the-Tigris: Three archives dating to the Hellenistic Period have been uncovered at the site of the ancient city of Seleucia-on-the-Tigris in Southern Mesopotamia, which served as one of the capitals of the Seleucid Empire until the Parthian conquest of Mesopotamia in the mid second century BCE. The sheer number of impressions represented by the material from these archives is quite staggering. The seals could be impressed into clay or bitumen in several formats. One of these formats, called ‘bullae’ by McDowell, consisted of a large clay ring which would have totally encircled the papyrus roll and into which seals would have been impressed. These bullae may have developed from the Mesopotamian use of clay envelopes when the use of scrolls superseded the use of cuneiform tablets. Seals could also be impressed into much smaller clay ‘appended 151 Garrison & Cool Root, 2001, pp. 33-34. 152 Cameron, 1948; Cameron, 1965; Garrison & Cool Root, 2001, p. 33. 53 sealings’ which consisted of a piece of clay wrapped around the loose end of string tied around the papyrus roll.153 The first two archives, classified as Archives A & B respectively, were uncovered first in two separate rooms of a large house. Archive A dates between 294 and 141 BCE. Archive B has a much shorter lifespan, dating to 188- 153/2 BCE.154 There were both inscribed public seals, many having to do with the salt tax and bearing Seleucid era dates, as well as smaller private seals represented by the bullae. One-hundred-and-fifty-six private sealings from these archives were published by McDowell, of which 144 bore Greek style images while only five bore Near Eastern style imagery, with five being considered Hellenized Near Eastern and two others being categorized as doubtful. Among the scenes were depictions of Seleucid royal portraiture, Greek deities such as Athena, Eros and Tyche as well as Greek mythological scenes, and animals such as lions and crabs.155 Besides these types, McDowell describes a total of 42 impressions that show images of symbols or objects. Seven of these impressions were larger in size and bore inscriptions naming specific bureaucratic offices, indicating they were made by official seals. This group included images of a tripod-lebes as well as Seleucid anchors associated with either horse-head protomes or vessels.156 The remainder are much smaller and un-inscribed with numerous different types, including a tripod, astrological signs like stars and crescents, a caduceus, a lamp, a ceremonial standard, a winged thunderbolt, herms, theatrical masks, both comic and tragic, vessels, pilei of the Dioscuri, a columns, a hand clasping an object, and vegetal motifs, notably 153 McDowell, 1935, pp. 2-3. 154 Plantzos, 1999, p. 31. 155 McDowell, 1935, pp. 15-24; Wallenfels, 1994, p. 2. 156 McDowell, 1935, I.A.1.a.1-2, I.A.1.c.1, I.A.2.a.1-4. 54 the palm tree.157 Another huge archive was later found in the 1960s by the Italian expedition to the site. This archive was found in the remains of a large, purposefully built archival structure which was destroyed apparently at the time of the Parthian conquest and from which some 30,000 impressed clay/bitumen-rings or “bullae” were eventually recovered. Many of the impressions are of large official inscribed stamps dealing with either the payment of or the exemption from the salt tax. Often these stamps would bear Seleucid symbols, such as the anchor or portions thereof, but mostly as ancillary symbols and not as types in themselves. The other impressions could either be the larger often inscribed official seals with images of official Seleucid symbols, such as the anchor or the tripod- lebes, bearing reference to the βιβλιοφύλαξ, as well as Seleucid royal portraiture, or they could be of the small, less-often inscribed variety with an extremely wide variety of types.158 The final publication of the archive has recently appeared in three massive volumes by Antonio Invernizzi and others, thereby cementing the preeminent importance of Seleucia in the study of Hellenistic glyptic.159 These three volumes divide the large number of impression scene types into official seals, including both the large stamps and the other mentioned above, images of divinities, and images of human figures, animals, vegetal motifs and objects. Images of symbols and objects are to be found both among the official seals of the first volume and the objects, grylloi and vegetal motifs of the third. There are 24 official seals, representing a total of 291 separate impressions that display symbols or objects. Mostly these involve the presence of the Seleucid anchor 157 McDowell, 1935, III.D.1.a.1-III.D.6.c.1. 158 Negro Ponzi, Invernizzi & Gullini, 1971, pp. 22-23; Invernizzi, 1968, pp. 19-124. 159 Invernizzi, Bollati, Messina & Mollo, 2004. 55 motif, with or without inscription and often associated with another symbol, such as a quadruped, a horsehead protome or a cornucopia, but there are also examples of inscribed tripods and even an inscribed column.160 Furthermore, there are 836 of the smaller, probably private, seals, used in a total of 4342 impressions, that show symbols and objects. These include images of masks, which are by far the most popular with 246 seals and 3523 impressions, animal protomes, vegetal motifs, composites, tripods, cornucopias, thunderbolts, vessels, pilei of the Dioscuri, ears, feet, hands, columns, Mesopotamian cultic symbols, lanterns, weapons, musical instruments, caducei, and crescents and Herculean clubs.161 Uruk: Not only does Mesopotamia possess the archives of Seleucia, but it is also blessed with the material from Uruk, which is actually the product of several distinct archives. Unlike Seleucia, which only possessed the bullae which sealed the documents and not the documents themselves, Uruk possessed both the bullae and tags of the long disappeared papyrus documents as well as cuneiform tablets in Akkadian, preserving both the sealings and the documents themselves, which allows for the connection of names to seals for more in-depth study as well as the chance to glimpse the interaction between the native Mesopotamian archival tradition and that of the incoming Greeks. The records mostly come from different temples such as the Reš and Irigal temples as well as the Anu Antum temple, which seems to have served as some sort of public record office. The corpus seems to cover most of both the third and second centuries BCE, with 160 Invernizzi, Bollati, Messina & Mollo, 2004, SU1-24. 161 Invernizzi, Bollati, Messina & Mollo, 2004c, M1-246, PA1-73, GyT1-68, Vg1-106, Og1-343. 56 the dates on the tablets ranging between 329-108 BCE.162 In general, the bullae could preserve as many as forty seal impressions each or as few as two, while the tablets held between one and twenty-eight, averaging about ten. Clay tags, however, usually bore only one impression.163 The sheer number of seal impressions is, like Seleucia, quite large. For instance, Wallenfels has published some 1100 seals used in 1523 impressions from some 154 tablets, representing some 22% of the total tablet corpus in the Yale collection.164 The Oriental Institute of Chicago has some 388 bullae from Uruk in its collection.165 The seal impressions from Uruk, like Seleucia, include examples of both the larger seals designated as being official and the smaller private seals. The official seals, which include tax stamps and royal portraiture seals, generally turn up on bullae, while the private predominate on the tablets. Among the private seals, the majority tend to be rings, though the impressions of four cylinders and ten Achaemenid Period style stamp seals do appear in the early documents predating 300 BCE.166 Relatively few of the sealings came from Hellenistic convex gems, which were quite popular at Seleucia. Most of the private seal rings seen on the tablets bear motifs which were derived from precedents on Greco-Persian seals, as well as seals from Achaemenid and earlier periods. For example, there are several examples showing the motif of a fish-apkallu, a type of sage and exorcist in Mesopotamian mythology. Interestingly, all but one of these is owned by a member of one of the seven hereditary exorcist clans at Uruk. The motifs on 162 Wallenfels, 1994, p. 4; Phantzos, 1999, p. 30. 163 Wallenfels, R. 1996, pp. 114-115. 164 Wallenfels, 1994, pp. 5 & 143. 165 Oelsner, 1996, p. 110. 166 Wallenfels, 1994, pp. 3-4. 57 the fewer gemstones, however, seem to have been derived more from Classical Greek and Hellenistic motifs and include heads/busts, full human figures in asymmetrical stances, dolphins, birds, crabs, flowers, and vessels. Some of the heads thus displayed bear some resemblance to the headdresses and features found on Seleucid portraits but are so conventionalized that they can no longer be truly called portraits.167 Another group of popular motifs on seals consists of representations of zodiac signs, usually in groups of three, representing astrological triplicities, or signs separated by 120 degrees of arc in the sky.168 There are a total of twenty-four examples in Wallenfels’ corpus that can be classified as symbols or objects in a set of eight iconographic categories. These categories include the vessels and flowers mentioned above as well as frontal bucrania, animal head protomes, trees, Mesopotamian cultic symbols, miscellaneous designs with raised borders, and pairs of gods’ mitres, which are in formal appearance similar to the category of pilei of the Dioscuri found at Kedesh.169 Wallenfels also mentions a further 10 previously published examples from Uruk that were not from the Yale collection, but which fall into the same general iconographic categories.170 As a whole, the material from Uruk tends to illustrate an amalgam of both Babylonian and Hellenic sealing and archival traditions, displaying much more Babylonian influence through the presence of the tablets than at Seleucia, but still bespeaking a good deal of Greek influence through the official archival and sealing praxis and the use of certain motifs. 167 Wallenfels, 1994, pp. 145-146 & 151. 168 Wallenffels, 1996, pp. 122-123. 169 Wallenfels, 1994, #1020-1042. 170 Wallenfels, 1994, pp. 134-137. 58 North Africa: Carthage: The region of North Africa has also produced a couple of archives. For instance, the city of Carthage has produced the remains of a fairly large archive which predates the destruction of the city by the Romans in 146 BCE. The actual date range of the archive seems to be quite wide, with examples dating from the late sixth century all the way to 146.171 The remains of the archive, consisting of individually impressed clay bullae which had been wrapped around the string binding rolled papyrus documents, was found in the ruins of the Temple to Baal and Tanit, the main deities of the city, and thus could be considered some sort of public archive.172 In total, some 4025 impressions were recovered of which some 1575 represent scarabs imitating the seals of Thutmosis III, like the one found at Elephantine.173 Here, their popularity might indicate that they had something to do with the temple administration.174 Otherwise, the choice of motifs seem to have been rather varied with examples of Greek, Greco-Phoenician, Punic, Etruscan, Italic, and Egyptian(izing) motifs, reflecting perhaps the cosmopolitan nature of the city.175 For instance, among those with ‘Greek’-style depictions were images of heads, and pictures of mythological personages such as Heracles, who was also identified with the Phoenician god Melkart, and Athena, who was herself linked with the goddess Tanit.176 Among the varied motifs at Carthage, there are 93 published examples that depict objects or symbols in several different categories; several of which include 171 Berges, 1996, p. 341. 172 Berges, 1993, p. 246. 173 Berges, 1996, p. 341; Plantzos, 1999, p. 23. 174 Berges, 1993, p. 250. 175 Berges, 1996, p. 346; Plantzos, 1999, p. 23. 176 Berges, 1993, pp. 256-257 59 Silenus masks, composites and janiform heads, vessels, floral motifs, hands, weapons, and others (including deity attributes like thunderbolts and caducei) are congruent with the categories at Kedesh.177 Cyrene: Finally, another Late-Hellenistic-Early Roman archive was uncovered at the site of the ancient city of Cyrene in North Africa. The archive was recovered from a building off the city’s agora which an inscription dating to the first century BCE identified as the Nomophylakeion of the city. This building was fronted by a later portico with an inscription that names the emperor Domitian. The complex was destroyed by a fire probably late in the reign of Domitian, and thus the archive has been dated to the first centuries BCE/CE. 178 In total, some 4000 clay bullae were recovered from this structure though only 421 were readable. These 421 bullae have a total of some 1311 separate seal impressions as most bullae had multiple impressions, with the average being around three.179 These impressions were mostly made with metal finger rings, judging by the shape of the impression, though there are some made by gems. Among the impressions, there is a wide variety of types. Many impressions show scenes of Greek gods including Apollo, who was the city’s patron deity, as well as Athena, Aphrodite, Nike, Tyche, Hermes, Herakles, Demeter, and Zeus who also all appear frequently. Others show individual portraits or people, possibly locals. Scenes of animals, including quadrupeds, dolphins, fish, insects and various birds, either alone or in group compositions, are also present in good numbers. Some impressions bear emblems, such as representations of the 177 Berges, Ehrhardt, Laidlaw & Rakob, 1997, #387-426, #778-826. 178 Madolli, 1964, pp. 41-42. 179 Madolli, 1964, pp. 39 &, 47. 60 goddess Roma, from the Romans, who first acquired the area and appointed a magistrate for it in 96 or 75 BCE. Another category of impressions consists of inscriptions, especially monograms. Of interest, in regards to the Kedesh symbol sealings are the impressions that show various symbols and objects of which there are 125 impressions. These include warships, vegetal motifs, composites, vessels, masks, cornucopias, and images of hands.180 MAP 2.1: The Archives in the Ancient World The Archival Practices of the Kedesh Archive in Light of the Other Archives: On the surface, reconstructing who used the Kedesh archive and how it was used is a daunting task. Unlike the archive at Uruk, where the documents in question were written on clay and therefore preserved along with the impressions of the seals, the 180 Madolli, 1964, pp. 57-61. 61 original documents at Kedesh were destroyed in the same inferno that baked the sealings, thereby preserving them. Thus all information regarding the types of documents in the archives and the principals involved in those documents has been lost. At the same time, however, the fact that the Kedesh archive existed within a larger universe of archives might offer some hope. These other archives offer points of comparison that can help explain in general how the archive at Kedesh functioned and who was using it. A good place to begin is the question of “public” or “official” seals, which were used under the authority of some governmental body, department or official, and “private” seals that were at least potentially the property of a single owner. The nature of the PHAB itself indicates that the Kedesh archive existed under official auspices within the Seleucid administration of the region. And yet, this does not necessarily dictate the sort of documents that might be there. The archive in the public building at Seleucia was greatly focused on documents related to the salt tax, whereas the Nomophylakeion of Cyrene may have functioned more generally as a public depot for any registered private act.181 The sealings themselves can be more useful in this situation. It is often possible to differentiate seals belonging to the “official” sphere from others. As noted already in the discussion of Seleucia, official seals are generally much larger than others. They also often bear motifs that are associated with specific political entities, such as by reusing royal iconography from coins. Another marker of official seals is that they are often 181 Madolli, 1964, pp. 46-7. 62 inscribed in order to identify explicitly the office or political agency which they represent.182 These features show up on several of the sealings from Kedesh. For instance, all of the published inscribed seals appear to be non-private or official in nature.183 Their legends refer the offices of specific officials, a governmental body of Tyre, as well as the harbor of Sidon and the site of Kedesh itself.184 Furthermore, and more important for this work, the Kedesh archive also possesses six sealings whose impressions are quite large and display either an anchor or an anchor with a horse’s head, both of which belong to Seleucid royal iconography. These sealings will be discussed further below in the section on anchors in the catalogue. The distinction between “official” or “non-private” and “private” seals is not always cut and dry. There are cases of ambiguity, where the iconography or size might suggest that a seal would belong in one category while every other characteristic might point to the other. For instance, there is the question of the use of royal portraits on small un-inscribed seals, examples of which do appear at Kedesh.185 McDowell saw the use of these images as only expressing general authority of the government on their own without additional elements like an inscription to be specific.186 Boussac, on the other hand, links the presence of such portraits on Delos to the well attested presence of 182 Boussac, 1993, p. 680; McDowell, 1935, pp. 26-29. 183 Ariel & Naveh, 2003, p. 61. 184 Ariel & Naveh, 2003, #1-2, 5, 7. The first of these seals shows a sign of the goddess Tanit and gives the title of an official as “He who is over the land” in Phoenician. The second shows a caduceus and has a bilingual Phoenician/Greek inscription that names the judges and common assembly of Tyre. The third shows a head of wheat and a bunch of grapes and gives the name of the site in Greek. The last of these seals shows a female figure seated on the prow of a ship with a Greek inscription that might be translated as “in the harbor of the Sidonians.” 185 Ariel & Naveh, 2003, p. 61; Herbert & Berlin, 2003, p. 329. 186 McDowell, 1935, p. 32. 63 “Friends of the King” on the island, who may have acted in such a function only occasionally.187 This possibility of individuals using a seal motif to identify and allegiance with, or themselves as occasional agents of, a greater power is especially interesting considering the presence of the anchor and dolphin motif on some of the Kedesh sealings which might indicate an affiliation with a specific commercial group, the Poseidoniasts of Berytus, who were active in the second century. I will explore this possibility fully within the discussion of the motif in the catalogue below. In any case, the question of the status of royal portraiture on seals shows that there is a certain ambiguity between the concepts of “private” and “non-private” or “official” seals and that there may not be a hard and fast dividing line between the two groups. For instance, McDowell also raises the possibility that those seals which are larger but do not bear official iconography might belong to some sort of semi-public institution, like a temple.188 This is an interesting possibility considering one seal in the Kedesh archive (THU2) shows a winged thunderbolt, the symbol of Zeus, and is at 14 mm. substantially larger than the others. Still, there is little besides the large size, such as a legend, to indicate the possibility that it is the seal of a temple. If we disregard the seal impressions of ambiguous status, the Kedesh archive has 22 seal impressions that can be classified as “official.” This represents only 1% of the 2043 total sealings recovered from the archive. This number puts it closer to the Delos archive, which was found in a private house and had only 10 “official” seal impressions 187 Boussac, 1993, p. 680. 188 McDowell, 1935, p. 26. 64 than the public Seleucia archive, which was found in a public building where the preponderance of the bullae have stamps dealing with the salt tax.189 Another thing to consider is the physical nature of the sealings and how this may reflect on the nature of the Kedesh archive. At Kedesh, the sealings uniformly have one impression each. This differentiates the archive from such places as Delos, Seleucia, Elephantine and Selinus, where multiple impressions per sealing do frequently occur.190 In another case, Cyrene has only 22 sealings out of 421 that bear a single impression each, and in each of these 22 cases the seal impressions are larger and appear to be public or “official” in nature.191 Rather, the practice at Kedesh is closer akin to places like Edfu where each sealing bears a single impression.192 One effect that the Kedesh practice would have it to physically place a greater limit on the number of participants (principals and witnesses) that could seal any one document, as each sealing participant would have to impress a separate piece of clay on its own string without damaging any other possible impressions. Theoretically it would be possible to place more than one piece of clay on the same string, but this would make the handling of the document afterwards much more difficult and make it much more likely that the still unbaked and friable clay sealings get torn off in the process of handling. It would therefore seem that the documents of the Kedesh archive were acts that required only a few participants seal them to be in effect. As such, the 2043 bullae 189 Boussac, 1993, p. 680; Invernizzi, 1968, p. 71. 190 Boussac, 1993, p. 678; McDowell, 1935, p. 2; Invernizzi, 1968, p. 70; Salinas, 1883, p. 291. 191 Madolli, 1964, p. 47. 192 Plantzos, 1996, p. 307; Milne, 1916, pl. IV. 65 that have been recovered might represent the remains of at least several hundred actual documents. Isolated finds of sealed and intact papyrus documents from Hellenistic Egypt may also provide some hints as to the possible nature of the Kedesh archive documents. In the Egyptian material, both private contracts drawn up by notaries and receipts for goods or payment generally have a limited number of sealers with each impression having its own separate sealing. The first type generally only bears the seal of the notary as a sign that the act has been registered.193 The second type consists of a record of payments or goods received in which the people getting paid or receiving the goods seal the document as a means for the payer to prove that the payments had already been made.194 On one hand it is unclear to what extent documentary practices within Egypt paralleled those in the Levant, and it remains impossible to prove the exact nature of the documents in the Kedesh archive itself. On the other hand, either or both of these types of documents would conceivably fit well into an archive housed in a regional administrative centre such as the PHAB. And so, the nature of the Kedesh archive is not altogether impenetrable. In comparison with other archives, the seals themselves through various aspects might illustrate into which broad category of user, be it private or non-private, that might have used it. The structure of the sealings themselves also might provide clues as to the nature of the documents they sealed. 193 Vandorpe, 1996, pp. 235-7. 194 Vandorpe, 1996, pp. 237-8. 66 Chapter 3: The Catalogue Section 1: Introduction and Explanation of Catalogue A brief explanation is necessary in order to clarify the structure and operation of the catalogue in the following section. The catalogue is divided into chapters based on iconographic categories, which are sequenced more or less alphabetically. These categories consist of anchors, bucrania, caducei, columns, composites, cornucopias, ears, flowers, ligatures, lyres, masks, pilei of the Dioscuri, shields, stars, thunderbolts, vessels and a final category for those seals whose images show objects of an indeterminate, or unknown nature. Each chapter follows the same procedure and set up. Each section starts with a description of the iconographic motif itself as it appears at Kedesh. This discussion includes the number of that motif and any possible variants that appear in the Kedesh sealings. There is also a discussion of the nature of the symbol with its history as well as any possible meanings or associations which it might have. The Distribution: Afterwards, there is a discussion of the distribution of the motif across the ancient Mediterranean. Each of the noted occurrences has both a known date range and geographic provenience. Those occurrences of the motif, whose locations are not well known, such as gems of unknown provenience in museums, may be mentioned, especially if they show a marked similarity to one of the Kedesh seals, but they are not counted towards the overall picture of the distribution. The occurrences of the motif are organized 67 first by time, with those that fall into the sixth-fourth centuries (or roughly the Persian period) coming first and those of the Hellenistic period coming second. In each period, the occurrences are subsequently broken down by medium with sealings from other archives coming first, then numismatic examples, followed by other media. The examples of the motif in each medium are also broken down by geographic region (see below). Occurrences of any variants of a particular motif are also discussed after the discussion of the main motif. Following the listing of the various separate occurrences, the overall picture of the distribution of the motif is discussed with particular attention being paid to areas and periods of concentration for the motif. The Catalogue Proper: Next, the chapter proceeds to the catalogue of the sealings themselves. The various sealings are grouped together into uses of specific individual seals, each with its own catalogue entry. I grouped sealings together into seals on the criterion that their depictions bear such a marked similarity to one another so as to make it highly probable that they were originally made by the same seal. Each entry is set out in the same fashion. The seal’s title, consisting of three letters based on the type of motif and its number within the sequence of seals of that motif, is given. Afterwards, the number of impressions and individual Kedesh bullae numbers are. The number of each seal within the sequence is based on the original Kedesh catalogue number given to its first bulla. For instance the first bulla number of ANC1, the fist seal of the anchor category, is K99 0002, while that of ANC2 is K99 0065. The entry subsequently contains the maximum preserved dimension of the seal’s 68 image, a note as to the general shape of the seal and a description of the image as it can be reconstructed from the various sealings. Next, each catalogue entry has a set of three images. These consist of a photograph of one of the impressions of the seal, a reconstruction drawing to scale of the original seal as can be deduced from the various impressions, and then an enlarged drawing at 4:1 scale of the seal so as to allow easier inspection. The Charts and Maps: After all of the seals of each motif category have been catalogued, the next portion of the chapter consists of a series of charts and maps that help to illustrate and break down the information discussed in the distribution section in each chapter. The exact number of charts and maps will vary in each chapter based on the number of variants of a motif that are present at Kedesh as well as their general popularity in the ancient Mediterranean. Very popular motifs, for instance, will have additional detail maps of areas within the Mediterranean for the sake of clarity. Still, each motif or variant that is present at Kedesh will have at least one chart and one map for each of the two main periods involved. The Charts: Each chart statistically compares the distribution of a motif by grouping its occurrences by both media and into geographic regions in order to show any trend towards prevalence in any one or several areas, media or periods. Here the occurrences are based on media so that an issue of coins with the symbol has the same numerical 69 weight as a piece of sculpture or a seal in an archive. The regions noted here are the same ones that were used in the distribution section itself and are described below. Geographic Regions for the Comparison of Motif Distribution: In the distribution section of a catalogue chapter, the various occurrences of a motif are grouped into a series of separate regions, which are numbered for the sake of convenience and ease of referral. The purpose of this endeavour is to divide up the wide geographic expanse of the Ancient World into more manageably sized portions in order to facilitate the comparison of the distributions of multiple motif categories. The method of this division is to create regions that are, wherever possible, both geographically discrete from other regions by the presence of distinguishing features and culturally distinct from surrounding regions by the peoples and polities that inhabit it. However, since this is not always possible, a certain amount of arbitrariness must remain. For instance, valid arguments might be made for specific areas being in one region or another or for regions being joined together or a single larger region being subdivided. Still, the regions as described below represent a consistent compromise between detail and precision on the one hand and breadth and generality on the other in order to provide a useful base for the comparison. 1. The Levant: This region consists of the littoral area of Syria Palestine. It extends from Cilicia in the north to Egypt in the south. On the other axis, it extends from the coast inland up to the region of the Euphrates River and the Syrian Desert. It also includes the island of Cyprus. The population groups in this region are fairly diverse and include Phoenicians, especially along the coast, Jews and Samaritans 70 in the interior regions of Palestine, and ethnic Greeks in Cyprus and in the Hellenistic centers such as Antioch, as well as others. This is the region that encompasses Kedesh itself along with its archive as well as the archives that were found at Alexandria ad Issum, Wadi ed-Daliyeh, Jerusalem and Paphos. 2. Egypt: This region encompasses Egypt. It is centered on the Nile valley, including the delta and the region of the Fayum and nearby oases, such as Siwa. The population here is somewhat more homogeneous and includes Egyptians and Greeks, the latter especially in Greek centers such as Alexandria and Naukratis. This region contained a couple of archives, namely Edfu and Elephantine. 3. Southern Greece: This region consists of the extreme south of the Balkan Peninsula and most of the Aegean islands, including Crete. As such, it encompasses a good deal of Greece proper, including the major urban centers of southern Greece like Athens and Corinth. The population here was largely ethnic Greek, though other groups are attested on occasions, such as the Italians and Phoenicians on Hellenistic Delos. The region also contains the archives that were found both at Delos and at Callipolis. 4. Western Anatolia: This region includes the western portion of Anatolia, notably the Aegean littoral and the associated nearby islands as well as the valleys leading up to the Anatolian plateau. It encompasses the areas of Aeolia, Ionia, Caria, Lydia, Mysia, Bithynia, and Pisidia. Its populations included not only ethnic Greeks, but also peoples, such as the Lydians and Carians who had long- term close contact with them. This region also contains one archive in the form of the Daskyleion material from the Persian period. 71 5. Northern Greece: This region consists of most of the Balkan Peninsula north of southern Greece proper. It encompasses areas such as Thessaly, Macedonia, Thrace, Epirus, and Illyria as well as the Danube valley, with their populations and ethnic groups. 6. The Black Sea: This region consists of the land that surrounds the Black Sea. It includes the northern coast of Anatolia and the Crimea. It includes numerous Greek foundations as well as the Hellenistic Pontic Kingdom and peoples such as the Scythians in the north. 7. Central Anatolia: This region lies sandwiched between Region 6 to the north, Region 4 to the west and Region 10 to the south and east. It consists of the areas of central Anatolia, including the plateau and the southern coast along with the eastern highlands. As such it encompasses the lands of Phrygia (and Galatia in the Hellenistic period), Armenia, northern Mesopotamia, and Cilicia. This region also encompasses the archives at Artaxata and Doliche. 8. Italy: This region consists of the boot of Italy from the Po valley southwards. As such it includes several large important urban centers including Rome and Tarentum. It was inhabited by a variety of different peoples, such as the Etruscans and the Romans, other Italian peoples and Greeks, especially in the south. 9. Sicily: This region consists of the island of Sicily as well as some smaller islands to the north and south: the Lipari Islands and Malta respectively. The region contains several prominent Greek foundations such as Syracuse as well as 72 Phoenician settlements such as Motya. The archive of Selinus is located within this region. 10. The East: This very large region consists of those lands lying to the east of the greater Mediterranean basin. It includes several distinct geographical units, including Mesopotamia, Arabia, the Iranian plateau and Bactria. As such, it includes a great number of different peoples and polities at different times. Of note within this region are the Persian Imperial Capitals, such as Persepolis and Susa, the old Mesopotamian cities, and Greek foundations such Seleucia and Ai Khanoum. The region also contained several ancient archives, namely Persepolis, Nippur, Uruk, and Seleucia, as well as the non-archival collections of glyptic finds from places such as Ur. 11. North Africa: This region consists of the coastal areas of North Africa running westwards from Region 2. It also includes other regions of the Punic west, namely the islands of Sardinia and Corsica and islands further west like Ibiza. The region included both Greek foundations, in the form of Cyrene, and Phoenician/Punic ones such as Carthage. Both Cyrene and Carthage also have published remains of ancient archives. 12. Europe: This region consists of the rest of Europe outside of the Mediterranean basin. It is geographically quite large and peopled with a variety of different groups, but does not, on the whole, possess many comparanda with the motifs of this work. There are no published ancient archives for this period from this region. 73 Please note that, as no occurrences of any motif have been localized to Saharan Africa, this region has been left blank for our purposes and has not received a number. Map 3.1: A Map of the Regions in the Catalogue The Maps: The maps are standardized and generally fairly small in scale, showing most of the ancient Mediterranean and Near East. In such cases as the number of occurrences in a region would hinder the clarity of the map, detail maps of certain regions, specifically of the Aegean (S. Greece, W. Anatolia and N. Greece) and Italy (Italy and Sicily) will be provided. The occurrences are localized on each map in a shape that denotes its medium 74 and a color, ranging from red to dark green, which denotes the similarity of that occurrence to the Kedesh seals, with those that are green being closer than those that are red. The criteria for the color graduation are listed below. For the sake of clarity, only one occurrence is noted for any one site in any one time period on the map itself. The occurrence that appears is the one that shows the closest similarity to the Kedesh exempla. If more than one instance of a motif from the same site shows roughly the same similarity to the Kedesh images, then the one that appears on the map will be based on the medium, starting with glyptics and moving to numismatics before going on to sculpture and so on. Colour Coded Degrees of Similarity for Occurrences of a Motif: • Dark Green: The image in question represents a (near perfect) match to one or more of the examples from the Kedesh corpus. The image may not represent the use of the same seal (and, in the cases of images in different media, it can not) but the image presents the same motif as at Kedesh with all of the same elements. Examples include the composite hippocamp at Delos carrying a riding Eros, or plain Seleucid anchors. • Light Green: The image is very similar to one or more of the Kedesh symbol sealings. Some of the minor elements may be altered between the image and the Kedesh examples, but the two are overall highly similar. An example of this includes the motif of an anchor and dolphin where the dolphin is turned the opposite way. 75 • Yellow: the image represents the same motif category as the Kedesh examples but the images may not look particularly similar in themselves. Various elements may be changed between the various examples. Examples of this include the grand corpus of New Comedy slave masks which are each different one from another but follow similar trends. For the purposes of the charts, only those occurrences of yellow or higher color will be considered instances of the motif in question itself. • Orange: The image represents a similar motif category to the Kedesh examples. There are small but significant differences between the two groups. Certain minor elements may be added or missing. Examples include occurrences of columnar obelisks in regards to depictions of columns themselves. Another possibility is the appearance of other composite birds in comparison to the composite eagles at Kedesh. • Red: The images here are roughly similar to the motif categories at Kedesh. The images either have or are lacking significant elements as compared to the Kedesh examples. This colour included those times when there are multiple related versions of a motif, such as double cornucopias. Examples include an animal standing on a thunderbolt in comparison to thunderbolts, bucrania with or without an object between the horns and caducei with wings. 76 Section 2: Anchors Anchors appear as the main subject on nine of the Kedesh sealings, representing a total of four distinct seals. These impressions can be divided into two distinct types. The first type, represented by two seals (ANC1 and 3), can first be distinguished by its size. In each case, the seal itself was too large for the bulla so that only a portion of the image was ever impressed into the clay. In each case the portion of the image that was impressed was the more recognizable crown and flukes of the anchor. Parts of the anchor, usually the shaft and top of the anchor but including parts of the flukes on occasion, always surpass the edge of the bulla, and the edge of the seal is not seen in any of the Kedesh examples. This type has two variants. In one seal (ANC3), at least, the anchor is also accompanied by the presence of the head of a horse placed along the shaft of the anchor to the left. The presence of this symbol on the other seal of the type (ANC1) can not be confirmed, but it is possible that the horsehead is simply lacking due to the fact that the seal surpasses the size of the bulla. The size of the seal suggests that it served some official purpose. The anchor indicates the Seleucids. According to Appian, the first Seleucus had a signet ring bearing the image of an anchor.195 This personal symbol later became a royal symbol within the Seleucid kingdom.196 Indeed, McDowell thought that the anchor was the symbol of the Seleucid treasury, as indicated by its appearance on large sealings that note the payment or remission of certain taxes at Seleucia.197 On the other hand, Invernizzi sees the anchor motif as referring to a number of Seleucid governmental departments, due 195 Appian, Syr. 56. 196 Newell, 1978, p. 44. 197 McDowell, 1935, I.A.1.C.1, I.A.2.a.1-4 77 to the variety of its appearance on sealings at Seleucia.198 Thus, it could very well be that the two variants present at Kedesh, distinguished by the presence or absence of a horsehead, may represent correspondence from two separate departments or offices of the Seleucid administration. The second type of Kedesh sealing with an anchor, represented by two examples (ANC2 and 4), is distinctly different in both depiction and character from the previous group. Here, the image is much smaller. The whole of the anchor can mostly be seen, along with traces of the edge of the impressing seal. A dolphin wraps itself around the shaft of the anchor in both cases. In once example, the head of the dolphin is pointing down towards the fluke of the anchor while in the other it is pointing towards the top. Such a scene is known elsewhere from gems. The presence of the dolphin entwined around the anchor does not fit well with the standard Seleucid symbolism. The small size also differentiates it from the official Seleucid seals above. Thus, considering the size and motif, it would appear that this type does not represent official Seleucid correspondence, but more likely the seal of a particular individual. In general, the idea of the anchor as a symbol existed at least as far back as the early-6th century BCE. The anchor on its own occurs as a motif on an early Attic Black- figure olpe from the first half of the 6th century that was found on Delos, and on an Attic Black-figure neck amphora from the second half of the same century.199 It also appears as a shield device in a scene on an early Attic Red-Figure lekythos from the last decade of 198 Invernizzi, Bollati, Messina & Mollo, 2004, p. 29. 199 Hägg, 1988, p. 232, FIG.3; Beazley, 1956, #15.34 78 the sixth century.200 While technically, the anchor here was part of a larger scene of armed combat, it does show that the painter thought it appropriate to use it as a symbol on a warrior’s shield. The anchor does appear as the type on a fourth century (400-375 BCE) diobol issue from Apollonia Pontica in Thrace.201 However, it is only in the Hellenistic period that the anchor becomes popular as a symbol, appearing on coins from numerous polities and expanding into various niches, such as mosaics. A great deal of this popularity may have to do with the adoption of the symbol by Seleucus and its promotion by his dynasty, thereby increasing the number of times and the places in which it appeared. Distribution: The distributions of each type and variant are somewhat different one from another and will be treated as distinct but related sets. To begin, there is the motif of the plain anchor by itself. This variant, being the simplest, has the widest distribution of all three. It is also the only variant that appears prior to the Hellenistic period. As noted above, during the 6th-4th centuries, the plain anchor motif appears in one numismatic example from coastal Thrace (Black Sea) and two Attic black figure vessels, one of which comes from Delos (S. Greece). During the Hellenistic period, the occurrences of the motif multiply considerably, especially in regards to the Seleucid Empire. In glyptics, the motif is present on a large- scale official seal from Seleucia-on-the-Tigris (The East) with one known impression and 200 Moore, 1997, #829. 201 ANS Online Catalogue, 2004.14.40. 79 two sealings from Callipolis (S. Greece). As with Kedesh, the two Callipolitan examples are differentiated by size, and, therefore, origin. The larger of the two is probably the Seleucid official insignia, representing correspondence between the Seleucids and the leaders of that city. The second is much smaller and probably represents a private individual.202 In numismatics, the anchor by itself appears fairly frequently on lower denomination coins. Apollonia Pontica (Black Sea), for instance, continued to use the anchor as a type on a Hellenistic Bronze Issue (300-200 BCE).203 The anchor motif also appears in Italy, specifically on bronze coins from Tuder in Etruria (300-1BCE), bronze coins of Etruscan type but indeterminate mint from the same period, as well as bronze Republican coins from Paestum.204 Unsurprisingly, considering its use as a dynastic symbol the anchor as a coin type was fairly popular for small coins within the Seleucid Empire itself. These occurrences tend to appear in The East, especially in and around Mesopotamia itself, but also happen at sites in The Levant. The minting of coins with the plain anchor type was especially popular under the first two kings of the dynasty. Specifically, under Seleucus I and Antiochus I (312-261 BCE), anchors appeared on coins from Seleucia-on-the-Tigris, Susa, Dura-Europos, and Carrhae.205 In the mid-third century, during the interregnum after the death of Antiochus II (246-244 BCE), Apamea also minted coins with an anchor as a reverse type as did Nisibis under Seleucus II (240-230 BCE).206 In certain cases, such 202 Invernizzi, Bollati, Messina & Mollo, 2004, SU 19; Pantos, 1985, #95-96. 203 ANS Online Catalogue, 1955.31.1, 2004.14. 204 ANS Online Catalogue, 1944.100.22, 1944.100.71989, 1944.100.72029-30, 1949.100.22, 195,1954.263.186-88, 1954.263.195, 1954.263.268, 1967.153.208, 0000.999.32579-80; Poole, 1963b, Paestum 63-66. 205 Newell, 1978, #45-46, 58, 60, 291, 297; Newell, 1977, #879, 886. 206 Newell, 1977, #815-817, 1144. 80 as at Seleucia and Carrhae, the obverses of these coins would have a horned horsehead as a type, but the two motifs did not appear together on the same face.207 In the second century BCE, the anchor appeared as a type on Seleucid coins from Antioch under Demetrius II (146-140 BCE) and Antiochus VII (139-129 BCE) as well as at Jerusalem under Antiochus VII.208 Alexander Zebina (128-123 BCE) also minted copper coins with an anchor as a reverse type.209 Several of the successor states to the Seleucids in the same two regions, including the Hasmoneans in The Levant and the Parthians, Characene and Elymais in The East continued the tradition of minting coins with plain anchors as types starting in the second century BCE.210 In other media, a threshold mosaic of an anchor by itself appears on Delos in a House in the Theatre District and on one side of a Late-Hellenistic token from Athens, both in S. Greece.211 In the region of W. Anatolia, the plain anchor motif also occurs on Cnidian amphora stamps of the 3rd-early 2nd centuries BCE.212 The motif also occurs on a lead sling bullet from Dor (The Levant), dated to 139/8 BCE.213 Overall, the plain anchor motif had a fairly broad distribution in the Hellenistic period. It was especially popular in numismatics and in the regions controlled by the Seleucids (The Levant and The East), who used the motif as a dynastic symbol. The motif, however, also had fairly strong 207 Newell, 1977, #886; Newell, 1978, #46. 208 ANS Online Catalogue, 1941.131.1084, 1944.100.76533-76534; 1944.100.77922-77929; 1949.163.975, 1952.142.479, 1984.56.125, 1984.66.125, 1985.69.3, 1999, 32.57. 209 Gardner, 1878, Alexander Zebina 29. 210 Mørkholm, 1991, p. 178; Hill, 1965, Alexander Jannaeus 1-8 & 61-88; Wroth, 1964a, Orodes I 224- 232. 211 Bruneau, 1972, #270 (House IIIS of the Theatre District); Lang & Crosby, 1964, L76b. 212 Grace, 1934, #139-146. 213 Stern, 2000, fig. 142. 81 showings in other regions, such as Greece and Italy, and could appear in other media, such as the Delian mosaic. The next set with the anchor accompanied by a horsehead protome has a very different and more focused distribution from the preceding set. Specifically, it only appears within glyptics on large-scale seals. Within glyptics it actually appears at more places than the anchor by itself. These occurrences include fourteen official seals from Seleucia (The East) with a total of 224 impressions, one impression from Uruk (The East), one published seal from Delos, and another from Callipolis (both S. Greece), where it appears in an unusual sub-variant that contains a bird perched upon the anchor along with the protome.214 In each of these cases the appearance of the motif seems to represent some form of correspondence between Seleucid officials and these specific places. Thus the distribution of the motif is more focused being present only in archives that, other than Kedesh, are found in S. Greece and The East. In addition, the imbalance in the use of the plain anchor motif versus the anchor and horsehead at Seleucia, with one impression versus 224, is interesting in regards to the apparent parity of impressions at Kedesh, though admittedly the sample size here is much smaller. This disparity at Seleucia between the two variants might indicate that, whatever Seleucid official organ each variant was used to represent in glyptics, the organ associated with the anchor and horsehead was predominant at Seleucia while the one represented by the plain anchor was more active at Kedesh. The last set of the anchor motif has the most restricted distribution of all three. 214 Boussac, 1998, fig. 15; Invernizzi, Bollati, Messina & Mollo, 2004, SU 3-16; Pantos, 1985, #94; Wallenfels, 1990, #80. 82 The dolphin and anchor motif does not appear in published examples from other Hellenistic archives or on coins. However, the precise motif does appear in four mosaic panels in private homes on Delos in S. Greece.215 Another example of the anchor and dolphin motif comes in the form of an amphora stamp of the fabricant Μενέσρατοϛ from Rhodes in W. Anatolia that dates to 146-108 BCE. Interestingly one example of this stamp comes from a Rhodian amphora handle that was recovered from Beersheba in southern Israel.216 The mercantile nature of Hellenistic Delos, and the appearance of the dolphin and anchor motif on mosaics in houses there, as well as the nature of the Rhodian amphora as a vessel for the trade of commodities moves the motif out of the political contexts associated with the use of the other two anchor motif variants and into the realm of trade and commerce. The motif might also, considering its appearance at Kedesh and the find of the Rhodian amphora stamp at Beersheba, be one especially appropriate in association with the trade between the Aegean and the Levant. Indeed, the association may go even further than that. One of the houses at Delos where the anchor and Dolphin appears as a mosaic is the House of the Trident. The house gets its name from a mosaic emblem that appears in the portico right in front of one of the entrances that leads via a vestibule to the street.217 The dolphin and anchor motif appears as an emblema in the portico facing a different vestibule with access to the exterior. Interestingly, both emblemata are given relatively equal status of presentation, being the first things seen on the portico floor for people entering from the outside. In addition, the dolphin wrapped around a trident is a well known reverse type of coins from Berytus 215 Bruneau, 1972, #68 (Ilot des Bijoux), #166b( house B of the Inopus Quarter), 228.a (House of the Trident), & 261.b (House IIIN of the Theatre District). 216 Coulson, Mook, Rehard & Grace, 1997, #23; Finkielszteijn, 2001, p. 148, pl. 14.12. 217 Westgate, 2000, fig. 7. 83 dating from the late 2nd-early 1st centuries BCE.218 Furthermore we know of an association of merchants and shippers focussed around the worship of Poseidon, whose attributes were the trident and the dolphin, from Berytus active on Delos in the 2nd century BCE: the Poseidoniasts of Berytus.219 These two factors: the close association of the trident and the anchor and dolphin motif in the House of the Trident and the fact that the motif appears in three other mosaics at Delos but not elsewhere, both point to the anchor and dolphin motif potentially being one of importance to the Poseidoniasts. Certainly, both the anchor and the dolphin would be appropriate images for an association formed from sea traders. The presence of the motif on several seals at Kedesh would likewise represent the interaction of some members of this group with the archive in some fashion. In addition, considering this and the fact that some of the amphorae at least ended up in The Levant, the Rhodian amphora stamp of Μενέσρατοϛ may represent amphorae that were originally meant for sale to members of the group or even a consignment of vessels to be carried in their hulls.220 In the end, the appearance of the anchor at Kedesh fits in well with the site’s overall context. The presence of ANC1 and ANC3 represents some interaction with Seleucid officials on some level. This is completely in keeping with an administrative site that was under Seleucid control. The anchor by itself appears consistently in regions under Seleucid control and in media, such as coins and large official seals that indicate that influence. The presence of the horsehead is somewhat different in that, though it still appears to be Seleucid, it does not appear in media outside of glyptics. The other two 218 Hill, 1910, Berytus 16. 219 Meijer, & Von Nijf, 1992, p. 57. 220 The clincher here would be the presence of Μενέσρατοϛ stamps at Kedesh. However, considering the date so close to the end of the functional life of the archive, this is unlikely. 84 anchor seals (ANC2 and ANC4) are more enigmatic and idiosyncratic. This motif does not seem to be linked with the Seleucid dynasty. Instead, the motif appears in Delian mosaics and in a Rhodian amphora stamp, suggesting a commercial not political context. Catalogue: ANC1 Number of Sealings (Bullae): 3 (K99 0002, K99 0011, K00 0362) Maximum Preserved Dimension: 16 mm. Shape: large, flat seal Description: An anchor. In all three cases only the lower portion of the anchor with the tines, shaft and ring are preserved. The shaft and tines are both flat and broad. There are no concrete traces of a horse’s head visible within the impression. ANC2 Number of Sealings (Bullae): 2 (K99 0065, K00 0471) Maximum Preserved Dimension: 10 mm. Shape: convex oval seal Description: Dolphin entwined around the shaft of an anchor. The head is towards the tines. Traces of the bottom and crossbar of the anchor are visible. It is similar to ANC 7, but the tines of the anchor are small and do not surpass the head of the dolphin. 85 ANC3 Number of Sealings (Bullae): 3 (K99 0123, K99 0236, K99 0586) Maximum Preserved Dimension: 20 mm. Shape: large, flat seal Description: An anchor. The lower portion is preserved, showing the flukes, ring and shaft. The Shaft and flukes are narrower than ANC1. There is a horsehead protome facing left to left of shaft. 86 ANC4 Number of Sealings (Bullae): 1 (K00 0250) Maximum Preserved Dimension: 11 mm. Shape: flat oval seal Description: A complete anchor with crossbar, tines, shaft and ring. Around it twines the form of a dolphin with its snout pointed up. It is similar to ANC2, but the anchor is of a different shape and the dolphin is facing the opposite direction. Chart 3.1: The Distribution of Plain Anchors in the Persian Period               87 Map 3.2: The Distribution of Plain Anchors in the Persian Period Chart 3.2: The Distribution of Plain Anchors in the Hellenistic Period                   88 Map 3.3: The Distribution of Plain Anchors in the Hellenistic Period Chart 3.3: The Distribution of Anchors with Horsehead Protomes in the Hellenistic Period                 89 Map 3.4: The Distribution of Anchors with Horsehead Protomes in the Hellenistic Period 90 Chart 3.4: The Distribution of Anchors with Dolphins in the Hellenistic Period Map 3.5: The Distribution of Anchors with Dolphins in the Hellenistic Period              91 Section 3: Bucrania At Kedesh, there are a total of eight impressions of bucrania, or more specifically bucephalia, or bull’s heads, as all of the examples appear to be fleshed and not just the bones of the bull’s cranium. For our purposes here, however, the terms bucranium shall be used to denote either a bull’s head or skull shown frontally without the presence of a neck or body, as it often is in scholarship. The eight bucrania impressions at Kedesh represent the use of a total of six separate seals. In each case, the head of the hull is displayed frontally with the horns curving upwards and the ears of the bull spreading out horizontally below them. In certain cases (BUC2-4), the bull’s muzzle is flanked by sacrificial bands or ribbons which hang down from the horns. In two cases (BUC2 and 5), the bull’s head is apparently wreathed, since sprigs of vegetation spring out from behind the horns on either side of the head. Also, in certain cases (BUC1, 4 and 6), there is some sort of object positioned on or above the head between the horns. In the case of one seal (BUC6), with a single impression, the object is a bit ambiguous, but appears to be a ship’s prow and ram or possibly a rudder. In two other seals (BUC1 and 4), representing some four impressions in total, the object is a bird in profile to the right. The seal BUC1 is also interesting, in that the bird appears to be a falcon and has what looks to be the crook of the Egyptian god Horus sprouting from its breast. In addition, if we were to remove the horns and ears of the bull from the image, we would be left with a Horus-falcon sitting on top of a trumpet shaped flower with a series of lines indicating the flower’s petals. The motif of Horus in various forms placed upon a flower has a long history in Egyptian art from which it passed into the Phoenician. For instance, a fifth century BCE scarab from Tharros on Sardinia shows a falcon in the 92 crown of Upper and Lower Egypt sitting on a lotus flower.221 A similar scene also appears on an 8th century ivory carving discovered at Nimrud.222 Kedesh itself has another sealing (K99 0015) that shows the infant Harpocrates (the Hellenized version of Horus) seated on a bloom with his hand characteristically to his lips (see figure 3.1). It appears therefore that BUC1 is a modification of a well known motif. Considering the small amount of modification involved, namely the adding of four short lines to create the horns and ears, the seal itself may have originally consisted of the Horus on the flower motif and was then subsequently changed to that of the bucranium bearing a bird. In so doing, the seal-owner/user at the time might have had a variety of reasons. One possibility is that the modifications lessened the immediate appearance of the Egyptian- influenced iconography, perhaps as a way of ingratiating himself with the Seleucids who had captured Kedesh and the surrounding region from the Ptolemies after the Battle of Paneion in 200 BCE. Certainly, Wallenfels puts the origin of the motif in Mesopotamia and The East, with antecedents dating back into the Bronze Age.223 On the other hand, it is also possible that the modifications were meant to change the scene to refer to another, still Egyptian, deity represented by a bovine such as Apis. 221 Boardman, 1997, fig. 5/27. 222 Acquaro, 1988, #80. 223 Wallenfels, 1990, p. 697. Specifically, Wallenfels points to Kassite cylinder seals, Iranian metal work and Achaemenid seals that bear images of birds perched atop the facing heads of bull figures. 93 The bucranium itself as a symbol is polyvalent in the meanings that it can convey. In the Near East, the bull was a symbol of power and strength, especially under the Achaemenids who used it as a symbol of political power. This idea would later be taken up by various Hellenistic kings, such as Seleucus I and Demetrius Poliorketes. Furthermore, in Greece, the bull’s function as a typical animal for sacrifice also meant that the bucranium can be taken as a symbol of a sacrifice, especially one that has been successfully performed.224 As such, they can be also taken as a symbol of general piety or of a good relationship to the gods. In line with this idea is the presence of sacrificial bands or fillets on some of the Kedesh examples and elsewhere. On a more specific level, the bucranium may represent specific deities who were linked with the bull or had one as their attribute. For instance, both the Phoenician Ba’al and the Greek Poseidon were associated with bulls. The latter is especially interesting here as BUC6 with its possible ship’s prow above the bucranium might suggest the Greek sea god, but it is impossible to ascertain. Likewise, as noted, the Horus-falcon of BUC1 may be referring to the Egyptian Apis bull. 224 Webb, 1996, p. 30. 94 Distribution: The use of the bucranium/ bull’s head motif has an incredibly long history, appearing in many places at different times and in different media. In glyptics the motif is especially long lived, dating back to the Bronze Age in the Aegean. For instance, a Mycenaean sardonyx from the Argive Heraion shows a frontal Bull’s head with a double ax between its horns.225 An Archaic period stone from Melos (S. Greece) also displays a frontal bull’s head with downturned horns and a nose ring.226 A Classical Period chalcedony from Patrae in the same region is especially interesting for this discussion as it shows a bucranium flanked by pendant fillets tied to the end of its horns.227 And the motif continued well into the Roman period, where it appears on glass pastes.228 In the realm of glyptic, examples of the bucranium motif from the 6th-4th centuries BCE appear in two of the archives. Selinus (Sicily) has five sealings that show bucrania. One of these shows a plain bucranium with fillets while the other four all have something between the horns, including a wreath, stars and a lion’s head.229 Carthage (N. Africa), for its part, has four sealings that show bucrania, all of which date from the fourth-third centuries. In each of these cases there are objects between the horns. In two of the examples, the object consists of two horseheads placed back to back, while a third sealing has a thunderbolt and the fourth a jug.230 Besides these examples, there are also the previously mentioned gems from Melos and Patrae that date to this period. In both of 225 Furtwängler, 1900, pl. II, p. 42. 226 Boardman, 1970, pl. 275. 227 Boardman, 1970, pl.588. 228 Brandt & Schmidt, 1970, pl. 181, #2066; Platz-Horster, Schulter & Zazoff, 1975, pl. 176, #1285 229 Salinas, 1883, CCCXXXIII-CCCXXXVI, CCCXXXVIII. 230 Berges, Ehrhardt, Laidlaw & Rakob, 1997, #107-110, 832.2 & 833.1. 95 these cases, the bucranium is plain, lacking an object between its horns. The bucranium also makes frequent appearances on coins, starting very early. For instance, before switching to the more famous “Owls” the city of Athens (S. Greece) minted coins bearing a bucranium as a type.231 Other places where the bucranium appeared as a type on coins from the sixth through fourth centuries include Aegina (404- 350 BCE), Cydonia (400-300 BCE), Eretria (6th-4th centuries BCE), Moda (400 BCE), Phaestus (431-300 BCE), Phocis (6th-4th centuries BCE), Polyrhenium (4th-3rd centuries BCE), Praesus (before 400 BCE), and Same (370-189 BCE) in S. Greece; Assus in the Troad (400-241 BCE) and Lycia (500-460 BCE) in W. Anatolia; and Abdera (before 400 BCE), Craneia (431-330 BCE) and Macedonia (530-480 BCE) in N. Greece.232 None of these coins displayed any object between the horns of the bucranium. In other metalwork from the same period, the plain bucranium appears on a set of sling bullets that were originally from Megalopolis and Gortyn in S. Greece that bear the name Cleander in the genitive and that have been dated to the mid fourth century BCE.233 Likewise, the plain bucranium motif occurs on a Praenestin mirror of the early fourth century.234 Besides coins and metalwork, the bucranium also makes appearances as a symbol in ceramic vessels of the period. The plain bull’s head is used as a device on shields 231 ANS Online Catalogue, 1967.152.266. 232 ANS Online Catalogue, 1909.999.44, 1941.153.419, 1941.153.421-3, 1944.100.19676-8, 1944.100.19680-7, 1967.46.52, 1976.216.1, 1992.54.298-309, 2008.1.14; Gardner, 1963a, Cranium 48-55, Same 34-36; Head, 1963b, Phocis 1-51, 78-90, 94-104, Eretria? 19-20, Eretria 13-14, 33-39; Head & Gardner, 1963, Abdera 31; Plant, 1979, #1033, #1036, #1039-40, #1429, #1433, #1750, #1760, #2581, #2722b; Wroth, 1963, Cydonia 12-13, Phaestus 3, 21-26, Polyrhenium 1-8, Praesus 7-8, 11; Wroth, 1964c, Assus 4-7. 233 Foss, 1975, #6-8, #14. 234 Eldritch, L1918, p. 270. 96 carried by figures in some Attic Black-Figure and Red-Figure scenes, as well as a decorative or stand alone motif in Attic (S. Greece) and Apulian (Italy) Red-Figure.235 A vessel in the shape of a bull’s head, dating to the sixth century BCE, has also been found on Rhodes in W. Anatolia.236 The motif also appears on a lamp from the Athenian Agora (S. Greece), dating from the second quarter of the 4th through early third centuries BCE.237 The bucranium motif, both with the head by itself and with an object on top between the horns, appears on sealings from a number of the Hellenistic archives. Besides the earlier examples from Carthage that last into the third century, the bucranium appears on four published sealings from Paphos in The Levant, on six sealings at Callipolis in S. Greece, one sealing at Cyrene in N. Africa, one published example from Artaxata in C. Anatolia, some fifty-nine separate sealings at Seleucia, representing the use of fifty separate seals, and four seal impressions at Uruk both in The East.238 Most of these examples show a straightforward frontal bull’s head, perhaps with flanking fillets. Objects placed above the head do appear occasionally at some of the archives. The Seleucene examples are particularly cogent here as they seem to present similar variants to those found at Kedesh. At Seleucia, the examples of bucrania can be characterized as simple bucrania, bucrania with a symbol above the bull’s head, which is usually a crescent in that corpus, and bucrania supporting a bird in profile between the 235 Moore, Philippides, & von Bothmer, 1986, #260, #687, #719, #741; Moore, 1997, #358, #526, #1408; Trendall, 1970, figs.1, 3a-b; Trendall, 1984, figs.7, 8b, 9a-b. 236 LIMC, Apis, #10. 237 Howland, 1958, #267. 238 Kachatrian, 1996, fig. 38; Michaelidou-Nicolaou, 1979, pp. 414-5; Nicolau, 1978, pl. CLXXVII.12; Pantos, 1985, #1-6, #8, n.103; Maddoli, 1965, #773; McDowell, 1935, pl.II n.8; Invernizzi, Bollati, Messina & Mollo, 2004, PA 23-73; Wallenfels, 1990, #1055-1058. 97 horns. All three variants can be with or without ribbons hanging down from the horns.239 The lone published example from Artaxata is also topped by a star along with other symbols, while one Paphian example has the sun-disk of the goddess Hathor and the other three are plain with inscriptions of cities. Finally, the Uruk examples include simple bucrania as well as examples topped by birds and one example over which a crescent moon appears. During the third through first centuries, some of the same polities that had minted coins with a bucranium either with or without fillets as a type continued to do so while different polities also started the practice. For example, the motif is found on a couple of Seleucid issues from Seleucus I at Sardis (W. Anatolia) and Seleucus II at Seleucia (The East).240 Besides Seleucid issues, bucrania also appear on coins from Cythera (250-146 BCE), Lappa (200-67 BCE), Phocis (339-146 BCE), Polyrhenium (4th-3rd centuries BCE), Praesus (300-200 BCE), and Same (370-189 BCE) in S. Greece; Assus in the Troad (400-241 BCE), Caystriani (2nd-1st centuries BCE) and Cyzicus (300-200 BCE) in W. Anatolia; Corcyra (300-48 BCE) in N. Greece; Chersonesus (300-200 BCE) and Paphlagonia (under Pylamenes) along the Black Sea; Metapontum (Hellenistic), Rubi (Hellenistic), Tarentum (after 281 BCE), Lucanian Thurium (Hellenistic), and Vestini (Hellenistic) in Italy; as well as the Arabian Himyarites in The East.241 The vast 239 Invernizzi, Bollati, Messina & Mollo, 2004, p. 159. 240 Newell, 1977, pl.I n.23; Newell, 1978, pl.XVII n.6. 241 ANS Online Catalogue, 0000.999.32914, 0000.999.32918-20, 0000.999.32930, 0000.999.32934, 0000.999.32940, 1935.24.13, 1935.36.805-8, 1935.36.810, 1935.36.813, 1935.999.325, 1944.100.14370-1, 1944.100.69494-5, 1944.100.69815-9, 1944.100.72167-70, 1944.100.72171-5, 1944.100.72177-81, 1944.100.72184-8, 1944.100.72189-93, 1944.100.72195, 1944.100.72202-10, 1944.100.72212, 1944.100.72214-7, 1944.100.72219, 1944.100.72221, 1944.100.72228-31, 1970.251.1, 1974.226.52, 1977.158.838, 1978.5.1-2, 1991.2.128, 1999.13.58; Gardner, 1963a, Cythera 11, Same 34-36; Gardner, 1963b, Corcyra 207-214, 434-446, 449-468 & 479-481; Head, 1901, Caystriani 5-9; Head, 1963b, Phocis 105-113; ; Plant, 1979, #1032, #1034-5, #1037-1038, #1041-1043, #2560, #2563, #2583, #2617, #2622, 98 majority of these bucrania do not display anything above the head between the horns. The coins from Cythera and Caystriani, however, show a bull’s head with objects above between the horns. The Cythera example has an ivy leaf, while at Caystriani the bucranium is in the form of a lyre with the two horns forming the uprights and the strings between them. The Himyarite examples, which date to the second century, are also highly unusual in that they show a bull’s head with antelope horns rising from the top and flanking a set of feathers, giving an effect not unlike the headdress of the goddess Isis. In other Hellenistic media, plain bucrania without object appears a silver vessel from a third century Etruscan tomb near Clusium (Italy).242 In addition, bucrania both with and without fillets are used as motifs in painted and molded decoration for vessels during the Hellenistic period at Athens, Corinth and Gialova in S. Greece as well as Samaria in The Levant.243 In none of the cases listed above do the bucrania display some sort of object between the horns. The bucranium also appears in plastic decoration on Hellenistic braziers from several sites including Ashdod in The Levant; Naucratis in Egypt; Athens in S. Greece; Calymna, Ephesus, and Halicarnassus in W. Anatolia, and Lanuvium in Italy.244 Most of these examples lack an object between the horns, but one of the examples from Halicarnassus has a rosette. The plain bucranium also appears on Hellenistic lamps of the second and first centuries BCE from Alexandria and an unidentified site in Egypt; Pergamon in W. Anatolia; Capua and an unidentified site in #2666; Poole, 1963b, Vestini 1, Rubi 2-3, Tarentum 432, Metapontum 39-44, Thurium 139; Wroth, 1963, Polyrhenium 9-12; Wroth, 1964b, Pylamenes 2-3, Wroth, 1964c, Assus 4-7. 242 Eldritch, 1918, #13.2864. 243 Crowfoot, 1957, Q285. Edwards, 1986, #10 & #15; Rotroff, 1997, pp. 56-79, #603, #1571, #1672, #1683. 244 Rahmani, 2001, Ha 3-5, Ep 1, Ky 3-4, At 1, La 1, Na 3. 99 Italy; and Carthage in N. Africa.245 In addition, the plain bucranium also appears on the stamps of Cnidian (W. Anatolia) amphora handles of the third and second centuries BCE.246 The plain bucranium also appears at the site of Marisa (The Levant) in the form of a terracotta appliqué, possibly for a wooden sarcophagus.247 Bucrania also become a very popular motif to use in architectural decoration, usually in sculpture but also painting and mosaic, beginning in the third and lasting into the second centuries BCE. The plain bucranium motif appears in sculpture in many buildings in W. Anatolia in this period, including on the Temple of Apollo at Chryse; the Megalesion, the Temple of Demeter and the propylon of the Sanctuary of Athena at Pergamon; the Ephebeion at Priene; the Sanctuary of Artemis and the Agora at Magnesia. Elsewhere around the Aegean, the motif appears on the Altar of Poseidon and Amphitrite at Tenos; on private houses, the building of the Poseidoniasts, and the theater on the island of Delos in S. Greece; and on the Propylon of Ptolemy II and the Arsinoeion on Samothrace in N. Greece.248 There is also at Delos a mosaic frieze with bucrania in a garland with masks.249 Outside of the region of the Aegean proper, the plain bucranium is used as a motif on two Hellenistic tombs, one sculpted and the other painted, at Sveshtari and Kasanlik respectively, in Bulgaria (N. Greece) and as architectural sculpture on a couple of lararia at Pompeii (Italy).250 From these occurrences, it is apparent that the use of the bucranium motif in general at Kedesh is part of a much larger and fairly popular constellation of motif-use. 245 Bailley, 1975, Q607, Q626, Q704-05; Mlynarczyk, 1997, Type G; Schaefer, 1968, Q39. 246 Grace, 1934, #108-20 & #202-17; Lenger, 1957, #90 & #115-8. 247 Erlich & Kloner, 2008, #203. 248 Webb, 1996, pp. 29-30, 55-56, 60, 93, 95, 101, 133, 136-141, & 148-9. 249 Bruneau, 1972, mosaic 68.M (pp. 160-165). 250 Boyce, 1937, #61, #211; Verdiani, 1945, fig.7; Webb, 1996, p. 32. 100 During the 6th-4th centuries BCE, the motif appeared in several media, but clustered predominantly in the region of S. Greece with only a few examples appearing in other regions. In the subsequent period, the plain bucranium continued to appear in multiple media. The region of S. Greece still produced the largest number of occurrences of the motif, but other regions, namely W. Anatolia and Italy now each produced several instances of the motif as well. Elsewhere the motif appears much less frequently, but isolated occurrences appeared in almost every region. This is in marked contrast to the distribution of the variant that has an object, be it a bird, or an astronomic symbol or something else, between the bull’s horns. This version appears much less frequently than the plain bucranium, and in only a few types of media. Indeed, except for a couple of numismatic examples, the bucranium with object variant is limited to glyptics. The locations of these examples are also of interest in that they tend to lie in areas outside the traditional Greek homeland such as The Levant (Paphos), C. Anatolia (Artaxata) The East (Seleucia and Uruk) and N. Africa (Carthage). Indeed the Carthaginian examples consist of nothing but bucrania topped by objects. All of these facts indicate that the variant is much more limited in its diffusion, and possibly with a source of impetus outside the Aegean proper where its cousin was so popular in the Persian and Hellenistic periods. The presence, in two different forms, of this variant in the Kedesh sealings also fits in well with this pattern. At the same time, the exact object atop the head of BUC6 is unparalleled elsewhere and, while birds do appear atop bucrania at Seleucia, a bird with the Horus crook does not. Thus, these two Kedesh examples are still somewhat idiosyncratic, even within this variant. 101 Catalogue: BUC1 Number of Sealings (Bullae): 2 (K99 0075, K99 0104) Maximum Preserved Dimension: 12 mm. Shape: flat oval seal Description: A bird on a frontal bull’s head. The bird standing to the right has wings and tail drawn by three fanning lines from body. To the right of the body is a possible Horus crook. The bull head is broad with short, curved horns, ears and a large bulbous nose. The forelock and snout are defined by several prominent vertical lines. BUC2 Number of Sealings (Bullae): 1 (K99 0084) Maximum Preserved Dimension: 9 mm. Shape: flat oval seal Description: A frontal bull's head with pendant fillets and wreath. The snout is broad and the nose is small but bulbous. The horns are short and point diagonally up. The forelock 102 is not well defined. BUC3 Number of Sealings (Bullae): 1 (K99 0097) Maximum Preserved Dimension: 11 mm. Shape: flat oval seal Description: A frontal bull's head/ bucranium with pendant fillets. The image is not well impressed, with the tip of the muzzle missing and with the left horn and ear outside the impression. The snout is long and narrow. Traces of the two fillets can be seen hanging down. The up-thrusting right horn is visible. BUC4 Number of Sealings (Bullae): 1 (K99 0576) 103 Maximum Preserved Dimension: 14 mm. Shape: flat seal Description: A bird on a frontal bull's head. The bird faces right and is ill defined and incomplete in the impression. The bull's head is stylized, with a long and narrow snout, globular eyes, and bulbous nose and forelock. The horns and ears are a set of four lines in the shape of an 'X'. There is an extraneous “horn” rising from the center of the head in a straight line, which might be the breast of the bird. BUC5 Number of Sealings (Bullae): 2 (K00 0074, K00 0211) Maximum Preserved Dimension: 10 mm. Shape: slightly convex oval seal Description: A frontal bull's head. The left side is less well preserved in both impressions than the right. The snout is 'U'-shaped. The right horn is long and thin and curls upwards. There are traces of a wreath above the horns and of ears below. There are also traces of 104 ribbons or fillets hanging from knots located below the ears. BUC6 Number of Sealings (Bullae): 1 (K00 0476) Maximum Preserved Dimension: 10 mm. Shape: convex oval seal Description: A frontal bull's head topped by symbol. The bull's head is lacking its nose and portions of its right horn. The forelock is prominent. The snout is long and narrow. The horns are short and curve upwards, while the ears go straight out from the snout. The symbol is somewhat unclear, though it appears to be the ram and prow of a ship. It consists of a single vertical line with one long and four short horizontal lines attached. The long horizontal line ends in a loop. 105 Chart 3.5: The Distribution of Plain Bucrania in the Persian Period Map 3.6: The Distribution of Plain Bucrania in the Persian Period                    106 Map 3.6a: Detail Map Showing the Aegean Region 107 Chart 3.6: The Distribution of Bucrania with an Object Between the Horns in the Persian Period              108 Map 3.7: The Distribution of Bucrania with an Object between the Horns in the Persian Period 109 Map 3.7a: Detail Map showing the Aegean Region 110 Chart 3.7: The Distribution of Plain Bucrania in the Hellenistic Period Map 3.8: The Distribution of Plain Bucrania in the Hellenistic Period                     111 Map 3.8a: Detail Map showing the Aegean Region 112 Map 3.8b: Detail Map showing Italy 113 Chart 3.8: The Distribution of Bucrania with an Object between the Horns in the Hellenistic Period                114 Map 3.9: The Distribution of Bucrania with an Object between the Horns in the Hellenistic Period 115 Map 3.9a: Detail Map showing the Aegean Region 116 Map 3.9b: Detail Map showing Italy 117 Section 4: Caduceus One sealing from Kedesh displays a simple caduceus. Another two, previously published sealings from Kedesh also display the caduceus. These two sealings, however, are inscribed in both Aramaic and Greek, giving the title of a group of officials (judges or špṭm) in the common assembly (κοινοδήµιον) of the city of Tyre and giving the year in two separate eras, one of which was the era of the city of Tyre.251 There is no trace of any inscription around the caduceus sealing here (CAD1). The field of the impression here is also fairly concave, which is unlike the more or less flat fields of the inscribed impressions. These facts indicate that CAD1 comes from a different seal than the ones that made the inscribed impression. The caduceus in CAD1 takes the form of a simple staff topped by a pair of entwined snakes that form a circle topped by a pair of comma shaped “horns” for the snakes’ heads. The symbol of the caduceus had many potential valences in the Hellenistic period, stemming from a couple of sources. Certainly the most widely known source comes from Greek contexts, where the caduceus (or kerykeion) was the staff of office for heralds and, more famously, the main attribute of the god Hermes. This last association was an ancient one by the time of the Kedesh archive. It appears, for example, with the god on a scene from a Corinthian pinax, dated to the second quarter of the seventh century BCE.252 Hermes was a god concerned with guiding travelers, including the souls 251 Ariel and Naveh, 2003, pp. 64-70. 252 Chittenden, 1947, p. 100, pl.XX, a. The fragment shows Poseidon and Amphitrite in a chariot followed by a bearded figure carrying the caduceus. Most of the figure is not preserved, but Chittenden identifies him as Hermes by the beard and caduceus. 118 of the dead and protecting flocks as well as commerce.253 The caduceus thus had associations with protection, both through Hermes and through the inviolability of heralds, and generally of increase and prosperity through the god’s focus on flocks and commerce. Beyond the Greek tradition, the caduceus also made independent appearances in Phoenician and Punic contexts. The use of the caduceus in these contexts appears to have developed connotations independent of the Greek. For instance, the caduceus may have had lunar associations among the Phoenicians.254 In addition, starting in the fifth century and becoming very popular in the third, the caduceus was frequently associated with the schematized female form of the symbol of the Phoenician goddess Tanit. This phenomenon is especially apparent at Carthage and other sites of the Punic west where the symbols often appear together on carved stone stelae. In third century cases from Cirta, for example, the Tanit symbol is shown holding a caduceus.255 In certain cases, the two symbols might actually become conflated the one with the other, such as the Tanit symbol appearing with horns, as in the case of a lead weight which might be from Tyre.256 In such contexts, therefore, the caduceus would also have associations with fertility, through the nature of the goddess, in addition to lunar associations. Such an association with fertility is born out by a 3rd-2nd century sealing from the archive in Carthage that couples a cornucopia with a caduceus.257 253 Ariel and Naveh, 2003, p. 69; Chittenden, 1947, pp. 89-91. 254 Lipinski, 1995, pp. 203-9. 255 Moscati, 1988, p. 306. 256 Dunand & Duru, 1962, Fig.76. 257 Berges, Ehrhardt, Laidlaw & Rakob, 1997, #819. The authors here see the caduceus in this sealing acting as a symbol of the goddess in itself, in order to explain the association between the two items. 119 And so, the caduceus as a symbol could represent a variety of generally favorable concepts, depending upon the context of its use. Distribution: The caduceus used as a symbol, plain and by itself, appears with middling frequency during the Persian and Hellenistic periods. It is certainly more popular in some media than others. In many cases, the caduceus motif is altered by/combined with other iconographic elements in a scene that differentiate it from the simple version of the motif at Kedesh. During the 6th-4th centuries, there are fewer examples of the caduceus motif overall. For instance, it only appears in a couple of the archives. The archive at Carthage in N. Africa has a single sealing from the 4th-3rd centuries that shows a single plain caduceus.258 The motif also appears on sealings in other related forms, either winged or in association with other symbols. For example, there is the previously mentioned Carthaginian example where the caduceus is next to a cornucopia. Selinus in Sicily has four sealings of this sort, one of which shows a winged caduceus, while the other three show a caduceus in association with a cornucopia, grain or a dolphin.259 The caduceus motif, plain and unassociated, does appear more frequently on coins during this period. Examples from the sixth through fourth centuries include issues from Pheneus (431-300 BCE), Sciathus (400-200 BCE) and Sybrita (4th century BCE) in S. Greece; Clazomenae (387-300 BCE), Lampsacus (4th-3rd Centuries BCE) and 258 Berges, Ehrhardt, Laidlaw & Rakob, 1997, #820. Here, the authors take the symbol to be standing in for the sign of Tanit, since the caduceus is placed rather oddly by itself on a short ground line. 259 Salinas, 1883, CCCXXV, CCCXLV, CCCLIX, CCCXCII. 120 Phocaea (400-300 BCE) in W. Anatolia; and Aenus (400-350 BCE), Macedonia (475- 435 BCE), and Sestus (before Alexander the Great) in N. Greece.260 Issues from the same period wherein the caduceus was shown either combined or associated with other symbols were produced at Thebes (387-374 BCE) in S. Greece, where the caduceus was physically combined with the club of Hercules, as well as at Aenus in N. Greece.261 The coins with the caduceus motif from this period therefore cluster in those regions that center on the Aegean. This is interesting considering that the glyptic examples of the period lie further west. Outside of coins, the appearance of the motif in the 6th-4th centuries is more erratic. There is also an example of a lead weight that is now in the Louvre which bears the image of a caduceus and a Phoenician inscription, which is dated paleographically to the first half of the fifth century, stating that it belonged to (the temple) of Melqart of Tyre in The Levant.262 During the Hellenistic period, the number of examples of the caduceus motif increases. In the realm of glyptics it does appear on the two previously noted sealings from Kedesh (The Levant) that bear inscriptions. The inscriptions here are bilingual, with part being in Phoenician and part in Greek, and give a date of 164 BCE using both the Seleucid era and the era of Tyre.263 This is especially interesting considering the lead weight, which is also associated with Tyre. Elsewhere, the motif appears in published examples on four sealings from Seleucia in The East, each representing a separate 260 ANS Online Catalogue, 1944.100.45940, 1944.100.46760, 1990.47.7; Gardner, 1963a, Pheneus 6, 17- 20; Head and Gardner, 1963, Aenus 31a, Sestus 9; Kroll & Walker, 1993, #810; Plant, 1979, #2300, #2303, #2307, #2309, #2316; Wroth, 1963, Sybrita 2. 261 Head, 1963a, Thebes 193-198; Head and Gardner, 1963, Aenus 29-30, 32-39. 262 Bordreuil & Gubel, 1988, II.4. 263 Ariel & Naveh, 2003, pp. 64-67. 121 seal.264 Seleucia also has one example in the form of a surface find sealing which shows a caduceus next to a column.265 Cyrene in N. Africa, for its part, has seven sealings that show a caduceus in association with other objects, such as clasped hands, cornucopia, grain and/or a crab.266 During the Hellenistic period, the caduceus on its own appeared on issues from Pheneus (300-240 BCE), Sciathus (400-200 BCE) and Syros (3rd-1st centuries BCE) in S. Greece; Clazomenae (after 300 BCE), Bithynia under Prusias I and II (183-149 BCE) and Lampsacus (4th-3rd centuries BCE) in W. Anatolia; Aenus (280-200 BCE) in N. Greece; Callatis (300-200 BCE) in the Black Sea region; Aspendus (c. 300 BCE) in C. Anatolia; Hipponium, Populonia, Republican Rome (280-276 BCE) and an unidentified central Italian mint in Italy; Calacte (275-212 BCE) and Menaenum (275-212 BCE) in Sicily; and Bactria under Demetrius I (200-190 BCE) in The East.267 As in previous centuries, several polities minted coins during the Hellenistic period that showed variants of the caduceus as a type, including winged caducei, a caduceus combined with the club of Hercules, or a caduceus incorporated with other items and symbols. For instance, the Seleucid king Antiochus II (261-246 BCE) minted an issue at the city of Bactra (The East) that showed a set of crossed caducei on the reverse.268 Other places with coins bearing caduceus variants include Chalcis sub Libano (1st century BCE) in The Levant; 264 Invernizzi, Bollati, Messina & Mollo, 2004, Og 293-6. 265 McDowell, 1935, III.D.1.d.i. 266 Maddoli, 1965, #887, #98c5-7, #994-6, #1111. 267 ANS Online Catalogue, 1944.100.751-2, 1944.100.779, 1944.100.15238, 1944.100.41961-2, 1954.263.100-5, 1954.263.153, 1957.52.101-7, 1969.83.401-6, 1969.83.466-8, 1979.45.13-14, 1981.999.2, 1992.34.1, 1993.30.2; BMC4, Syros 20, BMC5, Uncertain Central Italy Class III 14-17, Class IV 12-14, Various 50-1, Hipponium 4-6, Populonia 3, BMC6, Pheneus 21; BMC9, Sciathus 1-5; BMC13, Clazomenae 100-1; BMC15, Prusias II 19; BMC18, Calacte 6, Menaenum 4; Kroll & Walker, 1993, #819; Plant, 1979, #1089, #2299, #2301, #2316. 268 Newell, 1978, #716. 122 Knossos (with wings, 200-67 BCE) and Sparta (as Laconia, with club, 146-32 BCE) in S. Greece; Caystriani (with wings, 2nd-1st centuries BCE), Nicaea (united with club, 62- 59 BCE) and Samos (peacock on caduceus, 129-120 BCE) in W. Anatolia; Callatis (300-200 BCE) and Paphlagonia (with wings, 133-103 BCE) along the Black Sea; Metapontum (with wings) and a 1st century Republican Rome series (with an ear of wheat and fasces) in Italy; and Cephaloedium (with wings, 275-212 BCE) in Sicily.269 In other media, the use of a plain caduceus by itself as a symbol is again more sporadic. As previously noted, the caduceus appears along with other symbols, especially that of the goddess Tanit, on Punic stelae from Carthage as well as the one example from Cirta.270 The caduceus by itself appears as a motif on Hellenistic amphora handle stamps from different sites, including Cnidus (late-3rd-2nd centuries BCE) and Rhodes (2nd century BCE) in W. Anatolia; and Thasos (3rd century BCE) in N. Greece.271 The caduceus does occasionally also appear on Hellenistic mould made lamps, as in one second century example from the Athenian Agora and on Hellenistic lamps from Delos (both in S. Greece) as well as on second-first century lamps from Alexandria (Egypt).272 Also the caduceus appears on eleven Hellenistic lead tokens, either in the plain form or 269 ANS Online Catalogue, 0000.999.52947, 1944.100.15238, 1944.100.41498, 1947.97.321, 1957.52.101- 4, 1957.52.106, 1998.18.152, 2002.46.417-9, 2002.46.422-4; Edwards, 1933, #363, #367, #453; Gardner, 1963a, Laconia 14-21; Head, 1901, Caystriani 1-4; Head, Poole & Gardner, 1963, Cephaloedium 5; Kroll & Walker, 1993, #819, #951; Poole, 1963b, Metapontum 204; Wroth, 1963, Knossos 70; Wroth, 1964b, Nicaea 1. 270 Acquaro, 1988, #184, #186, #191-193, #202, #207. Each of the examples noted here dates to the Hellenistic period. In most of these cases, the stele shows the caduceus in association with other symbols. In #207, the lone example from Cirta (all of the others come from Carthage itself) the caduceus is shown in the hands of the schematized female figure of the Tanit symbol. In #186 and #191 the caduceus is shown next to, but not touching , the Tanit symbol, while in #186 and #192 the caduceus is shown associated with human hands. In two of the examples (#184 and #202), the caduceus is associated with other elements, namely a ship’s prow in one and flanking a winged uraeus in the other. Only in one Carthaginian example (#193) is the caduceus shown by itself on the stele. 271 Coulson, Mook, Rehard, & Grace, 1997, #9-10, #125-9; Grace, 1947, fig.8. 272 Brunaeu, 1965, #2199 & #2644; Howland, 1958, #657; Mlynarczyk, 1997, pp. 54-7 (type G). 123 in some variant from the Athenian Agora.273 The plain caduceus also appears in a wall painting next to an exterior cultic niche on the House of the Dolphins at Delos.274 And so, the plain caduceus by itself was not all that popular or widespread a symbol in Hellenistic glyptics. It only appeared in the published examples of a couple of the archives, though variants of the motif did appear more frequently. The motif and its variants do appear with more frequency in coins during the Hellenistic period. In regards to the plain caduceus motif, Hellenistic coin issues continue to appear frequently in S. Greece, as they did in previous centuries, but fewer such issues appear in W. Anatolia or N. Greece. Instead, the motif seems to have spread out, occurring on the occasional issue in the Black Sea, Anatolia and Sicily. Interestingly, in this period, the plain caduceus motif appears at the same number of mints in Italy overall, including Rome, as it does in Greece proper, indicating the motif’s strong popularity there, making Italy the region with the third highest number of occurrences. In other media of the Hellenistic period the caduceus otherwise appears in a consistent if sporadic fashion. For instance, the caduceus by itself or with associated symbols appears on Punic stelae from Carthage (N. Africa) and elsewhere. Thus, the overall image of the motif’s distribution is one of a motif with generally middling popularity that was originally focused in the Aegean area and spread out from there. In the Hellenistic period, the motif appears in several different regions, but only occasionally in any one region. However, occurrences of the motif do tend to cluster in Italy, S. Greece and W.Anatolia. Besides these clusters, certain occurrences are also of interest. The references to Tyre, both in the inscribed sealings 273 Lang & Crosby, 1964, L19, L43, L47a-b, L75, L76b, L170-3, L183. Of these, three show plain caducei (L19, L75 & L76b), while three others show winged caducei (L47a-b, L170-1). The rest show a caduceus in combination with other objects. 274 Bulard, 1926, pl.X.1. 124 from Kedesh and in the 5th century lead weight, are of special interest in regards to the Kedesh sealings. The specific re-occurrence of the motif over many centuries in reference to two public entities within the city, its main temple and its common assembly, may indicate a special prominence of the motif within that specific city. Catalogue: CAD1 Number of Sealings (Bullae): 1 (K99 0294) Maximum Preserved Dimension: 11 mm. Shape: elongated, possibly oval and slightly convex seal Description: A caduceus. The field is concave. The staff of the caduceus rises from the bottom of the field, where the bottom of the shaft is missing, and is topped by a pair of entwined snakes. The snakes form a circle with a pair of comma shaped “horns” for the snakes’ heads. The place at the top of the circle where the snake bodies would entwine is not well impressed. 125 Chart 3.9: The Distribution of Caducei in the Persian Period Map 3.10: The Distribution of Caducei in the Persian Period                126 Chart 3.10: The Distribution of Caducei in the Hellenistic Period Map 3.11: The Distribution of Caducei in the Hellenistic Period                    127 Section 5: Columns Two sealings from Kedesh show images of freestanding architectural columns. These sealings represent the use of two distinct seals. In each case the column shaft is shown as smooth with no fluting. In each case, the column is topped by a Doric capital with both echinus and abacus visible. In no case here is the base preserved in the impression, and, to judge by the size of the preserved impressions, it was not meant to be. One particular example (COL1) is especially interesting as it shows a column topped by a plain Doric capital that is in turn topped by a somewhat enigmatic object. This object is wide and curved with projecting horizontal ends, resembling nothing so much as an Egyptian style boat. What this object actually represents remains unclear. Perhaps it does represent a model of an Egyptian boat. Another possibility is that it is a version of a sundial topping a column as seen in Hellenistic and Roman stones.275 The other Kedesh example (COL2) is simpler in composition. It shows a smooth column with a simple two-stepped capital, which is apparently Doric. There is a prominence or bulge on the left side of the shaft whose nature is indeterminate. The Kedesh archive also has two other sealings that show related images (see figures 3.2 & 3.3). The first sealing (K99 0136) shows a single bird sitting atop a Doric column and facing right. A similar motif 275 Furtwängler, 1900, XXVIII.55 & 63 128 occurs on a sealing from Daskyleion in the fourth century.276 The second sealing (K00 0272) shows two confronted birds on a basin which is itself set atop a column or pillar. This last motif has exact parallels in the sealings from both Seleucia and Cyrene.277 However, the focus of the image in both Kedesh bullae (K99 0136 & K00 0272) is the bird or birds, rather than the column. As such, they fall outside the purview of this work. As an architectural member designed for bearing loads, a column by its nature denotes both strength and stability. Beyond this, depictions of columns might have further significance. In the Archaic period of Greece, freestanding columns served as tomb markers. Here, the columns may have been acting as stand-ins for the owners of the tombs, much as kouroi and korai statues did.278 As such, an image of a column on a seal might represent the tomb itself, as a means of protecting against it with sympathetic magic, or even as an alternate persona to the seal user, as with an image of a mask. Furthermore, the tombs marked by columns would have been aristocratic. The columns served not only to mark the tombs but also to monumentalize them and also serve as markers for the races of aristocratic funerary games, such as those recounted in Homer.279 276 Kaptan, 2002, DS118. 277 Invernizzi, Bollati, Messina & Mollo, 2004, AR 328-9; Maddoli, 1965, #1103. 278 McGowan, 1995, pp. 615-7. 279 McGowan, 1995, pp. 618-632. 129 In this light, a depiction of a column might also be making a reference to the same aristocratic and heroic status, be it real or pretended, on the part of the seal owner. Distribution: Images of columns by themselves or topped by various objects appear only sporadically in different media. There are a number of variations on the theme, many of which bear some resemblance to the rather simplistic Kedesh examples, but few with any really close relationship. For example, in Persian period glyptics Elephantine (Egypt) has a 4th century sealing that showed a column topped by a helmet crest with a sword hanging off of it as the design on a shield.280 Selinus (Sicily) has two sealings that show columns or columnar altars by themselves.281 Daskyleion (W. Anatolia) also has a sealing of comparable theme that shows a bird sitting on a column.282 Furtwängler also notes a related theme on a gem from Melos, dated from the 5th-2nd centuries BCE, wherein two columns are shown standing together, which he relates to the twin Dioscuri.283 In addition, a fourth century coin of Euboean Chalcis in S. Greece shows the head of Hera sitting atop an Ionic capital, while a fourth century coin of Apollonia in N. Greece shows a columnar obelisk with a stepped base.284 In the Hellenistic period, there are a on the whole a few more examples of the column motif. In glyptics, Seleucia (The East) has nineteen sealings, representing the use 280 Rubensohn, 1907, #4. 281 Salinas, 1883, CCCXCIII & CCCCXXXVIII. 282 Kaptan, 2002, DS118. 283 Furtwängler, 1900, XIII, XXIX.44. The first example comes from Melos and Furtwängler dates it to the Classical-Early Hellenistic periods, while the second is dated from the Hellenistic to the Early Roman periods. 284 Gardner, 1963b, Apollonia 2;Morkholm, 1991, #242. 130 of nine separate seals, which show columns or column-like altars.285 Some of these have Ionic capitals and bear cylindrical objects of an indeterminate nature.286 Another shows a smooth shafted column with a simple base and capital, similar to COL2.287 Among other Hellenistic examples, Carthage (N. Africa) has a sealing that shows a dog seated upon a short Ionic column.288 At Susa (The East), a sealing from the Greco-Parthian period shows an altar in the form of a short Ionic column.289 Columns also appear on engraved gems. Of note is the previously mentioned type, which appears on several gems, where the image shows a semicircular sundial on a column capital, usually Corinthian or Ionic.290 Other variants include a type therein the column is topped by or hung with arms, similar to the Elephantine example.291 Other examples in gems include an old Italic sardonyx with a bird perched upon a grape vine draped column and a 2nd-1st century BCE Campanian-Roman paste that shows a garlanded columnar altar topped by a bucranium.292 In numismatics, the Hellenistic coins of Byzantium as well as Apollonia, Oricus and Ambracia in N. Greece all show columnar obelisks on stepped bases that are similar in form and appearance to the Kedesh columns.293 In other media, lead tokens bearing a horsehead protome on a column, dating to the third century, occur at Athens (S. Greece) 285 Invernizzi, Bollati, Messina & Mollo, 2004, Og246-251, 265, 315; McDowell, 1935, III.D.1.d.1, III.D.5.a.1-2. 286 Invernizzi, Bollati, Messina & Mollo, 2004, Og 246-247. 287 Invernizzi, Bollati, Messina & Mollo, 2004, Og 248. 288 Berges, Ehrhardt, Laidlaw & Rakob, 1997, #118. 289 De Mecquenem, 1927, #92. 290 Furtwanlger, 1900, #XXVIII.55, 63, XXIX.44; Vollenweider, 1979, #533-534, 536. Furtwängler dates his examples from the Hellenistic to the Early Roman periods, while Vollenweider dates hers to the mid second century BCE. 291 Plantzos, 1999, #649; Vollenweider, 1979, #537. Vollenweider dates her example to after 168 BCE while Plantzos simply gives a Hellenistic date for his. 292 Furtwängler, 1900, #LXIV.50; Maaskant-Kleibrink, 1978, #156. 293 Gardner, 1963b, 49-53, 73-79, Oricus 1-4, Ambracia 1-14; Head & Gardner, 1963, Byzantium 37. 131 and columns bearing amphorae appear as relief decoration on a mould-made bowl at Samaria (The Levant).294 Altogether, the column motif as seen at Kedesh is exceedingly rare. Only one appearance in Sicily for the period of the 6th-4th centuries and one appearance in The East for the Hellenistic period come close to the images on the Kedesh sealings. The appearance of the motif at Kedesh is thus highly idiosyncratic. Catalogue: COL1 Number of Sealings (Bullae): 1 (K99 0656) Maximum Preserved Dimension: 10 mm. Shape: flat, possibly oval seal Description: A column supporting an enigmatic object. The column has a fairly broad and smooth shaft. It supports a Doric capital with two bands (the top one appearing more rounded). On top of the capital, the column supports an unidentified crescent shaped object with a projecting horizontal end at least on one side. This object may possibly represent a boat or a sundial. 294 Crowfoot, 1957, fig.62.20; Lang & Crosby, 1964, L135a-b, L136a-b. 132 COL2 Number of Sealings (Bullae): 1 (K99 0944) Maximum Preserved Dimension: 9 mm. Shape: small convex oval seal Description: A Doric column, with a smooth shaft topped by a simple two stepped capital. On the left-side of the shaft, there is a triangular prominence whose nature is unclear. Chart 3.11: The Distribution of Columns in the Persian Period              133 Map 3.12: The Distribution of Columns in the Persian Period Chart 3.12: The Distribution of Columns in the Hellenistic Period              134 Map 3.13: The Distribution of Columns in the Hellenistic Period 135 Section 6: Composites One category of Kedesh sealings displays subject matter which can be referred to as Grylloi or Composites. These seals display images which are themselves constructed out of several smaller but complete discrete units, such as animals, birds, cornucopias or masks. The composition of these elements is done in a somewhat irrational way, such as by transposing a face or mask in the place of a torso, using a cornucopia as a bird’s tail, or by using the form of a bird to represent the beard of a man. It is this irrational ordering that distinguishes “Grylloi” as a class from standard composite mythological creatures of the Near East and Greece such as sphinxes. The combination of multiple discrete elements allows for a wide variety of specific images within the Composites category, though they can be generally grouped into one of two iconographic sub-categories or groups: Composite Heads and Composite Animals. The first group consists of images of human heads formed from multiple elements. For example, one common type consists of male heads in profile where the beard, and the hair take the form of a bird with its wings- and sometimes neck-folded.295 Another possibility within this sub-category would dispense with the birds and have two or more heads joined together, in many cases back- to-back like the Roman god Janus. In some versions of this type, the various heads would share specific body parts like ears and eyes.296 The second group or sub-category of Composites consists of many elements going together to form the image of an animal, often a bird or a sea animal like a dolphin. One version of an image that falls into this category and which was popular on Roman gems consists of a rooster, shown in profile 295 Boardman, 2003, p. 125; Examples of this type have been found on sealings at Ur: Collon, 1996, #717- 725; Seleucia: Invernizzi, Bollati, Messina & Mollo, 2004, GyT 1-7; and Uruk: Wallenfels, 1990, #7.. 296 Examples of this type can be found on Persian period Samarian coins among others. See Meshorer, Y. & S. Qedar, Samarian Coinage. Jerusalem, 1999, # 138-145, #210. 136 with a head formed by a horse-headed protome and a body from a Silenus mask with a ram’s head for the wings.297 A great amount of variation, however, on the subject matter and on specific types does exist. As a group, Composites probably had strong amuletic or apotropaic connotations. The discrete components of each image often have apotropaic values in themselves in that they are often symbols of good luck, prosperity or aversion of evil. For example, cornucopias, symbols of abundance make regular appearances, as do masks which serve as symbols of aversion. As such, the combination of these elements in one place can be seen as an attempt to strengthen the talismanic affect of each separate element. Furthermore, the images themselves in their totality can also represent symbols of power or good fortune, such as eagles with their connection to Zeus or roosters with their links to the sun, thereby reinforcing the impact of the aggregate elements.298 At Kedesh, there are a total of nine impressions displaying composites, representing the use of eight distinct seals. These seals fall into three types. The first Kedesh type, represented by three seals (COM1, COM2 and COM6), consists of an eagle shown frontally with its wings spread and its head turned to one side. The breast of the eagle takes the form of a mask or human face. The face is also shown frontally and is beardless though the exact sex can not generally be determined. Comparanda for this particular type are not widely attested, indicating its lack of popularity outside of Kedesh. An engraved sard and a number of glass paste replicas in Berlin, however, do 297 Roes, 1935, p. 233; This type can also be found at Seleucia: Invernizzi, Bollati, Messina & Mollo, 2004, GyT 34-43; and was very popular in Roman gems from the 2nd century BCE onwards: Maaskant-Kleibrink, 1978, #1081-1084 & #1086-1095. 298 Blanchet, 1921, 50. 137 reproduce the motif exactly though their provenience is lacking.299 The second type of Kedesh composite is represented by four seals (COM3, COM4, COM5, and COM8) with five total impressions and consists of a hippocamp formed by a horse-head protome connected to a bald human head in profile with a long tail following behind. In two Kedesh examples, the tail may possibly be formed by cornucopias, while, in one of these examples, the hippocampus is being ridden by a small Eros. A single sealing/seal (COM7) depicts a composite head. This sealing shows an image of a female mask connected back to back with the mask of a silenus. Distribution: Composite motifs, especially composite heads, have a long artistic history in the Mediterranean and Near East. They were especially popular in the sphere of glyptic from the Persian period all the way to the late Roman.300 During the sixth-fourth centuries, composite heads, including janiform heads, appeared in several archives. Janiform heads appear on five sealings from Selinus (Sicily), on two sealings found in a 5th-4th century grave at Ur (The East) and on six sealings dated to the 5th and 4th centuries at Carthage (N. Africa).301 Besides these archives, janiform heads also appear on a Classical period bronze ring found at Olynthus (N. Greece); and two contemporary green jasper scarabs from Tharros (N. Africa).302 The janiform head motif also appears in coins during the 6th-4th centuries from 299 Furtwängler 1900, pl. XXVI.71; AGD II, 1788-1804. 300 Invernizzi, Bollati, Messina & Mollo, 2004, p. 175; Boardman, 2003, pp. 123-132; Roes, 1935, pp. 232- 235. 301 Berges, Ehrhardt, Laidlaw & Rakob, 1997, #419, #421-26; Collon, 1996, fig.12.O&P; Salinas, 1883, CLIII-IV, CLXX, CLXXX, &CCXII. 302 Boardman, 1970, pl.417, fig. 257 &270.1; Barnett & Mendleson, 1987, #7.23 & 9.33. 138 various places. The places that produced coins with composite heads include Ashdod (400-300 BCE), Gaza (Philisto-Arabian coins, 5th-4th centuries BCE) and Samaria (4th century) in The Levant; Lampsacus (400-330 BCE) and Tenedos (6th-early 4th centuries BCE) in W. Anatolia; Amphipolis (336-323 BCE) and Thasos (411-400 BCE) in N. Greece; Histria (4th century BCE along the Black Sea; Cilicia (4th century BCE) in C. Anatolia; and Syracuse (5th-4th centuries) in Sicily.303 Besides seals and coins, janiform heads also appear on a fourth century clay token from the Athenian agora in S. Greece.304 The motif also occurs in plastic Attic vases of the late sixth through fifth centuries BCE, wherein the entire vase takes the form of the janiform head.305 During the Hellenistic period, janiform heads continue to make appearances in the various archives. Delos, for example, in S. Greece has at least four sealings with the motif, though the exact number hasn’t yet been published.306 The janiform head motif also appears on one sealing from Edfu in Egypt, one sealing from Artaxata in C. Anatolia, on thirty sealings (representing nineteen seals of which eleven show a female mask facing right and a silenus mask facing left) at Seleucia in The East; as well as on two sealings from Cyrene and on one third century sealing from Carthage in N. Africa.307 Likewise, the janiform head motif continued to be popular in Hellenistic 303 ANS online catalogue, 1944.100.56920-2, 1983.51.57, 1983.51.59, 1993.141.54, 1998.115.4, 1998.115.81, 2002.46.524, 2008.51.5; Gerson, 2001, fig. 5-27; Head & Gardner, 1963, Istrus 1-14, Thasos 51-52; Hill, 1900, Cilicia 5-7; Hill, 1965, Philisto-Arabian 1-2 & 30; Meshorer & Qedar, 1999, #112, #138-145, #210; Mildenberg, 1992, pl.VII.3a-b, pl.VIII.4, pl.VIII.6, pl.IX.9; Plant, 1979, #729, #744-7, #1627; Wroth, 1964b, Tenedos 1-8, 11-21 & 27-33. 304 Boardman, 1979, fig.270.1, Lang & Crosby, 1964, C7. 305 Beazley, 1929, pp. 38-78. 306 Boussac, 1988, fig.7a-d. 307 Berges, Ehrhardt, Laidlaw & Rakob, 1997, #418; Invernizzi, Bollati, Messina & Mollo, 2004, GyT50- 68; Kachatrian, 1996, fig.32; Madolli, 1965, #637 & #639; Milne, 1916, #33. 139 numismatics. The motif appeared on issues from Lampsacus (4th-3rd centuries) and Tenedos (2nd-1st centuries BCE) in W. Anatolia; Amphipolis (187-31 BCE) and Thessalonica (1st century BCE) in N. Greece; Chersonesus (300-200 BCE) along the Black Sea; an uncertain central Italian mint, Canusium (209-208 BCE), Luceria (209- 208 BCE), Rhegium (283-89 BCE), Rome (3rd-2nd centuries BCE), Uxentum, and Volterra in Italy; Catana (275-212 BCE), Panormus (after 212 BCE) and Syracuse (after 212 BCE) in Sicily; and Spain (1st century BCE) in Europe.308 In other Hellenistic media, janiform heads appear on lead tokens from the Athenian agora (S. Greece).309 And so, the janiform head motif during both the Persian and Hellenistic period appeared across a vast geographical space. In the 6th-4th centuries, many regions had at least a couple of occurrences of the motif. The regions of The Levant and N. Greece 308 ANS online catalogue, 0000.999.23, 0000.999.94, 0000.999.247, 0000.999.321, 0000.999.543-4, 0000.999.546-7, 0000.999.555-6, 0000.999.558-60, 0000.999.56153, 1001.2.3, 1001.1.10605-6, 1001.1.12089, 1001.1.12832, 1001.1.2536, 1895.13.1, 1905.999.5, 1906.236.694, 1906.263.689, 1908.89.4, 1908.115.15, 1927.38.1, 1927.999.26, 1933.49.74, 1937.158.569, 1941.131.15, 1941.131.43, 1941.131.94-6, 1941.131.1032, 1941.131.1036, 1943.102.7, 1943.102.9, 1943.127.1, 1943.127.6, 1943.999.3, 1944.100.42-52, 1944.100.69, 1944.100.84-5, 1944.100.179, 1944.100.211, 1944.100.219, 1944.100.248, 1944.100.251, 1944.100.253-5, 1944.100.257-65, 1944.100.267-73, 1944.100.277, 1944.100.292, 1944.100.584-5, 1944.100.594-5, 1944.100.728, 1944.100.734-7, 1944.100.745, 1944.100.14382-3, 1944.73758, 1944.100.81001, 1947.2.25-6, 1948.8.1-2, 1948.19.32, 1949.98.276, 1949.100.27, 1949.100.29, 1952.63. 1-10, 1952.142.1, 1954.18.25-6, 1954.18.29, 1954.203.4, 1954.263.30-41, 1954.263.58-66, 1954.263.78-82, 1957.172.8, 1965.24.1, 1969.83.56, 1969.83.79-85, 1969.83.129-30, 1969.83.273-5, 1969.83.279-88, 1969.83.290, 1969.83.313-9, 1969.83.360-1, 1969.83.368-78, 1969.83.385-7, 1969.83.459, 1969.83.485, 1969.83.513, 1969.83.518-9, 1969.83.532, 1976.9.1, 1976.102.1, 1976.102.5-6, 1976.102.8-12, 1978.39.1, 1978.64.287-8, 1978.64.302, 1980.27.1, 1981.16.48, 1981.164.1, 1984.88.1-12, 1986.76.44-6, 1986.161.6, 1986.161.84-6, 1986.161.90, 1987.35.6- 7, 1987.35.10, 1987.79.1, 1988.50.1, 1990.3.50, 1991.1.178, 1992.41.8, 1992.121.81, 1994.58.2, 1995.11.8-9, 1995.11.11, 1995.11.1674, 1995.11.1859, 1995.11.1863, 1995.16.1, 1998.85.5, 1998.85.11, 1998.85.15-6, 1998.85.29, 1998.85.33, 1998.108.6, 1998.133.1-2, 1998.133.4-6, 1998.133.15, 1998.133.30, 1999.13.9, 1999.13.52, 2002.46.3-18, 2002.46.88-9, 2008.63.136; Head, Poole & Gardner, 1963, Catana 91-2, Panormus 29-30; Syracuse 722; Plant, 1979, #8, #742-3, #746, #2469, #2609; Poole, 1963a, Volterra 1-15, Volterra Class II 1-17, Volterra Class III 1-2, Roma 1-3, Uncertain Central Italy Class III 1-3, Uncertain Central Italy Class IV 1, Uxentum 4-5, Rhegium 90-94; Wroth, 1964b, Tenedos 27-33. 309 Lang & Crosby, 1964, L44, L79, L208, 140 had the most occurrences with three each, predominantly coins but also including the ring from Olynthus. This predominance is not very great as four other regions have two occurrences each. The distribution of the Hellenistic period is much different. Here, the regions of Italy and Sicily predominate with seven and three appearances respectively. All of these occurrences are coins. The most other regions only have two occurrences of the motif. The preponderance of the motif in Italian coins may be because of the influence of Rome, where the use of Janus and janiform heads on coins happened frequently. At the same time, the closest examples to the actual depiction of the janiform head at Kedesh come from the sealings of Seleucia in The East. And so it seems that, while the motif appears most frequently in the west, the inspiration for the Kedesh example may not necessarily derive from there. Comparatively, the other two Kedesh composite types are much less popular. They have almost no geographically and temporally locatable published parallels. Only the seal COM7 which has an Eros riding a composite hippocamp has good parallels from twelve seals at Delos (Fig.1) in S. Greece.310 Delos also has published examples of composites of other sea creatures, including dolphins and a Capricorn being ridden by Erotes.311 At Seleucia in The East, on the other hand, there are two seals with one impression each that show an Eros riding a plain, non-composite hippocamp.312 A red jasper gem from the 2nd-1st centuries BCE, now in Munich, also shows a composite hippocamp, while a Roman period jasper in Bucharest shows an Eros riding a composite 310 Stampoulidis, 1992, #749-761. 311 Stampoulidis, 1992, #605, 764. 312 Invernizzi, Bollati, Messina & Mollo, 2004, Er249-250. 141 hippocamp, neither of which have definable geographical findspots.313 Other composite sea creatures, notably dolphins, also appear on gems well into the Roman Period.314 The composite eagle motif represented by COM1, 2 and 5 has a parallel in a sard and several glass pastes in Berlin and Munich, but seems to be lacking any other parallel from the published examples from known archives or other geographical context.315 Composite birds in general, however, especially those images of composite roosters, do make the occasional appearance. The third century terracotta plaque from Stratos in Acarnia (S. Greece) is but one example.316 Similar composite roosters have been found on 9 seals at Seleucia (The East) with a combined total of twelve impressions, and on six seals from Delos (S. Greece) where an Eros rides them.317 Composite birds, again mostly composite roosters but including rarer examples of peacocks and crows, appear on gems and rings from the late-Classical period and last well into the Roman, where 313 Brandt, 1968, #502; Gramatopol, 1978, #606. 314 Platz-Horster, Schulter & Zazoff, 1976, #209-210, 1115 (1st century BCE gems in Hannover); Furtwängler, 1900, XLVI.17 (a Greco-Roman gem in Berlin).. 315 Furtwängler, 1900, XXVI.71; AGD1.2, #1447-1451. 316 Daux, 1965, 760-761 317 Invernizzi, Bollati, Messina & Mollo, 2004, GyT 34-42; Stampoulidis, 1992, #743-748. 142 they are especially popular.318 And so, the Composite Animals at Kedesh are highly idiosyncratic variant examples within a much larger and well attested iconographic category. Of the two Kedesh types, the Composite Hippocamp is well attested at the archive at Delos, at least in the version where an Eros is riding the hippocamps in COM6, but does not otherwise appear in any other published examples from other archives, or in other media. One of the Kedesh examples therefore has a link through a parallel with the Aegean, but the link can not be demonstrated for every example of the motif at Kedesh. The Composite Eagles, on the other hand, do not have any geographically locatable parallels whatsoever. The presence of the motif at Kedesh in three separate seals, therefore, is highly unusual. Indeed, it might be that the specific Composite Eagle motif is a predominantly local variant within the larger Composite Bird category. If this is the case, the gems and pastes that also show the same motif might also have some link, possibly their origins, in this same region. Catalogue: COM1 318 See Boardman, 1970, #1060 (a 4th century ring in a Swiss collection); Brandt, 1968, #389, 505 & 506 (Hellenistic gems in Munich); Brandt & Schmidt, 1970, #1447-1451, 1894-1898, 1906 (Italian-Republican gems in Munich with #1447-1451 representing composite eagles proper); Brandt, Gerke, Krug & Schmidt, 1972, #2354, 2749, 3383 & 3583 (Roman gems in Munich); Forbes, 1978, #133 (Red Jasper from 1st centuries BCE/CE from Egypt); Fossing, 1929, #1575-1582 (Greco-Roman gems of composite birds including composite roosters and peacocks); Furtwängler, 1900, XXIX.59 (Hellenistic-Early Roman), XLVI.29, 33, 36-37 (Greco-Roman gems); Gerke, Scherf & Zazoff, 1970, Kassel #85-87 & Braunschweig #176 (Roman gems from 1st-2nd centuries CE); Gramatopol, 1978 #586-595 (Roman gems with composite roosters in Bucharest); Maaskant-Kleibrink, 1978, #1081-1084 & 1086-1094 (Roman gems in the Hague dating from 2nd century BCE or later); Platz-Horster, Schulter & Zazoff, 1975, Hannover # #1114, 1119, 1616 & 1618, Hannover #68 (gems dating from late-1st century BCE-2nd century CE); Spier, 1992, #192 (a mid-1st century BCE composite peacock in the Getty Museum); Vollenweider, 1979, #384-392 (1st century BCE gems and pastes in Geneva). 143 Number of Sealings (Bullae): 1 (K99 0088) Maximum Preserved Dimension: 11 mm. Shape: convex oval seal Description: A composite eagle. The eagle is shown flying frontally with its wings spread and its head looking up to the left. The breast of the eagle takes the form of a staring face or mask of a woman or a youth with the tail hanging below the chin. COM2 Number of Sealings (Bullae): 1 (K99 0330) Maximum Preserved Dimension: 9 mm. Shape: convex seal Description: A composite eagle. The eagle is shown frontally with spread wings, which partially surpass the impression. The head of the bird also surpasses the impression. The breast of the eagle take the form of a frontal face or mask whose details are somewhat indistinct in the impression, though traces of eyes, mouth and nose can be seen. The tail 144 of the eagle spreads out below the chin of the mask. COM3 Number of Sealings (Bullae): 1 (K99 0600) Maximum Preserved Dimension: 11 mm. Shape: flat oval sea Description: A composite hippocamp facing right. The head is a horsehead protome. The body is formed by a male face in profile with a pointed beard on its chin. The tail is formed by a cornucopia with fruit or a branch of some kind. COM4 Number of Sealings (Bullae): 2 (K99 0906, K00 0062) Maximum Preserved Dimension: 13 mm. Shape: convex oval seal 145 Description: A composite hippocamp facing right. The hippocamp is formed by a horsehead protome, which has traces of a bridle, sitting atop a beardless head in profile. The tail is long and snaky, running behind the creature. COM5 Number of Sealings (Bullae): 1 (K99 1050) Maximum Preserved Dimension: 9 mm. Shape: slightly convex seal Description: A somewhat fragmentary composite hippocamp. The image follows the same pattern laid out in the other cases of composite hippocamps at Kedesh. The head is in the form of a horsehead protome looking right, while the body is in the form of a beardless mask looking down and to the right. The tip of the face’s nose and the hippocamp’s tail are both missing from the impression. 146 COM6 Number of Sealings (Bullae): 1 (K99 1075) Maximum Preserved Dimension: 10 mm. Shape: flat seal Description: A composite eagle. This seal has the same general setup as COM1 and COM2, with a beardless human face taking the place of the eagle’s breast. Here, however, the eagle looks towards the right. The eagle is better modeled than the previous examples while the human face is less so. The eyes and the brow of the face are fairly clear, but the nose and mouth of the human face is summarily done. COM7 Number of Sealings (Bullae): 1 (K00 0102) Maximum Preserved Dimension: 12 mm. Shape: possibly round convex seal Description: A janiform composite head. The head consists of a female mask facing right with a high strip of hair arching forwards over the forehead. Connected to the back of the mask is another mask of a bald Silenus looking left. The Silenus mask is not as well impressed and only traces of the lower face, including the mouth, nose and eyes, survive. 147 COM8 Number of Sealings (Bullae): 1 (K00 0550) Maximum Preserved Dimension: 10 mm. Shape: round flat seal Description: A composite hippocamp facing right and ridden by a small, winged Eros. The Hippocamp has a horse head with a body formed from the head of a bearded Silenus in profile right. Two tufts of hair on the bald head of the Silenus form a sort of forelimbs for the hippocamp. The tail is formed by both the beard of the Silenus and by a small cornucopia with fruit. 148 Chart 3.13: The Distribution of Janiform Heads in the Persian Period Map 3.14: The Distribution of Janiform Heads in the Persian Period                149 Chart 3.14: The Distribution of Janiform Heads in the Hellenistic Period Map 3.15: The Distribution of Janiform Heads in the Hellenistic Period                    150 Map 3.16: The Distribution of Composite Eagles in the Hellenistic Period Chart 3.15: The Distribution of Composite Hippocamps in the Hellenistic Period              151 Map 3.17: The Distribution of Composite Hippocamps in the Hellenistic Period 152 Section 7: Cornucopias Seven bullae from Kedesh bear images of cornucopias, representing the use of six distinct seals. In each case, the horn of the cornucopia forms an “S”-shaped curve, arching down and to the right. Generally, the cornucopia is depicted as full with a large up-thrust cone at the top flanked by round dots to indicate the fruit. Five of the represented seals (COR 1-2, 4-5, and 7) also display bunches of grapes or other pendant fruit hanging off the rim to either side. Four out of the seven seals (COR 2-4, and 7) also have a representation of a band or ribbon, often referred to as a diadem, being tied around the shaft of the horn, roughly at the midpoint. In COR 2 and COR 4, the impression still preserves the image of the knot in the ribbon. The prong end of the horn is sometimes stylized or decorated as in COR 6, though in most of the Kedesh examples, the end is missing due to breakage or the lightness of the impression. The Roman poet Ovid briefly discusses the mythological story behind the cornucopia in his work the Fasti, in all probability basing the story on earlier traditions.319 According to him, the nymph Amalthea used her she-goat to nurse the infant Zeus. When the goat broke one of its horns against a tree, the nymph filled it with fruit and carried it to the young god, whereupon the horn became associated with abundance.320 In the Hellenistic period, the cornucopia as a symbol of abundance and prosperity was associated with the iconography of many different deities as an attribute denoting the ability to provide these same benefits. Among those deities that were commonly 319 Price & Kearns, 2003, p. 21. In fact, Ovid seems to have connected at least two earlier versions of the story, one involving Amalthea as the nurse of Zeus, the other involving the nymph Amalthea and a horn of plenty. 320 Ovid, Fasti V, 111-28. 153 associated with the cornucopia were Tyche, Demeter, and, in Carthage, Tanit.321 Other deities possibly associated with the motif include Ploutos and Eirene.322 The use of the cornucopia by itself as a symbol, as here, may, therefore, refer back to one of the deities that had it as an attribute, or, more generally, it might simply represent a general symbol of prosperity and abundance. Distribution: Unlike some other symbols in the Kedesh corpus, the cornucopia does not have an exceedingly long history as a symbol. The cornucopia as a stand alone symbol does make a couple of appearances prior to the Hellenistic period. These occurrences are usually from the fourth century and can be problematic. For instance, cornucopias are present on some fourth century fragments of Panathenaic amphorae from the Athenian agora (S. Greece), but these were probably the attributes of some figures who were not preserved in the image, such as Ploutos or Eirene.323 In glyptics, the archive of Selinus (Sicily) has three sealings with the motif that might date to this period.324 The single cornucopia also appeared as a type on a series of bronze coins from the island of Melos (S. Greece) that date from the fourth through the first centuries BCE and from Naples (Italy) dating from the late fourth-third centuries.325 During the Hellenistic period, the distribution of the cornucopia motif changed 321 Thompson,1965, pp. 39-40; Berges, Ehrhardt, Laidlaw & Rakob, 1997, #819. 322 Moore, Philippides & von Bothmer, 1986, #287-8. 323 Moore, Philippides & von Bothmer, 1986, #287-8. 324 Salinas, 1883, CCCLXXI-CCCLXXIII. The examples from Selinus must date sometime between the fires of 409 and 250BCE, but it is difficult to be more precise. Considering the nature of the rest of the evidence, it is possible that they date to the later portion of that period, namely in the Hellenistic period. 325 Plant, 1979, #1944; Wroth, 1963, Melos 22-25 . 154 dramatically. Indeed the cornucopia used by itself as a symbol really explodes onto the scene of the ancient Mediterranean, appearing in many media, especially coinage, at many different places during the third through first centuries. Examples of the solitary cornucopia, with or without attached ribbon, can be found at several archives, including 21 sealings at Seleucia on the Tigris (The East) , representing 19 distinct seals, at least one published sealing from Delos (S. Greece), and one sealing from Cyrene (N. Africa).326 Carthage (N. Africa) also has one sealing that shows a cornucopia, which has been dated in the 3rd-2nd centuries BCE. The image of this sealing combines a cornucopia with a caduceus and an unclear object in the same field.327 Likewise, the cornucopia motif appears on various gems. Of note are two Hellenistic garnets, one in the collection of the Staatliche Museen in Berlin, the other in an unknown collection which display solitary cornucopia with knotted diadems, both of which curve down to the right.328 In the realm of numismatics, the solitary cornucopia is a motif used by many powers for a great variety of issues. Among these powers, the Ptolemies were the ones who seemed to have helped to greatly popularize the motif during the Hellenistic period. Indeed, the later author Athenaeus ascribed the creation of the motif to the 3rd century Ptolemaic king Philadelphus who used it in a statue of his queen.329 The existence of the earlier Melian coins invalidate this claim, but the king may have had a great and active role in the spread of the motif. The Ptolemies used the cornucopia and its variant the double cornucopia (or dikeras) extensively as reverse types on their coins. Indeed, the dikeras became a symbol of the Ptolemaic queen Arsinoe II in the third century and 326 Invernizzi, Bollati, Messina & Mollo, 2004, Og 13-31; Boussac, 1988, fig.27; Maddoli, 1965, #1002. 327 Berges, Ehrhardt, Laidlaw & Rakob, 1997, #819. 328 Plantzos, 1999, #638-639. 329 Athenaeus, Deipnosophistai XI, 497. 155 remaining associated with her after her death.330 The Ptolemies minted coins with the dikeras type in Phoenicia in The Levant; at Alexandria and unspecified Egyptian mints in Egypt; and in Cyrenaica in N. Africa.331 At the same time, the Ptolemies also minted types with single cornucopias at Berytus, Cyprus, Phoenicia, Sidon, Tripolis and Tyre in The Levant; Alexandria, and unspecified Egyptian mints in Egypt; and Cyrenaica in N. Africa.332 After the turn of the second century BCE, when the entirety of the Levantine coast fell under the sway of the Seleucids, the single cornucopia continued to be used as coin types in the area, both as civic coins and as Royal Seleucid issues, especially under Demetrius I and II. Together, these coins were minted at the sites of Antioch, Apamea (2nd century BCE), Carne (188-137 BCE), Damascus (1st century BCE), and Ptolemais- Ake (2nd century BCE) (The Levant).333 In the same general region, double cornucopia appeared on coins from Gaza and Marathus and in Judaea under the Hasmoneans, all starting in the second century BCE.334 Elsewhere, various cities during the Hellenistic Period including Melos (4th-1st centuries BCE) and Paros (3rd-1st centuries BCE) in S. Greece; Adramyttium (2nd-1st centuries BCE) and Laodicea (after 133 BCE) in W. Anatolia; Illyrian Apollonia (229-100 BCE), Byzantium (after Alexander), Corcyra (300- 330 Thompson, 1966, p. 12. 331 ANS Online Catalogue, 0000.999.48312, 1944.100.78820, 1944.100.78931, 1954.255.9, 1955.190.37, 1957.191.46-9, 1964.79.62, 1967.152.682-3, 1980.109.104, 1991.60.15-6; Mørkholm, 1991, #294; Poole, 1883, Arsinoe II 1-40, Ptolemy VII 132-133; Cleopatra VII 2-4. 332 ANS Online Catalogue, 1944.100.76255-6, 1944.100.76358, 1944.100.76363, 1952.142.458, 1953.171.1710, 1955.190.35, 1964.79.60, 1967.152.562-3, 1967.152.626, 1991.60.17, 1991.78.79; Mørkholm, 1991, #307, #316, #322-3; Plant, 1979, #1934, #1936, #1938, #1940-2, #1944, #1946, #1948- 9, #1953; Poole, 1883, Ptolemy III 102-105, Berenice II 1-8, 15-16, Arsinoe III 1-5, Ptolemy V 50-51, Ptolemy X 42-48. 333 ANS Online Catalogue, 1923.150.362, 1944.100.70941-7, 1944.100.71184-9, 1944.100.75344, 1944.100.75349-53, 1944.100.75361-9, 1944.100.75373, 1944.100.76535, 1948.140.10, 1952.142.400, 1961.154.264, 1961.179.87-8, 1963.216.3, 1967.152.618, 1977.158.668-9, 1999.32.190; Hill, 1910, Carne 4-6, Akko 1-7, 10; Wroth, 1964a, Apamea 11, Damascus 1. 334 Hill, 1910, Marathus 34-41; Mørkholm, 1991, #648; Hill, 1965, Gaza 1-3, John Hyrcanus1-47, Judas Aristobulus 1-3, Alexander Jannaeus 11-24, Antigonus Mattathias 1-55, 57-59. 156 229 BCE), Hephaestia (after Alexander), Kabyle (under the Gaulish king Kavaros c.219 BCE) and Thasos (after 280 BCE) in N. Greece; Amasia (under Mithradates Eupator, 120-63 BCE), Amissus (under Mithradates Eupator, 120-63 BCE), Panticapaeum (200- 175 BCE) and Sinope (under Mithradates Eupator, 120-63 BCE) along the Black Sea; Iguvium (3rd-1st centuries BCE), Larinum (late-3rd century BCE), Naples (340-220 BCE), Paestum (268-69 BCE), and Thurii (194-89 BCE) in Italy; and Aetna (after 210 BCE) in Sicily also issued coins with solitary cornucopia, with or without diadems as the reverse types.335 Doubled cornucopia also appeared on the coins from Troezen (after 322 BCE) in S. Greece; Laodicea and Lebedus (after 190 BCE) in W. Anatolia; Philomelium (after 133 BCE) in C. Anatolia; and Hipponium (182-89 BCE) in Italy.336 The cornucopia motif also makes and appearance on other media. A cornucopia appears in a Thasian amphora stamp of the early-3rd century from the Athenian agora.337 At Athens and Delos (S. Greece), the fragmentary remains of terracotta cornucopia from the Hellenistic period have been recovered.338 The cornucopia and dikeras both also make appearances as decorative motifs West Slope painted vessels at Athens, starting in the third century, and possibly as the result of Ptolemaic influence.339 Athens also produced a series of one-eighth mina weights and lead tokens that bore the image of a single 335 Gardner, 1963b, Apollonia 58-59, Corcyra 205-205, 235-239; Head, 1906, Laodicea 31-49; Head & Gardner, 1963, Panticapaeum 39-40, Byzantium 34-35, Kavaros 3, Hephaestia 11 ; Head, Poole & Gardner, 1963, Aetna 7-9; Kroll & Walker, 1993, #863; Poole, 1963b, Iguvium 4, 8-10, Larinum 11, Neapolis 255-262, Paestum 8-11, 39-45, 67-69, Thurium 150-151, Copia 1-4; Wroth, 1963, Melos 22-25, Paros 28; Wroth, 1964b, Amasia 2, Amissus 65-67, Sinope 45-46. 336 Head, 1906, Philomelium 3-6; Head, 1964, Lebedus 1; Kroll & Walker, 1993, #972; Plant, 1979, #1945, #1947. 337 Grace, 1934, #7. 338 Thompson, 1963, #32; Thompson, 1965, #8-14; Thompson, 1966, 13, #21. The example in the last source is actually a double cornucopia, or dikeras. 339 Rotroff, 1991, #3, #30, #41-3, #48-51. 157 cornucopia.340 Another weight from Byblos (The Levant) from the 2nd-1st centuries BCE bears the image of two crossed cornucopias framing an Isiac headdress.341 Finally, Hellenistic lamps bearing an image of a cornucopia, usually with the lug taking the form of one, have been recovered from Athens and Corinth in S. Greece; and Cosa in Italy, while lamps from the second half of the second century to the first century BCE from Alexandria in Egypt have the motif on their nozzle.342 Taken together, all of these examples indicate that the examples of the cornucopia motif in the Kedesh sealings represent a direct connection with the swelling popularity of a fairly new motif. The symbol appears on the seals from a few of the archives, most notably the very large corpus from Seleucia, where it is most frequently attested. In coins, it is even more widespread. Here, this motif had begun to be popularized in the third century by the Ptolemies, who controlled most of the Levant, including Kedesh at the time, and continued grow in popularity into the second century with its adoption by various cities and by the Seleucids. Indeed, Kedesh was at the very ground zero for the explosion of the motif as many of the cities of Syria-Palestine, both along the coast and inland, minted coins bearing cornucopia at some point during the third and second centuries. In this light, the coins from Ptolemais-Ake, which date to the mid second century, are especially interesting due to their geographical and temporal proximity to the Kedesh archive. Therefore, the appearance of this symbol on the Kedesh sealings is very representative of the very time and place of the Kedesh archive itself. 340 Lang & Crosby, 1964, LW28(countermark), LW51-53, L49-50 (cornucopia between two ears of wheat), L72, L177. 341 Kushnir-Stein, 2002, #4. 342 Broneer, 1930, #302; Howland, 1958, #541, #544, #610, #634, #636, #698, #700, #704; Mlynarczyk, 1997, pp. 54-7 (type G); Rickman Fitch & Wynick Goldman, 1994, #206 (150-70BCE). 158 Catalogue: COR1 Number of Sealings (Bullae): 1 (K99 0018) Maximum Preserved Dimension: 11 mm. Shape: convex, possibly oval seal Description: A cornucopia to curving to the right with hanging fruit. The bottom portion is missing. The horn is fairly wide, and the fruits are stylized or indistinct. COR2 Number of Sealings (Bullae): 1 (K99 1116) Maximum Preserved Dimension: 9 mm. Shape: small, convex oval seal Description: A cornucopia curving to the right with fruit (including a pendant bunch of grapes) hanging out. The horn is fairly thick and tired with a diadem with knot to the left. The bottom of the horn is missing. 159 COR3 Number of Sealings (Bullae): 2 (K00 0371, K00 0677) Maximum Preserved Dimension: 11 mm. Shape: convex oval seal Description: A cornucopia curving to the right with hanging fruit and tied with a diadem. The depiction is somewhat angular and stylized. The bottom of the horn and the diadem ties are missing. The diadem is tied in an angular bow and the pendant fruit is depicted by clusters of parallel lines. The horn ends in a ‘Y’-shaped finial, perhaps representing an animal head. COR4 Number of Sealings (Bullae): 1 (K00 0414) Maximum Preserved Dimension: 8 mm. Shape: slightly convex oval seal Description: A fragment showing a cornucopia curving to the right with hanging fruit. The left side and bottom of the image are missing. The horn is fairly thick, and the fruit is stylized or ill defined. The fragment is small enough that it might possibly be a portion of 160 a larger scene. COR5 Number of Sealings (Bullae): 1 (K00 0434) Maximum Preserved Dimension: 11 mm. Shape: convex seal Description: A cornucopia curving to the right. The horn is complete in length, and there is decoration on the end of the horn. There is a diagonal crack across the shaft midway of the impression. COR6 Number of Sealings (Bullae): 1 (K00 0708) Maximum Preserved Dimension: 10 mm. Shape: narrow oval convex seal Description: A cornucopia curving to the right. The one example is not well impressed. The horn is long and thin. There are traces of pendant fruit and a diadem tied across the 161 center point. Chart 3.16: The Distribution of Possible Cornucopias in the Persian Period               162 Map 3.18: The Distribution of Possible Cornucopias in the Persian Period Chart 3.17: The Distribution of Cornucopias in the Hellenistic Period                  163 Map 3.19: The Distribution of Cornucopias in the Hellenistic Period 164 Section 8: Ears A human ear, either on its own or held by a hand, appears as a motif on three of the Kedesh sealings, representing the use of three separate seals. In each case, the impressions are of right ears, which would have been reversed on the seals themselves. The first seal (EAR1) shows a hand grasping the lobe of an ear between thumb and index finger. Another possibility is that the image represents the prow of a ship, with the top of the lobe representing an acrostolion. However, in such case, the depiction of the bent line behind it that is identified with a forefinger is highly unusual. It does not represent any structure present on other depictions of ships’ prows.343 In addition, the fact that the volute of the acrostolion would be facing aft would also be highly unusual. As such, it seems most probable that the first possibility is correct. The other two seals show human ears alone. The first of these seals (EAR2) is more finely done and appears to be sporting an earring at the base of the lobe, perhaps with the purpose of indicating that the ear belongs to a female. The second seal (EAR3) is blockier and more crudely done, lacking any sign of an earring. The ear as a symbol is by its very nature as a representation of a sense organ directly linked to the realm of human individuals themselves and to the human senses. This inherent nature in turn mediated how the symbol was used. For instance, ears often appear as votive dedications. For example, a set of gold foil ears were uncovered from the foundation deposit of the Archaic Artemis temple at Ephesus.344 Likewise, other replica ears have been recovered from sanctuaries at different Greek sites (see below). 343 See Forsythe-Johnston, 1985, Class 4, Hell 1, Hell 4-10. These examples consist of plastic vessels and various sculptural elements in the shape of ship’s prows. Each one is lacking any element behind the acrostolion that would account for the angular line of the forefinger here. 344 Hogarth, 1908, p. 108, pl.VII.48. 165 These votive dedications were thanks offerings, signalling either the cure of some ear- associated illness, such as deafness, or, in the case of dedications to deities not normally associated with medicine, simply that the deity in question has heard (and answered) the prayer of the individual.345 Likewise, the use of the ear as a symbol on a seal may also have represented some mediation between the individual and the supernatural world, by incorporating an amuletic function. This idea is best illustrated at the archive of Seleucia where seals bear images of ears that are combined with other elements to indicate a heightening of the senses. For instance, Seleucia has 48 seals that show an ear topped by a bird (see figure 3.5).346 This version of the motif is reminiscent of the Homeric idiom of “winged words” (ἔpiεα piτερόεντα)347 and, therefore, elicits the idea of eloquence, either already a characteristic of the bearer or as a desire of the bearer to have it. In addition Seleucia has 1 example each of a seal showing a hand holding an ear between thumb and forefinger similar to EAR1 (see figure 3.6).348 This particular version of the motif continues for a long time, appearing on gems into the Late Roman 345 Deonna, 1938, p. 216. 346 Invernizzi, Bollati, Messina & Mollo, 2004, Og179-216 & Og219-230. 347 For example, see Homer Il. 2.7 & Od. 1.122. 348 Invernizzi, Bollati, Messina & Mollo, 2004, Og245. 166 period, where it usually bears the Greek inscription “MNHMONEYE”, indicating its amuletic function as a memory aid.349 All of these examples point to the use of the ear in seals as a motif with amuletic functions that group around the nature of the ear as a sense organ. Indeed, the ancient Greeks also believed that evil could not attack only a person through the site or glance of the evil eye (οφθαλµὸς βάσανος), but also by the voice, for which the image of an ear would be protection.350 Therefore, the simple ears at Kedesh may likewise have had strong amuletic connotations, perhaps with the idea to provide the bearer with protection through heightened sensitivity to the outside world or the ear of a supernatural protector, echoing the function of some of the votive ears. Distribution: The ear by itself appears very sporadically as a motif in the Persian and Hellenistic Periods. It occurs in a few different media at a handful of different sites. The actual execution of these sporadic instances is often idiosyncratic and can vary widely in detail, as do the executions, for instance, of the two Kedesh examples. During the sixth through fourth centuries BCE, the corpus is dominated by plastic 349 For example see Gerke, Scherf & Zazoff, 1970, Kassel 94-96. The three gems listed are all carnelians and date to the second or third centuries CE. 350 Elworthy, 1958, p. 14. 167 representations, usually in the form of votive dedications at sanctuaries. For instance, in The Levant, there is an example of a limestone right ear from the temple in Golgoi, Cyprus, which dates to the fourth or third centuries and is inscribed in the Cypriot syllabary.351 The region also produced the only numismatic example of the motif in the form of a left ear on a rare Judaean fractional issue of the Fourth Century.352 The form is generally similar to EAR2 but it is facing the wrong direction and is lacking the earring. In S. Greece, the Asklepieion of Corinth has also produced a set of terra cotta votive ears from fourth century contexts. Here, there is a single example of a right ear on a plaque, two single ears (one right the other left) by themselves and two plaques which show both ears together.353 The region also produces an example of a fourth century relief plaque with two ears from the sanctuary of Asclepius at Athens.354 In W. Anatolia, there are the previously mentioned gold foil ears from the Ephesian Artmesia of which there were three with two coming from the base of the temple.355 Finally, in Sicily, another Persian period example consists of a carved ivory left ear, which dates to the fifth century BCE, and which was uncovered on the island of Malta.356 Interestingly, the ear is pierced as for an earring, like EAR2. In the Hellenistic period, there appears to be a somewhat greater variability in the media, with seals appearing in different places. The overall distribution is, however, still sporadic. In The Levant, the Golgoi example potentially remains. The same region has also produced an ornate bronze lamp from near Tel Aviv dating to 155 BCE that bears a 351 Karageorghis, 2000, #418. 352 Gerson, 2001, p. 112 (Y-2B). 353 Roebuck, 1951, p. 120, pl.33. 354 Bieber, 1910, pp. 5-8, pl.I. 355 Hogarth, 1908, p. 108, pl.VII.48. 356 Acquaro, 1988, #483. 168 pair of ears, one on either side of the vessel, as part of its decoration.357 In S. Greece, the island of Delos has produced several sculpted examples in both bronze and stone, in pairs and in singles with diverse findspots. These examples sometimes bear inscribed dedications, including to Aphrodite and Apollo.358 In The East, examples of the motif appear in the archives at Seleucia, where fifteen seals bearing the image of single ears.359 In addition, Seleucia also has the previously noted forty-eight examples with an ear topped by a bird.360 If anything, images of a hand holding an ear such as EAR1 are even rarer. Only Seleucia in The East has the previously mentioned single sealing showing the motif during the Hellenistic period.361 Considering the popularity of the motif in later roman glyptics, it seems very possible that the Kedesh and Seleucia examples represent the very beginnings of a motif that would later become popularly used for amulets in glyptic. It thus appears that the ear motifs appear only occasionally in the ancient world during the Persian and Hellenistic periods. When it does appear, it often bears links to the supernatural world, either because it was used as a votive dedication to a god or because it displays indications of amuletic properties. Catalogue: EAR1 Number of Sealings (Bullae): 1 (K99 0204) Maximum Preserved Dimension: 10 mm. 357 Sussman, 2006, p. 39. 358 Deonna, 1938, A263, A490, A550, A553, A556-8, A694-6842, A816-10070, A1858, A2244, A2396, & A2561. The findspots include the sanctuary of Apollo and its immediate environs, The Sarapeion, the Sanctuary of Foreign Gods, the Inopus, and different agoras. 359 Invernizzi, Bollati, Messina & Mollo, 2004, Og217-8, Og231-243. 360 Invernizzi, Bollati, Messina & Mollo, 2004, Og179-216 & Og219-230. 361 Invernizzi, Bollati, Messina & Mollo, 2004, Og245. 169 Shape: flat seal Description: A hand holding an ear. Both the hand and the ear are right sided. The hand has three fingers extended and a forefinger bent backwards against the lobe of the ear which merges with the back of the hand in the impression. Presumably, the thumb is meant to either be behind the lobe of the ear or perhaps alongside it, as it it not clearly shown in the image in the image. EAR2 Number of Sealings (Bullae): 1 (K99 0556) Maximum Preserved Dimension: 10 mm. Shape: convex oval seal Description: A right ear. There is an apparent round earring at the base of the ear lobe. The seal was cut originally into an oval convex gem. 170 EAR3 Number of Sealings (Bullae): 1 (K99 0720) Maximum Preserved Dimension: 10 mm. Shape: possibly oval flat seal Description: A right ear. This ear is thicker and blockier than the previous sealing. Chart 3.18: The Distribution of Ears in the Persian Period               171 Map 3.20: The Distribution of Ears in the Persian Period Chart 3.19: The Distribution of Ears in the Hellenistic Period              172 Map 3.21: The Distribution of Ears in the Hellenistic Period Chart 3.20: The Distribution of Ears with Hands in the Hellenistic Period              173 Map 3.22: The Distribution of Ears with Hands in the Hellenistic Period 174 Section 9: Floral Motifs Seven of the Kedesh sealings display images of flowers or related floral motifs, specifically a rosette. These represent a total of six distinct seals. The images in this category are highly variegated with some, such as FLO1 and FLO4, being more naturalistic and the others being more stylized. In the more naturalistic images, it may be possible to identify the species, or at least type of flower that is meant to be represented, while in the others it is only possible to identify the image as a flower or rosette. Of the two more naturalistic seals, FLO1 may represent some sort of crocus. The seal FLO4 on the other hand is a rose. Among the more stylized examples, there are two examples (FLO3 and FLO6) that display simple rosettes, both with eight petals. On the other hand, FLO2 is a highly stylized tulip-like flower, while FLO5 consists of a spray of five straight petals or leaves on a base of some sort. In essence, we can divide the Kedesh examples into two broad categories: flowers proper, which are seen from the side and which can be either more naturalistic or more stylized, and rosettes, which are seen face- on and which are generally more stylized. The possible meanings behind the use of the flower or rosette motifs are often hard to pin down. In part this is because they often function as ornament or decoration in the media that they appear. In addition, the rosette in particular appears so frequently across the artistic spheres of different cultures that it is almost ubiquitous in ancient art.362 As such, it can be difficult to see beyond a general decorative or ornamental quality inherent in these two motif categories. Still, in some cases, deeper allusions present 362 Van Buren, 1939, p. 99. In fact, Van Buren states, “The rosette is an artistic convention which originates spontaneously among almost every race in the world who have had an art of their own.” 175 themselves. First, in regards to flowers, certain flower types become linked to specific places. The most prominent case of this is the use of the rose by Rhodes in different media starting by the late fifth-century, including coins and the stamps on Rhodian amphora handles so that the rose became an emblem of that place.363 The use of the lily is likewise linked to Jerusalem through its re-use in the coinage of that place in the Persian period and again in the Hellenistic under both the Seleucids and Hasmoneans.364 For their part, rosettes also present specific allusions in certain contexts. For instance, the eight petal rosette was a symbol of the goddess Inanna in Sumerian art, and from there became a symbol of the goddesses of other cultures that were linked to her like Ishtar and from there Astarte.365 In other cases, various cultures in the Iron Age Near east, including the Neo-Hittites, the Assyrians and the Judaeans used rosettes with different numbers of petals, including eight, as royal insignia.366 At the same time, it is difficult to definitively link any of these specific allusions or meanings to the examples from Kedesh. The rose that appears on FLO4 may be a reference to Rhodes, but it need not be. Other places also used the rose on their coins, such as Hellenistic Cythnus, so the motif is not exclusively Rhodian.367 In addition, the fact that a clay lamp from nearby Dor, dated 200-50 BCE, has an eight petal rosette decorating the inside of its ring base makes it difficult to say that the rosettes from Kedesh 363 Grace, 1934, #71-95; Plant, 1979, #1816-20, #1823 & #1825. 364 ANS Online Catalogue, 1941.131.1084, 1944.100.62674-5, 1944.100.62774, 1944.100.77922-30, 1949.163.975; , 1952.142.479, 1984.56.125, 1984.66.125, 1999.32.57; Gerson, 2001, Y-2-A & Y-8. 365 Van Buren, 1939, pp. 99-107. Several frit rosettes were, for example found as part of the decoration of the Temple of Ishtar at Ashur, which Tukulti Ninurta built in the Neo-Assyrian period. 366 Cahill West, 1995, p. 251. 367 Plant, 1979, #1824 176 need necessarily refer to either Astarte or any form of royal power.368 All that we can say is that the possibility of such allusions exists in the depictions of rosettes and flowers. On a more general level, both rosettes and flowers represent blooming, and therefore living, plants and vegetation. As such, their use on seals at Kedesh may function as symbols of fertility and abundance, with the resulting prosperity, much in the same way as the cornucopia that also appear on the Kedesh symbol sealings. Distribution: There is a marked difference in the distributions and histories of the two Kedesh floral categories. For its part, the use of flowers as a symbol is comparatively rare. Naturalistic flowers first appear as symbols in the Persian/Classical period, though certain stylized versions date further back, such as the lotus flower. This particular motif appears early on in Egyptian art, as shown by a scarab of the 15th Dynasty (1650-1540 BCE) with the flower carved on its back.369 Later on it spread into the art of other Near Eastern cultures, appearing in such places as the 8th century Nimrud ivories.370 Within glyptics of the Persian period, specific parallels to the Kedesh examples occur at Carthage (N. Africa), where one fourth century sealing is similar to FLO1, and Daskyleion (W. Anatolia), where one sealing bears a rose similar to FLO4.371 Selinus (Sicily) has three sealings that show flowers, all of them appearing to be poppies.372 Elsewhere, a silver ring of the Classical period from Olynthus (N. Greece) shows a poppy bloom whilst a gold 368 Rosenthal-Heginbottom, 1995, Lamps. Type 11.3. 369 Ben-Tor, 1989, #39. 370 Acquaro, 1988, #72. 371 Berges, Ehrhardt, Laidlaw & Rakob, 1997, #797; Kaptan, 2002, #146. 372 Salinas, 1883, CDIII-IV, CDVI. 177 ring from Panticapaeum (Black Sea) from the same period shows a locust on a rose.373 In the case of coins, flowers, especially naturalistic ones in general and roses in particular, appear more frequently. During the sixth through fourth centuries, the rose appeared on coins from Paphos (350-332 BCE) in The Levant; Megiste (4th century BCE) and Rhodes (4th-1st centuries BCE) in W. Anatolia; Tragilus (400-350 BCE) in N. Greece; and Selinus (6th-Early 5th centuries BCE) in Sicily.374 A highly stylized lily also makes an appearance on fourth century coins from Judaea in The Levant.375 The Lotus flower appears on coins of Idalion (475-460 BCE) in The Levant.376 A sixth century issue from Stagira also provides an interesting point of comparison in that it shows five rose blooms arranged around a central dot, like a rosette, thereby combining the two categories.377 The issues from Rhodes are the most important for us as they were very plentiful and lasted far into the Hellenistic period, becoming a symbol of the city as the owl did for Athens. Outside of numismatics, flowers are again somewhat rare as symbols in the Persian period. They do make appearances on jewellery, such as some pendants and beads in the form of lotus flowers from Akhziv (The Levant), dating to 7th-6th centuries BCE, and from Tharros on Sardinia (N. Africa), dating to the 6th-4th centuries.378 The sites of Carthage and Tharros (both in the region of N. Africa) have also produced dies or moulds 373 Boardman, 1970, fig.255 & pl.703. 374 ANS Online Catalogue, 1909.999.110-1, 1925.172.74, 1929.115.11, 1941.153.870-4, 1944.100.48565- 95, 1944.100.48597-610, 1952.123.193, 1957.172.1733-4, 1967.152.467-8, 1974.26.1061, 1977.158.425- 7, 1980.109.132, 1983.51.569-72, 1983.51.579, 1984.65.90, 1991.78.63, 1992.4.50-1, 1992.32.103, 1992.121.42, 1994.5.1, 2000.17.203; Head, 1964, Paphos; Head, Poole & Gardner, 1963, Selinus 21-22; Plant, 1979, #1818, #1821-2. 375 Gerson, 2001, Y-2-A & Y-8. 376 Head, 1964, Idalion 5-19. 377 ANS Online Catalogue, 2004.18.3. 378 Acquaro, 1988, #624; Barnett, R. D. & C. Mendleson, 1987, 1.39; Dayagi-Mendels, 2002, Tomb ZR X, #22. 178 bearing the lotus flower and dating to the 6th century BCE.379 In addition, the island of Malta (Sicily) has produced a sculptural fragment of a lotus flower from the 6th-5th centuries. Altogether, the use of flowers as motifs by themselves is uncommon in the 6th-4th centuries. The region of N. Africa with its multiple use of the older and more stylized lotus flower motif has the most examples with five. Otherwise, the regions of The Levant, W. Anatolia, N. Greece, Sicily, and N. Africa all have between two and four occurrences each. The flower motif therefore seems to have been both fairly rare and at home in a variety of geographic areas during the Persian period. In the Hellenistic period, the motif continues to be rare, though the distribution of occurrences changes somewhat. In glyptics, Cyrene (N. Africa) has two sealings showing flowers, one that is similar to FLO1 and another that resembles FLO5, whilst Seleucia (The East) has twelve seals, characterized as palms or buds, which generally resemble FLO5.380 Uruk in the same region also has one sealing which shows a rose topped by some sort of quadruped, while Doliche in C. Anatolia has a very late Hellenistic sealing of a bird sitting on a rose.381 In Hellenistic numismatics, places that minted coins showing flowers, including roses and lilies, consisted of Jerusalem (a lily, both under the Seleucid Antiochus VII and the later Hasmoneans) in The Levant; Cythnus (rose, 2nd-1st centuries BCE) in S. Greece; Laodicea (lotus flower, after 133 BCE) and Rhodes (4th-1st centuries BCE) in W. 379 Acquaro, 1988, #228 & #587. 380 Invernizzi, Bollati, Messina & Mollo, 2004, Vg8-19; Madolli, 1965, #948 & #956. 381 Maaskant-Kleibrink, 1971, #79; Wallenfels, 1990, #1065. 179 Anatolia; and Phanagoria (rose, 1st c. BCE) in the Black Sea.382 In other Hellenistic period media, flowers make several if somewhat isolated appearances. First, the rose is a common motif on Rhodian (W. Anatolia) stamps for amphora handles of the 3rd and 2nd centuries BCE.383 Flowers appear as motifs on imported Delphiniform lamps at Cosa (Italy) as well as on 2nd-1st century mould-made lamps from Ephesus and Cnidus (W. Anatolia).384 The rose appears on a lead token from Athens (S. Greece).385 There are also two impressions on clay loom-weights from Hellenistic Gordion in C. Anatolia that show a nude female emerging from the bloom of a rose.386 If anything, the use of the flower motif is even more uncommon in the Hellenistic period as compared to the preceding one. The one region where the motif clusters is W. Anatolia, which has four occurrences, bolstered by Rhodes’ use of the rose as an official symbol. Elsewhere, the flower motif only has sporadic appearances in different places. On the other hand, as noted, the rosette appears much more frequently in ancient art. As a general motif, the rosette has a history dating back to the Bronze Age in both Greece and the Near East.387 In general, the use of the rosette, with varying numbers of 382 ANS Online Catalogue, 1927.165.20-1, 1941.131.1084, 1944.100.48611, 1944.100.48619-20, 1944.100.46822-4, 1944.100.48629-31, 1944.100.48635, 1944.100.77922, 1944.100.62674-5, 1944.100.62774, 1944.100.77923-30, 1949.163.975, 1952.142.479, 1984.56.125, 1984.66.125, 1999.32.57; Head, 1906, Laodicea 24; Head, 1963b, Phanagoria 4; Hill, 1965, John Hyrcanus I 48-56, Alexander Jannaeus 1-10; Mørkholm, 1991,#235, #548-52, #554-55; Plant, 1979, #1816-20, #1823-26; Wroth, 1963, Cythnus 6-7. 383 Grace, 1934, #71-95. 384 Bailey, 1975, Q172, Q177 & Q310-1; Rickman Fitch & Wynick Goldman, 1994, #188-198. 385 Lang & Crosby, 1964. L218. 386 Dusinberre, 2005, #85-6. 387 See Boardman, 1970, fig.20, and Carter, Hole, Baihrani, Spycket & Aruz, 1992, #22. The Boardman example consists of a stone prism from EMII-MMI Heraklion on which one side has been carved with a simple eight-petal rosette. The Susa example is an impressed Bulla dates from the late Uruk period (c.3300 BCE) which shows a series of eight-petal rosettes between intertwining snakes. 180 petals, continued to be a popular motif in the Ancient World for a long time in multiple media, as indicated by the finds of elaborate rosettes carved on bone disks from the seventh century at the Sanctuary of Artemis Orthia at Sparta.388 The Kedesh examples are, therefore, part of a long general continuum of use for a popular decorative motif. At the same time, the use of the rosette in glyptics is rather limited from the sixth century onwards, appearing sporadically. In the archives of the Persian period the rosette, here with four petals, appears as a motif on only two sealings from Carthage (N. Africa), dated to the 5th-4th centuries.389 Outside of glyptics, the rosette does appear somewhat more frequently in coins from different mints during the 6th-4th centuries. Often these are early coins as the motif appeared as a motif on the reverse punch early on. The mints that produced coins with rosettes include Tyre (with 8 petals, 410-377 BCE) in The Levant; Erythrae (with a varied number of petals, including 8, 6th-5th centuries BCE) in S. Greece; Cyme (with 8 petals, 350-320 BCE) in W. Anatolia; and Corcyra (stellate floral pattern, 6th-4th centuries BCE), Macedonia (with 16 petals, 510-480 BCE), and Maroneia (with 16 petals, 500-450 BCE) in N. Greece.390 The Tyrian issue is especially interesting here, given the geographical proximity of the city to Kedesh. Also of interest during this time period is the previously mentioned coin from Stagira that combines elements of flowers and rosettes into one motif. The rosette also appears with great frequency in the 6th-4th centuries as a 388 Richter, 1968, #72, #75, #80. 389 Berges, Ehrhardt, Laidlaw & Rakob, 1997, #799-800. 390 ANS Online Catalogue, 1925.14.21, 1944.100.46212-9, 1944.100.72662-3, 1948.77.17, 1954.185.17, 1956.28.212, 1974.95.139, 1977.158.349-50, 1998.17.1, 2008.39.6; Gardner, 1963b, Corcyra18-82 & 130- 143; Head, 1963b, Locri Opuntii 2-6, Head & Gardner, 1963, Maronea 4; Head, 1964, Erythrae 1, 3-6 & 18-30; Plant, 1979, #1245b, #1813-4; Wroth, 1964a, Cyme 15. 181 decorative motif in various media associated with architecture, such as architectural sculpture, mosaics, and wall paintings. Persepolis, for instance, in The East has bands of rosettes, generally with twelve petals, which flank the figures in many of the sculptural reliefs.391 In the same region, the Tomb of Cyrus at Pasargadae is also decorated with a large and very ornate rosette.392 Further west, the rosette also occurred in sculpture, such as in the eyes of fifth century Ionic capital volutes at Athens in S. Greece; Neapolis and Thasos in N. Greece and Histria along the Black Sea.393 The motif also occurs in some pebble mosaics from the same period, namely at Sicyon in S. Greece and Olynthus in N. Greece.394 In other media, the rosette motif appears fairly frequently on Attic (S. Greece) Red-Figure in the fifth and fourth centuries.395 Rosettes with eight petals also appear as stamps on jar handles from the Persian period at the Judaean sites of En Gedi, Jerusalem, Ramat Rahel and Tell en-Nasbeh in The Levant.396 The Persian period also saw the production of precious metal vessels that were decorated with rosettes in relief such as silver bowls that were found at Hieropolis in Egypt and Susa in The East, as well as a silver cup from Varbitsa (N. Greece).397 The rosette is also sometimes found on jewellery like earrings or inlay, such as from tombs at Tharros (N. Africa) and Vratsa (N. Greece) 391 Roaf, 1983, pl.XIIIb, XVa, XIXa, XX, XXIV, XXVa, XXVIb, XXVII, XXXIX, XLII, XLIV, XLVa. 392 Roaf, 1983, fig.147. 393 Merritt, 1982, p. 86. 394 Tsakirgis, 1989, pp. 411-412 (n.69). 395 Moore, 1997, #120, #129, #131, #1048. These rosettes consist of both dot rosettes (1048) and what are called “solid” rosettes (120, 129 &131). 396 Cahill West, 1995, p. 250, Table 1. Cahill West notes that these rosettes stamps, her Class V may date earlier to the Iron II period, but are generally dated to the Persian period because of their association with YHD and YRŠLM stamps in a fill from Ramat Rahel. 397 Bivar, 1961, 193-199; Muscarella, Caubet & Tallon, 1992, #170; Venedikov, 1977, pl.7. The bowls from Susa and Pithom (Hieropolis) as well as the cup from Varbitsa all date from the fourth century, while the bowl from Thmouis is dated to right around 300 BCE. 182 during the Persian period.398 Thus the rosette motif is well established in the 6th-4th centuries, appearing in many regions in a variety of media. The motif was especially popular in N. Greece, where it has the greatest number of occurrences and types of media. Other areas where it appears frequently are The Levant, which has the second most number of appearances as well as The East, S. Greece, and N. Africa, which each have similar levels of occurrences. In the Hellenistic period the rosette motif remains popular, though the exact geographic and media make-up shift around. In glyptics, twenty-one sealings from Seleucia (representing twelve separate seals) in The East and one sealing from Callipolis in S. Greece have rosettes, of which only eleven Seleucene sealings (from four seals) have eight petals.399 Elsewhere, there is a discoid seal of an eight-petal rosette, dating 330-150 BCE, which was found at Gordion in C. Anatolia, and both Susa (The East) and Artaxata (C. Anatolia) have Hellenistic sealings of cruciform or stellate decorative motifs that are similar to rosettes.400 Fewer coins appear to have used rosettes during the Hellenistic period. Those that do use the motif sometimes represent the continuation of an older tradition. Corcyra in N. Greece, for instance, continued to mint coins with rosettes into the third century.401 Other coins from the 3rd-1st centuries with rosettes came from an unknown central Italian mint, 398 Barnett. & Mendleson, 1987, #1.62, #2.26, #9.14, #20.34-5; Venedikov, 1977, pl.6. 399 Invernizzi, Bollati, Messina & Mollo, 2004, Vg 81-91; McDowell, 1935, TS.b.2.d.1; Pantos, 1985, #19. The examples with eight petals at Seleucia are Vg84 and Vg87-9. FLO3 is especially close to Vg88 while FLO6 is closer to Vg90, which is characterized as having six petals, though the photograph seems to indicate eight. 400 Dusinberre, 2005, #72; Kachatrian, 1996, fig.42; de Mequenem, R. 1927, #101. 401 Gardner, 1963b, Corcyra 198-200. 183 and, at Metapontum, which is also in Italy, there were coins that showed five barley- corns arranged around a central dot as a rosette.402 On the other hand, the rosette is often found in sculptural friezes from the Hellenistic period, usually alternating with bucrania but sometimes by themselves or with phialae and altars. The motif occurs at places such as the Altar of Poseidon and Amphitrite on Tenos in S. Greece; the Temple of Athena at Ilium, the Temple of Demeter at Pergamon and the Altar of Artemis at Magnesia in W. Anatolia; and a Hellenistic tomb at Sveshtari and the Propylon of Ptolemy II and the Arsinoeion at Samothrace all in N. Greece.403 Likewise, painted friezes of alternating bucrania and four-petal rosettes appear in another Hellenistic Bulgarian tomb at Kasanlik.404 Rosettes appear as ornaments on Punic stelae in N. Africa as well, such as 3rd-2nd century examples from Carthage with eight petals and a 2nd-1st century example from Sulcis with multiple petals.405 The rosette motif in a frieze as a border device or as a panel of emblem is also very common in Hellenistic mosaics appearing at Athens, Delos, Eretria and Tainaron in S. Greece; Pergamon and Priene in W. Anatolia; Tarentum in Italy; Agrigento and Morgantina in Sicily; and Aï Khanoum in The East.406 Of course, it is within the realm of Hellenistic ceramics (vessels and lamps) where the rosette is extremely popular as a decorative or even filler motif. The motif appears extensively in Hellenistic painted pottery, especially West Slope style vessels, such as those found at Athens (S. Greece) and on a trefoil jug from Rhodes (W. 402 Poole, 1963b, Uncertain Central Italy 39-42, Metapontum 45, 178-9. 403 Webb, 1996, pp. 29, 32, 47-48, 95-96, 133, & 148-149. 404 Verdiani, 1945, Fig.7-12. 405 Acquaro, 1988, #183, #185 & #521. 406 Bruneau, 1972, #93, #95, #194, #210, #214, #217, #261a, #267d, & #306a; Tsakirgis, 1989, #7, #13-14, 411-412 (n.69). 184 Anatolia).407 It was also a very common motif on mould-made vessels and other Hellenistic fine wares, including vessels from Antioch, Dor (including examples with 8 petals, 2nd century BCE) Samaria (including some 8-petal examples, 2nd century BCE) and Tel Anafa (including examples with eight rosettes, late 2nd-1st centuries BCE) in The Levant and at Athens and Corinth (both with examples of eight petals), Delos, and Sparta in S. Greece; Pergamon and Priene in W. Anatolia; and Tarsus in C. Anatolia.408 In addition, Hellenistic lamps that bear rosettes as a decorative motif occur at Dor (including examples with eight petals, 3rd-1st centuries BCE) in The Levant; Alexandria (the second half of 2nd-first centuries BCE) in Egypt; Athens, Corinth and Delos in S. Greece; Ephesus (including examples with eight petals, 2nd-1st centuries BCE), Cnidus (including examples with eight petals, 2nd century BCE), Mytlinene (with eight petals, second half of 2nd century-first half of 1st century BCE), and Pergamon (late-4th-1st centuries BCE) in W. Anatolia; Tsamourli (2nd-1st centuries BCE) near the Black Sea; and Cosa in Italy.409 407 Rotroff, 1997, #32, #74, #78, #116, #136-7, #156, #161, #167, #190-5, #198-200, #202-9, #212, #215- 6, #218, #240-2, #246-7, #250, #252, #265, #268, #269, #302-3, #306, #311, #315, #334, #338, #341, #354, #362, #364-9, #408-9, #411-6, #423-4, #427-30, #433, #435-8, #440-3, #460, #463-4, #478, #490-1, #580-1, #601-2, #605, #608, #820, #822-4, #830-1, #833, #835-7, #1106, #1148, #1151-2, #1171, #1177, #1180, #1216, #1220, #1224-5, #1235-6, #1265, #1268-9, #1271, #1275-1280, #1283-4, #1319, #1329, #1336, #1366, #1635, #1660, #1671, #1673, #1675, #1679, #1682, & #1685; Schaefer, 1968, Abb8.2 . Included in Rotroff are both dot-rosettes and the regular kind, while the Schaefer examples have six petals. 408 Conze, Bertlet, Philippson, Schuchhardt, & Graber, 1913; Cornell, L. “A Note on Molded Bowls.” Tel Anafa II,i. Ann Arbor, 1997, MB13-4 & MB 56-8; Courby, 1922, fig. 76.4; Crowfoot, 1957, Q1344, Q0346c & Q0350; Edwards, 1986, #5, #9-10, #16, #38, #40, #46, & #56-7; Goldman, 1956, fig. 130.F; Rosenthal-Heginbottom, 1995, Relief Bowls #28, #34-5 & #61; Rotroff, 1982, #4, #6-8, #14-5, #19, #20, #22, #26-7, #30-2, #35, #40, #42, #49, #51-2, #54-6, #58-60, #62-3, #67-9, #71-74, #76-8, #81, #83, #86-7, #89, #92-3, #96-100, #102, #106-11, #114-8, #120-6, #133-7, #139, #141, #144, #146, #150, #152, #154, #157-60, #162, #167-8, #170, #172, #174, #176-7, #182-3, #186-7, #189, #191, #193-4, #200, #202, #204- 5, #207, #209, #212, #215-6, #219, #224-5, #228, #230-2, #236, #238, #240-1, #243, #248-9, #252-3, #255, #257-8, #262-3, #265, #274-6, #287-8, #292, #298-304, #305, #315-6, #323, #327, #330, #332, #336-8, #340-4, #346, #352, #357-9, #361, #365, #368, #371-3, #377, #381, #383, #397, #400, #402-3, & #406-9, #411; Waage, Elderkin & Stillwell,1948, fig. 9.30 & 15.23; Warner Slane, 1997, FW322, FW340, FW346, FW359, & FW388; Wiegand, Schrader, Winnefeld, Zahn, Krumer, & Wilberg, 1904, #27 & #28; Woodward & Hobling, 1925, fig.2.i.. 409 Bailley, 1975, Q168, Q177, Q301, Q303, Q305, Q308, Q310-1, Q314, Q330, Q470; Broneer, 1930, #303, #320, #333, #341, #357, #365; Bruneau, 1965, #2051-7; Howland, 1958, #583, #589, #591, #594, 185 Hellenistic braziers bearing rosettes, mostly with eight petals, occur in W. Anatolia at Calymna, Cnidus, and Halicarnassus and in The Levant at Dor.410 An eight petal rosette also appears on an amphora handle stamp from the Athenian agora that is probably from Rhodes (W. Anatolia) as well as a six petal rosette on a stamp from Cnidus in the same region.411 Rosettes also appear occasionally in various metal objects in the Hellenistic period. Rosettes with multiple petals occur on an early Hellenistic silver bowl from Thmouis (Egypt) and on an ornate bronze second century lamp found near Tel Aviv (The Levant).412 At Athens, in S. Greece, rosettes with six petals appear on Hellenistic lead tokens from the Agora.413 A rosette with eight petals, similar in form to FLO6, also appears on a lead weight from Marisa (The Levant) that bears a date in the Seleucid era equivalent to 143/2 BCE.414 Finally, rosettes with eight petals also make appearances in Scythian jewellery, such as a gold necklace from a 2nd century tomb at Ryzhanovka (Europe) and an elaborate gold buckle from a 1st century tomb near Kurzhips (Black Sea).415 And so, during the Hellenistic Period, the rosette motif appeared in many different media, but was most popular as a decoration on vessels of various types. The motif occurs most frequently during this period in the regions of S. Greece and W. #602, #606, #619, #638; Mlynarczyk, 1997, type G; Rickman Fitch & Wynick Goldman, 1994, #170, #188, #190-1, #284, #289, #309, #311, #1090; Rosenthal-Heginbottom, 1995, Lamps, Type 11.3, 13.6 & 16.3; Schaefer, 1968, M3, Taf55. 410 Rosenthal-Heginbottom, 1995, Braziers #13; Şahin, 2001, Ha 2-5, Ha 7, Ha 59, Cn1, Ky 2, Ky 7. Of these examples, the example from Dor, the Cnidian, one Calymnan (Ky 1) and three examples from Halicarnassus (Ha 2-4) have eight petals while the others have a different number. 411 Grace, 1934, #200 & #273. 412 Bivar, 1961, 300; Sussman, 2006, 39-50. 413 Lang and Crosby, 1964, L219 & 220. 414 Kushnir-Stein, 2002, #2. 415 Minns, 1913, 180 & 224, figs. 74 & 127. 186 Anatolia, and The Levant, buoyed in part by the motif’s popularity on the ceramics there. The rosette appears in the other regions consistently if less frequently so that no region lacks at least one example of the motif. As a whole, the presence of floral motifs at Kedesh, both flowers and rosettes, is slightly unusual. Neither flowers nor rosettes appear commonly in the glyptics of the Persian or Hellenistic periods, though they are not unknown and specific examples from Kedesh are similar to examples found at the other archives. Outside of glyptics, the pictures diverge between the two categories. For its part, the use of flowers as a symbol is altogether rather sporadic and limited. Flowers appear occasionally in a few media, such as coins. The rose is perhaps the most common, appearing on the coin issues of several places, especially Rhodes to which the flower is especially associated. Rosettes, on the other hand, are much more common in many media from ceramics to sculpture. They seem to especially cluster in and around the Aegean, though they appear with regularity elsewhere, including the Near East. Of note with regard to the Kedesh examples are parallels in the sealings of Seleucia, which are very similar in appearance and a coin from Tyre, which is close geographically. Catalogue: FLO1 Number of Sealings (Bullae): 1 (K99 0313) Maximum Preserved Dimension: 10 mm. Shape: small flat rectangular or oblong seal Description: A flower in a long and narrow impression, apparently done by an all metal ring. It consists of a round bud of closed and pointed petals on a bent stem with at least one leaf and a possible stamen or pistil above the petals. The general shape of the flower 187 would seem to indicate that it is meant to represent some species of crocus. FLO2 Number of Sealings (Bullae): 1 (K99 0326) Maximum Preserved Dimension: 12 mm. Shape: convex, probably oval seal Description: A highly stylized flower that is hard to interpret due to summary nature of the image. There are several converging lines forming triangular a triangular body which seems to represent a vaguely tulip-like flower, perhaps with stamina or pistils emerging from between the petals. FLO3 Number of Sealings (Bullae): 1 (K00 0245) Maximum Preserved Dimension: 10 mm. Shape: flat oval seal 188 Description: An eight petal rosette made with sixteen separate rays around a central button made up of two pieces. There are traces of a cable border. Three quarters of the impression is preserved. FLO4 Number of Sealings (Bullae): 2 (K00 0283, K00 0543) Maximum Preserved Dimension: 12 mm. Shape: flat, probably round seal Description: A naturalistic rose. A total of five petals are shown closed up. Below the flower there are traces of leaves and a long stem. There are also traces of a border along the edges of the impressions. FLO5 Number of Sealings (Bullae): 1 (K00 0575) Maximum Preserved Dimension: 8 mm. 189 Shape: small flat oval seal Description: A flower or some sort of plant. There are five lines of different lengths that form a floral spray emerging from a base of two diagonal lines. The edges of the seal seem mostly intact. FLO6 Number of Sealings (Bullae): 1 (K00 0604) Maximum Preserved Dimension: 8 mm. Shape: small flat round seal Description: A Rosette. There are eight round petals that irregularly surround a round or oval middle. 190 Chart 3.21: The Distribution of Flowers in the Persian Period Map 3.23: The Distribution of Flowers in the Persian Period                  191 Chart 3.22: The Distribution of Flowers in the Hellenistic Period Map 3.24: The Distribution of Flowers in the Hellenistic Period                 192 Chart 3.23: The Distribution of (Eight-Petal) Rosettes in the Persian Period Map 3.25: The Distribution of (Eight-Petal) Rosettes in the Persian Period                     193 Map 3.25a: Detail Map Showing the Aegean Region 194 Chart 3.24: The Distribution of (Eight-Petal) Rosettes in the Hellenistic Period Map 3.26: The Distribution of (Eight-Petal) Rosettes in the Hellenistic Period                             195 Map 3.26a: Detail Map Showing the Aegean Region 196 Section 10: A Ligature A single example from the Kedesh symbol sealings displays the image of a kind of knot or ligature. The knot displayed takes the form of a single band forming the outline a square or rectangle. At the single surviving corner in the impression, and presumably on each of the corners that previously existed and of which there remain traces, the band loops around outside the bounds of the square in order to pass on to form the next side of the square. Distribution: This motif is highly unusual and there almost no parallels. Yet, one of the sealings from the Hellenistic archive at Artaxata (C. Anatolia) shows a very similar ligature. The Artaxata exemplar, however, consists of three bands, instead of just one, and contains a small bird as an additional element.416 Catalogue: LIG1 Number of Sealings (Bullae): 1 (K00 0355) Maximum Preserved Dimension: 8 mm. Shape: slightly convex, probably oval seal Description: A ligature. This knot takes the form of a single band forming the outline of a square with loops at each corner. Only the upper right-hand loop is complete in the impression. 416 Khachatrian, 1996, fig.41. 197 Chart 3.25: The Distribution of Ligatures in the Hellenistic Period              198 Map 3.27: The Distribution of Ligatures in the Hellenistic Period 199 Section 11: Lyres Three of the Kedesh sealings bear impressions depicting stringed instruments, specifically lyres or kitharas. These three impressions represent the use of three distinct seals. Between the three examples, there is a certain amount of variation in the depictions. The first seal (LYR1) displays the tortoiseshell sounding box of a chelys, or turtle-shell lyre. In the case of the second seal (LYR2), however, the bottom portion of the instrument is missing from the impression and it is not possible to tell if the seal represented a lyre or a kithara with its different shaped sounding box. In both of these cases, the depictions were done in a naturalizing fashion that would seem to indicate a familiarity with either the instruments themselves or their iconography in Greek art. The third seal (LYR3) differs from the first two in that is a much more simplified and stylized depiction that seems not to have taken direct inspiration from either the instruments themselves or their iconography. This third seal is also markedly different from the other two due to it being much smaller in size than the others. Classically, the lyre, specifically the kithara, was the attribute of the god Apollo. He is, for example shown holding it in sixth century Attic Black-Figure.417 The lyre was also used as a symbol on its own from an early period as illustrated by a sixth century Attic Black-Figure phiale and a coin from Delos from the same time.418 The lyre as a symbol often continued its association with Apollo. Thus, many of the coins that used the lyre as a reverse type have the head of Apollo on the other, such as with a fifth 417 Moore, Philippides, von Bothmer, 1986, #571. 418 ANS Online Catalogue, 2004.18.4; Moore, Philippides, von Bothmer, 1986, #1431. The Delian coin shows an obverse with a seven-string kithara and a four-part incuse square on the reverse, while the fragments of the phiale show two friezes that alternate drinking horns and lyres. 200 century coin from Acanthus, in the collection of the American Numismatic Society.419 Besides, the association with Apollo, the lyre motif can also be linked with other divinities, such as Dionysus, as well as the pleasurable aspects of life and leisure by its nature as a musical instrument.420 Distribution: Used as a symbol, the lyre seems to have different levels of commonality depending on the media used. In the case of glyptics, the motif is somewhat uncommon. It does make appearances, just not as often as other symbols that are also found in the Kedesh corpus. For instance, during the sixth-4th centuries, the motif does not make an appearance on the sealings of any archives. Some carved gems from this period, however, indicate that the motif did appear, if infrequently, in the glyptic repertoire. One such example is a mottled late 5th century scaraboid that shows a kithara, now in the Ashmolean Museum.421 While glyptic examples of the motif during the Persian period are rare, numismatic examples are more common. As noted, a lyre appears on a sixth century coin from Delos. Issues of the Fifth and Fourth centuries bearing lyres, including chelyes and kitharas, appear at Delos (before 480 BCE), Megara (400-338 BCE), Melos (4th-1st centuries BCE), Pellene (370-280 BCE), Sicyon (370-250 BCE) and Zakynthus (394- 357 BCE) in S. Greece; the island of Lesbos (440-350 BCE) and its cities, Methymna 419 ANS Online Catalogue, 1944.100.10234. 420 Either case can be inferred from its association with the drinking horns in the case of the friezes noted on the previously mentioned phiale. 421 Boardman, 1970, pl.614; Boardman & Vollenweider, 1978, #111. 201 (420-377 BCE) and Mytilene (4th-3rd centuries BCE), Colophon (5th-4th centuries BCE), Hamaxitus (400-310 BCE), Iasus (412-408 BCE), Pordosilene (after 450 BCE), and Teos (4th-3rd centuries BCE) in W. Anatolia; Acanthus (424-380 BCE), Aetolian Apolonia, Bottice (392-379 BCE), Chalcidice (392-358 BCE), Olynthus (400-348 BCE), Thracian Sestus (4th-3rd centuries BCE) and Zone (350-300 BCE) in N. Greece; Rhegium in Italy; and Adranum (345-275 BCE) and Halaessa (345-275 BCE) in Sicily.422 These mints tend to cluster in a couple of places, namely around the Chalcidian Peninsula in the north of the Aegean (N.Greece) and along the eastern coast of the Aegean. Outside of the Aegean region, the motif only appears in the issues from one mint in Italy and a couple in Sicily. During the Hellenistic Period, the lyre motif seems to have become more popular, appearing more frequently. In glyptics, examples of the motif appear in the published sealings of three of the archives. Specifically, it appears at Callipolis and Delos in S. Greece and Seleucia in The East. At Seleucia, lyres, including both chelyes and kitharas, appear on seven sealings, each one representing a distinct seal.423 At Delos, two sealings bear impressions from the public seal of Colophon that depicts a kithara along with an inscription naming the city. A similar seal impression was also found at Callipolis.424 Both kinds of instrument also make the occasional appearance in gems of 422 ANS Online Catalogue, 1944.100.10234, 1957.172.1623, 1989.40.1-2, 1992.4.28, 1992.54.40-3, 1992.140.11, 1998.115.70, 1999.3.2, 1999.4.22; Gardner, 1963a, Pellene 5, Sicyon 82 &162, Zakynthos 29-32; Gardner, 1963b, Apollonia 2; Head, 1963a, Megara 1 & 4; Head, 1964, Colophon 1-8, 11, 13-15, Teos 26, Satrapal (Iasos) 13; Head, Poole & Gardner, 1963, Adranum 1-2, 4, Alaesa 1; Kroll & Walker, 1993, #451 & #472; Plant, 1979, #2169-70, #2177, #2179-82, #2185, #2198; Poole, 1963b, Rhegium 43- 44, 84-86 & 108; Wroth, 1963, Melos 30-31; Wroth, 1964b, Hamaxitus 1-1, Lesbos 61, Mytilene 15-16, Pordosilene 1. 423 Invernizzi, Bollati, Messina & Mollo, 2004, Og 283-287. The seal Og 287 is more stylized and bears some resemblance to the depiction present in LYR3. 424 Boussac, 1992, SP6; Pantos, 1985, #78. 202 the Hellenistic period from various collections.425 In numismatics, the increase in popularity is even more apparent during the Hellenistic period. The number of polities issuing coins with lyres as a type increased greatly, with several of the older cities continuing the practice and many more appearing throughout the third through first centuries BCE. These places included Antioch (under Seleucus I) and Ptolemais-Ake (under Antiochus VII and Cleopatra and in 111-110 BCE) in The Levant; Anactorium (first half of 4th century BCE), Aptera (200-67 BCE), Athens (130-90 BCE), Cythnus (2nd century BCE), Delos (200-87 BCE), Eleutherna (3rd century BCE), Lappa (200-67 BCE), Leucas (a chelys ,after 167 BCE), Megara (307-243 BCE), Melos (4th-1st centuries BCE), Sicyon (370-250 BCE) and Thespiae (a chelys, 146-27 BCE) in S. Greece; Abydos (320-200 BCE), Aegae (2nd-1st centuries BCE), Alexandria Troas (2nd-1st centuries BCE), Bithynian Apamea (300-203 BCE), Bithynian royal issues ( under Prusias I and II), Calymna (3rd century BCE), Caystriani (in the form of a Lyre made from a bucranium, 2nd-1st centuries BCE), Chalcedon (280-early 2nd century BCE), Colophon (330-280 BCE), Cos (a chelys, 88-50 BCE), Halicarnassus (2nd- 1st centuries BCE), Lycia (168 BCE-43 CE), Methymna (330-240 BCE), Mytilene (4th- 3rd centuries BCE), Lydian Philadelphia (2nd century BCE), Sardis (under Antiochus II), Smyrna (3rd century BCE), Teos (4th-3rd centuries BCE), Termessus Minor (1st century BCE) and Tripolis (Apollonia, 1st century BCE) in W. Anatolia; Aetolian Apollonia (1st century BCE), Thracian Sestus (4th-3rd centuries BCE) and Zone (350-300 BCE) in N. Greece; Sarmatian Olbia (175-150 BCE) in the Black Sea; Canusium (a chelys, c. 300 425 Brandt & Schmidt, 1970, #664 & #2137-2139; Platz-Horster, Schulter & Zazoff, 1975, #745; Maaskant-Kleibrink, 1978, #233. 203 BCE), Capua, Naples (250-200 BCE), Rhegium (270-203 BCE), Rubi (a chelys, 3rd century BCE), Thurium, Tuder, Venusium, and Vibo Valentia (192-89 BCE) in Italy; Adranum (345-275 BCE), Calacte (275-212 BCE), Centuripe (after 241 BCE), Halaesa (3rd and 1st centuries BCE), Lilybaeum (after 241 BCE), Malta (2nd-1st centuries BCE), Menae (275-212 BCE), Panormus (before 254 BCE), and Syracuse (317-289 BCE) in Sicily; Bactra, Ecbatana (both under Antiochus I) and Seleucia on the Tigris (under the Seleucid usurper Molon) in The East; and Cyrene (308-277 BCE) in N. Africa.426 Among this group, the various Seleucid issues with lyres are especially notable considering the association between the instrument and the dynastic patron deity Apollo. Altogether, the various Hellenistic issues tend to cluster in a few geographic regions. The greatest single concentration of issues lies in W. Anatolia, especially along the eastern coast of the Aegean. The region of S. Greece also has multiple appearances, continuing the earlier tradition. Interestingly, the motif seems to have fallen from popularity as a coin motif in the northern Aegean, as N. Greece only has two occurrences. The regions of Sicily and Italy also have numerous occurrences of the lyre motif on coins. 426 ANS Online Catalogue, 1940.77.49, 1941.36.105-6, 1941.131.537, 1944.100.15129-30, 1944.100.20253-8, 1944.100.41924-5, 1944.100.41957, 1944.100.48506-10, 1944.100.56972-3, 1944.100.71190-1, 1944.100.73608-17, 1944.100.73630, 1944.100.77892, 1944.100.78177-8, 1944.100.78184-5, 1944.100.78197-200, 1944.100.79493-8, 1944.100.79506-7, 1944.100.79509-10, 1944.135.126, 1944.135.134, 1945.18.2, 1947.99.85, 1953.171.129, 1953.171.853-8, 1956.28.366, 1958.206.76, 1960.170.359, 1970.142.106, 1970.142.506, 1974.140.112, 1986.78.910, 1992, 32.50, 1993.29.6, 1999.4.22; Gardner, 1963a, Pellene 5, Sicyon 82, 162; Gardner, 1963b, Apollonia 80-82, Anactorium 3, Leucas 121-131& 171-191; Head, 1901, Caystriania 5-9, Philadelphia 5-7, Tripolis 2-3; Head, 1963a, Megara 10-14, 44; Head, 1964, Smyrna 13, Teos 40-42; Head, Poole & Gardner, Adranum 1-2, 4, Alaesa 1, 9, 16, Calacte 5, Centuripa 7-8, Lilybaeum 2-3, Menaenum 10, Panormus 7, Syracuse 266-269; Mørkholm, 1991, #129 & 252; Kroll & Walker, 1993, #111, #420, #451, #606, #646, #649, #827-9, #903; Newell, 1977, #922, #1392, #1397, #1401, #1407; Newell, 1978, #525-7; Mørkholm, 1991, #361; Plant, 1979, #2150-54, #2156-2161, #2163-8, #2171-6, #2180, #2183-4, #2186-9, #2194-5, #2197; Poole, 1963b, Tuder Class I 1, 12-18, Capua 19-20, Canusium 1-3, Venusia 3, Thurium 145, Vibo Valentia 31-32, Rhegium 43-44, 84-86; Wroth, 1963, Aptera 17, Lappa 3, Cythnos 1-3, Delos 4-8, Melos 30-31; Wroth, 1964b, Abydos 48, Aegae 15, Alexandria 29-36, Apamea 12-13, 15-17, Chalcedon 22, Methymna 10, 12-15, Mytilene 28-95, 145-152, Prusias 1. 204 Elsewhere, the motif appears more sporadically with an occasional issue in The Levant, The East (both under the influence of the Seleucids) and a single occurrence in both the Black Sea and N. Africa. In the Hellenistic period the lyre motif also begins to make sporadic appearances in other media. In sculpture, for instance, it does show up as a repeated motif flanked by heraldic griffins at the Temple of Apollo at Didyma (W. Anatolia), dating to the first half of the second century.427 The lyre also appears on a Late-Hellenistic lead token that was recovered from the Athenian agora (S. Greece).428 Both of these places were within the geographic regions where the coin issues tended to cluster. Thus the appearance of lyres on three of the Kedesh sealings is slightly unusual. The lyre motif, while not unknown in Hellenistic glyptics was not especially popular and appears infrequently. The lyre motif was much more popular in numismatics, where numerous polities used it on various issues, especially along the western coast of Anatolia and in Sicily. Kedesh itself is outside the regions where the issues of lyre coins tended to cluster. As such, it would seem that the motif played only a small role in the symbolic repertoire of Kedesh, being present but not necessarily all that popular in the region. This position would fit well with the small numbers of examples at Kedesh. Catalogue: LYR1 Number of Sealings (Bullae): 1 (K99 0526) Maximum Preserved Dimension: 17 mm. Shape: convex, possibly oval seal 427 Webb, 1996, p. 105. 428 Lang & Crosby, 1964, L207. 205 Description: A lyre with a round sounding box that is probably a tortoiseshell (a chelys), curved arms and five strings. The work is finely done. One of the arms is damaged, seemingly the result of a fault in the stone of the seal. The seal itself appears to have been a convex oblong gem of good size. LYR2 Number of Sealings (Bullae): 1 (K00 0098) Maximum Preserved Dimension: 11 mm. Shape: flat seal Description: A lyre or kithara. Only the arms remain in the impression which is in fairly rough condition. The body is missing and the strings are not visible. The arms curve in the same fashion as the previous seal. 206 LYR3 Number of Sealings (Bullae): 1 (K00 0530) Maximum Preserved Dimension: 9 mm. Shape: convex, possibly oval seal Description: A stylized lyre-like stringed instrument. The right side is missing from the impression and the work is somewhat crudely done. The body is football-shaped and the arms are straight. There are six strings. Chart 3.26: The Distribution of Lyres in the Persian Period                     207 Map 3.28: The Distribution of Lyres in the Persian Period 208 Map 3.28a: Detail Map Showing the Aegean Region 209 Chart 3.27: The Distribution of Lyres in the Hellenistic Period Map 3.29: The Distribution of Lyres in the Hellenistic Period                      210 Map 3.29a: Detail Map Showing the Aegean Region 211 Map 3.29b: Detail Map Showing Italy 212 Section 12: Masks Some thirty-one of the Kedesh sealings show images of masks. These sealings represent the use of twenty separate and distinct seals. The mask category is, in fact, the largest single category amongst the Kedesh symbol sealings. The Kedesh masks fall basically into two sub-groupings: those that come from the world of Greek theatre and those from other traditions. The Greek theatrical masks are the largest class with 22 sealings (14 seals) that may be connected specifically with New Comedy and Satyr Plays. The second class is much smaller with only 10 sealings, representing the use of six separate seals. The class is also more nebulous than the first as it contains two masks (MAS8 and MAS14) that seem to come from a Phoenician/Punic artistic tradition as well as a set of three masks (MAS3, MAS5, and MAS16) whose depictions render their exact artistic ancestors unclear. The masks that show satyrs might belong to either class as they appear both in the Greek theatrical and the Phoenician mask traditions. The Theatrical Masks readily fall into the categories devised by Webster, especially in his work on New Comedy.429 At Kedesh, New Comedy masks are the most common, specifically the masks of slaves, which predominate with 8 seals (MAS1-2, MAS7, MAS9, MAS13, MAS17, and MAS19-20) and a total of eleven sealings. These masks are characterized by their large, grinning mouths with beard trumpets. Their facial features are often contorted, notably the eyebrows, and their noses are often small or broad and flat. They often have a roll of hair, called a speira, which may be peaked (see MAS1, MAS9, MAS17, and MAS19-20). Though distinguishing characteristics, such as hair colour, are not present, they appear to fall into either Webster’s type 22 (Leading 429 Webster, 1995; Webster, 1967. 213 slave) or 27 (Wavy-haired Leading Slave) types.430 One example (MAS2) may represent one of Webster’s type 26 (The Tettix), because of his bald head and crossed eyes.431 There is also a single seal (MAS6), with two sealings, that shows a pornoboscos from New Comedy (Webster type 8), identifiable from the shape of the face and the hair. It is similar in depictions to sealings of the same mask type found at Delos.432 The second largest group of theatrical masks, with 7 sealings and 5 seals, consists of masks from Satyr plays (MAS4, MAS11-12, MAS15, and MAS18). These masks represent either satyrs or the god Pan. They can be either bearded or clean shaven. They have pug noses and wild hair. Three of these masks (MAS4, MAS15, and MAS18) appear to show masks of the god Pan in that they have horns rising from the forehead. The other two (MAS11- 12) apparently lack the horns and show, instead, bearded satyrs or Sileni. Other Theatrical Masks include a single example (MAS10) with a single sealing. It shows a bearded old man, which could possibly also be a bearded satyr. The other sub-category of mask seals is much smaller, only containing five seals (MAS3, MAS5, MAS8, MAS15 and MAS17). The masks in this sub-category are not united by a single unifying theme. The first seal (MAS3) can be found on four sealings and lacks distinguishable details. It has small eyes and a closed mouth with lines of unidentifiable purpose rising from the top of its hair. The second (MAS5) is also enigmatic. It is more stylized than the first, being frontal with bulging eyes and cheeks with a closed mouth. It sports a strange headdress in the form of stalks of varying heights with bulbous ends, vaguely resembling plant shoots or flowers. This strange headdress maybe represents a stylized wreath or the snakes of a 430 Webster, 1995, pp. 26-29 & 32-34. 431 Webster, 1995, p. 32. 432 Boussac, 1997, fig.36. 214 gorgon’s head. In either case, the seal lacks close parallels. Another example (MAS16) shows a frontal mask or mask like face with a closed mouth that is missing any definite detail that would allow us to place it correctly. The shape of its nose and the traces of its hair may indicate that it is another example of a satyr mask, but we can not be certain from the impression. Finally, two examples (MAS8 and MAS14) alone have identifiable parallels, specifically in the tradition of Phoenician/Punic masks. The first shows a closed-mouth male with a heavy brow and a long beard. The second is likewise closed- mouthed but with smooth cheeks. The masks here with their closed mouths and triangular faces most closely resemble different types of masks found in different Phoenician/Punic contexts. These Phoenician masks have been discovered at a number of sites in the Levant, Cyprus and the Punic west, mostly from funerary contexts, but including sanctuaries especially in the east. These masks have antecedents dating back to the Late Bronze Age in the Levant. Many examples of Phoenician masks have been found in Iron Age contexts. The motif appears in the Punic west by 700 BCE, and many examples dating down into the 5th century have also appeared. During the 4th century, the Phoenician mask began to be supplanted in its normal funerary and religious contexts in Phoenician/Punic sites by masks from the Greek theatrical world, and very few examples of the Phoenician style masks have been found dating to the Hellenistic period. The Phoenician masks seemed to have functioned as apotropaic emblems, representing Phoenician gods or demons, especially those found in tombs, as many of the examples that have been recovered have been too small to actually be worn.433 The 433 Ciasca, 1988, pp. 354-358; Piccard, 1965, pp. 7-104; Karageorghis, 1996, pp. 813-821; Stern, 1995, p. 447. 215 Phoenician masks and protomes come in a variety of types. Cintas has classified the Carthaginian finds into six categories: three different categories of grotesque masks that are typified by various deformations of their open mouths and that often have wrinkles on their faces; a fourth category consisting of male masks and protomes that have closed mouths and are often bearded; a fifth consisting of satyr masks; and female protomes.434 Two Kedesh examples (MAS 8 and MAS 14) both fall into Cintas’ fourth category. It is important to note that satyr masks also show up in the same contexts as the Phoenician masks. The motif therefore appears in the Greek Theatrical world and in the Punic- Phoenician tradition. Its appearance in the latter is actually a conflation of two artistic motifs already present in Phoenician contexts. Specifically, Phoenician art already had a the motif of a bearded, horned demon with animal ears, circular forehead decoration and a wrinkled face that appears in the form of small amulets of frit or silver from seventh and sixth century contexts at Byblos and Cyprus in the east and Tharros and Carthage in the west.435 At the same time, Phoenician art is replete with depictions of the Egyptian god Bes, with his bearded face, bald head and pug-nose. At some point, these two motifs became linked with one another.436 This linkage means that the satyr mask motif, either borrowed from the Greeks or developed independently, would fit perfectly into the particular conceptual space of both the demon and Bes within Phoenician art without having to necessarily differentiate between the motifs. The satyr mask motif could 434 Cintas, 1946, pp. 32-55. 435 Culican, 1976a, fig.1 436 Culican, 1976a, pp. 21-24. For examples of Bes in Phoenician art see Acquaro, 1988, #66 (6th century Amulet from Larnaka), #278 (6th-4th century Amulet from Carthage), #465 (4th-3rd century Plastic Vase in the form of Bes from Sicily), #499-500 (2nd century statues of Bes from Cagliari), #555 (4th century statuette of Bes from Cagliari), #686 (5th-4th century Amulet of Bes from Sulcis), #699 (7th-4th century Amulet of Bes from Tharros), #810 (4th century Die with image of Bes from Ibiza), and #862 (4th century Amulet of Bes from Ibiza). 216 therefore represent either one of the other motifs, both of them together, or its own creature entirely, as it did in Greek art. The fact that satyr masks occur in the same Phoenician contexts as other Phoenician masks, such as funerary, indicates that this motif also served an apotropaic function there. This function was not limited simply to its use in Phoenician ritual. Instead it seems to reappear again and again in different places around the Mediterranean. Indeed, the exaggerations of facial features and expressions give the image of satyr masks in general an appearance similar to other grotesque heads that are used for apotropaic functions by various cultures around the world, such as the well known gargoyles on European churches.437 A good example where the satyr mask has an apotropaic function is the popular use of the motif on ceramic vessels that provide light and warmth, namely lamps and braziers.438 Here, the satyr mask takes the role of the evil or harmful supernatural force that is frightened off by the light of the vessel.439 At the same time, we also note that the masks of the New Comedy slave share many of the exaggerations of form and expression, such as the exaggerated leer and brow that would make them suitable for the same apotropaic functions.440 Coincidentally, they also appear frequently on lamps in the Hellenistic period.441 As such, the New Comedy slave mask would also 437 Sütterlin, 1989, pp. 65-74. Sütterlin here discusses the presence and function of both lewd gestures and grotesquely exaggerated features on sculpture from Medieval Europe. She links these to widespread patterns of depiction across many cultures and time periods, which are used for apotropaic purposes. She notes that many of these gestures or expressions, such as the showing of the tongue, rise from cross- cultural behaviours involved with human aggression and, as such, get used in art to fend off potential harm from whoever displays them. 438 See for instance: Sussman, 2006, pp. 39-50; and Şahin, 2001, pp. 91-132. 439 Sussman, 2006, p. 43. 440 Interestingly, these are the same general exaggerations of features that appear on the wrinkled or grotesque Phoenician masks. 441 For instance see Bailley, 1975, Q273-4, Q300, Q316, Q321, Q324, Q326, Q328, Q330, Q612, Q618. These lamps come either from Cnidus or Egypt. 217 serve in a similar protective function as satyr masks. On the other hand, these two mask types represent more than simply grotesque faces. The satyr mask represents a mythological creature who is a companion of the wine god Dionysus and a member of that god’s thiasos. The New Comedy slave mask by its presence refers to a popular form of entertainment and of leisure in the Hellenistic period. Thus the two masks can also refer to the idea of a good life free of cares and work. In this regard, we may not be dealing with a question of one valence being more prominent over another in any given use of these motifs. Rather, we should consider that the two motifs have the potential for several valences simultaneously with the exact preponderance of meaning being determined by the actual user, which would hinder us from reconstructing a specific valence for any one use of the motif. Distribution: Masks as a general motif were ubiquitous in the Classical and Hellenistic Periods. They were extremely popular over a wide geographical expanse and a large number of media, including seals/gems, vessels, terra cottas, sculptures, mosaics, and paintings. The one medium in which they are uncommon is coins, where they rarely appear. Each of the three groupings of mask motifs from the Kedesh sealings (New Comedy slave, satyr and Phoenician) has a different distribution, both in time and space, and will be treated separately here. Of the three, the New Comedy slave mask appears most frequently and widely. Starting with the advent of New Comedy in the Hellenistic period, the New Comedy slave mask motif appears all over the Greek world. It started in 218 the Greek heartland, especially in Athens which was both a producer and consumer of objects bearing the motif, and expanded outwards from there. Occurrences cluster in the area of the Aegean and in Italy, where New Comedy was also very popular. The New Comedy slave mask motif appears in numerous media. In glyptics, the motif appears on 2451 sealings from Seleucia (representing the use of 62 separate seals) in The East, hundreds of sealings from Delos, two sealings from Callipolis (both S. Greece), and two more from Cyrene (N. Africa).442 A lone sealing showing a slave mask has also been found in Egypt.443 As might be expected the archives such as Elephantine, Carthage and Selinus that were founded before the Hellenistic period do not appear to have examples of the New Comedy slave mask motif, though they do have examples of other Greek theatrical masks. Interestingly, this absence of the slave mask motif occurs even if the archive continued to function well into the Hellenistic period, such as Carthage and Selinus, indicating perhaps a failure of the motif to expand into those particular niches. The motif is also commonly found in the form of miniature terra cotta masks and heads from a variety of sites and regions, all dating from the 3rd to the 1st centuries BCE. These places include Cyprus, Kharayeb, Marisa, and Mersinaki in The Levant; Alexandria, the Fayum, Naukratis, and undefined Egyptian findspots in Egypt; Athens, Corinth, Delos, Knossos, Melos, Olympia, Pitsa, Thebes, and undefined Greek findspots in S. Greece; Aeolis, Clazomenae, Iasus, Larissa, Mytilene, Pergamon, Priene, Smyrna, and Troy in W. Anatolia; Doriskos, and Thasos in N. Greece; Amisus, Callatis, Mesambria Pontica, and Panticapaeum in the Black Sea; undefined Asia minor and 442 Invernizzi, Bollati, Messina & Mollo, 2004,M 33-M 90; McDowell, 1935, III.3.a.1, 4-7; Boussac, 1997, p. 148; Boussac, 1988, fig. 68; Madolli, 1963-1964, #640 & 643; Pantos, 1985, #80-81. 443 Webster, 1995, 3DJ1. 219 Gordion in C. Anatolia; Cales, Campania, Capua, Locri, Polychoron, Segni, Tarentum, Tarquinia, Tuscana, and Vulci in Italy; Centuripe, Ieta, Lipari, Morgantina, and undefined Sicilian findspots in Sicily.444 Other sculptural appearances of the motif include a decorated Corinthian capital in Petra (The Levant); a fragmentary frieze from the Stoa of Attalos in Athens (S. Greece); terra cotta masks from the wall decoration of a house in Priene, a relief from the theatre in Pergamon (W. Anatolia), a terracotta plaque of masks from Amphipolis (N. Greece); a mask relief from Ostia, mask reliefs from Pompeii, a relief in Rome, terra cotta antefixes from Tarentum, and a relief from an undisclosed Italian site (Italy); a limestone mask of a cook from Ai Khanum (The East).445 The motif is also very commonly found during the same period as a decoration on ceramic vessels, including lamps, appearing as painted decoration, as an appliqué or as part of a moulded decoration. The sites and regions where such objects have been found include Antioch, Cyprus, Samaria, and Tell Arqa in The Levant; Alexandria, Karanis, Naukratis and undefined Egyptian findspots in Egypt; Aegina, Argos, Asine, Athens, Corinth, Delos, Delphi, Helleniko, Knossos, and Syros in S. Greece; Cnidus, Myrina, Mytilene, Pergamon, Rhodes, Samos, Smyrna, and Troy in W. Anatolia; Marvinci in N. Greece; Kepoi in the Black Sea; undefined Asia Minor findspots, Olbia and Tarsus in C. Anatolia; Bari, Cales, Campania, Civita Castellana, Clusium, Corchiano, Etruria, Metapontum, Naples, Polychoron, Pompeii, Rhegium, Rome, Sybaris, Tarentum, Tuscana, Volterra and undefined Italian findspots in Italy; Heraclea Minoa, Lipari, 444 Erlich & Kloner, 2008, #182; Webster, 1995, 1AT17, 1AT28, 1AT30, 1AT59-61, 1AT63, 1BT12, 1CT2-3, 1DT6, 1DT8-9, 1DT11, 1DT18-23, 1E5, 1KT5, 1NT11, 1ST37-41, 1ST67, 1TT16-19, 2DT4-5, 2DT7-10, 2DT12, 2DT17-18, 2DT22-24, 2ET10, 2KT3, 2RT21, 2RT23, 2RT28-30, 2RT32-34, 2ST6-7, 2TT7-10, 3CT2-3, 3DT43, 3DT45, 3DT51, 3DT67, 3DT77-78, 3DT82, 3DT84-87, 3DT91.1, 3DT97-99, 3DT102, 3ET3-5, 3ET9, 3ET12-13, 3NT9, 3RT9-10, 3RT17-20, 3RT23, 3ST8-12, & 3TT5-7. 445 Webster, 1995, 1AS1, 1BT5, 1TT20, 2HS1, 3AS2-3, 3DS2, 3DT90, & 3HS1 . 220 Marineo, and Tyndaris in Sicily; and Carthage and Mahdia in N. Africa.446 Besides ceramic vessels, comic slave masks appeared as decoration on precious metal vessels such as a bronze twin spouted lamp from Priene in W. Anatolia and a silver phiale from Tarentum in Italy.447 Likewise, the motif also appears as jewellery decoration in places such as Corinth and Delos in S. Greece; Abdera in N. Greece; Neapolis Scythica and Panticapaeum in the Black Sea; Herculaneum, Pompeii and Stabiae in Italy; and Vaison in Europe.448 New Comedy slave masks as separate motifs, distinct from characters, also appear in media such as sculpture, mosaics and wall-paintings, notably mosaics and a painting from Delos (S. Greece), a mosaic from Rhodes, a mosaic from Palace V in Pergamon (W. Anatolia); mosaics from Pompeii, a mosaic from the Via Ardeatina in Rome, a mosaic in Tusculum (Italy); and a mosaic from Malta (Sicily).449 Finally, there are also many examples of scenes, usually terra cotta figurines but also in other materials, wherein a person or divinity holds an identifiable slave mask. These have been found at Athens, Canusium, Capua, Cyprus, Delos, Fasano, Italy, Myrina, Paphos, Rome, Tanagra, Tarentum, Thasos, and Veii.450 There are also a couple of coins in this group, namely one from late second century Athens and one from 446 Bailley, 1975, Q102, Q273-4, Q300, Q316, Q321, Q323-4, Q326, Q328, Q330-1, Q612, & Q618; Webster, 1995, 1CV1, 1DV1, 1EV1, 2AV1-7, 2AV10-11, 2AV13, 2AV18-21, 2AV33, 2AV35-39, 2BV5- 7, 2CV1, 2CV4-8, 2DV6-7, 2DV13-14, 2EV1, 2EV3-4, 2HV1, 2NV3, 2NV8, 2 RV1, 2RV5, 2RV7-8, 2SV2, 2TV2-3, 2TV5-8, 2XV1, 2XV3-8, 3AV1-5, 3BV1, 3DL9-14, 3DL16-28, 3DL31-39, 3DL41-43, 3DV1-2, 3DV7-12, 3DV17-18, 3EL3-7, 3EV1-4, 3HL1-2, 3HV1, 3NV5, 3NV10, 3RV1-2, 3SV1-2, 3TV1-3, 3XV3, 3XV8-10,& 3XV12; Mlynarczyk, 1997, fig. 93.3 (Type J-Prime); Schaefer, 1968, E63, Q21, Q23, Q51. 447 Webster, 1995, 1TA1, 2DB1. 448 Webster, 1995, 1CB1, 1DA5-6, 2BA1, 3DB1, 3XB9-12, 3XB16-17. 449 Webster, 1995, 2DM1, 3DM4-6, 3DP1, 3NM2, 3RM1,& 3SM1. 450 Webster, 1995, 1BT17, 1BT19-25, 1BT27, 1DT29, 1KA1, 1TA2, 1TT23-24, 2DT33, 2KT4, 2NT1, 2RT16, 2RT37, 3DT107,& 3DT109-116; Stampoulidis, 1992, #57-58,, #60-63, #70-76,& #79. 221 Republican Rome.451 The satyr/Silenus mask motif also appears fairly frequently, though not with the same kind of popularity as New Comedy slave masks. They also appear earlier than the New Comedy slave masks, making appearances in different media during the 6th-4th centuries BCE. Furthermore, satyr masks are the one mask variant to make appearances in numismatics. In Persian period glyptics both Carthage in N. Africa and Selinus in Sicily have satyr masks on some of their sealings with ten and fifteen sealings respectively.452 In other contemporary media, the satyr mask appears on a couple of coins from Lesbos (W. Anatolia) and Catana (Sicily) from the 5th and 4th centuries.453 Satyr masks also appear fairly frequently in the form of terra cotta masks and heads. Persian period miniature terra cotta satyr masks in Phoenician/Punic contexts have been found at Tharros and Sulcis on Sardinia, as well as Carthage and Ibiza in N. Africa.454 Other places of the same period where masks were found include Italy, Sparta (S. Greece) and Samos (W. Anatolia).455 The motif also appears on a 5th century amphora handle stamp from Mende in N. Greece.456 In the Hellenistic period, the number of occurrences of the motif and the types of media in which it appears both increase. In glyptics, Seleucia in The East had 167 sealings with satyr masks (representing the use of 122 separate seals), while in S. Greece 451 Webster, 1995, 3AC1 & 3RC1. 452 Berges, Ehrhardt, Laidlaw & Rakob, 1997, #398-407; Salinas, 1883, CXXX-CXLV. 453 Head, Poole & Gardner, 1963, Catana 43; Wroth,1964b, Lesbos 76. 454 Ciasca, 1988, p.365; Cintas, 1946, pl. XIII; Piccard, 1965, I.17-19, II.5-6, III.5-7. 455 Culican, 1976b, pp. 71-72; Webster, 1967, IT11. 456 Webster, 1967, XA1. 222 Delos has an unspecified number and Callipolis has three sealings.457 In addition, the site of Pergamon in W. Anatolia has also produced a stamp for decorating pottery from the Hellenistic period that bears an image of a bearded satyr mask.458 In the Hellenistic numismatics, coins bearing satyr masks appear at Metapontum (Italy) and Dionysoupolis (C. Anatolia, 2nd-1st centuries BCE) during the Hellenistic Period.459 Hellenistic examples of miniature masks in terra cotta appear in Kharayeb and Marisa in The Levant; Alexandria and unspecified findspots in Egypt; Athens and Corinth in S. Greece; Smyrna and Troy in W. Anatolia; Abdera in N. Greece; Tarentum and unspecified findspots in Italy; and Butera, Centuripe, Lipari, and Morgantina in Sicily.460 The satyr mask motif also appears in other sculpture during the Hellenistic period. For instance, Athens in S. Greece has produced a sculpted marble with a papposilenos head as well as two marble satyr masks from friezes, all from the late 2nd century BCE.461 In W. Anatolia, Pergamon has a marble relief of satyr and paposilenos masks decorating the door of the theatre and a marble tripod decorated with a wreathed satyr masks and Miletus had a similarly decorated tripod, all dating to the 2nd century BCE, while the theatre at Letoon, dating to the early 1st century BCE, had satyr masks among the various carved metopes of masks.462 457 Boussac, 1997, p. 146; Invernizzi, Bollati, Messina & Mollo, 2004, M91-213; Pantos, 1985, #82, #206, #209. The satyric masks at Delos form part of a corpus of 1000 impressions that show masks of different types, of which the largest single group come from New Comedy. 458 Schaefer, 1968, Abb. 5.1. 459 Head, 1906, Dionysoupolis 2; Poole, 1963b, Metapontum 176;. 460 Erlich & Kloner, 2008, #183; Webster, 1967, AT9, AT35-36, AT39-41CT3, ET4, ET25, ST1, ST14-17, ST24-26, ST28-30,TT4, XT1-2, ZT3, ZT10, ZT13-14, 461 Webster, 1967, AS12-14. 462 Webster, 1967, ZS6-7,& ZS9. Webb dates the sculpture of the Pergamene theatre to the late 1st century BCE (Webb, 19996, p.68), while Webster puts it during the late 2nd century. 223 Like the New Comedy slave masks, satyr masks also appear in various forms on ceramic vessels, including lamps and braziers in the Hellenistic period. The motif appears on vessels especially braziers and lamps from areas and sites such as Ascalon, Ashdod, Bamboula, Caesarea, an unspecified findspot in Cyprus, Dor, Ptolemais-Ake, and Salamis in The Levant; Alexandria, Fayum, Naukratis and unspecified findspots in Egypt; Athens, and Delos in S. Greece; Calymna, Cnidus, Cos, Halicarnassus, Pergamon, Priene, Rhodes and unspecified findspots in W. Anatolia; Canusium in Italy; and Centuripe in Sicily.463 A bronze lamp with a lid bearing a wreathed satyr mask on the lid for the fill hole and dating to around 155 BCE has also been found at Ha-Qirya in Tel Aviv (The Levant).464 The motif also appears occasionally during the Hellenistic period in other items of precious metals, such as jewellery, at Bolsena and Tarentum in Italy, and Mahdia in N. Africa.465 The satyr mask motif also emerge in mosaics and paintings during the Hellenistic period, including Letoon, Miletus and Pergamon in W. Anatolia; Boscoreale, Imola, and Pompeii in Italy; and Malta in Sicily.466 Satyr masks are also occasionally carried in the hands of terra cotta figurines, such as with a fifth century example from Megara and a 2nd century one from Panticapaeum.467 463 Guz-Silberstein, 1995, krater type KR15, fig. 6.13; Rahmani, 1984, #1-9, #13-22; Rosenthal- Heginbottom, 1995, Braziers # 3-10, Lamp Type 13.1;Şahin, 2001, Ha8-40, Ha42-55, Cn1, Mt1-3, Ky5-7, Ky9-10, Rh1, Ka1-3, At2, Ce1, Ba1, Sa1, Eg1, Eg3-5, Na4-14, X1; Schaefer, 1968, E62, Q21, Q23, Q51; Webb, 1996, 68-69, 123; Webster, 1967, AV41-42, AV44-51, AV65EA1, DL1-2ET8, EV2-3, ZL3-7, ZL10, & ZV1. 464 Sussman, 2006, 39-50. 465 Webster, 1967, FB1, IB2-3, &TJ1. 466 Webster, 1967, IM1, JM2, NM3, NP2-3, NP27, ZS6-7 & ZS9; Webster, 1995, 3AS3. 467 Webster, 1967, AV25 & XT10. 224 Thus during the Hellenistic period, the satyr mask motif is fairly widespread, appearing in many different media and regions. It is most popular in W. Anatolia, but makes strong appearances in the Levantine ceramics and in a variety of media in Italy. The Phoenician mask motif appears less frequently and in fewer places overall, though it has the longest history of the types, with its Late Bronze Age predecessors. For the purposes of comparison here, I will only be looking at those examples whose dates fall at least partly from the sixth century onwards and which fall into the grotesque or male mask groups as the satyr masks have already been discussed and as female protomes fall far enough outside the appearance of the Kedesh examples as to be treated as separate motifs. Not surprisingly, only the Carthage (N. Africa) archive also has evidence of the Phoenician mask motif being used on seals in the form of two seals showing grinning masks from the fifth and fourth centuries.468 Otherwise the motif generally appears only as actual masks, usually in terra cotta, though a gold male funerary mask from the 5th-4th centuries has been uncovered from somewhere in Phoenicia.469 Male Phoenician masks from the 6th-4th centuries, bearded and not, have been found in terra cotta at Akhziv, Amathus, Dor, Kourion and Tell Keisan in The Levant; Sparta in S. Greece, Motya in Sicily; Bizerte, Carthage, Ibiza, Monte Sirai and Sulcis in N. Africa; as well as Cadiz in Europe.470 One mask from Utica and another from Carthage in N. Africa dates to the very beginning of the Hellenistic Period at the end of the fourth century.471 Of all of these masks, an example from Ibiza is extremely close to MAS8, while MAS16 looks very 468 Berges, Ehrhardt, Laidlaw & Rakob, 1997, #395-396. 469 Ciasca, 1988, p. 354. 470 Ciasca, 1988, pp. 355-6, 362, 364 & 366-7; Cintas, 1946, pl. 53; Cullican, 1976b, pp. 71-72; Paraire, 1980, I.122; Piccard, 1965, pp. 18-19, 30-36 & 39; Stern, 1995, pp. 447. 471 Piccard, 1965, pp. 18 & 29. 225 similar to a mask from Kourion.472 Meanwhile, terra cotta grotesque masks from the 6th- 4th centuries have been found at Kourion and Tell Keisan in The Levant; Sparta in S. Greece; Spina and Tarentum in Italy; Motya in Sicily; and Carthage, Ibiza, San Sperate and Tharros in N. Africa.473 And so, the Kedesh mask seals present an interesting picture. On one hand, the New Comedy slave masks that form the single largest sub-group at Kedesh are an incredibly widespread motif that was popular in a wide variety of media throughout the ancient world, especially in Greece proper and south-Italy. Similarly, the satyr masks show many of the same characteristics of distribution, though they have a longer distinct history and do appear alongside both Greek theatrical masks and Phoenician masks. It is the presence of these Phoenician masks, available in only two examples at Kedesh, which is perhaps the most interesting. On one hand, they represent the culmination of a tradition dating back centuries in the Levant and other regions where the Phoenicians colonized. On the other hand, these masks were, during the Hellenistic period, being supplanted by masks derived from Greek theatre. Kedesh, therefore, in its proportions of masks, is mirroring in its sealings what is going on at the same time in other media and at other sites in traditionally Phoenician regions. Catalogue: MAS1 Number of Sealings (Bullae): 2 (K99 0063, K99 0839) Maximum Preserved Dimension: 11 mm. 472 Ciasca, 1988, p. 366; Piccard, 1965, fig.40 & 48. 473 Ciasca, 1988, pp. 357-360; Cintas, 1946, pl. IX-XI; Cullican, 1976b, pp. 71-72; Paraire, 1980, I.121; Piccard, 1965, pp. 11-16, 31-33, 35-36 & 38-39. 226 Shape: convex oval seal Description: A frontal comic slave mask. The mask has a peaked speira or hair-roll, a pug nose, an arched brow and a large grinning mouth trumpet. This may be an example of a Leading Slave type. MAS2 Number of Sealings (Bullae): 2 (K99 0072, K99 0134) Maximum Preserved Dimension: 8 mm. Shape: convex oval seal Description: A grotesque frontal mask, probably of a comic slave. This mask has a very broad leering grin with teeth, a broad flat nose, and eyes that are misaligned. It also bears some sort of vegetal wreath, possibly grapes. It also appears to be bald, with a high domed head. MAS3 Number of Sealings (Bullae): 4 (K99 0089, K99 0101, K99 0250, K99 0503) Maximum Preserved Dimension: 12 mm. 227 Shape: flat seal Description: A mask facing slightly right. This mask has small eyes and smooth, expressionless features. It is surrounded by a solid hood of hair from the top of which rises many lines to an indefinable purpose. MAS4 Number of Sealings (Bullae): 2 (K99 0491, K00 0214) Maximum Preserved Dimension: 12 mm. Shape: slightly convex oval seal Description: A mask of a Pan facing 3/4 right. The Pan is youthful with a furrowed brow and two horns rising from the center of the forehead. The nose is wide and flat. The hair falls back from the horns. There are traces of an ear with hair behind it. MAS5 Number of Sealings (Bullae): 2 (K99 0629, K00 0324) Maximum Preserved Dimension: 10 mm. 228 Shape: convex seal Description: A frontal mask or mask-like face. The face has bulging eyes, round chin and cheeks with a straight a nose. From the top of the head are several projection resembling stocks of plants, a wreath, or the snakes of a gorgon. MAS6 Number of Sealings (Bullae): 2 (K99 0730, K99 1176) Maximum Preserved Dimension: 10 mm. Shape: convex oval seal Description: A mask facing 3/4 right. The impressions are both fairly poor, perhaps indicating wear on the seal itself. There are traces of the left eye, the gaping, grimacing mouth and the roll of hair. The overall shape of the face and of the hair makes this image probably a depiction of a pornoboscos from New Comedy, like one found at Delos. MAS7 Number of Sealings (Bullae): 1 (K99 0838) Maximum Preserved Dimension: 10 mm. 229 Shape: slightly convex oval seal Description: A comic slave mask facing 3/4 right. The nose is broad and flat. The grinning mouth trumpet has lines indicating the beard. The brow is scowling and the hair forms a speira that arches over the top of the head. MAS8 Number of Sealings (Bullae): 1 (K99 0855) Maximum Preserved Dimension: 12 mm. Shape: slightly convex oblong seal Description: A frontal bearded mask. This mask has a long straight nose over a beard indicated by vertical lines. The brow is heavy and the eyes are large dots/pellets. The top of the forehead ends in a straight horizontal line. Part of the left-side of the forehead and the mouth are damaged in the impression, but the mouth appears closed. The mask may represent either an example of the mask of an old man or a mask of a bearded Dionysus 230 type, as the mouth is closed. MAS9 Number of Sealings (Bullae): 3 (K99 0869, K00 0389, K00 0502) Maximum Preserved Dimension: 10 mm. Shape: slightly convex oval seal Description: A comic slave mask facing 3/4 right. The brow is heavy with a slight notch over the right eye. The eyes are two pellets immediately below the relatively straight brows. The nose is broad and flat over a wide, grinning mouth trumpet. The hair is in a speira that arches over the head with a peak over the brows. MAS10 Number of Sealings (Bullae): 1 (K99 0973) Maximum Preserved Dimension: 11 mm. Shape: convex, probably oval seal 231 Description: A bearded and wreathed mask of an old man facing front. The lower portion of face is missing from the impression, but the mustache is visible. There are traces of an ivy wreath on the head but no traces of hair. Considering the presence of the beard and the wreath, it may be that this seal actually shows a bearded Silenus instead. MAS11 Number of Sealings (Bullae): 1 (K99 1117) Maximum Preserved Dimension: 8 mm. Shape: convex oval seal Description: A satyr head in profile right. The satyr has high hair over the forehead and a shaggy beard. He has pointed ears and a snub nose. The top and back of the head are missing from the impression. MAS12 Number of Sealings (Bullae): 1 (K99 1128) 232 Maximum Preserved Dimension: 10 mm. Shape: flat seal Description: A Silenus mask facing lightly right. The Silenus has pointy ears, round cheeks, bulging eyes, a pug nose, thick lips, a furrowed brow and a well defined, shaggy beard. MAS13 Number of Sealings (Bullae): 1 (K99 1188) Maximum Preserved Dimension: 10 mm. Shape: slightly convex oval seal Description: A frontal comic slave mask. It is badly impressed and missing the left side. There are traces of the wide, grinning mouth, the small nose and the heavy brow with hair above, possibly in a speira. MAS14 Number of Sealings (Bullae): 1 (K00 0158) Maximum Preserved Dimension: 10 mm. 233 Shape: flat oval seal Description: A frontal Phoenician mask. Its features are smooth and without expression. The top of the mask is fairly wide and heavy in proportion to the rest. The eyes are very large and there are traces of nose and mouth. MAS15 Number of Sealings (Bullae): 2 (K00 0325, K00 0343) Maximum Preserved Dimension: 10 mm. Shape: flat or slightly convex oval seal Description: A bearded Pan mask facing 3/4 right. The Pan has a wide-open, bearded grimace, like the trumpet beard on a comic slave mask. He also has a small pug nose and lowering brow. There are two horns rising from the forehead to the right. The hair is full and runs down to the left, ending in two long strands. 234 MAS16 Number of Sealings (Bullae): 1 (K00 0430) Maximum Preserved Dimension: 11 mm. Shape: slightly convex round seal Description: A frontal mask or mask-like face. The nose, cheeks and chin are all rounded. The nose is also rather wide. The mouth is closed. The brows are high and arched. There are traces of hair in the form of thick locks that arch away from the forehead. The form of the nose, the brow and the hair may indicate that this image is meant to represent a satyr, though it is difficult to be certain. The right-hand side is missing from the impression. MAS17 Number of Sealings (Bullae): 1 (K00 0480) Maximum Preserved Dimension: 8 mm. Shape: flat round seal Description: A comic slave mask facing front. The right side has been damaged and is missing. The cheeks are full. The mouth has striated lines for a beard trumpet. The eyes seem closed and flank a snub nose. There is what may be a wreath on the outside of the 235 speira, which may be peaked. MAS18 Number of Sealings (Bullae): 1 (K00 0517) Maximum Preserved Dimension: 12 mm. Shape: flat round or oval seal Description: A Pan mask facing right. The mask has a prognathic face and lopsided mouth beneath large eyes. The hair is long and possibly wreathed. Two horns emerge from the upper head. MAS19 Number of Sealings (Bullae): 1 (K00 0560) Maximum Preserved Dimension: 8 mm. Shape: flat, probably oval seal Description: A comic slave mask facing lightly to the right. The mouth is semicircular with prominent cheeks in the corners. The brow is thin but prominent over two small 236 pellets for the eyes and a larger one for the nose. The hair forms a speira that arches over the whole and hangs down to the left. MAS20 Number of Sealings (Bullae): 1 (K00 0732) Maximum Preserved Dimension: 9 mm. Shape: flat oval seal, traces of the setting to the right Description: A comic slave mask facing 3/4 right. The mouth trumpet of this mask is very broad and striated vertically to show a beard. The cheeks are fat, while the nose is small and flat. The brow above is very heavy and lowers down to below the eyes, which are small pellets. The hair is in a thin speira and has a prominent roll or peak in the center above the forehead. 237 Chart 3.28: The Distribution of New Comedy Slave Masks Map 3.30: The Distribution of New Comedy Slave Masks                    238 Map 3.30a: Detail Map Showing the Aegean Region 239 Map 3.30b: Detail Map Showing Italy 240 Chart 3.29: The Distribution of Satyr Masks in the Persian Period Map 3.31: The Distribution of Satyr Masks in the Persian Period                  241 Chart 3.30: The Distribution of Satyr Masks in the Hellenistic Period Map 3.32: The Distribution of Satyr Masks in the Hellenistic Period                     242 Map 3.32a: Detail Map Showing the Aegean Region 243 Chart 3.31: The Distribution of Phoenician Masks in the Persian Period Map 3.33: The Distribution of Phoenician Masks in the Persian Period                   244 Chart 3.32: The Distribution of Phoenician Masks in the Hellenistic Period Map 3.34: The Distribution of Phoenician Masks in the Hellenistic Period               245 Section 13: Pilei of the Dioscuri Twenty-three sealings from Kedesh depict sets of the two rounded helmets, or pilei, which are often associated with the Dioscuri. These twenty-three sealings represent the use of fifteen separate seals. These numbers indicate that the motif of the pilei of the Dioscuri is one of the most popular amongst the Kedesh symbol sealings, second only to that of depictions of masks. The defining characteristic of this type is the presence of two rounded cones or horseshoe shapes, which represent the helmets or pilei, sitting side by side. They often rest upon a horizontal base, representing the brims. Due to the positioning of certain impressions, however, traces of only one pileus may be actually preserved (PIL 12 and 13). From there various details may be added, though none are present in all of the seals. One common element is the presence of stars, another symbol associated with the Dioscuri, hanging directly above each pileus. Six seals at Kedesh (PIL 1, 3, 5, 8-9 and 14) display at least traces of such stars, which generally seem to be six-pointed. In certain cases (PIL 2, 7, 9-13, and 15) the quality or placement of the seal in the impression does not indicate the presence of stars, but does not necessarily counter- indicate their possible presence on the original seal. In two cases (PIL 4 and 6), however, the pilei are topped by elements which resemble plumage or foliage rather than stars. In seven of the Kedesh seals (PIL 3-5, 8, 10, 12 and 14), the pilei have bands midway up across their bodies of the caps and/ or projecting spikes on the side which seem to represent the presence of encircling wreaths. Two Kedesh seals (PIL 9 and 13) also show pilei with a pair of ribbons or fillets hanging down from the brim of each cap. 246 These represent perhaps the straps meant to secure the helmet to the head. One seal (PIL 3), which is represented by two impressions, has especially noteworthy iconography. This seal is unique in the present corpus in having the presence of a coiled serpent lying between the two pilei. In both impressions of this seal, the tongue of the serpent can be seen, below which hangs what appears to be a flame-shaped beard. The serpent is resting on what appears to be a columnar altar with a domed top and a garland draped around its shaft. The motif of the two pilei is usually seen as symbols of the Dioscuri, since they are often shown wearing similar head gear. In many examples, the Dioscuri are also showed topped by stars which would fit well with their astral associations. Coins especially depict the Dioscuri in this manner, including a fourth century bronze issue from Tyndaris, some third century Seleucid issues from Nisibis and second century issues of the Seleucid Antiochus VI, a third century silver issue from Bruttium in Italy, a second century gold issue of the Bactrian king Eucratides. 474 In some of these cases, the pilei on the heads are also shown as wreathed as they sometimes are in the Kedesh seals. However, the association between the pilei and the Dioscuri is not the only possibility in every case, as the symbol may be co-opted for the images of other gods, who may or may not be associated with the Dioscuri. For example, at Uruk one of the sealings that show a pair of conical caps has a large horn protruding out of the side of one of them. Wallenfels takes this to be a representation not of the head gear of the Greek Dioscuri but of the crowns of the local gods Anu and Antum who wear similar horned caps.475 474 Newell, 1977, pl. VII.1-3, 6-13 & 19-22; Plant, 1979, #1127-1131. 475 Wallenfels, 1994, pp. 135-136. 247 The Kedesh seal PIL3 may be another example of such. Here, the caps flank a coiled and bearded serpent. The bearded snake often appears in the context of Roman domestic cult, namely in lararium scenes, as the Genius Loci, or spirit of a specific place.476 The snake in PIL3 also sits on a garland draped altar which can likewise appear in lararia scenes. The Genius Loci often appears at Pompeii accompanied by the representation of two Lares. These double gods are shown wreathed and sporting high belted short tunics.477 They are sometimes shown as dancing, with one leg crossed behind the other.478 The iconography of these gods does elsewhere borrow on the rare occasion from the iconography of the Dioscuri.479 Conversely, in another Italic context, that of Etruscan mirrors, the Dioscuri are often portrayed similarly to the Lares with high belted short tunics and a cross legged pose similar to the dancing pose seen with some Lares.480 As such, PIL3 seems to be incorporating outside elements into the established Dioscuri iconographic motif of the pilei to represent specifically an Italic idea, namely the images from Roman domestic cult shrine. Presumably, the image would represent the personal household gods of the seal user. Thus the motif of the pilei of the Dioscuri may be have been used to represent gods other than the Dioscuri themselves. 476 Boyce, 1942, pp. 13 & 17-18. 477 See Boyce, 1937, pl.22.1 (#468). This lararium painting from the kitchen of a house in Regio IX, shows a bearded genius loci coiled around a short columnar altar. The two lares stand on either side, wearing wreathes and short and high belted tunics. 478 See Boyce, 1937, pl.30.2 (#211). This lararium painting from the House of the Vetii shows two lares in short tunics in a cross legged dancing pose flanking a togate genius on one ground line, while a bearded genius loci snake approaches an altar on a separate ground-line below. 479 Waites, 1920, p. 251. Specifically, Waites refers to a republican coin that shows two gods depicted as warriors, and therefore Dioscuric, but labeled as Lares. 480 Emmanuel-Rebuffat, 1988, #40. This mirror’s engraving shows a scene with the two Dioscuri on the outside, flanking a nude and a clothed woman. The Dioscuri are clothed in high belted tunics and are sporting wreathed pilei. They are leaning back against the sides of the composition medallion with their legs crossed, supporting themselves on short ionic columns or altars. The mirror is dated stylistically to the end of the fourth-early third centuries BCE. 248 Distribution: The motif of the pilei of the Dioscuri is fairly widespread, but not as popular in the Hellenistic World as some others from the Kedesh symbol sealing corpus. The appearance of the motif seems to have been an almost entirely Hellenistic phenomenon. There are, however, occasional occurrences in the sixth-fourth centuries, which appear only in numismatics. For instance, there is a bronze issue from Colophon in W. Anatolia, which is dated 400-350 BCE which shows two laureate pilei surmounted by stars, while in C. Anatolia there is a bronze issue from Soloi (385-333 BCE) that shows a pair of pilei with straps surmounted by stars.481 There are also a couple of coins that show a pair of pilei flanking some sort of item, as they do in PIL3, including a bronze issue from Tenedos (450-387 BCE) in the Troad (W. Anatolia) that shows the pilei flanking a double axe and a later bronze issue from the island of Melos in S. Greece that is dated to 4th-1st centuries BCE and shows the pilei flanking a cornucopia.482 Otherwise, the occurrences of the pilei motif all date firmly within the Hellenistic period or later. Despite the small size of the sample it is possible to note that both the motif variant of the pilei by themselves and the variant of the pilei flanking an object seem to be limited geographically to the Aegean and Anatolia (S. Greece, W. Anatolia and C. Anatolia). The first variant appears only in Anatolia (W. Anatolia and C. Anatolia) while the second seems to be focused more on the Aegean, appearing in S. Greece and W. Anatolia. Still, the small sample size here precludes attaching any hard significance 481 Head, 1964, Colophon 40-41; Hill, 1900, Soli 42-44. 482 Wroth, 1963, Melos 22-25; Wroth, 1964c, Tenedos 25-26. 249 to this distinction. This is not the case in the Hellenistic period. Indeed the motif seems to have veritably exploded onto the scene in the Hellenistic Period, appearing in several different media and in many different places across the Ancient World. Within Hellenistic glyptics the pilei on their own appear in multiple archives within the Seleucid Empire. Examples appear on eighty-one seals with a total of eighty-eight combined impressions at Seleucia and three seals at Uruk, both in The East.483 The exact specifications of the pilei at these archives often differ in subtle details from those at Kedesh. At Seleucia, for instance, the pilei are generally depicted on horizontal lines which can be seen to be clubs in some cases and no example at Seleucia has indications of straps. In eight seals, however, the pilei are surmounted by stars.484 In a single other seal, the pilei are topped by cruciform crests, which are analogous to the leaves that top PIL 4 and PIL7. Of the three exempla at Uruk, on the other hand, one has a single star above and between the pilei, while another has a large horn sprouting from one of the hats. Wallenfels does not generally associate these images with the caps of the Dioscuri, but rather with the crowns of the local gods Anu and Antum.Outside the Seleucid areas of control, the pilei motif also makes an appearance at Paphos where they are shown flanking a coiled snake sporting a pshent headdress, which caused it to be identified as a uraeus or possibly the agathodaemon (see figure 3.7).485 This sealing would be the closest direct parallel to PIL3, though there are 483 Invernizzi, Bollati, Messina & Mollo, 2004, Og 102-178, p. 187; McDowell, 1935, III.D.4.c1-5; Wallenfels, 1994, #1030-1032. 484 Invernizzi, Bollati, Messina & Mollo, 2004, Og 118, Og 124, Og 138-140, Og 144-146. 485 Nicolaou, 1978, #11. 250 a couple of minor differences as noted above. The use of the pilei motif is more widespread in the medium of numismatics, where it appears in on the issues of several Hellenistic polities. The use of the motif shows as much variety between the different political issues as there is amongst the Kedesh sealings. Among the bronze issues of Antiochus I from the mint at Tarsus in C. Anatolia, the pilei are shown as simple conical caps, lacking stars or other associated symbols.486 In most cases, however, the pilei are shown surmounted by stars, as in the case of the Kedesh sealings. The plume/palm crests on the seals (PIL 4 and 7) can also be compared to the appearance of pilei on coins of various denominations minted by the Greco-Bactrian king Eucratides (170-145 BCE) (The East), where they are also accompanied by palm branches.487 The same motif of the pilei accompanied by palms was continued in the same region by the Greco-Indian kings Antialcidas (115-95 BCE) and Archebius (90-80 BCE) at Taxila.488 The motif of the pilei, lacking a central symbol 486 ANS Online Catalogue, 1944.100.78037-8, 1973.50.2; Newell, 1977, #1299-1303. 487 ANS Online Catalogue, 1911.105.980-1, 1924.176.33, 1944.100.74426-30, 1944.100.74440-7, 1944.100.74449, 1944.100.78308, 1957.172.2001, 1973.56.1339-40, 1979.45.42-5, 1979.45.57-8, 1981.159.2, 1993.29.25-6, 1993.30.7, 1995.51.303, 1995.51.104-5, 1995.51.107-9, 1995.51.172-3, 1995.51.269, 1995.51.292, 1995.51.297-300, 1995.51.304-8, 1995.51.310-12, 1995.51.321; Cunningham, 1884, Eukratides 5-6 & 12-13; Plant, 1979, #1904. 488 ANS Online Catalogue, 1911.105.998-9, 1944.100.74731-45, 1944.100.74826, 1973.56.1383-6, 1973.56.1390, 1979.45.134-40, 1979.45.166, 1993.29.85-7. 251 that they flank, also appear on coin issues from Antioch (under Antiochus VII and Antiochus X) and Aradus in The Levant; Attica, Sparta and Syros (3rd-1st centuries BCE) in S. Greece; Apamea (190 BCE), Philadelphia (post 133 BCE) and Tabae (168- 27 BCE) in W. Anatolia; Dioscurias (119-63 BCE) and Panticapaeum in The Black Sea; at Tyndaris and Catane (both 275-212 BCE) in Sicily, and at Ecbatana (under Phraates II, 138-127 BCE) in The East.489 In many other cases, the pilei flank a central symbolic motif on the coins, as they do in (PIL3), though never a bearded snake. For instance, Bronze coins from the mint at Apamea (The Levant) from immediately after the death of Antiochus II (246-244 BCE) show plain caps flanking an inverse Seleucid anchor.490 Similarly, coins from Cypriote Salamis and Paphos (205-180 BCE) show the pilei flanking an eagle; the coins of Berytus (109-31 BCE) show the pilei flanking a dolphin wrapped around a trident, some from Tyre (205-180 BCE) show the pilei flanking a Ptolemaic eagle; while some coins of Tripolis (81-80 BCE) show the pilei flanking a rudder.491 The coins of the Ptolemaic queen Berenice II (248-221 BCE) from Alexandria (Egypt) also show a reverse of pilei flanking a cornucopia.492 Other issues with the motif of pilei flanking a central symbol appear at Melos and Sparta (with amphora, 250-146 BCE) in S. Greece; Adramytium (with cornucopia, 200-150 BCE) and Phocaea (with griffin’s head, 300-100 BCE) in W. Anatolia; Hephaestia (with torch, 489 ANS Online Catalogue, 1912.60.37, 1944.100.48364-7, 1944.100.70712, 1944.100.76703-4, 1944.100.76706, 1944.100.7617-20, 1948.19.2404-5, 1973.50.2, 1984.66.116, 1992.54.2390, 0000.999.20764; Chapouthier, 1935, #83; Gardner, 1963a, Laconia 6-13; Head, 1878, #549; Head, 1901, Philadeplia 19-23; Head, 1906, Apamea 32; Head & Gardner, 1963, Panticapaeum 41, Hephaestia 7-9; Head, Gardner & Poole, 1963, Catana 54-56, Tyndaris 16-17; Plant, 1979, #1905-06, #1909; Wroth, 1963, Syros 14-17. 490 Newell, 1977, #1145. 491 ANS Online Catalogue, 1944.100.70133-5, 1944.100.72527-8; 1944.100.78686-9, 1961.154.222, 1992.54.697-717; Hill, 1910, Berytus 11-13, 16 & 23-25. 492 ANS Online Catalogue, 1944.100.76256, 1967.152.626; Chapouthier, 1935, #77. 252 289-190 BCE) in N. Greece; Gorgippa and the Pontic cities of Amissus, Amasia and Sinope (with cornucopia), these last being minted specifically during the reign of Mithradates Eupator (119-63 BCE), in The Black Sea; Cappadocia (with eagle on sheathed sword, mid first century BCE), Pessinus (with lion, end of 2nd century BCE), Synnada (with poppy-head, post133 BCE) and Tavium (with amphora, 1st century BCE) in C. Anatolia; as well as Catane (with owl, 3rd-2nd centuries) in Sicily.493 The nature of the central symbol can vary greatly ranging from thunderbolts to palms to cornucopias. The pilei also make appearances in other media during the Hellenistic period. In S. Greece, for instance, they appear at Delos in paintings, a lead medallion, and relief and sculpture in the round all from the Hellenistic period.494 A sculpture from the third century sanctuary of Artemidoros of Perge on Thera shows the pilei surmounted by stars amidst other divine symbols.495 In regards to the motif of the pilei flanking a central object, there is an example of a Hellenistic lead tessera from Athens where the pilei flank an Isiac symbol.496 In W. Anatolia, a possibly Hellenistic marble relief from Pergamum shows a single laureate pileus surmounted by a star.497 Other sculpture consists of two votive reliefs from Balboura, dating from 200 BCE onwards, show the two pilei by themselves.498 The pilei also make an appearance on third century BCE Rhodian 493 ANS Online Catalogue, 1944.100.41855-7, 1944.100.42407-8, 1944.100.46768, 1956.78.7, 1965.210.1, 1970.142.297, 1971.193.68, 2, 2004.14.43; Chapouthier, 1935, #76, 79-82, 84-85, 87; Head, 1906, Synnada5-6; Plant, 1979, #1908, #1940-1942, #2390; Wroth, 1963, Melos 22-25; Wroth, 1964a, Deisturus I 1, Tavium 1-2. 494 LIMC, Dioskouroi, #157-8, 238, 247-248. 495 LIMC, Dioskouroi, #239. 496 Chapouthier, 1935, #93. 497 LIMC, Dioskouroi, #240. 498 Smith, 1997, D4, D43. 253 amphora handle stamps associated with the name Zenon, which have been found at Delos, Paphos and Cypriote Salamis.499 In regards to the iconography of PIL3, the lararium paintings of Pompeii (Italy) also are an important point of comparison for the overall scene structure, with twin Lares often flanking a central garlanded altar and/or genius loci, if not the specific description (see figure 3.8). The geographic distribution of the occurrences of the pilei motif in the Hellenistic period is of some interest. The pilei by themselves tend to cluster in the area of the Aegean (S. Greece and W. Anatolia) with the greatest number of occurrences appearing 499 LIMC, Dioskouroi, #244. 254 there in several different media. The East is also well represented with occurrences of the motif. The region has the same total number of occurrences as W. Anatolia, but lacks the variety of media. Instead, the occurrences are limited to either glyptics, with Seleucia and Uruk, or Numismatics, with issues from Ecbatana as well as Greco Bactrian and Indo-Greek issues. Outside these areas, the motif seems limited entirely to coins. Sicily, The Black Sea and The Levant each have two numismatic occurrences of the motif in the period. Central Anatolia is still present during this period, as it was in the preceding, but only has one occurrence which is the same number that it had previously. And so the motif seems to have really grown in popularity in both Greece and western Anatolia during the Hellenistic period as well as spread out into different regions, especially in The East, while retaining essentially the same presence in southern Anatolia that it had previously. Regarding the variant of the motif where the pilei flank a central item, the image is very different. Here, the grand majority of the occurrences are numismatic in origin. The single largest grouping of occurrences also comes from The Levant, where the only glyptic example, and the closest parallel to PIL3, occurs. The regions of C. Anatolia and The Black Sea are tied for the next largest amount of occurrences with four each.500 The regions of S. Greece and W. Anatolia are still present with three and two occurrences respectively, but not to anywhere near the same extent that they are with motif of the pilei by themselves. Otherwise, the pilei flanking an item seem to appear sporadically with the occasional occurrence in Sicily, Egypt and N. Greece. The picture that 500 See Chapouthier, 1935, p. 99. Chapouthier noticed a similar distribution for his iconographic study which examined all occurrences of the Dioscuri, not just their pilei, flanking a female character, goddess, or feminine symbol. Where the distributions differ was largely in the greater presence of Greece proper, Macedonia, and Egypt within his study. 255 therefore emerges from the distribution of the motif indicates a couple of further possible connotations to consider. First, the motif in general tends to cluster in two ways. In media, it is especially prevalent in the sphere of numismatics. Geographically it also tends to cluster in Anatolia and The Levant. Second, the specific version of the pilei flanking a snake does not quite fit into this paradigm. It is a purely glyptic phenomenon. Also, though the only other direct parallel also comes from The Levant, the Kedesh example of PIL3 shows elements that demonstrate Italian influence or ideas, specifically in regards to the depiction of the snake itself. Such influence is very interesting considering the dearth of examples of the motif variant within Italy itself. The PIL3 seal might therefore represent the adoption of a glyptic sub-variant of a motif that was especially prevalent in Anatolia and The Levant and its subtle modification to show Italian religious figures by someone who had ties both to this area and the Italian peninsula, such as an Italian trader active in the Eastern Mediterranean. Catalogue: PIL1 Number of Sealings (Bullae): 1 (K99 0017) Maximum Preserved Dimension: 10 mm. Shape: convex oval seal Description: Two pilei of the Dioscuri sitting side-by-side. Both are on wide bases with 256 dots on the apices, which sit atop conical bodies. They are each surmounted by a star. PIL2 Number of Sealings (Bullae): 1 (K99 0019) Maximum Preserved Dimension: 9 mm. Shape: flat oval seal Description: Two pilei of the Dioscuri sitting side-by-side. The two pilei have short, rounded bodies and sit on wide bases. The right sided pileus has traces of a wreath on one side. PIL3 Number of Sealings (Bullae): 2 (K99 0110, K99 1067) Maximum Preserved Dimension: 12 mm. Shape: large, slightly convex seal 257 Description: Two pilei flanking a snake. The two pilei have thick, two stage brims and traces of wreaths on the sides. They are surmounted by stars which rise from the apices of each short, conical body on a stalk. They sit flanking a coiled snake with a flame- shaped beard and protruding tongue. The snake in turn sits on the domed top of a columnar altar which is draped midway down with a garland. PIL4 Number of Sealings (Bullae): 4 (K99 0252, K99 0357, K99 0406, K99 0867) Maximum Preserved Dimension: 11 mm. Shape: fairly large, flat seal Description: Two pilei of the Dioscuri. The tips of the pilei are side by side, each surmounted by a spray of leaves. The bodies are conical and squat. There may be traces of a wreath along the side of one of the pilei. In all of the impressions, most of the original seal does not seem to have been depicted. PIL5 Number of Sealings (Bullae): 2 (K99 0415, K99 0999) 258 Maximum Preserved Dimension: 10 mm. Shape: convex seal Description: Two pilei of the Dioscuri. The two pilei have tall, ovoid bodies and are side by side on thick bases and are touching. They are surmounted by stars that rise from the apices of the pilei on short stalks. Each has a horizontal line across the centre. On the outer sides of each of the two are two short parallel lines that might represent wreaths. There also may be traces of a horizontal line below the both of them. PIL6 Number of Sealings (Bullae): 1 (K99 0577) Maximum Preserved Dimension: 10 mm. Shape: slightly convex oval seal Description: Two pilei of the Dioscuri. The impression is fairly indistinct. The bodies of the two pilei are high ovals sitting side-by-side. Each one is topped by a trace of a spray of feathers or leaves. One of the ovals appears to have a trace of a notch on one side. 259 There are traces of a possible two stepped base or brim below the two pilei. PIL7 Number of Sealings (Bullae): 1 (K99 0621) Maximum Preserved Dimension: 11 mm. Shape: flat oval seal Description: Two pilei of the Dioscuri. The two pilei sit side by side, though one lies mostly off the impression, which has also been damaged by a crack running from top left to bottom right. The body of the pileus in the center of the impression appears to be conical and has traces of straps hanging down on either side of the crack. PIL8 Number of Sealings (Bullae): 5 (K99 0842, K99 0843, K99 0956, K99 1000 K991220) Maximum Preserved Dimension: 11 mm. Shape: convex, probably oval seal Description: Two pilei of the Dioscuri. The two pilei sit side-by-side with their brims touching. Each has two long straps hanging down below the brim. They are both 260 wreathed, represented by two horizontal lines which travel across and surpass the tall bodies on the outer-sides. They are also both surmounted by six-rayed stars. One of the lower rays on the right-handed star bears a break or fault characteristic in all the impressions. PIL9 Number of Sealings (Bullae): 1 (K99 0867) Maximum Preserved Dimension: 13 mm. Shape: slightly convex seal Description: Two pilei of the Dioscuri. Each pileus has a brim and straps and is surmounted by a star. The right pileus and the stars are largely effaced from the impression. 261 PIL10 Number of Sealings (Bullae): 1 (K99 1093) Maximum Preserved Dimension: 11 mm. Shape: convex, probably oval seal Description: Two pilei of the Dioscuri. The two pilei sit side-by-side and have thick, two part bases or brims. The tops of the pilei are outside the impression. PIL11 Number of Sealings (Bullae): 1 (K99 1094) Maximum Preserved Dimension: 8 mm. Shape: slightly convex seal Description: Two pilei of the Dioscuri. The two pilei sit side-by-side and have two part bases or brims consisting of two parallel horizontal lines. The tops of the pilei are outside the impression. 262 PIL12 Number of Sealings (Bullae): 1 (K99 1170) Maximum Preserved Dimension: 9 mm. Shape: slightly convex seal Description: A pileus of the Dioscuri. Only an incomplete single pileus is preserved within the impression, which is unclear. It has a horseshoe–shaped body and a brim on a two step base and there are traces of a wreath in the form of horizontal lines further up the body. PIL13 Number of Sealings (Bullae): 1 (K99 1189) Maximum Preserved Dimension: 6 mm. Shape: flat oval seal Description: A pileus of the Dioscuri. Only the lower portion of a single pileus is present within the impression. The pileus has a thick base or brim and two hanging straps. 263 PIL14 Number of Sealings (Bullae): 1 (K00 0567) Maximum Preserved Dimension: 10 mm. Shape: flat seal Description: Two pilei of the Dioscuri. The two pilei have tall, conical bodies and sit side-by-side upon conjoined double rims. They have wreaths in the form of sprays of lines from the sides. Five-rayed stars surmount each pileus, with the left-hand star being somewhat off-kilter. The right hand pileus is partly outside the impression. PIL15 Number of Sealings (Bullae): 1 (K00 0712) Maximum Preserved Dimension: 7 mm. Shape: slightly convex seal Description: Two pilei of the Dioscuri. The two pilei sit side by side upon thick, connected triple bases or brims. The upper portions of both are missing as is the right side of the right-hand pileus. 264 Chart 3.33: The Distribution of Pilei in the Persian Period              265 Map 3.35: The Distribution of Pilei in the Persian Period Chart 3.34: The Distribution of Pilei Flanking Item in the Persian Period              266 Map 3.36: The Distribution of Pilei Flanking Item in the Persian Period Chart 3.35: The Distribution of Pilei in the Hellenistic Period                    267 Map 3.37: The Distribution of Pilei in the Hellenistic Period 268 Map 3.37a: Detail Map Showing the Aegean Region 269 Chart 3.36: The Distribution of Pilei Flanking an Item in the Hellenistic Period Map 3.38: The Distribution of Pilei Flanking an Item in the Hellenistic Period                   270 Map 3.38a: Detail Map Showing the Aegean Region 271 Section 14: Shields Four of the Kedesh sealings show images of shields. These four sealings represent the use of two distinct seals. In both cases, the shields are oval with a convex body and a flat, offset rim that runs around the entire circumference (as much as the circumference is preserved). The first example (SHI1) exists on only one sealing. It is ornate with two dolphins being depicted on the body of the shield and with the shield rim decorated with a hatched or cable pattern. The second seal (SHI2) has three examples. It is less ornate with a plain body and rim. It is, however, placed at the centre of a stylized vegetal wreath that encircles it. The excavators identified it originally as a cult stone or baetyl. This identification is certainly possible, though the presence of the distinct linear border of the central convex oval would argue for it being rather a shield with rim. Both shields appear to be variants of the Galatian shield or thyreos. This type of shield was oval with a defined rim and a central boss, which is absent in the two Kedesh examples. The type came into widespread use after the Celtic invasions of Greece and Asia Minor in the 3rd century BCE, and was probably also Celtic in origin itself.501 The shield type appears as part of the total panoply of a set of Hellenistic soldiers depicted on a series of three painted funerary stelae that were found at Sidon.502 In at least one of the cases, the owner of the stela was apparently from Lydia. It would thus seem that the thyreos was a common piece of equipment for mercenaries working in Seleucid or Ptolemaic controlled Levant.503 As a motif by itself, the shield appears fairly early. Certainly by the early fifth 501 Perdrizet, 1904, p. 240. 502 Perdrizet, 1904, #1-3. 503 Perdrizet, 1904, p. 240. 272 century, examples of the Boeotian shield began appearing as types on Boeotian coins.504 By the end of the century, other areas in Greece were minting coins with their own distinctive shields. The use of the motif of shields continued to grow in the fourth century, spreading into architecture in the form of sculpted reliefs. The trend continued into the Hellenistic period, brought along in part by the spread of Macedon, as is illustrated by the spread of the depiction of the Macedonian type shield, which was round and rimless and usually decorated with crescents, in the field of numismatics, including Seleucid coins.505 The shield as a motif can have very specific connotations. As a weapon, a shield is conceptually linked with the sphere of war and its related ideas, such as victory or martial prowess. Certainly actual shields were dedicated to celebrate victories, such as the Athenians dedicating golden shields at Delphi to commemorate the Battle of Marathon.506 Furthermore, as a defensive weapon, the shield also connotes the ideas of defence and protection. For example, one interpretation of the appearance of round shields on Hellenistic rock-cut reliefs in the region of Oinoanda is that they represent the protection to the cemetery afforded by the presence of a local hero.507 Furthermore, as I have noted in the introduction, the shield with its blazon paralleled the seal in that both were used to denote identity in different contexts, one in combat the other in legal documents. The appearance of a shield as an image on a seal could also therefore be a kind of play on this relationship. The use of the shield as an image could function either as an affirmation of the user’s identity, especially if it possessed a distinctive blazon, by combining expressions of identity together. On the 504 ANS Online Catalogue, 1999.4.10, 2006.20.1, 2006.20.5. 505 Newell, 1977, #796, #880-1, #930, #942, #946-8 506 Herodotus, V, 58-9. 507 Milner, 2004, p. 58. 273 other hand, the shield motif, especially if blank, might problematize the question of the user’s identity by placing another marker of identity between the individual and the act (that of sealing a document). Interestingly, of the two shields in the Kedesh sealings one has a distinct blazon while the other is blank. Distribution: In general, the shield motif appears frequently during the Persian and Hellenistic periods in several media. These media include glyptics, numismatics, sculpture in both stone and terracotta, as well as others. The particulars of the two Kedesh examples are, however, not nearly as popular. Neither the appearance of the two dolphins, nor the use of a wreath seems to appear with a shield when it is used as a symbol in itself. As such, it would appear that the Kedesh examples are unusual variants within a widespread iconographic category. In regards to the glyptics of the 6th-4th century BCE, the shield as a motif appears at a couple of archives. At Carthage (N. Africa), for instance there are three sealings that show round shields, two of which date to 5th-4th centuries BCE, while the third dates to the 4th-3rd.508 For its part, Selinus (Sicily) has one seal, present in multiple impressions, that shows a round shield bearing a gorgoneion.509 Shields in general also appear frequently as motifs on coins both in the Persian and Hellenistic period. In this medium, the occurrences can often be linked to political or ethnic groupings. For example, during the sixth through fourth centuries, those places 508 Berges, Ehrhardt, Laidlaw & Rakob, 1997, #811-3. 509 Salinas, 1883, CCCXLVI. 274 under the sway of Thebes minted coins using the Boeotian shield as a type. These places included Acraephium (475-450 BCE), Boeotia (federal coinage, 5th-4th centuries BCE), Coroneia (456-374 BCE), Halliartus (550-480 BCE), Mycalessus (387-374 BCE), Orchomenos (387-374 BCE), Plataea (387-372 BCE), Pharae, Tanagra (456-374 BCE), Thebes (later 5th-4th ccenturies BCE), and Thespiae (4th century BCE) all in S. Greece.510 At the same time, other places issued coins with different styles of shields, generally round, including Elis (with eagle attacking a serpent, 471-365 BCE) and Attic Salamis (the shield of Ajax, 4th century BCE)) in S. Greece; Maroneia (with eagle attacking a serpent, 435-410 BCE), and the Molossi (with thunderbolt, before 340 BCE) in N. Greece; and Tarsus (a figure-8 shield, 361-333 BCE) in C. Anatolia.511 In sculpture, the shield also occurs as a motif in various places, usually in relief. For instance, the Theban victory monument at Leuctra (371 BCE, S. Greece) has a series of round hoplite shields around its domed roof.512 Round shields also appear along with other armour, weapons and items on the walls of the mid-fourth century Tomb of the Reliefs at Caere (Italy).513 During the Hellenistic period, the shield motif continues to appear in seals, coins and sculpture as it had in the previous period. In Hellenistic glyptics, for instance, the shield motif appears at several archives. At Seleucia (The East), four sealings each from a separate seal show images of shields, three of which are oval and decorated with some 510 ANS Online Catalogue, 0000.999.53411-4, 1944.100.20096, 1944.100.20125, 1986.78.104-8, 1986.78.110-2, 1999.4.10, 2006.20.1-5, 2006.20.8-9; Head, 1963a, Boeotia Uncertain 17-27 & 30-37, Acraephium 4, Halliartus 11, Orchomenus 22-24, Pharae 3-5, Thebes 18-28, 43-6, 64-80, 91-100, 107-68; Kroll & Walker, #592; Plant, 1979, #1868, #2140-1, #2207-16, #2379, #2416, #2707, #2725. 511 ANS Online Catalogue, 1941.153.770, 2002.18.12; Gardner, 1963a, Elis 31-7; Gardner, 1963b, Molossi 3-5; Hill, 1900,Tarsus 79; Kroll & Walker, 1993, #640-2; Plant, 1979, #1505, #2217-8. 512 Markle, 1999, p. 241, fig.40. 513 Spivey, 1997, fig.81. 275 sort of emblem.514 Callipolis (S. Greece) also has 12 sealings that show shields, including both oval, Galatian shields and round shields that are decorated with emblems such as stars, a thunderbolt, a gorgoneion, an inscription and a snake.515 In one case, the shield is accompanied by a quiver or sword sheath. Carthage (N. Africa) for its part also has the sealing from the 4th-3rd centuries that bears a round shield next to a sword.516 In coins, the exact nature of the shields popularly depicted changes in the Hellenistic period from the Boeotian shield, popular in Boeotia under Theban influence, to the more ornate Macedonian, which Alexader began to make popular and which several of his successors used. This type of shield appears on the coins of various mints. This practice continued under several of Alexander’s successors at places that included Alexandria ad Issum (under Antiochus IV), Antioch (under Seleucus I and Antiochus I), Apamea (with tripod on boss under Antiochus I) and Cypriote Salamis (under Philip III and Demetrius Poliorcetes) in The Levant; Miletus (under Philip III) and Lydian Philadelphia (2nd century BCE) in W. Anatolia; Amphipolis (under Alexander and Demetrius Poliorcetes), Macedonia ( with various elements depicted under Philip III, Demetrius Poliorcetes, Antigonus Gonatas, Pyrrhus of Epirus, Demetrius II, and others into the 2nd century BCE), Pella (under Pyrrhus) in N. Greece; Cilicia (under Alexander) in C. Anatolia; as well as Carrhae (with gorgoneion under Antiochus I) and Dura- Europus (with anchor boss under Antiochus I) in The East.517 Other places also minted 514 Invernizzi, Bollati, Messina & Mollo, 2004, Og275-278. 515 Pantos, 1985, #99, #101-7, #109-111a. 516 Berges, Ehrhardt, Laidlaw & Rakob, 1997, #812. 517 ANS Online Catalogue, 0000.999.7050, 0000.999.7053-5, 1905.57.99, 1921.100.9, 1941.131.477, 1941.131.637, 1944.100.401, 1944.100.13657-8, 1944.100.13706-9, 1944.100.13743-5, 1944.100.13749- 53, 1944.100.13795-7, 1944.100.13799-806, 1944.100.13831-2, 1944.13834, 1944.100.14151-6, 1944.100.14182-4, 1944.100.14216-22, 1944.100.14231-2, 1944.100.14249-51, 1944.100.14253-68, 276 coins displaying other types of shields during the Hellenistic period. These places included the Boeotian League (a Boeotian Shield), Chalcis (with eagle and serpent, 196- 146 BCE), Polyrhenium (with a bull’s head, 330-280 BCE), Thebes (a Boeotian Shield 338-300 BCE), and Thespiae (338-315 BCE) in S. Greece; Pergamon (under Eumenes II) in W. Anatolia; Amissus (with Gorgoneion under Mithridates Eupator) and Cabeira (with Gorgoneion under Mithridates Eupator) near the Black Sea; and Selge (a blank round shield, 300-190 BCE) in C. Anatolia.518 The shield motif also continued to appear in relief sculpture during the Hellenistic period. In Macedonia (N. Greece), sculpted round shields of varying types appear frequently, mostly on tombs, at such places as Pella, Vergina, Spelia, Katerine and Dion, dating from the late-fourth through second centuries.519 In the region of W. Anatolia, the Lion Tomb at Cnidus from the late-fourth-early third centuries had a single shield in relief that was placed above the door to the tomb.520 In the same region, at the Ta Marmara heroon from the second century, the facade possessed shields in relief between the columns.521 At Pergamum, shields, including oval thyreoi, appeared in the weapons reliefs that formed the balustrade of the stoa in the Sanctuary of Athena, dating to the first half of the second century, and a single round shield appeared on the pedimental 1944.100.33773-7, 1944.100.34129-44, 1944.100.75028-38, 1948.19.454-9, 1951.64.3, 1951.116.42, 1952.75.14, 1952.142.25, 1953.30.16, 1953.171.1675-6, 1954.185.5, 1956.28.28, 1956.174.3-4, 1957.152.41, 1957.172.174, 1957.172.715, 1964.42.6, 1965.275.7-11, 1966.75.40, 1966.75.42-4, , 1967.152.201, 1974.148.1, 1977.158.832, 1978.52.183-4, 1980.109.125, 1987.152.15, 1987.152.31-3, 1987.152.35-6, 1987.142.37-40, 1987.152.42, 1992.17.1, 1999.4.20; Gardner, 1963b, Pyrrhus 36-9; Hill, 1900, Alexandria as Issum 1; Head, 1901, Philadelphia 1-4; Kroll & Walker, 1993, #497-8, #506; Mørkholm, 1991, #430, #439, #585, #603-4; Newell, 1977, #796, #880-1, #930, #942, #946-8; Plant, 1979, #130a, #1087, #1921-2, #2201-6. 518 ANS Online Catalogue, 0000.999.53411-3, 2006.20.3-4; Kroll & Walker, 1993, #597; Mørkholm, 1991, #615; Plant, 1979, #2199, #2283, #2563-4. 519 Markle, 1999, pp. 219-220 & 239. 520 Webb, 1996, p. 121. 521 Webb, 1996, p. 103. 277 tympanum of the late second century gymnasium or Asclepius Temple.522 The second century bouleuterion of Miletus also has a number of round shields in relief that were situated between the engaged columns.523 Other examples of shields used as motifs in sculpture in the same region include Hellenistic rock-cut reliefs in the area of Oinoanda.524 Sculptures of round shields also decorated the late second century bouleuteria at Saglassos and Ariassos in C. Anatolia.525 At Delos in S. Greece, decorated round shields (imagines clipeatae) appear on the pediment of the Temple of Isis from the second half of the second century and on the pediment of the Monument of Mithridates VI Eupator from the very end of the century.526 An early first century tomb at Suweida (ancient Soada) in The Levant had examples of various armaments spaced between engaged columns, including both round and oval shields.527 The motif also makes appearances in other media during the Hellenistic period. For example, the Tomb of Lyson and Callicles at Leucada, dating around 200 BCE, was painted with a Macedonian shield along with other arms.528 Round shields also appear on lead tokens from the Athenian Agora, dating to the Hellenistic period.529 In two of these cases, dating to the third century, the circular shield is accompanied by a wreath on its face. Small terracotta votive shields have also been found at Corinth and Athens.530 The Athenian example in particular, is very close to the Kedesh examples. It is a thyreos with 522 Webb, 1996, pp. 58 & 67. 523 Webb, 1996, p. 102. 524 Milner, 2004, #15 & #19. 525 Webb, 1996, pp. 130 & 132. 526 Webb, 1996, pp. 128-9 & 141-2. 527 Lawrence, 1957, p. 221, pl.110b; Markle, 1999, p. 238. 528 Markle, 1999, p. 239. 529 Lang & Crosby, 1964, L26-7, L222-3. 530 Burr Thompson, 1966, #9; Davidson, 1942, #48-62. 278 an oval body and defined rim, and a thunderbolt decoration. And so, the shield motif as a whole seems to be equally at home in glyptics as numismatics and sculpture, and even appears in other media as well. The use of the motif does tend to cluster in areas such as Macedonia, where the Macedonian shield was a favourite numismatic type, as well as in western and southern Anatolia. Round shields do appear to be represented more frequently overall in the various media, though the oval thyreos, as in the Kedesh examples, was a standard piece of military equipment for actual soldiers in the region. These shields appear in a much more limited fashion and only in glyptic or sculpture. Only three regions (The Levant, S. Greece and The East) have occurrences of oval shields by themselves as symbols. Of these regions, S. Greece has the most occurrences but not enough to justify calling it a cluster. All of the occurrences also date to the Hellenistic period only. This makes sense in light of the date of the introduction of the thyreos into the warfare of the Aegean and the Eastern Mediterranean at large. The Kedesh examples both also possess features, namely the lack of a boss or the presence of dolphins or a wreath, which distinguish them from both representations of thyreoi and of shields in general. Thus, it is that the Kedesh shield sealings both seem to display singular examples within the larger constellation of the use of the motif. Catalogue: SHI1 Number of Sealings (Bullae): 1(K99 0339) Maximum Preserved Dimension: 9 mm. Shape: ‘flat’ oval seal (the large boss of the shield would make most of the seal convex) Description: An oval shield decorate with dolphins. The shield has a convex oval surface with a raised cable border, representing the shield rim. The field of the shield is decorated 279 with two serpentine dolphins meeting beak to beak at bottom, and spreading diagonally outwards and upwards in mirrored ‘S’-curves. A portion of lower right of the seal is missing. SHI2 Number of Sealings (Bullae): 3 (K99 0902, K99 1105, K99 1230) Maximum Preserved Dimension: 10 mm. Shape: flat seal Description: An oval shield within a vegetal wreath. The body of the shield is convex and oval. Around the body there is a raised linear border, representing the rim of the shield. Encircling the whole is a simplified or stylized vegetal wreath created by a series of 'V's of stylized leaves. 280 Map 3.39: The Distribution of Oval Shields in the Persian Period: 281 Map 3.39a: Detail Map Showing the Aegean Region 282 Chart 3.37: The Distribution of Oval Shields in the Hellenistic Period Map 3.40: The Distribution of Oval Shields in the Hellenistic Period               283 Map 3.40a: Detail Map Showing the Aegean Region 284 Section 15: Stars Two of the Kedesh sealings, each representing the use of a distinct seal, show an image of a star by itself. In each case the star is depicted rather simply with rays created by the incision of separate straight lines. One of the stars (STA1) has nine rays while the other (STA2) has eight. In both cases, one of the rays is somewhat longer than the others and stretches in to form the center of the star. One of the seals also has an oval raised border surrounding the star. The star motif has a very long history of depictions. As a filler motif in larger scenes, it appears on early Mesopotamian cylinder seals, such as an Early Dynastic example in shell from Ur.531 It continued in the same vein for a long time, as a Neo Assyrian cylinder from Nineveh attests.532 In both of these cases, the star depicted has eight rays, as in STA2. Stars continued to serve as parts of larger images through the Hellenistic period, appearing for example with the pilei of the Dioscuri on sealings from Kedesh itself. The idea of the star as a symbol goes back at least to the ninth century when an eight-rayed star appeared on a Moabite stamp seal with a crescent moon.533 The star also functioned as a symbol early on in the Greek world, since the motif appeared as the devices on shields carried by characters on Athenian black-figure in the sixth century.534 The long history of the star motif with its multiple associations provides a plethora of possible connotations to its use as a symbol, especially at Kedesh. One 531 Collon, 1987, fig. 83. 532 Collon, 1987, fig.739. 533 Bordreuil, 1986, #60. 534 Moore, Philippides & von Bothmer, 1986, #34 & #228. The first example is a one piece amphora, while the second is a Panathenaic amphora. Both examples date to 570-560 BCE. 285 possibility is that the star was meant to be a symbol of the Phoenician goddess Astarte, since the star of Venus was one of her attributes.535 On the other hand, the star motif might be referring to other possible deities, such as the Dioscuri, since stars are often present in their depictions, including cases with their pilei in the Kedesh sealings. Admittedly, the presence of only a single star in each case here would make the identification with the Dioscuri problematic, but a sealing from Uruk with a single star over two caps would indicate that it is still a possibility.536 Interestingly, both the Dioscuri and Astarte were associated with sea navigation and the protection of seafarers. The use of the star as a motif on the sealings from Kedesh, might, therefore, also bear connotations of navigation and seamanship. Again, the motif might represent a solar divinity such as Helios, since both stars and the rayed heads of Helios appear on Hellenistic amphora handle stamps from Rhodes, which was a center of the god’s worship.537 More generically, stars might not refer to specific deities, but rather to the idea of divinity in general. Certainly, stars were symbols associated with multiple different divinities. In addition, they could themselves be thought of as divine. Plato, for instance, equates the stars with eternal and visible divine forces associated with the keeping of Time.538 It may be, therefore, that the star as a symbol might not necessarily refer to any specific god in particular, but rather to the gamut of heavenly supernatural forces that could exist. And so, the star motif could have several possible associations that span the range from the very specific to the more generalized. 535 Stager, 2005, p. 440. In addition, seals of the Iron IIc period in Israel show the Assyrian cognate goddess Ishtar associated with astral signs, including an eight-rayed star identified with Venus. See Keel & Uehlinger, 1998, pp. 292-4, figs 286-8. 536 Wallenfels, 1990, #1066. 537 Grace, 1934, #57 & #94. 538 Plato, Timaeus, 38-39. 286 Distribution: The distribution of the star motif is heavily affected by the dual nature mentioned above. While instances of the star shown as a part of a larger image, either as a filler or as an associated portion of the whole, might abound, actual usages of a star by itself as it appears on the Kedesh sealings is much rarer. In addition, when stars do appear as symbols by themselves, they are often different in depiction from the Kedesh exempla, having different numbers of rays or being more ornate. In the period of the 6th-4th centuries BCE, coins dominate the occurrences of the star motif as a symbol. Mints from this period that issued coins with stars as a type include Itanus (with 8 rays, 5th-4th centuries BCE), Knossos (with eight rays, 500-220 BCE), Locris Opuntia (16 rays, 387-338 BCE), Melos (4th-1st centuries BCE) and Orchomenos (387-374 BCE) in S. Greece; Caria (ornamental with 4 rays, under Hecatomnus 395-377 BCE), Colone (with 8 rays, 400-310 BCE), Gambrium (with 12 rays, 4th-3rd centuries BCE), Miletus (various configurations including 8-rays, 7th-4th centuries), Phocaea (with 8 rays, before 545 BCE), Thymbrai ( with 8 rays, 4th century) in W. Anatolia; Corcyra (with different configurations including 8-ray, 500-300 BCE) in N. Greece; and Syracuse (with 16 rays, 412-345 BCE) and an uncertain mint (with 16 rays, 345-212 BCE) in Sicily.539 In addition, Argos in S. Greece and Pitane in W. Anatolia have variants on the motif, with Argos (421 BCE) having a five sided stellate pattern and 539 ANS Online Catalogue, 1987.152.43-6; Head, 1963b, Locris Opuntii 2-6, 35-36 & 45-49; Head, 1964, Miletus 10 & 12-50, Phocaea 80-81; Head, Gardner & Poole, 1963, Syracuse 241-242, Uncertain Mint 2- 3; Plant, 1979, #539, #2145, #2698, #2700-1; #2703-5x, #2707; Wroth, 1963, Cnossus 22-23 & 32-33, Itanus 1 & 19-23, Melos 9; Wroth, 1964b, Colone 1-6, Thymbra 1-4. 287 Pitane (4th century BCE) a pentagram.540 In other media from the same time period, the single star as a symbol is very rare. A single star appears on a double spouted, folded lamp from Kerkouane in N. Africa, dating to the 4th-3rd centuries BCE.541 This star has eight simple rays done as straight lines, much as the Kedesh sealings. In the same region, a tomb from Tharros has produced a square ivory plaque from the sixth-fourth centuries that has an eye of Horus on one side and a star of sixteen rays on the other.542 And so, in the 6th-4th centuries BCE, the motif of the single star as a symbol by itself occurs predominantly in the medium of coins and in the areas of S. Greece and W. Anatolia, which together have more occurrences than the rest of the Ancient World combined. The region of N. Africa is interesting in that it has produced two examples of the motif, neither of which is a coin. In the Hellenistic period, the single star motif used as a symbol begins to make appearances in a greater variety of media, though numismatics remains prominent. For instance, single stars make occasional appearances in Hellenistic glyptics. There is a single sealing from Seleucia in The East, which shows a star by itself with eight rays spaced around a central dot and another sealing from Callipolis in S. Greece that shows an 8-rayed star in a linear frame.543 Besides these examples, there is also a garnet in the hermitage Museum that shows an 8-rayed star similar to the Kedesh sealings, especially STA1, but with pointed rays.544 Artaxata (C. Anatolia), on the other hand, has two 540 Plant, 1979, #1924 & #2709. 541 Acqaro, 1988, #368. 542 Barnett & Mendleson, 1987, 4.27. 543 McDowell, 1935, III.D.I.c.1; Pantos, 1985, #112. 544 Plantzos, 1999, #656. 288 sealings with related motifs, one with various cosmological symbols and a second with four 8-ray stars integrated into a cruciform pattern.545 In Hellenistic numismatics, several of the cities from the previous period that minted issues with stars as types continued to do so, while new mints also appeared. These mints include Judaea (under Alexander Janneus) in The Levant; Coressia (with 8 rays, after 300 BCE), Julis (with 8 rays, 3rd century BCE), Knossos (with 8 rays, 350-220 BCE), Melos (4th-1st centuries BCE) in S. Greece; Gambrium (with 12 rays, 4th-3rd centuries BCE) and Pergamon (with 8 rays, 197-159 BCE) in W. Anatolia; Corcyra (with eight rays, 300-229 BCE) Ouranopolis (with 8 rays, under Cassander) in N. Greece; Pontus (120-63 BCE) on the Black Sea; Iguvium (with 16 rays, Hellenistic Period) and Luceria (with 6 or 8 rays, Hellenistic Period) in Italy; an uncertain mint (with 16 rays, 345-212 BCE) in Sicily; and Cyrene (312-277 BCE) in N. Africa.546 Certain mints of the period produced variants of the motif, which usually consisted of other astrological symbols, such as a crescent moon, accompanying the star. These places include Cydonia (with crescent, 200-67 BCE) in S. Greece; Pergamon (two stars, 300-284 BCE) in W. Anatolia; Byzantium (with crescent, 1st century BCE) and Ouranopolis (with crescent, after 300 BCE) in N. Greece; Tomi (with two 5-rayed stars, pre-Roman period) along the Black Sea; and Venusia (three stars and three crescents, 268-217 BCE) in Italy.547 In other media, the star motif occasionally appears as a symbol on Hellenistic 545 Kachatrian, 1996, figs. 40 & 42. 546 ANS Online Catalogue, 1944.100.41455-8, 1944.100.41460-6, 1944.100.41467-71, 1944.100.41481, 1944.100.79491-2, 1945.21.13, 1949.19.1, 1953.171.426-35, 1970.142.254; Head, 1963b, Locris Opuntii 2-6, 35-36 & 45-49; Wroth, 1963, Cnossus 22-23 & 32-33, Itanus 1, 19-23, Coressia 57-59, Melos 9; Poole, 1963b, Iguvium 7, 10-13 & 40-42; Gardner, 1963b, Corcyra 45-54, 72-82, 130-143 & 198-200; Mørkholm, 2001, #75; Plant, 1979, #2702, #2706; Reifenberg, 1947, #14-15. 547 Head & Gardner, 1963, Tomi 3; Plant, 1979, #1666, #2696, #2710, #2714, #2718; Wroth, 1963, Cydonia 23. 289 lamps. Besides the previously mentioned Kerkouane lamp, other cases include examples from Alexandria (Egypt) and Delos (S. Greece), where eight-ray stars were engraved on the bases.548 The motif of single 8-ray stars also appears occasionally in Hellenistic amphora stamps from Cnidus and Rhodes in W. Anatolia.549 Finally, a star does appear as a symbol on a Hellenistic lead token from the Athenian Agora (S. Greece).550 A second token from the agora has a star associated with a crescent moon.551 And so, the distribution of the star motif in the Hellenistic period behaves in a mildly different fashion from the previous period. The regions of S. Greece and W. Anatolia still dominate with the highest number of occurrences of the star as a symbol, but isolated instances of the motifs are much more widespread, occurring almost everywhere. The motif also makes appearances in more media outside of numismatics than it did previously. Catalogue: STA1 Number of Sealings (Bullae): 1 (K99 0059) Maximum Preserved Dimension: 5 mm. Shape: small, slightly convex round seal Description: A Star. The star is done very simply with nine linear rays radiating outwards. One of the rays is longer than the others and stretches inwards to form the center of the 548 Bruneau, 1967, #4284; Mlynarczyk, 1997, fig.93. The Alexandrian lamp dates to the second-1st c. BCE. 549 Grace, 1934, #57 & #195. The Cnidian example probably dates to the first century BCE, while the Rhodian dates to 220-180 BCE. 550 Lang & Crosby, 1964, L20. 551 Lang & Crosby, 1964, L178. 290 star. STA2 Number of Sealings (Bullae): 1 (K99 0846) Maximum Preserved Dimension: 9 mm. Shape: flat oval seal Description: A star. The star has eight simple, linear rays spreading outwards. One of the rays reaches inwards to from the centre. Around the star is a raised border that is oval in form. The star itself is elongated slightly to fit the oval shape of the seal. 291 Chart 3.38: The Distribution of Single Stars in the Persian Period Map 3.41: The Distribution of Single Stars in the Persian Period                  292 Map 3.41a: Detail Map Showing the Aegean Region 293 Chart 3.39: The Distribution of Single Stars in the Hellenistic Period Map 3.42: The Distribution of Single Stars in the Hellenistic Period                    294 Map 3.42a: Detail Map Showing the Aegean Region 295 Section 16: Thunderbolts Among the Kedesh Symbols Sealings, there are a total of eight whose impressions bear the image of a thunderbolt. Each impression appears to represent a separate and distinct seal. In each case, the thunderbolt consists of a central area, often depicted as a node or differentiated body, with tines or branches fanning out both above and below. The exact number of branches on either side of the central area varies by seal, but in the majority of them there is more prominent central branch which forms the main lengthwise body of the thunderbolt. The other branches emanate from the central node and flank this central branch. Five out of the eight (THU1-2, 4, 6 & 8) are also depicted with wings sprouting from the sides of the central body. In a sixth example (THU7) there are round nodules which run out from the side of the central body and which may be wings, but it is not certain. In another example (THU3), the thunderbolt is flanked on either side by uncertain objects, resembling either chains or branches of a garland which runs along the length of the thunderbolt. The size of THU2 is abnormally large at 14 mm. According to McDowell this might indicate that it was the seal of some sort of semi-public institution like a temple.552 The idea of THU2 representing the seal of a temple is certainly a conceivable possibility considering the link between the thunderbolt and Zeus. However, the execution of THU2 is somewhat summary, especially in the wings, which would seem odd if the seal was that of an institution like a temple that would conceivably have more resources at its disposakl to purchase a seal than an individual. As such THU2 might just be an abnormally large seal used by an individual. 552 McDowell, 1935, p. 26. 296 The thunderbolt as a visual object is widespread in both time and place. It starts early in the form of the attribute of Zeus. A possible very early example of this relationship exists in the form of a lidded pithos from Knossos, dated c. 700 BCE, which shows a bearded male figure with the thunderbolt.553 A somewhat more explicit example comes from an early sixth century pyxis by the C painter that shows a scene of the birth of Athena, wherein Zeus is depicted as seated with a thunderbolt in one hand.554 Here, the identification of the figure as Zeus is strengthened by its inclusion in a known myth. Another early and explicit example of the association of Zeus and the Thunderbolt comes from the scene of the Wedding of Peleus and Thetis on the Francois Vase, dated 575 BCE, by Kleitias.555 Here, the figure of Zeus carries a thunderbolt and bears an inscription identifying him by name. This association between Zeus and the thunderbolt would continue in art for a very long time, as indicated by a white chalcedony from Tunisia, dated to the second century CE that shows the enthroned figure of Zeus holding a thunderbolt and sceptre and fronted by an eagle.556 The use of the thunderbolt by itself as a symbol, presumably in reference to Zeus, is also long lived, and the motif appears early. Certainly, by the fifth and fourth centuries BCE, the motif appears in symbol form as shield emblems in scenes on Attic Black- and Red-Figure pottery.557 Distribution: As a symbol, the thunderbolt in its two variants (winged and plain) appears in 553 LIMC. “Zeus.” #12-13 & #16. 554 Beazley, J. D. The Development of Attic Black-Figure. Berkeley, 1986, pl.20.2. 555 Beazley, 1986, pl.24.3. 556 LIMC. “Zeus”, p. 322; Spier, 1992, #263. 557 Moore, Philippides & von Bothmer, 1986, #270, #976, #979-982, #985-988, #1037, #1110, #1117 & #1148; Moore, 1997, #765 & #1362. 297 many different places. Plain thunderbolts appear in the Persian period on one sealing from Carthage (N. Africa) and another from Selinus (Sicily).558 Another sealing from Selinus also shows a plain thunderbolt flanked by the club of Herakles.559 In the same period, the plain thunderbolt also makes several appearances on coins from different cities. Mints that issued coins dating from the fifth through fourth centuries BCE and bearing the plain thunderbolt motif as a type include Elis (6th-4th centuries BCE), Ithaca (4th century BCE), Mycalessus (387-374 BCE), Pisa (c.365 BCE) and Praesus (pre-400 BCE) in S. Greece; Assus (400-241 BCE) in W. Anatolia; Amphipolis (under Philip II) , Epirus (under Alexander I in 498-454 BCE and again in 342-326 BCE), Macedonia (under Philip II and Alexander III), Thracian Maroneia (435-410 BCE), Molossis, and Pella (under Philip II) in N. Greece; Croton (400-390 BCE) and Tarentum (under the Epirote king Alexander I, 498-454 BCE) in Italy; Agyrium (345-275 BCE), Catane (415-402 BCE), Halaesa (c. 340 BCE), and Syracuse (367-344 BCE) in Sicily; and Cyrene (375-308 BCE) in N. Africa.560 These mints cluster in the regions of S. Greece, N. Greece (especially in Epirus and Macedonia) and in Sicily. From there the numismatic use of the motif seems to have spread out in the next period into areas of Italy and the Near East, in this last case following the Macedonian conquerors. In the Hellenistic period, the plain thunderbolt motif appears on sixteen sealings, 558 Berges, Ehrhardt, Laidlaw & Rakob, 1997, #821; Salinas, 1883, CCCXLVIII. 559 Salinas, 1883, CCCXLIX. 560 ANS online catalogue, 1941.153.770, 1944.100.12071, 1944.100.12076, 1944.100.12503-7, 1944.100.18805-6, 1944.100.18808-9, 1944.100.37752, 1944.100.37755, 1945.30.1, 1947.97.97, 1964.42.10-11, 1964.42.16, 1967.152.186, 1967.152.242, 1977.158.98-9, 1984.83.22, 1991.96.1, 1992.32.58, 1992.32.68-9, 1995.20.1, 1997.9.89, 1998.113.2, 2002.18.12, 2002.46.526, 2008.2.17; Gardner, 1963a, Elis 1-50, 54-70 & 79-80, Pisa 1; Gardner, 1963b, Alexander I of Epirus 1-8, Molossi 3-4; Head, 1963b, Mycalessus1-3; Head, Gardner & Poole, 1963, Agrigentum 130, Agyrium 9, Alaesa 4-5, Catana 8, 45-48 & 50-52, Syracuse 313-319, 405-408 & 413-415; Mørkholm, 1991, #586; Plant, 1979, #1623, #2218, #2335, #2338, #2340, #2343, #2349, #2351, #2366, #2377, #2379; Wroth, 1963, Praesus 16. 298 representing the use of 12 distinct seals, from Seleucia in The East, and one sealing from both Callipolis in S. Greece and Gitane in N. Greece.561 These last two examples actually are impressions of the official seal of the Epirote League, identified by its large size and the presence of an ethnic inscription. Here the thunderbolt is surrounded by an oak wreath, similar to the garlands flanking THU 3’s thunderbolt.562 By the Hellenistic Period, numerous polities all over the Ancient world, including both many of the older examples as well as newer ones, were minting coins displaying plain thunderbolts as types. These mints include Antioch (under Antiochus I), Phoenicia and Seleucia Pieria (2nd century BCE) in The Levant; an unspecified mint in Egypt; Cythera (250-146 BCE), Elis (271-191 BCE), Ithaca (4th century BCE), and Sparta (pre- 250 BCE) in S. Greece; Aphrodisias (1st century BCE), Assus (400-241 BCE), Bithynia (under Prusias I), Lesbos (330-280 BCE), Philadelphia (2nd century, before 133 BCE), Plarasa (1st century BCE), and Poemanon (1st century BCE) in W. Anatolia; Amantia (230-168 BCE), Epirote Apollonia (post 100 BCE), Epirus (3rd-2nd centuries BCE), Macedonia, Oricon (230-168 BCE), and Phoinice (238-168 BCE) in N. Greece; Caelia, Capua, Consentia, Croton, Hyria (217-89 BCE), Locri, Luceria, Pandosia (238-168 BCE), Rome, Rubi, Tarentum, Thurii and an unidentified central Italian mint in Italy; Agrigentum (279-241 BCE) Catane (275-212 BCE), Centuripe (post 241 BCE), and Syracuse (4th and 3rd century issues) in Sicily; Bactria and Parthia in The East; Cyrene and Cyrenaica in N. Africa; and the Spanish city of Carteia (2nd-1st century BCE) in 561 Berges, Ehrhardt, Laidlaw & Rakob, 1997, #821-823; Invernizzi, Bollati, Messina & Mollo, 2004, Og 32-51; McDowell 1935, p. 114; Salinas, 1883, CCCXLVII-CCCXLIX; Pantos, 1985, #97; Preka- Alexandris, 1989, fig.12. 562 Pantos, 1985, #97. 299 Europe.563 Besides the use of the thunderbolt motif by itself, there are also numismatic examples that combine the thunderbolt with another motif. The most common of these is where the thunderbolt is under the claws of a bird of prey. The Ptolemies, for example, favoured the use of an eagle standing on a thunderbolt for their coin reverse, and minted them at Ake-Ptolemais, Ascalon, Berytus, Byblos, Cyprus, Damascus, Gaza, Joppa, Judaea, Kition, Paphos, Salamis, Samaria, Sidon, Tripolis and Tyre in The Levant; Alexandria and an unspecified mint in Egypt; Tarsus in C. Anatolia, and Cyrene and Cyrenaica in N. Africa.564 Similar eagle-on-thunderbolt motifs also appeared on an issue from the city of Aradus in The Levant; on a Hellenistic Spartan coin in S. Greece; on a late fourth century coin from Amphipolis and another from Larissa and on Epirote League coins after 232 BCE in N. Greece; on issues of the Pontic cities of Amastris, Dia, Gazioura, Pharnacia and Sinope under Mithradates Eupator in the Black Sea; as well as on a third century stater from Tarentum in Italy.565 Athens (S. Greece), for its part, begins minting small coins in the third century BCE that have one or more owls perched 563 ANS online catalogue, 0000.99.49322, 1940.77.171, 1940.88.1123, 1944.100.73122, 1944.100.74963- 76, 1944.100.79534-5, 1944.100.75039-44, 1944.100.79474, 1944.135.137, 1947.97.582, 1948.19.2320, 1952.112.193, 1953.171.1677, 1954.263.140, 1955.190.40, 1961.154.308, 1962.57.151, 1964.215.1, 1969.83.447, 1967.153.13, 1974.95.219, 1981.40.28, 1982.174.2, 1991.78.57, 1992.32.54, 1992.54.1253- 5, 1999.4.6, 2004.28.1; Gardner, 1963a, Cythera 10, Elis 125-128, 134-136, 139-140 & 148, Ithaka 7-9, Laconia 4-5, Pisa 1; Gardner, 1963b, Amantia 1-2, Apollonia 83, Oricus 5, Epirus 1-7 & 42-52, Pandosia 1, Phoenice 1-2, Pyrrhus, 20-23 & 40-47; Head, 1901, Apollonia 3-4, Philadelphia; Head, 1906, Abbaïtis 1-7; Head, 1963b, Prusias II 1; Head, Gardner & Poole, Catana 53, Centuripa 3-6, Syracuse 405-408, 413- 415, 428-429, 636-644 & 654-656; Hill, 1906, Olba 1; Mørkholm, 1991, p.180, #196, #216, #329, #417, #506, #524, #531, #586, #588; Plant, 1979, #1620, #2336, #2339, #2344-5, #2347-9, #2352, #2354, #2361, #2366-8, #2376; Poole, 1963a, Cyrene 4-9; Poole, 1963b, Asculum 9-13, Capua 17, Caelia 12-13, Consentia 1, Croton 110-112, Locri 2-5, 18-19 & 24-29, Luceria 36-39, Rubi 2-3, Tarentum 30-32, Thurium 146; Wroth, 1964a, Antioch 1, 5-8, 10, Seleucia Pieria 5-6, 15, 24; Wroth, 1964b, Lesbos (Methymna?) 1-9. 564 ANS online catalogue, 1944.100.75847, 1967.152.623, 1990.23.59, 1990.26.1, 1992.50.1, 1994.157.1, 1994.157.8, 1994.157.10-1, 1994.157.17, 1994.157.19, 1994.157.21-3, 1994.157.25, 1997.9.157, 1997.45.1-2, 1997.45.81997.157.12-3; Head, 1963b, Amastris 4-6, Dia 1-3, Gaziura1, Pharnacia 2, Sinope 37-41; Meshorer & Qedar, 1999, IC5; Mørkholm, 1991, #97-98, #130, #202, #291, #296, #301-306, #310, #312-313, #317, #319-320, #329. 565 ANS online catalogue, 1991.96.1, 1997.9.159; Mørkholm, 1991, #485, #506, #523. 300 on a thunderbolt.566 The bird on thunderbolt motif also appears on one of the sealings from Callipolis in S. Greece.567 Besides coins, the plain thunderbolt is used as a motif in several other media during the Hellenistic period. For instance, it appears as the reverse side of double sided Hellenistic clay tokens and on Hellenistic lamps from the Athenian Agora in S. Greece.568 Likewise the motif appears as decoration on Hellenistic braziers from Halicarnassus and Calymna in W. Anatolia that are now in the British Museum.569 In the same region, it also appears on a Cnidian amphora handle stamp from the mid second century BCE.570 Elsewhere, a plain thunderbolt appears as a sculpted decoration on the leg of a kline from an Alexandrian tomb in Egypt that dates to the first half of the third century.571 The plain thunderbolt was also used on sling bullets as attested by a third century BCE example from Anthemounda in N. Greece.572 Altogether, the original clusters of the motif in N. Greece and Sicily continue. New areas of concentration, namely Italy and W. Anatolia, have also appeared. The region of W. Anatolia now has the same number of total appearances of the motif as the old stronghold of N. Greece. The region of Italy, for its part, has even eclipsed all the other regions, having the most occurrences of the motif, all in the form of coins. Within this larger universe, the occurrence of winged thunderbolts is a separate subset. The winged thunderbolt motif is less popular and appears in fewer places overall 566 Kroll & Walker, 1993, #67, #81, #99, #100. 567 Pantos, 1985,#51. 568 Howland, 1958, #587, #600, #603, #606-09; Lang & Crosby, 1964, L74 &L75. 569 Şahin, 2001, Ha 1 and Ky 2. 570 Lenger, 1957, #91. 571 Venit, 2002, figs.50-1. The kline is located in the Mustafa Pasha Tomb 3. It shows a three-tined thunderbolt with a simple bar-like central nodule and no wings. 572 Bates, 1930, pp. 44-46. 301 than its wingless brethren. In many cases, especially in coins, a city that possesses examples of the winged version generally, though not always, has examples of the wingless. In the sixth-fourth centuries, the winged thunderbolt appears on a single example from Selinus in Sicily.573 In the same time period, winged thunderbolts appear on coins from Elis in S. Greece and Catane and Syracuse in Sicily all of which minted examples without wings.574 As such, the only real region with multiple occurrences is Sicily, though the motif does make an appearance in S. Greece. In the Hellenistic period, the winged thunderbolt motif does spread out a bit and appear more frequently. This growth is still more modest that that of its wingless cousin in the same period. Among the Hellenistic archives, both Edfu in Egypt and Seleucia in The East have examples of winged thunderbolts with a single sealing and 22 sealings, representing the use of eight separate seals respectively.575 In coins, the winged thunderbolt appears during the Hellenistic period at Antioch, Phoenicia and Seleucia Pieria (2nd century BCE) in The Levant; Axus (300-67 BCE), Elis (post 191 BCE) and Sparta (266-207 BCE) in S. Greece; Colossae (2nd-1st centuries BCE), Myndus (2nd c. BCE), Parium (2nd-1st centuries BCE) and Termessus Minor (1st centuries BCE) in W. Anatolia; Epirus and Macedonia (185-168 BCE) in N. Greece; Abbaitis Mysoi (190-133 BCE) and Selge (2nd-1st centuries BCE) in C. Anatolia; Locri, Rubi, Thurii (post 300 BCE), and Vibo Valentia (192-89 BCE) in Italy; Agrigentum, Catane (275-212 BCE), Centuripe (275-212 BCE), and Syracuse (late 4th-3rd centuries 573 Salinas, 1883, CCCXLVII. 574 Gardner, 1963a, Elis 19-30, 36-42, 44, 54 & 70; Head, Gardner, & Poole, 1963, Catana 8, 45-47 & 50- 52. 575 Invernizzi, Bollati, Messina & Mollo, 2004, Og 44-50; McDowell, 1935, p. 114; Milne, 1916, #14. 302 BCE) in Sicily; and Cyrene (under Ptolemy I, early 3rd century BCE) in N. Africa.576 In other media, a winged thunderbolt also appears as a decoration on a sandal strap from an acrolithic sculpted foot in Aï Khanoum in Bactria (The East).577 Similarly, Marisa in The Levant has produced an askos in the form of a sandal-wearing human foot of which the sandal’s buckles has the form of a winged thunderbolt.578 The winged thunderbolt also makes appearances on Hellenistic braziers found at Ptolemais-Ake and Dor in The Levant.579 The winged thunderbolt motif also appears at Dor in the form of lead sling bullets from a siege dated to 139/8 BCE.580 Altogether, the winged thunderbolt motif appears most often in the region of The Levant, where it has seven occurrences in three different categories of media. Otherwise, the use of winged thunderbolts, as opposed to wingless ones, appears most frequently in the regions of Sicily, Italy, and W. Anatolia with four occurrences each, all of them in coins. As such, the appearance of the thunderbolt motif as a symbol at Kedesh appears to be in keeping with a widespread and popular motif that has spread out its areas of concentration in northern Greece and Sicily into many fertile new areas during the Hellenistic period. The presence of thunderbolts at Kedesh is in keeping with the general glyptic universe, as it appears at several of the ancient archives, as well as the use of the motif in Hellenistic art in general. The predominance of winged thunderbolts at Kedesh 576 ANS online catalogue, 1944.100.74966-9, 1947.97.582, 1948.19.2320, 1961.154.308, 1962.57.151, 1981.40.28; Gardner, 1963a, Elis 148, Laconia 4-7; Gardner, 1963b, Epirus 49; Head, 1901, Apollonia 3- 4; Head, Gardner & Poole, 1963, Acragas 130, Catana 53, Centuripa 3-6, Syracuse 413-415, 428-429, 636- 644 & 654-656; Hill, 1900, Olba 1; Mørkholm, 1991, #329, #506, #531, #586; Plan, 1979, #2333-4, #2346, #2350, #2355-7, #2360, #2362-3, #2378; Poole, 1963a, Cyrene 4-9; Poole, 1963b, Locri 19, 24-29, Rubi 2-3, Thurium 146; Wroth, 1963, Axus 12-16; Wroth, 1964a, Seleucia Pieria 1-4. 577 Bernard, 1969, fig.16. 578 Erlich & Kloner, 2008, #215. 579 Rahmani, 1984, #26; Rosenthal-Heginbottom, 1995, Braziers #12. 580 Stern, 2000, fig.142. 303 certainly fits with the general popularity of the motif in The Levant, though the motif overall appears more sporadically than the plain thunderbolt. Catalogue: THU1 Number of Sealings (Bullae): 1 (K99 0016) Maximum Preserved Dimension: 10 mm. Shape: slightly convex, probably oval seal Description: A winged thunderbolt with a large central node, short wings and three tines on each side. THU2 Number of Sealings (Bullae): 1 (K99 0092) Maximum Preserved Dimension: 14 mm. Shape: large, flat oval seal Description: A winged thunderbolt with four tines flanking a thick central shaft and a raised border. The wings are summarily done. There is a linear border around the outer 304 edge. THU3 Number of Sealings (Bullae): 1 (K99 0107) Maximum Preserved Dimension: 11 mm. Seal: probably oval flat seal Description: A winged thunderbolt with five tines above and below the main body. The side tines branch out from individual nodes near the centre. Traces of wings flank the central node on each side. The thunderbolt as a whole is flanked on each side by uncertain objects. These objects are probably a garland of foliage that originally surrounded the thunderbolt. THU4 Number of Sealings (Bullae): 1 (K99 0845) 305 Maximum Preserved Dimension: 11 mm. Shape: flat oval seal Description: A winged thunderbolt with seven tines on either side of a small central node. The central tine is thicker than the lateral ones, and the outermost seem to be made of dots. The execution is very linear and summary. The thunderbolt was too big to be impressed properly on the bulla and portions along the edge are missing. THU5 Number of Sealings (Bullae): 1 (K99 1047) Maximum Preserved Dimension: 11 mm. Shape: slightly convex, probably oval seal Description: A thunderbolt? The seal is poorly impressed upon the bulla. Three possible tines clustered tightly around a central node. THU6 Number of Sealings (Bullae): 1 (K00 0215) 306 Maximum Preserved Dimension: 10 mm. Shape: flat oval seal Description: A winged thunderbolt with two tines on either side of thickened central shaft with a two piece central node. There is raised border along edge. THU7 Number of Sealings (Bullae): 1 (K00 0313) Maximum Preserved Dimension: 9 mm. Shape: small, flat and elongated oval seal Description: A thunderbolt with five tines on either side of a central node. The central tine or shaft is thicker than the ones flanking it. The central node is flanked by two round side nodes from which an additional tine on each side emerges flanking the central set. The side nodes may be meant to represent some form of wing. The proportions of the impression are very long and narrow. THU8 Number of Sealings (Bullae): 1 (K00 0490) 307 Maximum Preserved Dimension: 9 mm. Shape: slightly convex seal Description: A thunderbolt. The tines are haphazard and flank a thick central shaft with no central node. Jutting from the center are two perpendicular lines that are likely wings. Chart 3.40: The Distribution of Plain Thunderbolts in the Persian Period                   308 Map 3.43: The Distribution of Plain Thunderbolts in the Persian Period 309 Map 3.43a: Detail Map Showing the Aegean Region 310 Map 3.43b: Detail Map Showing Italy: 311 Chart 3.41: The Distribution of Winged Thunderbolts in the Persian Period Map 3.44: The Distribution of Winged Thunderbolts in the Persian Period                312 Map 3.44a: Detail Map Showing the Aegean Region 313 Map 3.44b: Detail Map Showing Italy 314 Chart 3.42: The Distribution of Plain Thunderbolts in the Hellenistic Period Map 3.45: The Distribution of Plain Thunderbolts in the Hellenistic Period                   315 Map 3.45a: Detail Map Showing the Aegean Region 316 Map 3.45b: Detail Map Showing Italy 317 Chart 3.43: The Distribution of Winged Thunderbolts in the Hellenistic Period Map 3.46: The Distribution of Winged Thunderbolts in the Hellenistic Period                    318 Map 3.46a: Detail Map Showing the Aegean Region 319 Map 3.46b: Detail Map Showing Italy 320 Section 17: Vessels Images of ceramic vessels appear on five of the Kedesh sealings. These five sealings represent the use of three distinct seals. The variability between the seals is high as each portrays a different type of vessel. As such they represent the use of separate sub-motifs within the same general motif. The first seal (VES1), for example, is represented by two sealings and portrays a kantharos, or possibly a table amphora, with high handles, a wide neck and ovoid body. The second seal (VES2) was also used on two sealings and presents the image of a Panathenaic amphora with a palm branch crossing behind it. The amphora itself has a triangular lid and base with angular handles and an ovoid body with high shoulders. The final seal (VES3) has only one impression and shows a lagynos with a lid and a wide, squat body. The use of images of different types of vessels as symbols has a fairly long history. Certainly vessels, especially amphorae, were appearing as symbols by the sixth century in different places and in different media. For instance, gold pendants in the form of transport amphorae appear at Tharros at this time.581 In Greece, amphorae also appear by the late sixth century as types on Attic and Cean coins.582 In the same period vessels were also appearing on gems, as indicated by a late sixth century carnelian scarab, now in the Cabinet de Médailles, that shows a volute crater.583 During the subsequent centuries, the vessel motif in different forms continued to appear with fair regularity especially in seals and coins, as indicated by a Classical period gold ring with a lidded table amphora in a wreath and a fifth century hemiobol from Mende that shows a 581 Pisano, 1988, p. 387. 582 ANS Online Catalogue, 1944.100.24137, 2004.18.5. 583 Richter, 1968, #112. 321 kantharos.584 This practice continued through the Hellenistic period and beyond. Interestingly, the Kedesh vessel impressions can be broken down into two thematic categories. One category (kantharos and lagynos) can be related to the drinking of wine and symposia. This set is part of a larger trend in glyptic from the late Classical and Hellenistic periods that references a good and pleasurable life by means of elements from symposia.585 Thus, Callipolis had one seal that showed a lekythos, while Carthage had a sealing that showed part of a larger symposium set with an amphora, oenochoe and ladle.586 In certain cases, such vessels can become quite elaborate, as in the case of a second century Etruscan agate that shows a crater decorated with a chariot and sphinxes, thereby enhancing the reference to the idea of physical luxury.587 Of course, kantharoi like VES1 can go even further in that the vessel is an attribute of the god Dionysus and is often used either by him or his followers. As such, a kantharos can not only potentially reference the symposium, but also the god therewith associated. Meanwhile, the other set from Kedesh (Panathenaic amphoras) is linked to an altogether more active ideal, namely athletic victory. Both the amphora and the branch were used as prizes in athletic contests, specifically here the Panathenaia. This thematic category also dates back to the Classical period. A fifth century sealing from Carthage, for instance, shows a Panathenaic amphora decorated with a wreath.588 Likewise the category remained popular during the Hellenistic period, as exemplified by a mosaic 584 ANS Online Catalogue, 1992.54.137; Boardman, 1970, pl.770. 585 Maddoli, 1964, 129; Pantos, 1985, 101. 586 Berges, Ehrhardt, Laidlaw & Rakob, 1997, #835.2; Pantos, 1985, #77. 587 Maaskant-Kleibrink, 1978, #83. 588 Berges, Ehrhardt, Laidlaw & Rakob, 1997, #780. 322 from Delos that shows a Panathenaic amphora with a palm similar to VES2.589 The vessels used for this category were not limited necessarily to Panathenaic amphorae. Other vessels could be used in this manner, as in reality other vessels could be used as prizes. For example, a bronze ring of the Hellenistic period from Gordion shows an extremely ornate amphora, whose decoration would seem to indicate that it was meant to represent a metal prototype, that is flanked by branches or fronds.590 Thus, depending on their portrayal, the vessels on the Kedesh sealings could be used to reference different themes and ideals. Distribution: The distribution of the vessel motif is specific to each subset and will be discussed separately. The sealings will each be compared based on vessel type with attention paid to the thematic function of the image. For example, the kantharos of VES1 is somewhat idiosyncratic in its depiction. It has a high body, distinct neck, and double lip, which differentiates it from many other depictions of kantharoi. The prototype of the Kedesh seal served the same functions as the prototypes of the other depicted kantharoi and probably, as such, referenced the same themes, but the vessel is formally distinctive. Images of kantharoi make appearances fairly early. In the sixth-fourth centuries the vessel makes the occasional appearance in the glyptics of the various archives. These appearances include one sealing from the late 4th-early 3rd century a Carthage (N. Africa) with an ivy leaf between the handles and a sealing at Daskyleion (W. Anatolia), neither 589 Bruneau, 1972, #218. 590 Dusinberre, 2005, #82. 323 of which are truly similar to the Kedesh example.591 Besides the archives, a kantharos also appears on one side of a two sided clay sealing, dating from the second half of the fourth century, which was found in the Athenian Agora.592 This last example was also dissimilar in form. Where the kantharos appears as a type more frequently during this period is in the sphere of numismatics. In many cases, this abundance seems to stem from the association of the vessel with the cult of Dionysus, since several of the issues have either Dionysus (or one of his followers) on the other side of the coin or have the kantharos associated with other Dionysiac emblems such as grapes or ivy. During the sixth through fourth centuries, the kantharos occurred on coins from Acraephium (456-446 BCE), Boeotia (387-374 BCE), Melos (4th century BCE), Naxos (with ivy leaf above and rapes on handles, 6th and 4th centuries BCE ), Thebes (with an ivy leaf or club, 446-395 BCE), Thespiae, and Zacynthus (pre-431 BCE) in S. Greece; Cebren (400-310 BCE), Methymna (5th-4th centuries BCE), Teos in W. Anatolia; Alopeconnessus (4th century BCE), Aphythis, Mende (460-423 BCE), Peparethus (400-200 BCE) and Thrace (under Ketriporis c. 356 BCE) in N. Greece; Nagidus (374-333 BCE) in C. Anatolia; and Naxos in Sicily.593 These examples generally do not closely resemble the kantharos of VES1 in form, though the examples from Melos are closer. In the 6th-4th centuries, the kantharos motif is thus most concentrated in the area 591 Berges, Ehrhardt, Laidlaw & Rakob, 1997, #786; Kaptan, 2002, DS130. 592 Crosby & Lang, 1967, C7. 593 ANS Online Catalogue, 1987.32.143, 1992.54.103-40, 1999.4.10, 2008.1.6; Gardner, 1963a, Zacynthus 7; Gardner, 1963b, Peparethus 1-3, Corcyra 55-58 & 83-88; Head, 1963a, Acraephium 4, Boeotia Uncertain 27 & 30-37, Thebes 78-80; Head & Gardner, 1963, Alopeconnesssus 1-3, Ketriporis 1-5; Head, Gardner & Poole, 1963, Teos 27-9, Melos 1, Naxos 1-12; Hill, 1900, Nagidus 374-333; Kroll & Walker, 1993, #901; Plant, 1979, #1835, #2105, #2108, #2113, #2119, #2133, #2135, #2141, #2148; Wroth, 1964b, Cebren 15, Methymna 11. 324 of the Aegean. The most examples of the motif come from the regions of S. Greece and N. Greece with W. Anatolia lagging only slightly behind in number of appearances. Outside of these three regions, the vessel appears only sporadically in areas like Sicily, N. Africa and C. Anatolia. This pattern is not surprising considering that the real vessel type was in origin a Greek drinking vessel and would most likely appear in areas dominated by Greeks at this period. Later in the Hellenistic period, the overall trend of appearances by specific media from the preceding centuries seems to continue. The kantharos makes a sporadic appearance in the glyptics of the period. Examples of the kantharos include not only the sealing from Carthage, but also one sealing from Seleucia (The East), which was also formally distinct from the Kedesh example.594 Again, the vessel appears more abundantly in the coinage of the period. The kantharos appeared on the coins of some places which had previously used the type and other places which had not. These places include Antioch (under Antiochus VI) in The Levant; Andros (3rd-1st centuries BCE), Arcesine, Athens, Melos (with wreath and ivy leaf, 4th-1st centuries BCE) and Naxos (3rd-2nd centuries) in S. Greece; Cius (with bunches of grapes), Cyme (post 190 BCE), Methymna (320-240 BCE), Nicaea, Sardis (pre-133 BCE) and Teos in W. Anatolia; Corcyra (3rd-1st centuries BCE), Issa (2nd century BCE), Peparethus (400-200 BCE) and Pharos (2nd century BCE) in N. Greece; and Hadria (with ivy leaf), an unidentified Italian mint and Tarentum in Italy.595 Of 594 Invernizzi, Bollati, Messina & Mollo, 2004,Og93. 595 ANS Online Catalogue, 1944.100.76542-8, 1953.171.1757, 1952.142.459, 1955.162.1, 1967.204.14, 1992.54.1825-6, 2004.14.27; Gardner, 1963b, Peparethus 1-3, Issa 10, Pharos 12-15, Corcyra 225-8, 241- 80 & 396-419; Head, 1878, Athens 608; Head, 1901, Sardis 45-6; Head, Gardner & Poole, 1963, Teos 32- 34; Kroll & Walker, 1993, #549, #821, #901-2, #942; Plant, 1979, #1043, #2109, #2115-6, #2123-4, 325 these, the examples from Tarentum, Methymna and Andros are closest in form to VES1. Outside of coinage, the use of the kantharos as a symbol does make the occasional appearance during the Hellenistic Period. It appears, for instance, at Athens (S. Greece) in the form of lead tokens from the Agora, one of which actually shows a vessel pretty close in form to VES1.596 The Kantharos also appears in sculpture on a stelae from Carthage (N. Africa), which is dated to the 3rd-2nd centiuries BCE.597 Again, the depiction of the vessel here is also close in form to that of VES1. Altogether, the kantharos as a motif tends to be concentrated in and directly around Greece proper, including the Aegean islands, and along the northern portion of the western Anatolian coast. There are a few outliers, both in the west and east, of which the Antiochene examples from the mid second century are the closest in time and place to Kedesh. In the case of amphorae, the picture of the distribution is somewhat different. Amphorae in general appear fairly frequently in a variety of places and media. The specific scene represented by VES2, however, with the Panathenaic amphora and the palm branch and its connotations of athletic victory is on the whole much rarer. In the glyptics of the Persian period, the amphora in general and the Panathenaic amphora with palm both appear occasionally. For instance, Carthage in N. Africa has a sealing that shows a Panathenaic amphora in a wreath dating to the 5th century BCE and another sealing from the second half of the fourth century that shows an amphora as part #2136, #2138; Poole, 1963b, Hadria 6-17, Uncertain Central Italy Class V 1-6, Various 5-6 & 27, Tarentum 28-9, 431-47, 464-71, 482 & 487; Wroth, 1963, Andros 7-8, Melos 11-19; Wroth, 1964b, Cius 20-3, 25, Cyme 95, Methymna 17-26, Nicaea 1. 596 Lang & Crosby, 1964, L200-2, L330b. Of these, L201 is closest in form to VES1, though the handles do not appear to be exactly the same. 597 Acquaro, 1988, #194. 326 of a symposium set along with an olpe and a ladle.598 Selinus in Sicily has three sealings that show single amphorae, of which one has a similar shape to VES2 and is accompanied by a palm frond.599 As with kantharoi, amphorae frequently appear as types on coinage. In fact, amphorae appear even more frequently on coins than kantharoi of the same period. In several cases, those places that produced coins with a kantharos often had similar issues showing an amphora. The use of the amphora also starts very early in coinage, appearing on sixth century coins. During the sixth through early-fourth centuries, the places that minted coins using an amphora include Athens (510-490 BCE), Boeotia (early 4th century BCE), Carthaea (6th century BCE), Cydonia (with bunches of grapes, 400-300 BCE), Halliarte (456-446 BCE), Histiaea (before 480 BCE), Lamia (with prochous and ivy leaf, 400-344 BCE), Locri Opunitii (387-338 BCE), Melos (4th-1st centuries BCE), Orchomenos (387-374 BCE), Pharae, Thebes (5th-4th centuries BCE), and Zacynthus (pre 431 BCE) in S. Greece; the islands of Chios (before 350 BCE), Lesbos (6th-4th centuries BCE) and Samos (439-344 BCE) in W. Anatolia; Abdera (411-385 BCE), Corcyra (occasionally with grape bunches, 6th-4th centuries), Issa (4th century BCE), Larissa (4th century BCE), Mende (plain or with ivy branches, 424-348 BCE), Peparethus (400-200 BCE), Sestus (4th-3rd centuries BCE), Thasos (411-340 BCE), and Torone (490-480 BCE) in N. Greece; Nagidus (420-380 BCE) in C. Anatolia; and Vibo Valentia (as Hipponium with a torch, 330-325 BCE) in Italy.600 None of these coins show a 598 Berges, Ehrhardt, Laidlaw & Rakob, 1997, #780, #835.2. 599 Salinas, 1883, CCCLXII-CCCLXIV. The sealing CCCLXIII shows an amphora that has a ribbed body, though is otherwise similar in form to VES2, and a palm frond to the right. 600 ANS Online Catalogue, 0000.999.53414, 1944.100.10499-513, 1944.100.15460, 1944.100.24137, 1944.100.47290-1, 1948.19.369, 1957.40.2, 1987.78.104-8, 1986.78.110-2, 1991.93.5, 1992.54.141-2, 327 Panathenaic amphora crossed with a palm branch. In other media, as previously noted, amphorae appear as jewellery pendants of glass and precious metals at Tharros on Sardinia (N. Africa).601 Carthage in N. Africa has also produced a similar pendant in gold dated, to the 6th-5th centuries.602 And so depictions of the amphora motif cluster consistently during the 6th-4th centuries in the three regions that surround the Aegean, especially in S. and N. Greece. The appearances are also predominantly numismatic. Outside of these three regions, appearances are more sporadic on the whole. The region of N. Africa does have a couple of occurrences, interestingly in media other than coins. This being said, there is only one occurrence that resembles the depiction of VES2 and it comes from Sicily. In the Hellenistic period, the use of the amphora motif in general spreads out both into new geographic areas as well as into new media. The use of the specific Panathenaic amphora with palm branch, however, still appears only rarely. In glyptic, for instance, Seleucia in The East has 52 sealings (representing 47 separate seals) that show amphorae, including both transport and table amphorae, some of which are quite similar in form to VES2, though none have the palm branch.603 At Carthage (N. Africa), there are two sealings with single amphorae, of which one is a Chian transport amphorae (3rd- 1992.54.488-90, 2004.18.5, 2006.20.2, 2008.39.25; Gardner, 1963a, Zacynthus 1-6; Gardner, 1963b, Lamia 1-5, Melaniensis (Lamia) 1-3, Peparethus 4-7, Issa 8, Corcyra 24-6, 45-58, 72-6, 83-6, 102-116 & 130-168; Head, 1963a, Locri Opuntii 1-6 & 35-36, Boeotia Uncertain 17-26, 28-29 & 42-49, Halliarte 11, Orchomenos 22-24, Pharae 3-5, Thebes 18-28, 43-46, 64-77, 91-98 & 110, Histiaea 1-3; Head & Gardner, 1963, Abdera 33, Thasos 51-58; Hill, 1900, Nagidus 6; Kroll & Walker, 1993, #450; Plant, 1979, #1867, #2112, #2118, #2120, #2132, #2139-40, #2142-3, #2145, #2147, #2213; Wroth, 1963, Cydonia 14, Carthaea 14-16, Melos 29; Wroth, 1964b, Chios 40, Samos 120-125, Larissa 1-4, Lesbos 41, 46. 601 Acquaro, 1988, #612; Barnett & Mendelson, 1987, 1.39, 4.24, 6.29, 6.32-3, 10:25-6, 11.10; Pisano, 1988, 387. 602 Acquaro, 1988, #240. 603 Invernizzi, Bollati, Messina & Mollo, 2004, Og53-92, Og94-100; McDowell, 1935, III.D.4.a.1. Of these seals, Og53, Og54 and Og68 are all very similar in shape to the amphora in VES2. 328 2nd centuries) and the previously mentioned scene of the symposium set involving an amphora with olpe and a ladle (latter half of the 4th century).604 The Cyrene archive in the same region also produced three sealings showing single amphorae, with one being similar to VES2 in overall shape.605 Besides these, both Uruk in The East and Artaxata in C. Anatolia have one published sealing each that shows a set of multiple transport amphorae side by side.606 In addition, as previously mentioned, a bronze ring from the Hellenistic period was recovered from Gordion (C. Anatolia) that shows a table amphora flanked by branches or fronds.607 The shape of the amphora seems to be based on the Panathenaic type, but it has been elaborated so as to indicate that it was probably meant to represent a precious metal vessel, not a ceramic one. During the Hellenistic Period, the places minting coins that used an amphora as a type included Antioch (under Antiochus VI) in The Levant; Andros (4th-1st centuries BCE), Athens (220-83 BCE), Cydonia (with bunches of grapes 400-300 BCE), Heraea (240-146 BCE), Laconia (flanked by the pilei of the Dioscuri, 250-146 BCE), Melos (4th-1st centuries BCE), and Thebes (338-315 BCE) in S. Greece; Chios (3rd-1st centuries BCE), Lebedus (as Ptolemais, 266-203 BCE), Myrina (2nd-1st centuries BCE), and Teos (3rd century BCE and onwards) in W. Anatolia; Illyrian Apollonia (100 BCE), Macedonian Apollonia (3rd-2nd centuries BCE), Corcyra (with or without a bunch of grapes, 3rd-1st centuries BCE), Mende (350-300 BCE), Peparethus (400-200 BCE), Sestus (4th-1st centuries BCE), and Thrace (75-50 BCE) in N. Greece; Cromna (topped by a bunch of grapes, late 4th century BCE) in the region of the Black Sea; as well as 604 Berges, Ehrhardt, Laidlaw & Rakob, 1997, #781, #835.2. 605 Maddoli, 1965, #924, #926, #934. 606 Kachatrian, 1996, fig.17; Wallenfels, 1990, #1072. 607 Dusinberre, 2005, #82. 329 Canusium (flanked by a symbol and an oenochoe), Etruria, Metapontum, Populonia (with octopus emerging from the amphora, 217-215 BCE), and Uxentum in Italy.608 Some of the Hellenistic coins of Athens, dating from the late third to early first centuries BCE, show a Panathenaic amphora crossed with a palm branch very similar to the way VES2 is set up.609 The amphora motif also appears in various other Hellenistic period media. At Athens (S. Greece), the Panathenaic amphora lacking victory palm appeared on Hellenistic bronze and lead weights, and lead tokens.610 A pair of ornate and lidded amphorae also appears in the wall paintings of the Hellenistic Tomb 1 at Marisa (The Levant), where they flank an alcove.611 The image of an amphora also appears on Cnidian (W. Anatolia) amphora stamps of the 3rd and 2nd centuries.612 In addition, two mosaic panels from Hellenistic Delos (S. Greece), one in the House of the Masks and the other in the House of the Trident, as well as a wall painting from House H, insula VI, in the Theatre Quarter all show amphorae fronting palm branches.613 Altogether, the amphora motif in general is fairly widespread, and it appears in a 608 ANS Online Catalogue, 0000.999.19865-6, 0000.999.53411-2, 1895.33.5-6, 1909.999.108, 1913.999.21, 1916.192.214, 1934.999.569-606, 1940.77.127, 1941.131.812, 1944.100.3650, 1944.100.10368-70, 1944.100.10499, 1944.100.47186-251, 1944.100.47253-6, 1944.100.47258, 1947.80.4, 1947.97.432, 1948.77.23-4, 1949.100.10, 1952.142.142-3, 1953.171.767-81, 1957.172.1685, 1965.168.30, 1968.57.101, 1975.121.12, 1970.142.466, 1977.158.375-76, 1978.82.63-5, 1978.168.21-3, 1979.168.20, 1979.168.24, 1987.37.57, 2008.2.13; Gardner, 1963a, Laconia 6-13, 35-40 & 47, Heraea 28; Gardner, 1963b, Peparethus 4-7, Corcyra 240, 314-48 & 420-433; Head, 1878, Athens 618, 621, 629 & 631; Head, Poole & Gardner, 1963, Teos 37, Chios 46-91; Hill, 1900, Nagidus 6; Kroll & Walker, 1993, #76, #85, #98, #101, #108, #450, #572, #821, #943-5; Plant, 1979, #1405, #1695, #2107, #2110-1, #2114, #2117, #2121-2, #2125-8, #2130, #2146; Poole, 1963b, Uncertain Etruria Class V 1-2, Tuder 32-34, Canusium 1-3, Uxentum 3, Metapontum 194-5, Hippponium 3 & 12-16; Wroth, 1963, Cydonia 14, Andros 3-6, Melos 29; Wroth, 1964b, Cromna 9-12, Myrina 20-26. 609 Head, 1878, Athens 618 & 631; Kroll & Walker, #85, #98, #101, #108. 610 Lang & Crosby, 1964, BW4, LW17-33, L157-61, 611 Venit, 2002, fig. 153. 612 Grace, 1934, #130 & #179. 613 Bruneau, 1972, #217, #234; Bulard, 1926, pl.IX.D. Of these the one in the House of the Trident (#234) is definitely identified as a Pananthenaic amphora and depicts a chariot painted on the front. 330 variety of media. Appearances do continue to cluster in the three regions surrounding the Aegean. At the same time, a further cluster of appearances exists in the region of Italy, and isolated examples occur in several other regions. The exact motif presented by VES2 appears much more rarely and includes Gordion, Selinus, Delos, and Athens. The appearance of the motif on Athenian coins and on Athens-influenced Delos may indicate, not surprisingly considering the specific iconography, that the origin of the motif lies here. The presence of the motif on a mosaic in the House of the Trident, which also has one of the few examples of the anchor and dolphin motif and possible connections to the Poseidoniasts, is especially interesting given the possible connection between the owners and The Levant. The use of a lagynos as a motif, on the other hand, as in VES3 is highly idiosyncratic. This vessel type doe not seem to appear as a motif with any frequency. A lagynos, which is rather similarly in form to VES3, does appear on 36 sealings from Seleucia (The East) representing the use of a single seal, but other published parallels or comparanda are lacking.614 It would thus appear that this motif was used but rarely in the Hellenistic period and only in glyptic as part of the practice of referencing symposia/ the good life through the vessels used. Catalogue: VES1 Number of Sealings (Bullae): 2 (K99 0246, K99 0622) Maximum Preserved Dimension: 12 mm. 614 Invernizzi, Bollati, Messina & Mollo, 2004, Og52. 331 Shape: flat oval seal Description: A kantharos or Amphora with a high shoulder, double lip and high curving handles. Parts of the right handle and the bottom are missing in both impressions. The body is triangular with high shoulders. The neck is wide with a two stepped lip or lid. VES2 Number of Sealings (Bullae): 2 (K00 0314, K00 0315) Maximum Preserved Dimension: 12 mm. Shape: flat oval seal Description: A Panathenaic victory amphora and palm branch. The amphora stands on a high foot with an ovoid body and high, angled handles. The amphora is topped by a triangular lid. Behind the vessel a palm branch runs diagonally with the fronds on the 332 top-right. VES3 Number of Sealings (Bullae): 1 (K00 0584) Maximum Preserved Dimension: 10 mm. Shape: flat oval seal Description: A lagynos or table amphora with lid. This vessel has a wide ovoid body, short base, long neck and high arching handles. The lid is triangular. 333 Chart 3.44: The Distribution of Kantharoi in the Persian Period Map 3.47: The Distribution of Kantharoi in the Persian Period                   334 Map 3.47a: Detail Map Showing the Aegean Region 335 Chart 3.45: The Distribution of Kantharoi in the Hellenistic Period Map 3.48: The Distribution of Kantharoi in the Hellenistic Period                  336 Map 3.48a: Detail Map Showing the Aegean Region 337 Chart 3.46: The Distribution of Panathenaic Amphora with Palm Branch in the Persian Period              338 Map 3.49: The Distribution of Panathenaic Amphora with Palm Branch in the Persian Period 339 Map 3.49a: Detail Map Showing the Aegean Region 340 Chart 3.47: The Distribution of Panathenaic Amphora with Palm Branch in the Hellenistic Period                341 Map 3.50: The Distribution of Panathenaic Amphorae with Palm Branches in the Hellenistic Period 342 Map 3.50a: Detail Map Showing the Aegean Region 343 Map 3.50b: Detail Map Showing Italy 344 Chart 3.48: The Distribution of Lagynoi in the Hellenistic Period Map 3.51: The Distribution of Lagynoi in the Hellenistic Period              345 Section 18: Unknowns There are two sealings at Kedesh, representing the use of two separate and distinct seals, which seem to display in each case a symbol or object that is enigmatic and not easily identifiable. In the first case (UNK1), the object consists of a squat shaft rising from a wide base and topped by a triangular cap. Thinner vertical lines flank the central shaft on either side. In the second case (UNK2), the upper portion of the image is missing from the impression, thereby hampering the identification of the object. What remains consists of a lightly tapering vertical shaft that has a five pointed star in a circle of dots upon its face. In each case, there are possible identifications, though none are certain. For UNK1, the original excavators identified it as a phallus. The image, however, possesses the flanking lines on the side, which are hard to explain in this scheme, and it lacks any indication of testicles, which would reinforce the identification. Another possibility might be that the image represents some sort of column or a form of aedicula that forms a frame with roof, base and walls around an aniconic divine representation, like a baetyl. Interestingly, an inscribed Kedesh sealing bearing the title “Ba’al of Tyre” also seems to show the lower portion of something vaguely similar and is also possibly identified as a baetyl.615 However, the identification of UNK1 still remains unclear. In the case of UNK2, the identification may be even more nebulous. One possibility here would be that the object is a round column-like altar. Similar altars can be seen in New Comedy genre scenes, such as those found on Delian sealings, which show a fugitive slave taking 615 Ariel and Naveh, 2003, p. 72, fig.5. 346 refuge from his master upon an altar.616 Indeed, the impression bears a portion of something in the upper right which may be part of the slave, sitting on top of the altar. The one problem with this identification is that the space required would make the seal perhaps a bit too large. Distribution: As the identifications of the motifs for these impressions are uncertain, it is near impossible to discuss a true distribution of these motifs. Still, it will be possible to point out some possible parallels and comparanda. This is most easily done with UNK2. If this sealing does represent a New Comedy scene of a slave sitting on an altar, then it would represent part of a fairly popular motif that appeared in the Hellenistic period. The motif of the slave seated on an altar was a popular in the Delian archive (S. Greece), appearing on 49 separate sealings.617 The scene is also used on a Delian wall painting in the House of the Comedians, dating from the mid second to the mid first centuries BCE.618 There are also a couple of bronze statuettes showing the same scene and dating to the same rough period that come from Alexandria (Egypt) and Mahdia (N. Africa).619 Otherwise, the motif is known almost exclusively from terracotta figurines. These figurines have been found at places such as Kition in The Levant; Alexandria, Coptos, Fayum and Memphis in Egypt; Athens, Delos, Eretria, Sparta, and Tanagra in S. Greece; Cyme and Myrina in W. Anatolia; Amphipolis and Thessalonica in N. Greece; Odessus along the 616 Boussac, 1997, pp. 151-152, fig.26-27. 617 Boussac, 1997, p. 151, fig.26-8. 618 Webster, 1995, 3DP2.3. 619 Webster, 1995, 3XB1-2 347 Black Sea; and Capua, Tarentum and an unspecified site in Italy.620 The motif, therefore, is one that appears regularly, if somewhat occasionally, across the ancient Mediterranean. It was especially popular in the regions of Egypt and S. Greece which each had several occurrences. More specifically, the motif seems to have been particularly popular on Delos, where it appears on multiple seals and in several other media. With UNK1, on the other hand, the problems with the identification make it extremely difficult to compare it to anything. Still, it is interesting to note the existence of a series of fifth century Punic stone cippi from Tharros on Sardinia (N. Africa) which typically display three squat pilasters, of which the middle one sports a pyramidal capital, like the Kedesh sealing, with a solar disk and crescent on the face. Moscatti sees these cippi as representing versions of the tree pillar-like baetyls common on Carthaginian stelae.621 As such, they may represent support for the theory that UNK1 represents some sort of baetyl. Catalogue: UNK1 Number of Sealings (Bullae): 1(K00 0174) Maximum Preserved Dimension: 9 mm. Shape: flat seal Description: A phallus, column or aedicula? This image consists of a broad base topped by a smooth shaft rising to a triangular cap. Flanking the shaft are two vertical lines. Three horizontal lines run across the middle of the shaft, with one going off to the right 620 Webster, 1995, 1AT3-5, 1AT16, 1KT3, 1NT3, 2BT2, 2NV7, 2TT2, 3AT2, 3CT1, 3DT35, 3DT72-3, 3ET8, 3XT4-5. 621 Moscati, S. “Stone Reliefs.” The Phoenicians. 1988, pp. 300-303. 348 past the vertical line. UNK2 Number of Sealings (Bullae): 1(K00 00660) Maximum Preserved Dimension: 9 mm. Shape: flat seal Description: A decorated altar? There is an upright shaft with star in circle of dots within it. The upper portion is missing. Something unidentifiable lies in the upper right-hand corner. It may very well be the remains of a Slave-on-Altar scene from New Comedy. 349 Chart 3.49: The Distribution of UNK2 (as Scene of Slave on Altar) in the Hellenistic Period Map 3.52: The Distribution of UNK2 (as Scene of Slave on Altar) in the Hellenistic Period                    350 Chart 3.50: The Distribution of UNK1 (as Pyramid Capped Baetyl) in the Persian Period Map 3.53: The Distribution of UNK1 (as Pyramid Capped Baetyl) in the Persian Period              351 Chapter 4: The Geographic Distribution of the Kedesh Symbol Motifs The appearance of a specific motif amongst the Kedesh symbol sealings represents a single point in the use-constellation of that motif. Each use-constellation consists of all of the uses of a given motif defined by the geographic location of the use, the date of the use, and the medium in which it appears. Each use-constellation describes for a given motif where and in what media the motif popularly appeared. In some cases, such as with the New Comedy slave mask motif, the use constellation can be complex, with many occurrences of the motif across numerous geographic regions and in varied media. In other cases, such as with the ligature or composite eagle motifs, the use- constellations of the motifs are quite simple, consisting of only one or two occurrences each. Taken together, the various use-constellations of the Kedesh symbol sealing motifs express the overall universe of the symbolic visual culture understood by the residents of Kedesh. Each use-constellation combines with the others to produce an overall map of where these motifs are most commonly found. In essence, this aggregate picture will illustrate the regions of origin as well as those places with which the seal users of Kedesh likely interacted for the construction of their own symbolic visual culture. Perhaps the most defining characteristic of a use-constellation is temporal. The geographic regions and media in which a given motif is popular can shift over time. The same is true for the aggregate of the motifs that form the visual culture of the symbols found at Kedesh. As such, we need to study the overall geographic distribution of the motifs diachronically. I will do this in a simplified fashion by maintaining the same 352 period division that I used in the catalogue. First I will examine the overall geographic distribution of the motifs of the Kedesh symbol sealings in the Persian period, or more generally in the 6th-4th centuries BCE. Then I will examine the picture produced by the geographic distribution of the motifs during the Hellenistic period (late-4th through late 1st centuries BCE). The Visual Universe of the Kedesh Symbol Sealing motifs during the Persian Period: In combining the various use-constellations of the different motifs present at Kedesh during the 6th-4th centuries, certain patterns emerge (see Chart & Map 4.1). The motifs tend to cluster in specific regions and media. The first point is that the symbol motifs are focussed in the three regions that surround the Aegean (S. Greece, W. Anatolia and N. Greece). These regions have produced the grand majority of the occurrences of the motifs. Together, these regions account for 127 of the 219 total occurrences, representing 58% of the total. Second, the source of their numerical predominance lies mainly in coins, which overshadow all other media. The region of S. Greece has the highest cumulative number of occurrences for this period with a total of 52 separate appearances of the various Kedesh motifs. Of these 52 occurrences, 39 come from coins with seals, sculpture, and mosaics each having 2-4 occurrences. The motifs in which the region has numerous examples include plain bucrania with 14 occurrences, lyres and kantharoi with 6 occurrences each, and stars and plain thunderbolts with 5 occurrences each. Other Kedesh motifs, like caducei, winged thunderbolts and janiform heads also have a couple of occurrences each. Within the 353 region, certain specific places make frequent appearances. Athens here appears most often, with ten occurrences. This is not surprising considering both its poltitcal/artistic pre-eminence in the 5th century BCE, combined with the thoroughness of its excavation and publication. More intriguing is the fact that the island of Melos, which has had none of these advantages, comes in second with six occurrences. Besides these two, the sites of Delos, Elis, Praesus, Sicyon, Sparta, and Zakynthus all have two occurrences each, while every other place in the region has one. The region of N. Greece comes in a distant second in the number of overall occurrences by region. It has 38 occurrences of Kedesh motifs during the 6th-4th centuries, of which coins account for 30 alone. In other media, seals, sculpture and vessels each have two occurrences each, while mosaics and other media each have one. This region is most prominent in the motifs of the rosette, where it has eight examples, the lyre with seven occurrences and the thunderbolt and kantharos motifs with six and five occurrences each respectively. Besides these, the region also produced, during this period, three occurrences each of the caduceus and the janiform head. Unlike the region of S. Greece in the same period, N. Greece has few locations that reappear again and again with multiple occurrences. Most places tend to have only one occurrence. Specific sites within the region that have multiple occurrences include Macedonia with 4 and Amphipolis, Corcyra, Maroneia, Mende, and Thasos, which each have two occurrences. The region of W. Anatolia trails N. Greece only slightly with a total of 37 occurrences of Kedesh symbol motifs. Once again, coins dominate the distribution with 31 occurrences. The other six occurrences consist of three glyptic examples, two sculptural examples and one vessel. The region has the largest number of occurrences in 354 the motif categories of the lyre, which has eight occurrences, the star which has six occurrences, and the kantharos, which has four. Besides these three motif categories, the region produces three occurrences each of the caduceus, plain bucranium and flower motifs. Unlike the first two regions, no one site dominates W. Anatolia during this period. Instead, the appearances are diffused throughout the region. The sites of Assos, Colophon, Daskyeleion, Lampsacus, Lesbos, Methymna, Phocaea, Tenedos, and Teos each produced two occurrences of Kedesh motifs while other sites had single occurrences. Besides the regions that encompass the Aegean, there are three more regions that produced significant numbers (for our purposes here in this period, more than ten) of occurrences of the motifs that appear in the Kedesh symbol sealings. The first of these is Sicily. This region produced 27 occurrences of which 13 were coins. Of the remaining occurrences, eleven were from glyptic and the remaining three were sculptural. The region produced five examples of the plain thunderbolt motif and three each of the flower and winged thunderbolt motifs. In the case of other motifs, the region generally does not possess more than one or two occurrences. The archive at Selinus dominates the region, producing examples of many different motifs. Other prominent regional sites during the period include Syrcause with four occurrences, Catane with three and Malta and Halaessa with two occurrences each. The next most abundant region is N. Africa, with 26 occurrences of Kedesh motifs. The interesting point here is that none of the occurrences come from coins. Instead, nine come from seals, ten are sculptural, two are vessels, and five occurrences come from the ‘Other’ medium category and consist mainly of jewellery. The region is especially prominent in the motifs of Phoenician masks, producing 5 occurrences, 355 flowers, with another five occurrences, and satyr masks, where it produced four occurrences. Like Sicily, only a couple of sites dominate the region for occurrences of Kedesh motifs. Carthage, for instance, accounts for eleven of the occurrences by itself. This predominance is a result of both the presence of the archive and the presence of its extensively excavated cemeteries, where uses of Phoenician and Satyr masks were frequent. Likewise, the cemetery at Tharros on Sardinia has also produced seven occurrences of different Kedesh motifs, including masks and Janiform heads. Meanwhile, the sites of Sulcis and Ibiza both produced two occurrences each. The final region of note during the 6th-4th centuries is The Levant, the region in which Kedesh itself is located. This region has produced twenty-one occurrences of the motifs during this period, including nine coins, seven sculptures and four seals. The most numerous motifs for the region in this period include the Phoenician mask, with six occurrences, the rosette with five, flowers with four, and the janiform head with three occurrences. As with W. Anatolia, The Levant lacks any predominant site during this period. Rather, the motifs that appear at Kedesh occur as singletons at various sites in the region. Only Jerusalem, Tyre, Akhziv and Idalion have multiple occurrences, with two each. And so a distinct pattern emerges from the overall geographic distribution of the Kedesh motifs during this period. Overall, the Kedesh symbols are most prevalent in those regions that at the time were the focus of major Greek settlement, such as the Aegean and Sicily. They are also most prevalent in the medium of numismatics. This pattern makes sense for several reasons. First, several of the motifs that appear commonly during this period are attributes of Greek gods, such as the thunderbolt of Zeus, the 356 caduceus of Hermes and the lyre of Apollo. Just as the Athenians displayed the owl of Athena on their coins, these symbols were sensible motifs to use on coins where the cult of that god was prominent. Certainly, this is the case of the thunderbolt, both plain and winged, on the coins of Elis. Outside of these four regions, the regions of N. Africa and The Levant did not have a strong Greek presence. Rather, they belonged in the general Phoenician cultural context. The motifs that are popular in these two regions at this time consist of either symbol motifs with shared history/popularity between Greek and Near-Eastern cultures, such as the rosette which was ubiquitous across many cultures or the janiform head that likewise appears in many separate cultures, or symbol motifs that had their own deep connection to the cultures of the Levant and of Phoenicia in particular, such as the Phoenician masks which have antecedents dating back to the Bronze Age in Canaan. The fact that some of the motifs do occur in both Greek and Phoenician/Punic contexts also helps to explain the pre-eminence of the Aegean since the regions in that area also had examples of these motifs. And so, in the 6th-4th centuries, most of the motifs from the Kedesh sealings that occur by this period are at home in Greek contexts, especially the Greek homeland centred on the Aegean. At the same time, there is a distinct subset that is also at home in Phoenico-Punic contexts. This subset consists of motifs that are either shared with Greek contexts or, occasionally as with the Phoenician mask, are specific to the Phoenician contexts. 357 Chart 4.1: The Total Distribution of the Symbol Motifs in the Persian Period                  358 Map 4.1: Sites with Multiple Occurrences in Significant Regions (More than 10 Total Occurrences) during the Persian Period The Visual Universe of the Kedesh Symbol Sealing motifs during the Hellenistic Period: In the next period, the overall picture of the symbolic visual universe shifts markedly (see Chart and Map 4.2). Upon examination, one of the first things that is apparent is the dramatic climb in the total number of occurrences of the motifs. Whereas previously, the region with the most occurrences had a total of 52, the highest total now stands at 137, nearly triple the previous limit. In addition, while in the Persian period only five regions had more than twenty occurrences, now seven regions have more than forty 359 and only one has less than twenty. And so, the number of occurrences of the symbol motifs that have survived and been published has increased dramatically, while the motifs have also spread out geographically in general. A part of this has to do with a shift in the distribution of motifs across media. In the Hellenistic period, coins no longer dominate the visual landscape as they once did. Coins still persist in large numbers of occurrences, but they often no longer have the same dominance in any given region as they once did. Instead, the Kedesh motifs now also appear in large numbers on vessels, sculpture, and elsewhere. As such, the appropriateness of using symbols has apparently migrated from coins into different media. Furthermore, we should also note that several of the motifs that are present at Kedesh only begin to appear in numbers during the Hellenistic period. These motifs include pilei of the Dioscuri (plain and with object), cornucopias, New Comedy slave masks, and oval shields. Geographically parts of the overall numerical distribution remain fairly similar to the pattern set in the 6th-4th centuries BCE. The region of S. Greece once again has the most occurrences of the Kedesh motifs of any region with 141 separate uses of motifs that also appear on the Kedesh sealings. Of these, 51 consist of coins and related media (like tokens and weights), representing now only 37% of the total as opposed to 75% in the previous period. In other media bearing Kedesh symbol motifs, the region has produced 30 examples of vessels, 26 examples of sculpture, 17 glyptic examples, eleven mosaics or paintings and 2 that fall into the category of ‘other’ representing usually non seal jewellery. Of the motifs that the region presents, the most numerous are occurrences of the New Comedy slave mask, with 28 separate examples as well as plain bucrania with 16 360 examples, rosettes with 14 examples and lyres with 13. As in the previous period, certain sites dominate the region in the number of occurrences of Kedesh symbol motifs. The region also produced the only other examples of the composite hippocamp motif in the form of sealings from Delos. In regards to individual sites, Delos is now first with 30 occurrences, consisting in part of the finds from its massive archive, its mosaics, sculpture, vessels, and terracotta figurines. Athens now runs a close second with a total of 29 occurrences of the various motifs. These occurrences are spread between coins, tokens, sculpture and stone figurines. Callipolis, which has its own archive, comes a distant third with ten examples. Meanwhile, Corinth has eight occurrences of the symbol motifs, and Melos has six. Besides these five sites, several other sites have between two and five occurrences each, including Argos, Cythnus, Elis, Knossos, Sparta, and Syros. In contrast to the previous period, the region of W. Anatolia now ranks second in number of occurrences, with a total of 134 examples. This total is very close to that of the leading region, representing 95% of the total from S. Greece, but is almost three times the total of N. Greece, the previous runner up. As with S. Greece, coins and related media now play a much less significant role than they did in the previous centuries. In the Hellenistic period, W. Anatolia possesses 52 occurrences of Kedesh symbol motifs in coins and related media, representing 40% of the total, as compared to 84% earlier, as well as a significant number of occurrences in glyptics (13), vessels (35), sculpture (26), and paintings or mosaics (7). Especially numerous are the motifs of New Comedy slave masks with 21 occurrences, lyres with 18, satyr masks with 17 examples, rosettes with 16, and plain bucrania with 13 separate occurrences. Unlike S. Greece, no single site or sites in W. Anatolia dominate the region with their level of occurrences of Kedesh symbol 361 motifs. Of the sites with multiple occurrences, Pergamon ranks first with fourteen, Rhodes second with eleven, and Cnidus third with ten. Besides these three, however, many sites have between two and five occurrences each, including Adramytium, Apamea, Assos, Bithynia, Calymna, Caystriani, Clazomenae, Cos, Cyme, Ephesus, Halicarnassus, Lampsacus, Laodicea, Letoon, Magnesia, Methymna, Miletus, Myrina, Mytilene, Nicaea, Philadelphia, Priene, Samos, Sardis, Smyrna, Termessus Minor, and Troy. Thus occurrences seem to be more spread out in this region than they were in the earlier one. Interestingly, though, these sites with multiple occurrences do tend to congregate along the coast and on the coastal islands, with few examples appearing far inland. The region with the next highest level of occurrences of the Kedesh symbol motifs is Italy. This fact once again marks a departure from the pattern set in the 6th-4th centuries. The region of Italy has 112 occurrences of the motifs dating to the Hellenistic period. This number is a sixteen-fold increase over the seven that the region produced in the earlier period. As such, the region now ranks third in sheer number of occurrences. Out of the 112 Hellenistic examples, exactly half, or 56, are coins. The remainder consist of 19 sculptural examples, 28 examples of vessels, 6 mosaics or paintings and three examples from other media. The region is especially numerous in occurrences of the New Comedy slave mask motif with a total of 35 separate examples. This total gives the region the highest total for that particular motif, beating out even S. Greece and W. Anatolia. Out of the other motifs, Italy is especially strong in plain bucrania with 11 examples, plain thunderbolts with 13, lyres with 9 examples, satyr masks with eight, janiform heads with seven, and cornucopias with 6 examples. As with W. Anatolia, the occurrences in Italy tend to be diffuse, appearing individually or in small groups across multiple sites. 362 The site with the largest number of occurrences is Tarentum, with a total of twelve. Rome and Pompeii are tied for second place with 6 occurrences each. Besides these three, several sites have between two and five occurrences each, including Cales, Canusium, Capua, Clusium, Cosa, Iguvium, Locri, Luceria, Naples, Rheghium, Rubi, Thurium, Tuscana, and Vibo Valentia. Most of these sites are located in the southern half of the ‘boot’ of Italy, especially in the areas of Latium, Apulia, Campania, and Bruttium. The region of The Levant ranks fourth in number of occurrences. There are a total of 75 separate appearances of the Kedesh symbol motifs. Of these appearances, 33 are coins; 26 are vessels; 8 are sculptural; 4 are seals; and 4 are from other media. Out of the 75 total occurrences, The Levant is notably strong in the satyr mask and cornucopia motifs which have twelve and eleven appearances respectively. The region also has numerous examples of New Comedy slave masks and rosettes, which have eight occurrences each, as well as pilei of the Dioscuri with object, and winged thunderbolt, all of which have 6 occurrences each. The site within the region with the most occurrences is Antioch, which has nine occurrences. The site of Dor has the second highest number of occurrences with six. Other sites with multiple occurrences of the Kedesh symbol motifs numbering 2-5 consist of Apamea, Ashdod, Berytus, Cyprus, Jerusalem, Kharayeb, Marisa, Paphos, Phoenicia, Ptolemais-Ake, Salamis, Samaria, Seleucia Pieria, Tel Aviv, Tripolis, and Tyre. The overall pattern of the occurrences in the region, therefore, resembles that of W. Anatolia and Italy, where the occurrences are diffuse across numerous sites with only a couple of sites having any sizeable concentrations. Also, like W. Anatolia, the occurrences in The Levant tend to happen in sites on or near the coast with fewer happening in sites inland. 363 In all of the next three regions (N. Greece, The East, and Sicily) there are between 45-50 occurrences each, with the region of N. Greece having the most with 48 occurrences. As compared to other regions in which the depictions of the symbols found at Kedesh doubled, this number represents only a 26% increase for N. Greece over its number in the previous period. These 48 occurrences consist of twenty-seven coins, ten pieces of sculpture, five mosaics, and four vessels. In addition, no one motif appears here in a great concentration, but instead there are isolated examples of several. The top two most numerous motifs in the region are cornucopias and plain thunderbolts with eight and seven examples respectively. The motifs of plain bucrania, rosettes, New Comedy slave masks, and kantharoi all have four occurrences within the region. This pattern extends even further, as there are few sites with multiple occurrences. The site of Corcyra has the most with five occurrences. Amphipolis has four examples of Kedesh symbol motifs. The area of Macedonia and the site of Apollonia in Illyria both have three occurrences. Besides these, Abdera, Epirus, Hephaestia, Samothrace, Thasos, and Thessalonica all have two occurrences each. In both The East and Sicily, there are 46 occurrences of motifs that appear at Kedesh. Of the two, the region of The East is especially interesting. Its 46 occurrences represent a more than an eleven-fold increase over the previous period, when the region had only four occurrences. The Hellenistic occurrences consist of twenty-three examples of sealings or seals, nineteen of coins or related media, three sculptural examples, and one mosaic. Considering the occurrences by category of motifs, the region seems to produce small numbers of many different categories, rather than large numbers of only one or two. Those motif categories with multiple occurrences in the region include plain anchors (6), 364 plain pilei of the Dioscuri (6), plain bucrania (4), lyres (4), plain thunderbolts (4), bucrania with objects (3), anchors with horsehead protomes (2), caducei (2), cornucopias (2), New Comedy slave masks (2), and winged thunderbolts (2). The prevalence of the anchor motif and its horsehead variant is not surprising considering the function of those motifs as Seleucid official symbols. In addition, the region also produced the only other known examples of the hand holding an ear, and lagynos motifs, as well as the only other glyptic examples of the ear by itself. In regards to the regional concentration of occurrences by site or area, the image follows the pattern set by S. Greece, with one or two major concentrations, rather than the more diffuse spread seen in other regions. Seleucia on the Tigris, with its massive set of archives and mint, accounts for twenty three occurrences, exactly half of the regional total, by itself. Indeed, almost all of the motifs that are present in the Kedesh symbol sealings are present at Seleucia. Only the composite hippocamp, the ligature, the Phoenician mask, the pilei of the Dioscuri with an object, and the Panathenaic amphora with palm branch motifs are lacking from the archive at Seleucia. Besides Seleucia, only a few sites or areas have multiple occurrences, including Uruk, Bactria and Ai Khanoum with three each as well as Parthia and Ecbatana with two each. The 46 Hellenistic occurrences from Sicily, on the other hand, represent a significant increase of 70% over the preceding period. More than half (59%) of the occurrences are on coins with twenty-seven examples. The rest consist of ten pieces of sculpture, five mosaic or paintings and four vessels. The most numerous motifs in the region are the New Comedy slave mask, which has eleven occurrences, and the lyre, which has 9. Other motifs with multiple occurrences in the region include satyr masks 365 with six, thunderbolts (both plain and winged) with four each, janiform heads with three, as well as caducei, rosettes and plain pilei of the Dioscuri with two occurrences each. As with N. Greece, the spread of occurrences by site is diffuse in Sicily. Centurippe has the most occurrences with a total of six. Many other sites have between two and four occurrences each, including Agrigentum, Calacte, Catana, Halaessa, Lipari, Malta, Morgantina, Panormus, Syracuse, and Tyndarus. Interestingly, most of the sites that have multiple occurrences tend to be located on the eastern and northern sides of the island, which are closest sides to Italy, with fewer to the west and south. Considering the upsurge in motif use in Italy itself during the Hellenistic period, the residents of the Sicilian sites with the closest or most direct access to the peninsula may have continued to show interests in the symbol motifs. This would have been especially the case for the most popular motifs in the two regions, such as the New Comedy and satyr masks, janiform heads, plain thunderbolts, and lyres. The remaining regions all possess less than forty occurrences each. The region of Egypt, for instance has thirty-three occurrences of which a full fifteen come from Alexandria. N. Africa for its part has twenty five occurrences total, which represents a net loss of 1 occurrence as compared to the previous period. The occurrences of the region are almost completely taken up by Carthage and Cyrene, which both have ten examples each of which the lion’s share comes from the archives located in each city. The region of the Black Sea, on the other hand has 28 Hellenistic examples, which is a fourteen-fold increase over its total in the 6th-4th centuries BCE. Its breakdown by site seems to follow the more diffuse pattern set by W. Anatolia and elsewhere as the sites of Amisus, Amasia, Callatis, Chersonessus, Panticapaeum, Paphlagonia, and Sinope all have 366 between two and five occurrences each. The region of C. Anatolia, for its part, has a total of twenty occurrences, which is a more than six-fold increase from the previous period. Within the region, only the sites of Tarsus, Artaxata and Gordion have multiple occurrences of the Kedesh symbol motifs, with 4 and 3 each respectively. And so the overall geographic distributions of the motifs that appear on the Kedesh symbol sealings changes significantly in the Hellenistic period from the pattern set previously. In certain ways, the pattern is similar to the preceding one. The three regions (S.Greece, W. Anatolia and Italy) with the most occurrences still outnumber the other regions by a large amount, with 387 of the total 713 occurrences. The focus on the Aegean also still remains to a certain extent. The two regions with the most occurrences still consist of S. Greece and W. Anatolia. The symbols that make up the Kedesh corpus continue to be most popular in these two regions. The distribution by site within these two regions remains similar to what was going on previously with S. Greece having a few sites with numerous occurrences and W. Anatolia being more diffuse with many sites having a few occurrences. If anything, this particular pattern has become more pronounced with Athens and Delos in S. Greece representing a large percentage of the total regional occurrences by themselves. Even here, however, there is a change as W. Anatolia now has the second highest amount, displacing N. Greece, which has dropped significantly from second to fifth place in its relative standing. One of the most marked shifts from the Persian to the Hellenistic distribution is the rise of Italy, which is now third in the relative standings. Like S. Greece and W. Anatolia, the region also displays a well rounded break-down of occurrences by media, with no one medium really dominating the total. The sites where the motifs occur include 367 both cities that were of Greek colonial origin, such as Tarentum and Metapontum, and Italic cities like Rome and Pompeii. It thus appears that the motifs from the Kedesh symbol sealings, while still being popular in Greek contexts, have spread beyond them into several different cultural contexts, including Italic ones. Indeed, the seal PIL3 in particular seems to indicate the adoption of a motif otherwise popular solely outside the Italian peninsula, namely the pilei of the Dioscuri flanking a snake or object, and the adaptation of the motif to a specifically Italic context, namely that of Italic domestic religion with the lares and the genius loci. The region of The Levant also seems to be an example of the overall spread of the symbol motifs outside of their original Greek contexts. In the Hellenistic period, the region has the fourth highest number of occurrences. Furthermore, its 76 occurrences represent 53% of the total of the region with the greatest total (S. Greece). This is a significant increase over the previous period when the Levantine total was only 39% of the total of the frontrunner (21 versus 53 occurrences in S. Greece). There are even more Hellenistic occurrences in The Levant than in N. Greece, one of the regions in the Greek homeland of the Aegean. Indeed, The Levant beats out N. Greece, which is its nearest rival, by almost fifty percent in numbers of occurrences. These occurrences include numerous different motifs, both those that existed previously and those that first appeared in the Hellenistic period. In fact, The Levant is the leader in the number of occurrences in the cornucopia motif, which was first becoming popular in the Hellenistic period. This fact indicates that the region was not simply bringing in symbol motifs from outside, but also engaging actively in the production and/or popularization of certain motifs on its own. The region thus was a fertile ground for the use of symbols as a whole during the 368 Hellenistic period. On the other end of the spectrum, there are the regions of Sicily and N. Greece which have both dropped in the relative ranking of regions. The region of N. Greece fell the most precipitously from the second rank in the Persian period to the fifth in the Hellenistic. Unlike other regions, it experiences only a very meagre growth in the overall number of motif occurrences. In the case of Sicily, the growth in number of occurrences was greater overall, but was still not enough to keep the region from slipping from fourth to a tie for sixth. Indeed, part of the growth of occurrences in Sicily may be the result of interaction with the Italian peninsula, which was itself experiencing a massive upsurge in its own number of occurrences, since Sicilian sites with multiple occurrences tend to lie in proximity to Italy. Both Sicily and N. Greece are also conservative in regards to the distribution of occurrences by media, with the much higher relative percentage of numismatic occurrences, such as characterized the earlier period. The two regions also seem conservative in their choices of motifs. Generally the two regions favour older more established motifs with only one or two well represented motifs that first become popular in the Hellenistic period. For instance, Sicily has few examples of pilei or cornucopias overall, but it does have several occurrences of New Comedy slave masks. On the other hand, N. Greece has an overall lack of New Comedy slave masks or pilei, but it does have several instances of the use of cornucopias as a symbol. As such the image that emerges from these two regions in regards to the Kedesh symbol motifs is one of overall stagnation, as opposed to the innovation that we see in The Levant. Two more regions: Egypt and N. Africa, are also interesting for the unexpected dearth of occurrences that they offer. The region of Egypt was the heart of the empire of 369 the Ptolemies who also controlled Kedesh during the late 4th and 3rd centuries BCE. In this light, we would normally expect some form of congruency between the symbols that show up at Kedesh and those from Egypt. This does seem to happen occasionally, such as with the cornucopia motif, which was first promulgated by the Ptolemies. At the same time, however, there is also a general disconnect between Egypt and Kedesh. The symbols that form the Kedesh symbol corpus just do not show up all that frequently in Egypt. The ones that do, like the New Comedy slave masks and bucrania, also show up elsewhere in much greater numbers. In addition, there is a general lack of any symbol at Kedesh that seems to come from the Egyptian cultural repertoire. In fact, the one seal that was originally Egyptianizing, namely BUC1 which was originally an image of a Horus falcon on a lotus flower, was changed after manufacture so as to lessen its immediately Egyptianizing quality.622 Altogether it seems that the Kedesh symbol universe had very little relationship to Egypt, even though the Ptolemies had controlled the site for a century. The picture that N. Africa displays in regards to the Kedesh symbol motifs is very different. Whereas in Egypt the motifs have almost no history of representation, appearing in only 1 example in the Persian period, N. Africa had a relatively strong connection with the motifs. This connection was carried in part by a strong Phoenician cultural connection between The Levant and Carthage, as exemplified by the Phoenician 622 Why this change was done remains unclear. Certainly the fact that other sealings from Kedesh, namely K99 0015, show overtly Egyptianizing motifs would indicate that such motifs were not effectively banned by the incoming Seleucids. One possible explanation is that the two seals represent two different time periods in the use of the archive, with BUC1 standing right at the cusp of the change in power from Ptolemy to Seleucid and K99 0015 representing either an earlier point in time under the Ptolemies or a later point after things had settled down and the use of an Egyptianizing motif on one’s seal may not have carried as much political weight. 370 mask motif. In addition, during the 6th-4th centuries both N. Africa and The Levant have roughly similar levels of occurrences of the motifs. This pattern changes radically in the Hellenistic period. Out of all the regions, N. Africa is the only one that experiences a net loss in the number of occurrences of Kedesh symbol motifs. Meanwhile, The Levant experienced growth and innovation in regards to the motifs. The one motif that had connected the two regions in the previous period, the Phoenician mask, was also disappearing during the Hellenistic period, being replaced in places like tombs and sanctuaries by less geographically distinctive masks from the Greek theatrical tradition. At the same time, the region of N. Africa remained interested in the depictions of symbols in general.623 What changes is that N. Africa seems to lack interest in depicting the Kedesh motifs or depicting them in the same ways.624 As such, the image that emerges is not one of stagnation as we have seen in regards to N. Greece or Sicily, but rather of divergence. In effect, N. Africa as the result of its own historical and cultural forces is developing its art differently than the ancient Phoenician homeland of The Levant, which is undergoing its own development. And so the overall pattern indicates that, during the Hellenistic period, the Kedesh symbol motifs are still at home in the Aegean area, where they remain popular. However, they have outgrown this cradle, spreading to other, non-Greek areas, including Italy. In addition, new motifs emerge in different regions, most notably in The Levant, which has a strong showing in new motifs such as the cornucopia, indicating the presence 623 For example see Acquaro, 1988, #184-211. These consist of Hellenistic grave stelae from Carthage with a variety of symbols on them, including the sign of Tanit, the Caduceus, hands, palm tree, and incense burners. 624 So, for instance, the above examples often show caducei but they are often combined with other symbols like hands or the sign of Tanit, which is very common at Carthage but only appears in the Kedesh sealings in the inscribed official seal (Ariel & Naveh, 2003, #1) that is not part of this corpus. 371 of innovation on the part of the use of symbols. In other cases, certain regions like N. Greece and Sicily show a tendency towards conservatism or stagnation in regards to the use of symbols, while others such as N. Africa demonstrate a divergence into the use of other, different symbols. Chart 4.2: The Total Distribution of the Symbol Motifs in the Hellenistic Period                    372 Map 4.2: Sites with Multiple Occurrences in Significant Regions (More than 40 Total Occurrences) in the Hellenistic Period The Models of Distribution: Altogether, there appear to be two separate ways for occurrences to appear in a region, with each model indicating a different sort of penetration of the symbol motifs into that region. Most of the regions in the Hellenistic period show occurrences at many sites in the region. Few of these sites, however, have many occurrences. Rather the occurrences are spread out. As such, the motifs have a broad diffusion in the area, but show little depth in penetration. They are generally popular across the region, but not any particular place in particular. At the same time, however, certain regions, notably S. Greece and The East 373 show a different pattern. Here, the occurrences of the Kedesh motifs are especially popular at one or two sites and appear less frequently elsewhere. This pattern indicates a greater penetration of the motifs in those particular sites, but possibly less penetration elsewhere. This is especially the case of The East, where, outside of Seleucia, the motifs appear very sporadically. Indeed, the effect of these sites with a large number of occurrences each can be quite profound for the entire region. Specifically, the extremely large numbers of symbol motifs that also occur at Kedesh warp the overall distribution for their regions. For instance, without Athens and Delos, the region of S. Greece would rank third in the number of occurrences. Meanwhile, without Seleucia, there would be fewer occurrences in The East than in Egypt and N. Africa. To a certain extent the congruency that these sites display with the Kedesh symbol motifs is the happy result of both preservation and publication. At the same time, the congruency also transcends these arbitrary conditions. As Uruk, which has its own well published archive and yet few links to the motifs at Kedesh, indicates, a well preserved and published archival site need not produce many occurrences of the Kedesh motifs. In addition, there are other factors that might help explain why these three sites in particular are so congruent with the Kedesh symbolic corpus. In effect, these three sites represent nexuses of occurrence for the symbol motifs at Kedesh. They reflect some of the artistic influences that affected the visual universe of the motifs. 374 Seleucia, for instance, was a major cosmopolitan city in the Hellenistic period.625 It was a new, Greek foundation in the Hellenistic period. It was also one of the capitals of the Seleucid Empire which also controlled Kedesh and oversaw its archive during the first half of the second century BCE. As such, the archive at Seleucia was well placed to reflect what artistic motifs were current in the Seleucid Empire as a whole. Certainly, the Seleucene archive contains examples of the symbols that appeared in official Seleucid glyptic contexts, namely the anchor and the anchor with horsehead variant, along with some of the only parallels to certain motifs that also appear at Kedesh, such as the lagynos and the hand holding an ear. Yet, at the same time, the congruency between Seleucia and Kedesh is not complete. Several motifs appear at Kedesh that are not represented at Seleucia, thereby indicating that the Kedesh archive was not simply a satellite version of the Seleucene on a smaller scale. The site of Delos presents a different picture to that of Seleucia. Unlike Seleucia, it was not a politically dominant capital city. Rather, it was a major trade hub in the Hellenistic period.626 Like Seleucia, Delos also possessed a very large archive that has been published at least in part. This archive has provided us with the only other known examples of the composite hippocamp motif. It also produces many occurrences of the Kedesh symbol motifs in several different media, ranging from ceramic vessels to mosaics. Indeed, Delian mosaics provide us with some of the few examples of otherwise rare motifs, such as the anchor and dolphin and the Panathenaic amphora with palm branch. The nature of the island as a free port and trade hub also gives us a possible link 625 Sherwin-White & Kuhrt, 1993, pp. 172-3. Here, the authors note both the immense physical size of Seleucia as well as the combination of temples that follow either the native Mesopotamian or incoming Greek architectural traditions. 626 Strabo X.5.4. 375 between The Levant and Italy, which were both experiencing a growth in the use the Kedesh motifs, since the island was home to both Italian traders, as indicated by the Agora of the Italians, and Levantine ones, as indicated by the presence of the Poseidoniasts of Berytus and their building and possible houses. Indeed, the House of the Trident alone provides us with mosaic occurrences of the Panathenaic amphora with palm branch and the anchor with dolphin motifs that otherwise appear rarely. Furthermore, an inscription from Delos, dated 149 or 148 BCE, indicates direct contact between the Poseidoniasts and members of the Italian community on Delos.627 Specifically the inscription, which was set up by the Poseidoniasts, honours one Roman banker, Marcus Minatius, who aided the Society monetarily and, as a result, was to receive an annual procession in his honour as well as membership in the society and pride of place at their meetings. Athens is a different case entirely. The city did not have anywhere near the importance in politics or trade that it had once held during the Classical period. At the same time, it was culturally important to Greeks in general and to Hellenistic rulers who endowed it with various constructions, like the Stoa of Attalos.628 Athens also had at least some form of Levantine community during the Hellenistic period as indicated by the presence of the 3rd century bilingual grave stele of an Ascalonite at Athens.629 In addition, Athens controlled Delos politically, both earlier in the fourth century and starting again in 166 BCE. As such, the two sites represent two ends of a small connecting unit. This closeness is best represented by the presence of the otherwise rare Panathenaic amphora 627Meijer & van Nijf, 1992, p.57; Dürrbach, Roussel, Launey, Plassart & Coupry, 1926, #1520. 628 Thompson, 1953, p. 255. 629 Stager, 2005, p. 436. 376 with palm branch at both sites: in the coins at Athens and the mosaics of Delos. Considering all of these factors, it is thus unsurprising that both sites show such strong congruency with the motifs found on the Kedesh symbol sealings. And so, overall in the Hellenistic period, the symbols represented in the Kedesh sealings were broadly distributed to a (relatively) large number of sites with little concentration at any one site in particular. On the other hand, there are a few sites- Athens, Delos, Seleucia- at which most of those symbols motifs also appear. In addition, at these same sites, there appear motifs in common with Kedesh that are not found elsewhere, suggesting an even closer connection between these places and Kedesh. Athens and the Use of Symbols over Time: The level of congruency that exists between the symbol sealing at Kedesh and sites such as Athens allows us the opportunity to examine how the Kedesh symbol motifs fit into the wider artistic universe in specific media over time. We do this by tracking and counting the number of occurrences of Kedesh symbol motifs compared to occurrences of other types of motifs across several centuries. We can then compare the resulting picture with what we have seen emerging so far in our examination. This task is most easily accomplished using the coins from the Athenian Agora, published by Kroll and Walker in 1993. The benefits of this approach are an extremely large sample universe of more than 16 557 Greek coins630 that span many centuries from the sixth century BCE through the Roman period. Even by limiting ourselves to the last six centuries BCE, we will still have a sample universe of many thousands of examples, each one with a more or less restricted 630 Kroll & Walker, 1993, p. 1. 377 date range. In addition, the coins themselves were struck at many different places. As expected, the majority of the coins are Athenian, but other cities and polities are also well represented. As such, we will get a good image of what the overall numismatic universe was regarding the use of different types of motifs at a given site (Athens), one which has a very well demonstrated congruency with the Kedesh motifs, at any given time. Basically, we will get an image of with what categories of motifs someone in the Athenian Agora at a given time would have interacted and at what frequency. The methodology is straightforward. I went through Kroll and Walker’s publication and tallied up each coin whose dates fell within a given century, placing it in one of three categories: ones whose types display Kedesh symbols motifs; coins whose types display other symbolic motifs that fit the criteria laid out in the introduction of this work but do not appear at Kedesh, such as tripods, tridents, animal protomes or variants of Kedesh motifs like amphorae without palm branches and winged caducei; and coins whose types did not display any symbol motifs. Any coin whose dates fell into multiple centuries was counted in each century. The results are as follows. In the 6th century, out of a total of 13 coins, only one coin (7.7%) shows a Kedesh motif, seven (53.8%) show other symbols, and five (38.5%) show non-symbol types. Of these coins, the example of a Kedesh symbol motif consists of a bucranium, while the non-Kedesh symbols include a bulbous amphora and examples of four-spoke wheels.631 In the fifth century, the number of coins jumps to 130. None of these coins display a Kedesh symbol motif, though 57 (43.8%) showed other symbols, of which the greatest proportion (54 coins or 41.5% of the century’s total) consist of 631 Kroll & Walker, 1993, #1, #3-5. 378 Corinthian coins which display a trident as a type and which date from the late-5th through 3rd centuries BCE.632 The remaining 73 coins of the 5th century (56.2%) had non-symbol types. In the fourth century, the coin total again rises precipitously to 3318 of which 14 (0.4%) display Kedesh symbol motifs as types, 455 (13.7%) show other symbol motifs, and the remainder (2849 coins or 85.9%) have non-symbol motifs. The total number of third century coins is actually somewhat smaller, numbering 3195. Of this total, 46 (1.4%) show Kedesh symbol motifs, while 713 (22.3%) show other symbols and 2436 (76.2%) do not show symbols. In the second century, the picture changes markedly. Out of 3641 coins, there are 832 coins (22.9%) that display Kedesh symbol motifs, 476 coins (13.1%) that display other symbols and 2333 (64.1%) that display motifs other than symbols. This represents a large jump in the amount of coins with Kedesh symbols on them, fuelled in large part by the appearance in the second century of the Athenian coin type with a Panathenaic amphora with palm branch, representing 758 coins (20.8% of the total for the century).633 And yet, 74 coins with Kedesh motifs do not fall into this category, nearly doubling the total of 46 from this category in the third century. If we were to leave out the amphora with palm branch type, the resulting proportions of the coins would be 2.5% for Kedesh symbol motifs, 16.5% for other symbol motifs, and 80.5% for other types. In the first century, the distribution seems to reset itself somewhat. Out of a total of 5290 coins, 115 (2.2%) display Kedesh symbol motifs, 571 (10.8%) show other symbols, and 4604 (87.0%) show other non-symbol motifs. These results are summarised in the two charts below. 632 Kroll & Walker, 1993, #667. 633 Kroll & Walker, 1993, #85, #98, #101, #108. 379 The overall pattern that emerges from these results indicates several interesting trends. First, the proportion of the total that the two symbol categories represent generally decreases over time. The proportion of symbols decreases markedly from the sixth through fourth centuries. This drop may be exacerbated by the small sample sizes for the 6th and 5th centuries. However, considering the complete lack of Kedesh symbol motifs in the 5th century, it is reasonable to assume that perceived trend is generally real and that the use of symbols overall in coins was diminishing in those three centuries. This trend fits well with the pattern we have seen in the total occurrences by medium, which has shown us a diminishing overall importance of coins in the number of occurrences of Kedesh symbol motifs from the 6th-4th centuries to the 3rd-1st. In this overall trend, the 3rd and second centuries represent an anomaly. The proportion of symbol motifs here is greater than that of the fourth century, though still less than the 6th-5th. It would appear, therefore, that the numismatic presence of symbols as types increased in these centuries, before diminishing again markedly in the first century. In fact, the second century represents the greatest single proportional and numerical presence of Kedesh symbol motifs out of all six centuries, caused in large part by the appearance of the amphora-and-palm-branch type in the coinage of Athens itself. This fact illustrates an interesting point, that the behaviour of the Kedesh symbol motifs is different from that of the corpus of symbol motifs as a whole. The Kedesh symbol motifs do not diminish proportionately from the fourth century on. Instead they seem to increase, starting as a mere fraction of a percentage point in the fourth and growing steadily in the third, before peaking in the second, and then diminishing in the first century. The second century peak is present even if we don’t count 380 the Athenian amphora-palm coins, though their presence makes it much more apparent. It would thus seem that the presence of Kedesh symbol motifs was greatest in the numismatic universe of the Athenian agora contemporaneously with their presence in the archive at Kedesh. And so, the coins from the Athenian Agora indicate both that coins were a fairly conservative medium in regards to the use of symbols in general, favouring them more so earlier than later on. This fits generally with the relative importance of coins in regards to the overall occurrences we have seen above. At the same time, the coins show a reverse trend in regards to the Kedesh symbol motifs themselves. These motifs tend to increase their presence, peaking in the second century, contemporary with the Kedesh archive itself. Indeed, it is the presence of the Panathenaic-amphora-with-palm-branch coins at Athens which link to Kedesh very concretely, as the motif appears very rarely elsewhere, as well as provide a large portion of the coin presence. All of this helps to ground the Kedesh symbol motifs firmly in the second century BCE. 381 Chart 4.3: The Total Number of Coins in Motif Categories by Century at Athens Chart 4.4: The Proportion of Coin Motif Categories by Century at Athens                                                   !  " #    !  " $% !  " & & & & & & & '& & (& &                                            !  " #    !  " $% !  " 382 The Pattern of Hellenistic Geographic Distribution in Other Media Returning to Kedesh, it will also be useful to compare the geographic pattern of distribution for the Kedesh symbol sealings with that of other media at Kedesh. If other media reflect similar patterns of distribution, that provides support for delineating regions with which the Kedesh archive users interacted/communicated. One interesting point of comparison is with imported ceramic vessels.634 Here, the image that emerges is largely congruent with the geographic distribution of the Kedesh symbol motifs. First of all, there is a general lack at Kedesh of imports dating to the third century, during the time when the Ptolemies were in charge of the site. Nor is there much that can be associated with Egypt directly. In fact, the only possible ceramic import from Egypt consists of one plate, and its origin is not entirely certain. This all fits well with the lack of motifs or occurrence of motifs from Egypt. On a more positive note, Kedesh does have numerous second century imports from the western coast of Asia Minor. These come from several sites ranging from Pergamon in the north to Rhodes in the south. These include Ionian mould-made bowls, whose fabric looks Ephesian, to Rhodian amphorae with stamped handles. Pergamon, for its part is present in the form of hemispherical bowls and table amphorae. Besides finewares and transport amphorae there are also cooking vessels like pans and braziers from the same region. Altogether, this fits with the situation of motif occurrence for the Kedesh symbol sealings, which has numerous occurrences spread out across several sites in W. Anatolia. There are also several baking pans at Kedesh that come from the Aegean 634 The information for the following discussion comes from personal communication with Peter Stone (October, 2010), who is working on the Hellenistic pottery from Kedesh as the basis for his upcoming PhD dissertation from the University of Cincinnati. I would like to thank him most sincerely for all the help and instruction on the material that he gave me on this topic. Any errors are, of course, my own. 383 area in general. Several imports from Italy also appear in the ceramic corpus. Kedesh has, for example, both Campana A ware and a Caline phiale. Both these wares come from the area of Naples, which is also a prime area for occurrences of the Kedesh symbol motifs. The wider Levant is also well represented by ceramic imports at Kedesh. Specifically, there are imports from Cyprus and from northern Syria, in the area of Antioch, which is also the single site with the most occurrences of Kedesh symbol motifs in the region. Imports from North Syria include BSP vessels, mould-made bowls and a middle Hellenistic red-slip fineware (Coastal Fine North). This last has a home range that spreads up into Cilicia, specifically to Tarsus, which is one of the few sites with multiple occurrences of symbol motifs in C. Anatolia. From The East, especially Mesopotamia, there are a few sherds of Parthian glazed ware, which echoes the few occurrences of symbol motifs from the region. There is also a general dearth of ceramic imports from Macedonia, Sicily and Carthage at Kedesh. This again matches the distinct symbolic corpora from Sicily, N. Greece and N. Africa as compared to Kedesh. The one place where the pattern of ceramic imports does diverge from the pattern of motif distribution is in the region of S. Greece. No ceramic imports from this region have yet been identified, though it is the prime region overall for motif occurrences. Indeed, there was no ceramic export industry within S. Greece during the Hellenistic period. Thus whatever interaction took place between the Kedesh seal users and the S. Greece artistic tradition, it was not paralleled by trade in the ceramics from that region. 384 Interestingly, Delos is also the repository for many imports from W. Anatolia.635 As such, both Kedesh and one of the major centres of S. Greece were the recipients of much of the same ceramic influx. So, with the exception of S. Greece, the pattern of ceramic imports at Kedesh essentially mirrors that of the geographic distribution of occurrences for the Kedesh symbol motifs. With the one notable exception, those regions that have a high number of occurrences of the Kedesh symbol motifs also have ceramic imports present at Kedesh. Trade and the Routes of Motif Communication The overlap between the ceramic imports and the geographic distribution of the symbol motifs suggests that trade, as opposed to political influence or the migration of peoples, was the primary means by which symbol motifs moved between Kedesh and the larger world. Trade consists of both the commercial movement of goods, which show up in the archaeological record, as well as the parallel movement of people, namely traders, which do not. Certainly, the commercial movement of people and goods is congruent with the picture of occurrences of motifs in many separate regions and in many separate media. Some of these media, like coins, ceramic vessels, lamps, braziers, terracotta sculptures, and even jewellery, played an active part in the trade as objects of exchange. Other media, like mosaics and sculpture, which were less mobile, would have played a more passive role, being experienced by the people engaging in commercial activities in various places. Certainly, the general congruency between the regions of numerous occurrences 635 For instance, see Brunneau, 1967, #1830, #2199, #2433. These lamps, according to Brunneau, come from either Cnidus or Ephesus. 385 and the ceramic imports tends to support the theory of trade as the primary means of communication for the symbolic motifs. For instance, the general dearth of third century imports during Ptolemaic control of the site and of Egyptian imports in general reflects the low numbers of motifs coming directly from Egypt. On the other hand, W. Anatolia is especially prevalent in both ceramic imports and occurrences of Kedesh motifs. Likewise the large congruency between the Kedesh symbol motifs and Delos points in a similar direction towards trade. Delos was a major trading hub in the second century BCE. It connected important trade routes from the Italian west to the Asiatic east. We have also noted the presence of the Poseidoniasts of Berytus on the island, who may have played a role connecting Kedesh in The Levant to Delos itself, especially considering the mosaics in the House of the Trident. This fact raises an interesting possibility in regards to the Kedesh symbol sealings. While it is generally impossible to say with certainty where the user of any particular seals actually came from, given that they are un-inscribed, the Kedesh seal(ing)s do hint at the possibility of nearby and incoming people both interacting with the archive. For instance, the seal PIL3 with its modification of an otherwise Levantine version of the pilei flanking an object motif to reflect specifically Italic domestic cult points to the possible presence of an Italian who interacted with the archive on at least two occasions, as indicated by the two separate sealings. Likewise, the dolphin with anchor motif may be linked, as previously noted, to the Poseidoniasts of Berytus, who were themselves trading at least as far away as Delos. As with PIL3, ANC2 appears in two separate sealings, indicating the interaction of the seal owner/user with at least two documents in the archive. If these seals do, as suggested, represent the presence of outside 386 traders at Kedesh, then they may have been purchasing local products, such as agricultural surplus, or even bringing in the ceramic imports. On the other hand, the seal BUC1 with its post-manufacture modifications to make it less Egyptian looking seems to only make sense in the context of the change from Ptolemaic to Seleucid control of the region at the beginning of the second century. The owner, therefore, could have come from nearby; even the direct environs of Kedesh itself (see also the discussion below on the Kedesh symbol artist). The images of large official seals bearing images of plain anchors and anchors with horsehead protomes reflecting Seleucid official glyptic on Kedesh sealings certainly points to some political influence within the Kedesh symbol motifs. Still, trade seems to have predominated overall as means of motif diffusion, at least in the second century BCE when the ceramic imports are congruent with the regions of greatest occurrence. Summary: Each appearance of a given motif represents a point in the use-constellation of that motif. By studying these appearances we can see the inhabitants’ overall symbolic visual culture. We can suggest where the motifs come from, how they developed over time and from which regions they were eventually picked up by the users of the Kedesh archive for use on their seals. The pattern that emerges shows that, during the Persian period, most of the motifs were at home in Greek areas, but that there was a distinct subset that was characterized by Phoenician or Punic contexts. In the Hellenistic period, the pattern shifts so that the motifs spread out from the Greek heartlands into new and fertile grounds, such as Italy and The Levant. At the same time, the diffusion was not 387 everywhere equal and some regions picked up on the use of the Kedesh symbols more than others. This same pattern of distribution is also displayed in other media, thereby showing a congruency across multiple media and concepts. The coinage of the Athenian Agora reflects in part the temporal pattern displayed by the occurrences by demonstrating the overall decrease of coins in displaying symbol motifs in the Hellenistic period, while at the same time locating the actual Kedesh symbol motifs firmly within the second century. On the other hand, the Hellenistic pottery imports at Kedesh reflect the geographic pattern of occurrences with imports coming from many of the prime regions for motif occurrences. This geographic pattern along with other specific factors appears to favour trade as the primary means of communicating motifs. 388 Chapter 5: Observations on the Nature and Possible Meanings of the Symbols of the Kedesh Sealings Beyond the question of geographic influence, the symbols that appear on the Kedesh sealings offer other avenues to explore the users of the archive in general. Each symbol on the sealings represents the conscious choice of someone who had a document in the archive. In effect, the corpus of the symbol sealings is a sort of fossil of the aggregate mindset of the users of the archive. In examining the motifs that were chosen within the context of what was going on in the larger Hellenistic world, we will illuminate some of the ways that the users interacted with and reacted to larger artistic traditions. The Relationship between the Kedesh Archive and the Larger Artistic World of the Hellenistic Period: One of the first points to notice in the choice of symbols in the Kedesh archive is their overall popularity in the Hellenistic period. With a few exceptions, the symbols that form the corpus of the Kedesh symbol sealings appear frequently in various parts of the Hellenistic world. The regions of geographic concentration do not matter as much as the overall commonality of the symbols in general over the entirety of the region. For example, while the motif categories of New Comedy slave masks, pilei of the Dioscuri, thunderbolts, cornucopias, and caducei all have their own individual regions of concentration, they all appear in the Hellenistic period in great numbers. Some of them, like the cornucopias, pilei of the Dioscuri and the slave masks only really became popular during the Hellenistic period, appearing infrequently or not at all during the preceding 389 period. All of this tends to illustrate that the users of the Kedesh archive were well integrated into the larger Hellenistic artistic world. Unlike the Uruk archive, which had a conservative bent and used seals that hearkened back to earlier Mesopotamian glyptics, the users of the Kedesh archive did not, for the most part, limit themselves to older forms and motifs. Instead, the archive users knew what motifs were popular and/or current in the larger world and were using them for their own personal identification in legal matters. One of the best illustrations of this is the New Comedy slave mask motif. It occurs at Kedesh in no less than eight seals, thereby representing the presence of presumably eight individuals using the motif for their seals. The date of the Kedesh archive in the first half of the second century BCE coincides with the rapid growth in popularity of the art form itself. The presence of the motif at Kedesh is even more interesting considering the fact that there is no town with a theatre nearby. Thus the individuals who used the seals chose the motif for itself and its own popularity in multiple media rather than because they were enamoured with or had even seen a New Comedy play, though it is certainly possible that they did. It is also apparent that the users of the Kedesh archive were well enough integrated into the larger Hellenistic world that they were able to pick up very quickly on motifs in distant places. The best illustration of this phenomenon is the amphora-with- palm motif that appears in one seal at Kedesh (VES2). As noted, the motif appears on Athenian coins and Delian mosaics of the second century BCE, though it is otherwise rarely attested.636 The Kedesh archive users were thus able to become aware of a 636 Kroll & Walker, #88, #98, #101; Bruneau, 1972, #234. 390 relatively new motif that appeared in very specific contexts and then adapt it and use it themselves at Kedesh before the archive went out of use in 145 BCE. This is not to say that all of the archive users were on the forefront of artistic taste in their choices of seals. A couple of the seals indicate conservative choices, using older motifs. The best illustration of this phenomenon is the presence of the Phoenician masks (MAS 8 and MAS16) at Kedesh. By the Hellenistic period, the use of this motif was on the wane. The contexts where the masks had been found in previous centuries, such as funerary, were now occupied by masks derived from Greek prototypes. The use of Phoenician masks on these seals therefore represents a conscious choice to hearken back to a motif from earlier times and contexts, one with deep cultural ties to the area. At the same time, the appearance of the motif in only a couple of examples indicates conservative choices on the parts of individuals rather than any sort of artistic isolation from the surrounding world. The Symbols at Kedesh and the Efflorescence of the Use of Symbols in the Hellenistic Period: The integration of the Kedesh archive into the larger Hellenistic artistic world, at least in regards to symbols, also reflects a separate widespread phenomenon. Specifically, the number of symbol sealings and the number of different motif categories help to illustrate a growth of the overall use of symbols in the Hellenistic world. First, the number of sealings at Kedesh that show symbols is fairly significant, representing around 6% of the 2043 sealings recovered. Indeed, some of the motif categories within the larger corpus, such as those of masks and pilei of the Dioscuri represent 1% of the total number 391 of sealings each. Furthermore, there are fifteen separate motif categories within this corpus, many with their own variants and subtypes, all of which indicates that the use of symbols was a vibrant and varied portion of the larger glyptic universe, in which the users of the Kedesh archive participated in the first half of the second century BCE. The proportion of symbols on the sealings from Kedesh places it in the middle of the various archives in regards to symbol use. On one hand, symbols only represent 1.6% of the total impressions at Uruk, appearing in only 24.637 On the other hand, at Seleucia, there are 4652 sealings bearing images of symbols in the large public archive, representing 15.5% of the 30 000 sealings. The motifs that form the Kedesh symbol sealings also show a marked increase in use during the Hellenistic period. As noted, a few motif categories such as cornucopias and pilei of the Dioscuri only began to appear in numbers during the Hellenistic period. In the case of plain pilei of the Dioscuri, for instance, the total number of appearances in the 6th-4th centuries BCE is two compared to 26 for the third through first centuries. In some cases, the Hellenistic period also sees the motif spread out into media in which it had not previously appeared. Thus, for the plain pilei, which had appeared only in coins during the sixth-fourth centuries BCE now appear in media such as glyptics and sculpture. Even motifs that have a long history, such as thunderbolts, lyres, and bucrania show a similar phenomenon of growth in numbers and media. Thus the use of lyres as symbols changes from a total of twenty-three occurrences limited exclusively to numismatics in the 6th-4th centuries to sixty occurrences that, while still predominantly numismatic in nature, include examples in glyptics and sculpture during the Hellenistic. 637 Wallenfels, 1994, pp. 134-7 & 143. 392 Thunderbolts show a similar pattern. In the preceding period, the motif of the plain thunderbolt that is used as a symbol appears fairly frequently with a total of 20 occurrences. These occurrences consist mostly of coins, but the motif also appears at two of the archives. In the Hellenistic period, occurrences of the motif increase drastically to a total of 51. The motif now also appears in different media, including vessels and sculpture. The winged thunderbolt variant likewise sees a manifold jump in representation in the same period, going from 4 occurrences in the Persian period to 29 in the Hellenistic. With its long history, the plain bucranium motif behaves in a slightly different fashion. In the 6th-4th centuries BCE it is well represented with 23 occurrences divided amongst several media, including glyptics, numismatics and vessels. However, even here, the Hellenistic period saw the total number of occurrences more than double to 57, with a greater spread into media such as sculpture and mosaics. And so, in the Hellenistic period we see in the Kedesh symbol sealings a reflection of a phenomenon consisting of both the creation of new symbol motifs as well as of an overall increase in the use of previously existing motifs, which have also been spreading into previously unattested media. Altogether, it is clear that symbols were becoming more widely used and accepted as opposed to previously. This phenomenon does not seem to have been limited geographically, but rather appears to have encompassed much of the ancient world during this period. The growth in the use of symbols can be explained by their utility as visual expressions of a number of conceptual categories that characterize the thought and worldview of the Hellenistic age. These will be discussed under two large rubrics: allegorical and amuletic uses. 393 Allegory One factor behind this rise lies in the contemporaneous rise in the use of visual allegory in general, as Onians discusses in his book on Hellenistic art and thought.638 According to him, the interest in visual allegory, in expressing a message other than what was apparently first intended, grew out of early fourth century Greek intellectual development, namely in the works of Plato and Aristotle on language and senses, which encounters with other cultures, notably the Egyptians with their complex and highly visual hieroglyphic writing system, had encouraged. The result was an effort on the part of Greek intellectuals to bridge the gap in visual and verbal communication, to create in effect a kind of visual language, one where images could express a wide variety of information as words do in verbal language. One aspect of this idea comes in the rise in the use of personifications, wherein more or less abstract concepts are expressed as anthropomorphic divine beings, of which the most famous of Hellenistic example is the Tyche of Antioch by Eutychides. Onians notes that the identification of these personifications was based on pose and attribute so that Eutychides’ statue of Tyche, for instance, was geographically set in place by the presence of the mural crown and by her being seated over another personification of the river Orontes.639 The figures of the seasons, on the other hand, were identified in art by carrying the right representative fruits.640 This phenomenon forced the artists of the time to study new poses and objects in 638 Onians, 1979, pp. 96-118. 639 Onians, 1979, p. 99. 640 Onians, 1979, p. 100. 394 order to communicate the identity of the growing list of personifications.641 One result was that the identity of a figure came to rest in its attributes. The symbols at Kedesh and elsewhere reflect this phenomenon. The growth in the use of symbols is part of the overall artistic growth in the Hellenistic period of visual allegory and the need to use and create objects in visual media to express the various allegorical messages. Many of the Kedesh motifs are attributes of different deities: thunderbolts for Zeus, cornucopias for Tyche, pilei for the Dioscuri, the caduceus for Hermes. In representations of these figures, they would be the means of communicating the identity to the viewer. On the sealings they have transcended the need to actually have the figure. Instead, the attributes represent the beings.642 The variants on the motifs appear to be the natural result of artistic exploration of objects for allegorical purposes. Onians also notes that the Hellenistic interest in a visual language gets expressed in Hellenistic poetry. He expressly points to the epigrams of two second century poets, Antipater of Sidon and Meleager, where the objects displayed on tombstone carvings are discussed for what they mean about the deceased.643 In one poem, Antipater describes an un-inscribed tombstone with nine dice carved in relief, representing three different named 641 Onians, 1979, p. 101. Another prime example of the use of personifications and attributes comes from the third century Grand Procession of Ptolemy Philadelphus, for which see Rice, 1983. This grand pageant was filled with allegory, including many personifications and their attributes. It started and ended with processions of the personified morning and evening stars (p.28) that thus functioned as temporal markers for the procession. Other temporal personifications that are mentioned (see pp. 49-51) include Penteteris (The Five-Year Span) with persea crown and palm branch, Eniautos (the Cyclical Year) with a cornucopia, and the four Horai with their representative fruit. 642 This idea of anthropomorphic figures being potentially unnecessary in divine representation is also reflected in the recount of the third century Ptolemaic Procession. For instance, see Rice, 1983, pp. 116-8. Here, Rice discusses a point in the procession where five chryselephantine thrones pass by. Four of the thrones bear divine attributes, while the fifth bears a royal crown. Rice takes the empty thrones to represent the sovereignty of the absent gods (and king) whose arrival is meant to be imminent and whose attributes should serve as identification, though the actual attributes, such as the cornucopia, are those used by several gods and are thus more generalized. 643 Onians, 1979, pp. 110-1. 395 throws: the “Alexander”, the “Youth”, and the “Chian”, from which he interprets that the tomb belongs to a youth called Alexander who was originally from Chios.644 In other poems in this genre, Antipater goes on to describe tombs where various objects such as whips, bows and cups are used to illustrate the characteristics of the deceased.645 Meleager for his part describes a (fictional) tomb of Antipater himself as bearing depictions of a cock with a sceptre, a die, and a palm, which he takes to mean that Antipater was a man who made himself heard (the cock), that the sceptre was an emblem of his profession as a poet, that he was originally born in Tyre (the palm) and that he died in his cups (the die).646 Both poets therefore see in the types of objects that appear on those tombstones, or for our purposes appear in the Kedesh sealings, as able to communicate a great deal of information concerning the individual to whom they belong, just like written language.647 The poetry of Meleager and Antipatris reflected what was happening in regards to actual tomb markers. One prime example of this consists of a funerary stela of one Shem/Antipatris the Ascalonite from Athens, dating to the third-second centuries BCE. The stela bears dedications in both Greek and Aramaic, in which the owner has a Phoenician name in the Aramaic and a different Greek name in the other, and a Greek epigram to explain the imagery of the stela. The imagery itself is highly unusual and shows definite use of visual allegory. The scene shows the body of the deceased on a bier over which the figure of a friend fights with a lion for possession of the body. Behind and above the figure of the friend, there is the depiction of a ship’s prow. Jennifer Stager, who 644 The Greek Anthology, Volume 2. #427. 645 Onians, 1979, p. 111; The Greek Anthology, Volume 2. 1953, #423-6. 646 The Greek Anthology, Volume 2. #428. 647 Onians, 1979, p. 111. 396 published the stele, notes that the epigram fails to adequately explain the imagery. She takes the lion and ship’s prow to be referring to the cult of Astarte, the Phoenician goddess whose attribute was the lion and who had dominion over matters such as seafaring and death. As such she sees the stela as representing the monument of a bilingual Phoenician who died on a sea voyage.648 Interestingly, Shem/Antipatris, Antipater and Meleager all came from second century Phoenicia, from Ascalon, Sidon and Tyre respectively. They were thus neighbours geographically and temporally to the users of the Kedesh archive, and probably shared general cultural and intellectual contexts. Onians himself notes that the Phoenicians were very well placed to explore and make use of visual allegory as a means of communication, since they had a long history of trading not only with the Greeks, but also with the Egyptians who had long used a pictorial script, namely hieroglyphs, with strong allegorical and enigmatic tendencies.649 It is not surprising then to find the users of the Kedesh archive playing their role in this artistic development of the Hellenistic period. The Economy Beyond this trend in artistic development towards visual allegory, other factors no doubt played some part in the rise of symbol use in the Hellenistic period. One possible set might be political/ economic in nature. The Hellenistic period saw the emergence of many new polities, both on the level of the large kingdoms like the Seleucids, the Ptolemies, and the Attalids as well as on the level of new civic foundations 648 Stager, 2005, pp. 427-449. 649 Onians, 1979, p. 113. 397 like Antioch and Alexandria with the concomitant rise in the number of mints and civic as well as imperial issues. Even long established cities might mint coins in order to pay off various expenses, such as the hiring of mercenary soldiers. The expanded operations led to the production and circulation of more and different coins. Within this phenomenon, the use of symbols played a significant role, which would in turn spread the motifs. Indeed, most of the symbol motifs at Kedesh are well represented in coins of the Hellenistic period. Interestingly the numismatic material from the Athenian Agora shows a small resurgence in the proportional use of symbol motifs in the 3rd and 2nd centuries after a decline from the 6th-4th centuries BCE, with the actual Kedesh symbol motifs peaking dramatically in the 2nd century. Furthermore, while the proportion of coin Kedesh motif occurrences within the whole was generally less in the Hellenistic period than in the previous period, the actual count of coin occurrences was much higher. One reason behind the numismatic use of symbols in general lies in their communicative flexibility. A thunderbolt, for example, could stand in for Zeus, but it did not necessarily indicate a specific cult of Zeus in a specific city, and it could also refer to various other aspects, such as the power and might of the god. Similarly, a cornucopia might refer to a goddess like Tyche or more generally to the prosperity and fertility of the city and its territory. Thus the symbol motifs could be imported and moulded to different purposes, offering a kind of conceptual leeway that would make them attractive to polities that did not have a long minting tradition with their own repertoire of images to use. However, the conceptual malleability of the symbols also allowed specificity if the polity wanted it. This is the case with amphorae in that, while many places might have minted coins with an amphora on them, Athens in the 2nd century used the Panathenaic amphora 398 with palm on its Hellenistic coins to refer back to a specific local festival.650 Interestingly, this new motif spread out from there, as far as Kedesh with VES2, to come to represent more generically athletic victory. The conceptual leeway available in symbols clearly proved attractive to the great dynasties of the Seleucids and the Ptolemies. Both controlled geographically large empires containing multiple ethnic groups, each with their own histories, stories and worldviews. By using symbols such as the anchor (for the Seleucids) or the cornucopia (for the Ptolemies), these powers could communicate potent messages through their coins that could be understood immediately by many people and/or explained quickly to others. Indeed, conceivably the conceptual leeway available in the symbols meant that the audience of the coinage would not have to know all of the associations of a given symbol in order to understand at least a part of the message that was being conveyed. A good example of this phenomenon is the cornucopia itself. A subject of the Ptolemies need not, for instance, know about the story of Amalthea or its use on a royal statue in Alexandria necessarily to understand that the horn represented fertility and abundance with its overflowing fruits, and therefore the prosperity associated with the royal figures on the obverse sides of the coins. Interestingly, the Ptolemies and later the Seleucids used the motif at multiple mints in and around Phoenicia proper, which, Onians notes, was especially primed for the use of visual allegory. In the case of another motif, a person in the deserts of Bactria would not have to know anything of the nautical associations of the anchor to associate it with the dynasty on whose coins it appeared. 650 Head, 1878, #618, #621, #629, #631; Kroll & Walker, 1993, #98, #101, #108. 399 For their part, the Seleucids, especially the later ones, tended to favour the use of symbols on lower denomination bronze and copper coins as opposed to silver and gold coins where they appeared much more rarely.651 These smaller coins would have had a swifter rate of turnover than their more precious cousins, being used for a variety of small-scale transactions between the minting polity and individuals and between individuals themselves. Meanwhile, the larger denomination precious metal coinages stood a better chance of being kept for longer periods of time, collected to pay for large- scale expenses or hoarded for a rainy day. The motifs that appear on the lower denomination currencies therefore passed more easily through numerous hands and thereby communicated the symbol motifs with their inherent conceptual flexibility and multiple meanings to a larger audience. We have already seen, however, that coinage forms only one part of the overall mix of media that use symbols in the Hellenistic period. Indeed, symbols spread into many different media, where they had not appeared previously. Their appearance within these media often shared many of the same as those described in Hellenistic numismatics. Thus, in the third century, the bucranium appeared in sculptural form on the Stoa of Antigonos Gonatos on Delos as well as the Arsinoeion and Propylon of Ptolemy II on Samothrace.652 The same conceptual leeway and the general acceptability of the symbols promoted by various Hellenistic powers and cities likewise made them attractive for 651 Gardner, 1878, Sel I 41-2, 71a, Ant I 41-4, Sel II 31, Ant III 35, Ant IV 83-4, Sel IV 26, Dem I 36-42, Al I 15, Al I & Cleopatra 85, Dem II 25, Ant VII 46-8, 69-70, Al II 30-1, Ant VIII 29, Ant IX 16-9, 32-4, Ant X 3-6, Dem III 7. These coins represent examples of royal issues that bear symbols found at Kedesh, namely the anchor, the bucranium, the cornucopia, the lily, the ship’s prow, and the thunderbolt. Of these examples, only the anchors of Seleucus I (Sel I 41-2), the cornucopia of Demetrius I (Dem I 36-42) and the thunderbolt of Alexander I Balas (Al I 15) were in silver. All of the rest were done either in bronze or in copper. 652 Webb, 1996, pp. 139 & 148-139. 400 individual use. Thus the cornucopia becomes popular in Attic West-Slope ware of the third century, possibly as the result of Ptolemaic influence.653 Likewise, in the third- second centuries, the caduceus, attribute of Hermes the god of commerce, appears on Thasian, Rhodian and Cnidian stamps on commercial amphora handles.654 And so, the rise in the use of symbols can be explained by a number of complementary trends and forces that came together during the Hellenistic period. There is the development of intellectual and artistic interest in using allegory forcefully in visual media as a communicative tool. This helped smooth the reception of symbol motifs, by making them more acceptable: now it was understood that the part could legitimately stand in for the whole, such as a god’s attribute standing in for the god himself. Furthermore the cosmopolitan nature of the period and the rise of many new minting polities, with their need for motifs that had the necessary conceptual leeway for communication, also aided in the spread of the symbols by making them available to a wider audience. Kedesh itself is well placed in regards to both of these larger trends. As a Levantine site, it was in the very midst of one of the prime areas in the development of the use of visual allegory and the ideas of a visual language. At the same time, being at one point a Ptolemaic possession and then a Seleucid one, Kedesh was also well situated in a cosmopolitan milieu amidst new foundations and polities to experience the numismatic developments that were happening around it. Beyond numismatics, the rise in acceptability of using symbols and their inherent flexibility also made them attractive for 653 Rotroff, 1991, #3, #30, #41-3, #48-51. 654 Grace, 1934, #9-10 & #125-9. 401 use in other media and situations. The symbols of the Kedesh sealings are in certain ways thus reflections of all of these complementary trends. Amuletic uses Onians notes that visual allegory was popular in sepulchral contexts, as the poems of Antipater and Meleager and the stele of Shem/Antipatris the Ascalonite all indicate, because allegory with its hidden meanings and mystery in effect gave it a kind of mystical or magical potency which made it appropriate for such liminal contexts as tombs.655 Symbols may be readily understood as possessing a kind of mystical power, since a single characteristic may reappear frequently in different motifs. Essentially, many categories of symbols, especially amongst the Kedesh motifs, express certain apotropaic amuletic functions, and could be useful in both averting harmful influences and promoting beneficial ones for the wearers of the seals. On one hand, this aspect of seals is not surprising. Considering their close association with an individual and his or her identity and their ability to bind an individual to transactions or property, seals themselves innately possessed amuletic functions which might equal or even surpass their administrative function.656 Indeed the ancients were quite explicit on this fact. Indeed, 655 Onians, 1979, p. 114. 656 See Ben-Tor, 1989, p. 10; Collon, 1987, p. 119; Boardman, 1997, p. 80; Platt, 2006, p. 235. Ben-Tor points out that the primary function of the scarab was as an amulet, which could occasionally function as a seal. Boardman also notes that the Greek seals could function as amulets, a function which he ascribes to eastern origins and which he sees for the Classical and Hellenistic periods as being secondary to their use as seals. Collon, for his part, notes that cylinder seals in the ancient Near East possessed an important protective function that paralleled their more mundane legal function as a seal. Meanwhile, Platt points out that in the Graeco-Roman world seals continued to have similar function, with special protective properties being ascribed to the choice of stone and image. 402 Pliny, for instance, derisively notes the contemporary beliefs that some people had in regards to the magical properties of some engraved stones, saying: The Magi falsely claim that the amethyst prevents drunkenness, and that it is this property that has given it its name. Moreover, they say that, if amethysts are inscribed with the names of the sun and the moon and are worn hanging from the neck along with baboon hairs and swallows feathers, they are protection against spells. Again they assert that, however they are used, amethysts will assist people who are about to approach a king as suppliants, and that they keep off hail and locusts if they are used in conjunction with an incantation which they prescribe. Moreover, they have made similar claims on behalf of the ‘smaragdus,’ provided that it is engraved with an eagle or a scarab beetle.657 The Kedesh symbol sealings show an additional level of preoccupation with the aversion of harm so that the choice of the symbol that appears reinforces the inherent protection offered by the seal. This amuletic function existed in parallel to the use of the seal to make documents for legal purposes, as well as any other function of the seal, such as for jewellery. The Kedesh symbols seek to avert harm and promote the benefit of the seal wearer through the use of specific methods that I have named deflection, repulsion, allegiance or wish. Individual symbols can express any of these methods, or even combinations of one or more methods, all in their efforts to increase the protection available to the wearer. 657 Pliny, Naturalis Historia 37.124. 403 Deflection The idea behind the deflection method is that the symbol acts as an alternate target for the harmful influences that would normally attack the wearer. Instead, the evil is deflected from the seal owner and drawn to attack the symbol itself. Perhaps the best examples of this idea at Kedesh are the New Comedy slave mask seals, in which deflection is inherent in their very nature. As an object the actual mask is a disguise which allows the user to assume another identity, usually in the context of theatre. This alternate identity in turn forms another character or persona, by which the wearer of the mask can act differently than one would normally do and even transgress the norms of behaviour. Thus, the alternate persona can become, in effect, the scapegoat for the person wearing the mask, allowing him to transgress the boundaries of normal behaviour without becoming the target of the normative retributions for those transgressions. These same ideas may underlie the appearance of masks on seals. The image of a mask can create an alter ego, thereby obscuring the target’s identity for the harmful influence and, hopefully, induce it to attack the image of the mask itself as an alternate, instead of the owner. Indeed, the mask itself may have been in this light inherently apotropaic since one of the possible reasons for theatre actors to have worn masks in the first place was so as to disguise themselves and thus confound the effects of any possible ‘evil eye’ in the audience.658 In this light, it is interesting to note the popularity, both at Kedesh and elsewhere, such as Delos and Seleucia, of the comic slave mask on seals. In the New Comedy popular during the Hellenistic Period, the clever slave characters were popular stock 658 Elworthy, 1958, p. 158. 404 characters. Their frequent appearance on seals can be seen on one hand as a continuation of this popularity. On the other hand, the various clever slave characters are also the ones in New Comedy who transgress the perceived standards of behaviour the most by plotting against their masters and other free men. Also, as slaves, they represent the characters who can most freely be attacked or threatened. For example, a slave can be beaten or whipped, or threatened with such actions. Indeed, some of the New Comedy scenes from Delos show the actor as slave hiding on an altar as sanctuary from his master’s threatened beating.659 It is a simple transference from the slave as target for his master’s violence to the image of the slave through the mask as the target for the harm that would normally affect the seal user. In addition, images of a generally extraordinary or marvellous nature also had prophylactic qualities against supernatural harm because they would attract naturally attract the attention and gaze of the malefic forces away from the person.660 As such, the idea of deflection might also play a role in the seals that show composites. Composites are by their very nature visual puns, combining multiple elements in odd ways. They are thus meant to be interesting and visually arresting images that would hopefully attract the attention of baleful gazes. Repulsion Another amuletic function that the symbols on the Kedesh sealing use is to present an image that would not merely deflect the attention of the harmful influence from 659 Boussac, 1997, 151, fig.26-8. 660 Hildburgh, 1948, p. 164. 405 the owner but to repulse or frighten away the influence as well. This effect can be done in several fashions. In regards to the bane of the evil or envious eye, one of the most well known methods is to outstare it. Hence, eyes appear as wards in various ancient cultures like the Egyptians, Greeks and Etruscans.661 In this vein, the images of masks come again into play. The images of masks, which are generally shown frontally or nearly so, act not only as alternate personae for the owner but as alternate faces as well, out-staring the baleful influences which would seek to do the seal owner harm. The images of composites which have a staring face as the body of an eagle would serve a similar function. Those seals which display bucrania functioned in a similar fashion as they display the frontal view of a bull’s head, staring out perpetually against evil. Another version of the idea of the affronted face would be the appearance of ears as symbols on the Kedesh sealings. The presence of an ear would have acted to ward off the harm from evil speech as an eye would for evil glances.662 An interesting, if much later, version of this is the appearance of both protective eyes and ears on the harbour defences of Valetta, built by the Knights of Malta.663 Besides the staring face, the idea of repulsion can be found in what might be generally thought of as images of power, which protect the wearer by placing him behind the aura of the image. Again, here we have those Kedesh seals showing bucrania, which rely on the power of the bull to express protection for the wearer. Here again, the image is shown frontally, placing the offensive weaponry of the bull, namely its head and horns 661 Moss & Cappanari, 1976, p. 3. For examples of the phenomenon of eyes appearing on Greek ships, see also Forsythe Johnston, 1985, Arch. 30, Arch. 32, Clas. 1, Clas. 4, Hell. 1, & Hell. 2. These examples are all terra cotta vessels or lamps dating from the early archaic through Hellenistic periods in the form of warships or warship prows with warding eyes or even boar’s heads. 662 Elworthy, 1958, p. 14. 663 Moss & Cappanari, 1976, p. 12. 406 between the seal user and harm. In the same category of repulsion, we should also consider the composites showing hippocamps. The hippocamp represents a potent mythical sea creature with potential ties to the afterlife.664 Its representation therefore represents an attachment to a source of power which the owner can co-opt and use to repulse harm. Also, seals that display a shield could stand in for a real shield protecting the wearer or owner from intangible harm. Similarly, the seals that present the image of a single star might be a sort of generic profession of power, in that the star provides illumination, guidance, and, hence protection for the wearer. Allegiance I term the third amuletic method used in the Kedesh sealings as allegiance. Here, the seal owner is using the image on his seal to put himself under the protection of a specific higher power, specifically a god. This is done in the symbol sealings by displaying a symbol of the particular god from whom protection is being sought, or whose protection is being advertised. Interestingly, the choice in displaying a symbol of a god whose protection is being sought or advertised seems to favour a deity with whom a personal connection on an individual level could be made over those of a more civic nature. The most striking example of allegiance within the Kedesh symbol sealings is the seal PIL3, for which there are two impressions, and which shows two pilei topped by 664 Holloway, 1986, 447-452. In this article, Holloway relates the appearance of a youth riding a hippocamp in the pedimental frescoes to the scene of the death of Troilos elsewhere in the tomb by identifying the youth and hippocamp as the voyage of Troilos to the afterlife. Holloway also notes the general apotropaic characteristic of hippocamps, appearing as they do in the same contexts as gorgons in Etruscan tombs. 407 stars, usually associated with the Dioscuri, but which here flank the bearded snake of a genius loci. As previously noted, this layout has the hallmarks of Italic lararia. If the pilei do here represent the lares accompanying a genius of place, than the seal would seem to represent a sort of portable lararium, representing that most personal set of divinities for an ancient Italic person: those of his hearth and home. The seal wearer would, in effect, be placing himself under the continuous protection of his own household gods. The images of the pilei in general maintain this kind of personal link. The Dioscuri, whose attributes the pilei were, were themselves deities that were seen to possess special and intimate connections to their worshippers. According to myth, they spent half their time in the land of the dead with the shades of mortals and half with the other gods on Olympus. They were also identified with the Kabiroi or Great Gods of Samothrace, who were the focus of a mystery cult and were seen as being the great patrons and benefactors of sailors. As such, their protection as noted by their symbols, normally their pilei, would be especially appropriate to display upon one’s seal, especially by those very same seafarers. This close individual and protective connection might also help to explain why there are 15 seals with pilei at Kedesh, making it the second largest category overall next to masks. Other divine symbols, beside the pilei, also exist. For instance, there are seven examples of sealings that show the thunderbolt of Zeus. There is also one example bearing the caduceus of Hermes. In both of these cases, the cults of these gods were often more civic in nature and the intimate connection with the seal owner is not as apparent as in other cases. This might help explain the lower numbers of their symbols on Kedesh seals. However, certain aspects of these deities do lend themselves to direct and personal 408 connections with the person seeking their protection, thereby explaining their presence. As king of the gods, Zeus represents the most potent single force that could be called upon for protection. At the same time, one of his well known cult epithets was that of Soter (Saviour), indicating his most active interest in protecting a variety of different groups of people. Interestingly, the composite eagle, as in the sealings COM1-2 and COM5, is also probably making reference to the Zeus, since the eagle was one of his attributes, and to his protectiveness. The presence of the mask in the eagle’s breast, for instance, would have acted as an alternate persona of the seal user, thereby concretely putting him under the protection of Zeus himself. Hermes, on the other hand, was both guide of the dead to the underworld (Psychopompus) and the patron of commerce and commercial success, functions which stem from different aspects of his role as a guide to and protector of mortals.665 His interests are therefore both intimately connected to individuals and well placed to act for their benefit. There is also a possibility that the caduceus need not refer specifically to Hermes. The caduceus is sometimes found associated with the symbol of the Phoenician goddess Tanit. In certain of these cases, the two symbols are actually combined into one. Berges, among others, has therefore postulated that the caduceus may be in fact another version of the Tanit symbol.666 As such, its appearance here may be a reference to either Hermes, or Tanit, with her local origin and links to fertility, or perhaps both at the same time, thereby multiplying the protective power associated with the image. 665 Chittenden, 1947, p. 90. 666 Berges, Ehrhardt, Laidlaw & Rakob, 1997, #820. 409 As this last example indicates, in certain cases, the images referencing gods need not be altogether clear and/or specific. For instance, the corpus possesses seven total sealings of cornucopias, representing the use of six separate seals. The cornucopia is an attribute often associated with certain goddesses like Tyche and Demeter, both of whom possess the intimate connections, one through personal luck and the other through her patronage of mystery cults, which would make them ideal as protective goddesses. There is, however, no way of identifying which one is represented, if indeed the specificity is required. The seal wearer might want to reference either Tyche or Demeter or both together depending on the circumstances. At the same time, the bull is a symbol of Ba’al, who is himself linked to Zeus and Poseidon. The presence of what might be either a ship’s ram or rudder above the bucranium in BUC6 would indicate the latter in that seal. Likewise the images of single stars might also be a generic reference to divine protection, such as to Venus/Aphrodite/Astarte, without necessarily needing to be specific. In each of these cases, however, the specificity of explanation is not there to make us certain. It may be one explanation or another, or possibly all of the explanations at the same time. 667 Wish The final amuletic method or mechanism, which can be termed as a wish, takes this sort of generality into account. Here the idea is to counteract any potential harmful influence by using images of general benefit, such as wealth, leisure or fertility to promote the benefit of the wearer. Such images represent more generic symbols whose 667 A more modern example of the phenomenon of allegiance being used apotropaically would be the use of images of important local saints by rural Italians in the 20th century to protect against evil forces, especially the evil eye. See Moss & Cappanari, 1976, p. 11. 410 presentation informs a desire to possess similar attributes or to exist in the type of beneficial atmosphere that the symbol or object indicates. In essence, it is an attempt to use sympathetic magic to cause like to produce like.668 For instance, while the cornucopia as noted above can be considered a symbol of a specific deity, it is also in itself a symbol of fertility and abundance which would be beneficial to the owner of a seal with its representation. Likewise the representations of flowers, of which there are five seals with six total impressions, may represent this idea of fertile prosperity which the seal owner hopes to gain or maintain. Images of vessels may also point to the good life and prosperity. There are two basic groupings in the category. The first consists of the motif of the Panathenaic amphora with palm branch, representing athletic victory with its resulting glory and honour. The second grouping has images of vessels associated with drinking wine, namely a kantharos and a lagynos. These vessels are linked to the world of the symposium and to leisure. Their representations, therefore, might have sought to evoke such qualities in the life of the user. Along the same lines, there are the images of lyres. Such instruments produced beautiful music, which might be enjoyed during times of leisure. The image of the lyre might be an evocation of such qualities in the life of the user. On the other hand, the seals with images of ears display directly an apparatus of the human senses. The image of the ear is tied ideologically to the individual’s hearing in different connotations. As such, the image of the hand holding an ear (EAR1), which in 668 Ellworthy, 1958, p. 70. Ellworthy here is predominantly discussing the use of sympathetic magic to cure diseases, such as in the use of the “Hair of the dog that bit you.” He does, however, note that the use of like to produce like is a standard practice in protecting against the evil eye. 411 later glyptics usually has an inscription “MNHMONEYE”, is here likewise being used as an exhortation to the user to remember. The images of single ears could also represent a wish for alertness or eloquence, with the increase in security and prosperity such traits would bring, or even simply a wish to hear and be heard. An extreme version of the idea of the wish is the composite eagle. The presence of the mask linking the alternate persona of the seal wearer to the protection of the god could go even further. The presence of the face, notably beardless, in the center of the eagle may also be a reference to the myth of Ganymede, the young Trojan prince whom Zeus in the form of an eagle abducted and brought to Mt. Olympus and immortality. Furtwängler identifies the image of the sard in Berlin that closely parallels the composite eagles of the Kedesh sealings as a representation of the same myth.669 In fact a more explicit scene of the myth also appears on another sealing from Kedesh (see figure 5.1). The composite eagle in and of itself could represent a scene of apotheosis and therefore function as a reference and wish for the easy life of the blessed immortal gods. 669 Furtwangler, 1900, XXVI.71. 412 The Combination of Amuletic Methods As already noted, the conceptual leeway inherent in the symbols themselves means that a single image need not use only a single amuletic method to protect the owner from harm. Instead, it could use a number of methods in conjunction, each one being equally valid and important as another. The best illustration of this phenomenon within the Kedesh sealings lies with the composite hippocamp COM8 in which an Eros rides a hippocamp formed from a horsehead protome for a head, a satyr masks for a body and a cornucopia for a tail. Here, the very image of the creature with its ties to the underworld and potency in myth points to the use of repulsion. The figure of Eros who, as the powerful god of love and desire, has direct ties to the lives of individuals upon the creature’s back bespeaks of the use of allegiance. Furthermore, the specific elements that form the hippocamp also use different amuletic methods. First, the satyr mask itself invokes different methods. First, the mask uses deflection, with the mask functioning as an alternate target to the seal user. The satyr mask also functions as a scary or powerful image to ward off danger, thereby using repulsion. Again, the satyr, as a traditional companion of Dionysus, may be making reference to the cult of that god and, therefore, use the method of allegiance. Finally, the cornucopia on the tail, with its connotations of fertility and prosperity, uses the wish method as well as possibly allegiance in reference to one of the deities such as Tyche, Tanit or Demeter (or possibly all three) that are associated with it. As such, this one image uses a variety of tools in an effort to bring as much protection to bear for its owner as possible. For the grand majority if not all of the symbol motifs that appear on the Kedesh sealings, at least one if not several methods to ward off harmful influence and promote the 413 benefit of the user of the seal may be at work. One of the guiding principles behind the choice of a symbol for a seal at Kedesh may well have been a preoccupation with the potency of the seal as an amulet. Summary: As artifacts of a conscious choice on the part of someone who left a document in the Kedesh archive, the motifs that appear in the Kedesh symbol sealing help to illustrate several points of the artistic and intellectual context in which the archive existed. First, the motifs are, for the most part, those that are current and popular in various places during the Hellenistic period. The users of the archive knew what was going on in the larger artistic world and were well integrated into it, while the few conservative examples using older motifs reveal the existence of some individuals with specific local cultural ties. Second, the use of symbols at Kedesh was part of a general Hellenistic trend which saw the growing popularity of using symbols. This trend is reflected not only in the appearance of new symbols during the period but also in the increased use of older symbols and their spread into new media. This rise of symbol use was fuelled by several factors, including a concurrent growth in popularity of visual allegory and the rise in general numbers of minting entities that helped to make the use of symbols more widely acceptable. Third, within this larger growth in popularity, the users of the Kedesh archive at least sought to use symbols for specific purposes further than simply as an identifier of an individual. Specifically, the users chose the various motifs seemingly to ascertain the potency of the seal in warding off harmful influences such as the evil eye. 414 Chapter 6: Conclusion The symbols that appear on the 125 sealings (representing the use of 91 seals) from the Kedesh archive and that are presented in this work are not simply ornament or decoration. The various motifs, be it an image of a New Comedy slave mask, a thunderbolt, an image of the pilei of the Dioscuri, or one of the other motifs that form the corpus represent the same decision on the part of someone, in the specific context of Hellenistic Kedesh, to use that particular motif as an identifier of his or her self in a document, just as with any other motif that appears in the archive. As such, the symbol motifs encapsulate a great deal of information for us to uncover. This includes information both on the transmission of the motifs as well as on their use and meaning. For each motif, its presence on sealings at Kedesh represents one point in a larger use-constellation composed of all of the appearances of that motif in various places and media through time. In combining these use-constellations we create a pattern of how and where the totality of the motifs is used over time, essentially being the visual culture universe of the symbols at Kedesh. The pattern that emerges shows that the dichotomy of Greek and Phoenician is too simplistic overall to describe the overall richness and mutability of the distribution. Different motifs appear in differing strengths in various regions. The pattern does demonstrate a shift of concentration from the sixth through fourth centuries to the third through first. In the earlier period, the symbols occur primarily in the Aegean region with a subset popular in the Levant and North Africa. Most of the occurrences from this period come from coins. In the later period, the number of occurrences skyrockets, spreading into new media such as ceramics, sculpture and 415 mosaics. Parts of the Aegean continue to be important areas of motif concentration, but now new regions emerge as major contenders in the use of these motifs. Among these regions are the Italian peninsula, especially in the southern areas, and the Levant, which has grown noticeably in regards to its use of symbols. Indeed in the case of certain symbols, such as the cornucopia, the region appears to be at the forefront of innovation and development. Furthermore, the ties that once bound the region to that of North Africa have, by this time, fallen into abeyance and the two regions have diverged in their use of symbols, with the Levant taking one path, and North Africa in general and Carthage in particular taking another. At the same time, there is very little connection between the Kedesh symbol motifs and Egypt, though the Ptolemies who controlled that region also controlled Kedesh during the 3rd century. This pattern as a whole is largely borne out again by the imported ceramics at Kedesh. The site has numerous imports from the northern Levant and the west coast of Asia Minor as well as some from Italy, but little from nearby Egypt. This congruency between the ceramic imports and the occurrences of the Kedesh motifs also suggests that trade played an important role in the communication of the motifs between regions. The material from Delos seems to support this possibility in that it was both a Hellenistic trading hub and likewise demonstrates an enormous congruency with the Kedesh motifs, including producing occurrences of motifs like the anchor and dolphin and the composite hippocamp that are rare elsewhere. Indeed, the presence of the anchor and dolphin motif at Delos also suggests the possibility that a specific group, possibly the Poseidoniasts of Berytus, who were active both in the Levant and Delos, may have interacted directly with the Kedesh archive. 416 Beyond simply indicating geographic distribution, the symbols that appear at Kedesh also demonstrate many other interesting points by their very presence. First of all, the motifs at Kedesh in general are those that are both current and popular over large areas of the ancient world. Some of the motifs really first appeared during the Hellenistic period. Only a couple of motifs, such as the Phoenician mask in particular point to older traditions. In fact, these motifs seem to have been chosen on an individual level specifically for their conservative bent and hearkening to an earlier age. Indeed, a few motifs, such as the hand holding the ear, presage motifs that would be popular later on in glyptics, but which only had a few Hellenistic examples. Thus the users of the Kedesh archive appear to have been in general well integrated into the developments of the larger artistic world and even be in the forefront of artistic developments in some cases. Second, the symbols that appear at Kedesh were part in a general increase in symbols that occurred during the Hellenistic period. Certainly, the sheer number of occurrences of the symbol motifs increased dramatically, during this period. In part, this rise was fuelled by the increased artistic interest in visual allegory that rose from early 4th century Greek intellectual developments and interactions with the Near East and was also expressed artistically in such areas as the increased use of anthropomorphic personifications like the Tyche of Antioch by Eutychides. The general cosmopolitan nature of the Hellenistic period and the rise of many new powers in search of convenient means of expression aided in this process by improving the acceptability of the use of symbols overall in several media. Third, almost all of the motifs that appear at Kedesh have strong amuletic qualities. The motifs express these qualities by using one or more specific methods. In 417 many ways it appears that the motifs were chosen so as to enhance the inherent amuletic qualities of the seal and protect the owner/user. This function acted in parallel to its legal function as an identifier of an individual in a document. Altogether, the symbols that appear on 125 of the Kedesh sealings are not just simple images with little in the way of scholarly value or avenues for scholarship. They have expanded our understanding of the site by anchoring it, its archives and the users thereof in complex networks of relationships that span many regions and centuries. They help to illustrate how the users of the Kedesh archive interacted with world around them on several levels that include time, identity, geography and the supernatural. 418 Thesis Bibliography Acquaro, Enrico 1988 “Catalogue.” In The Phoenicians, edited by S. Moscatti, pp.583-755. New York: Abberville Press. Acquaro, Enrico & Sabotino Moscati, editors 1996 Alle soglie della Classicità: Il Mediterraneo tra tradizione e innovazione. Studi in onore di Sabatino Moscati. Volume Secondo: Archeologia e arte. Rome: Instituti Editoriale e Poligrafice Internazionali. Ariel, Donald T. & Joseph Naveh 2003 “Selected Inscribed Sealings from Kedesh in the Upper Galilee.” Bulletin of the American Schools of Oriental Research 329: 61-80. Avigad, Nahman 1976 Bullae and Seals from a Post-Exilic Judean Archive. Qedem: Monographs of the Institute of Archaeology, the Hebrew University of Jerusalem, Volume 4. Jerusalem: the Hebrew University of Jerusalem, Institute of Archaeology. Babelon, Ernest 1910 Traité des monnaies grècques et romaines. Bologna: Forni Editore. Bailey, Donald M. 1975 A Catalogue of the Lamps in the British Museum I: Greek, Hellenistic and Early Roman Lamps. London: British Museum. Bal, Mieke & Norma Bryson 1991 “Semiotics and Art History.” The Art Bulletin 73.2: 174-208. Barnett, R. D. & C. Mendleson eds. 1987 Tharros: A Catalogue of Material in the British Museum from Phoenician and Other Tombs at Tharros, Sardinia. London: British Museum. Bates, William Nickerson 1930 “Two Inscribed Slinger’s Bullets from Galatista.” American Journal of Archaeology 34.1: 44-46. Beazley, John Davidson 1929 “Charinos.” Journal of Hellenic Studies 49.1: 38-78. 1956 Attic Black-figure Vase-Painters. Oxford: Clarendon Press. 419 1986 The Development of Attic Black-Figure. Berkeley: University of California Press. Ben-Tor, Daphna 1989 The Scarab: A Reflection of Ancient Egypt. Jerusalem: The Israel Museum. Berges, D. 1993 “Die Tonsiegel aus dem karthagischen Tempelarchiv.” Mitteilungen des Deutschen Archaeologischen Instituts, roemische Abteilung 100: 245- 268. 1996 “Der Fundkomplex griechischer Siegelabdrücke aus Karthago.” In Archives et sceaux du monde hellénistique, edited by Marie-Françoise Boussac and Antonio Invernizzi, pp.341-348. Bulletin de correspondence hellénique, Supplement 29. Athens: École française d’Athènes. Berges, D., W. Ehrhardt, A. Laidlaw, F. Rakob 1997 Karthago, Band II: Die deutschen Ausgrabungen in Karthago. Mainz: Philip von Zabern. Bernard, Paul 1969 “Quatrième campagne de fouilles à Aï Khanoum (Bactriane).” Comptes rendus à l’Académie des Inscriptions et Belles-Lettres: 313-355. Bieber, M. 1910 “Attische Reliefs in Cassel.” Mitteilungen des Kaiserlich Deutschen Archäologishen Instituts, athenische Abteilung 35: 1-182. Bivar, A.D.H. 1962 “A Rosette "phialē" Inscribed in Aramaic.” Bulletin of the School of Oriental and African Studies, University of London 24.2: 189-199. Blanchet, A. 1923 “Recherches sur les ‘grylles.’ À propos d’une pierre gravée, trouvée en Alsace.” Revue des Études Anciennes 23: 43-51. Boardman, John 1970a Greek Gems and Finger Rings: Early Bronze Age to Late Classical. London: Thames and Hudson. 1970b “Pyramidal Stamp Seals in the Persian Empire.” Iran 8: 19-45 1983 “Symbol and Story in Geometric Art.” In Ancient Greek Art and 420 Iconography, edited by Warren G. Moon, pp. 15-36. Madison: University of Wisconsin Press. 1997 “Greek Seals.” In 7000 Years of Seals, edited by Dominique Collon, pp.74-87. London: British Museum Press. 1998 “Seals and Signs. Anatolian Stamp Seals of the Persian Period Revisited.” Iran 36: 1-13. 2003 “Disguise and Exchange in Eastern Imagery.” In Culture through Objects: Ancient Near Eastern Studies in Honour of P.R.S. Moorey, edited by Peter Roger Stuart Moorey, Timothy Potts, Michael Roaf, and Diana L. Stein, pp.123-132. Oxford: Griffith Institute. Boardman, John & Diana Scarisbrick 1977 The Ralph Harari Collection of Finger Rings. London: Thames and Hudson. Boardman, John & Marie-Louise Vollenweider 1978 Ashmolean Museum, Oxford. Catalogue of the Engraved Gems and Finer Rings, 1: Greek and Etruscan. Oxford: Clarendon Press. Boussac, Marie-Françoise 1988 “Sceaux déliens.” Revue archéologique 2: 307-340. 1992 Les sceaux de Délos 1: Sceaux publics, Apollo, Hélios, Artémis, Hécate. Athens: École française d’Athènes. 1993 “Archives personelles à Délos.” Comptes-rendus des séances de l’Académie des inscriptions et belles-lettres 137.3: 677-693. 1997 “Masques et acteurs de théatre sur les sceaux de Délos.” Pallas 47: 145- 164. Boussac, Marie Françoise & Antonio Invernizzi, editors 1996 Archives et sceaux du monde hellénistique. Bulletin de correspondence hellénique, Supplement 29. Athens: École française d’Athènes. Bordreuil, Pierre 1986 Catalogue des sceaux ouest-sémitiques inscrits de la Bibliothèque Nationale, du Musée du Louvre et du Musée biblique de Bible et Terre Sainte. Paris: Bibliothèque Nationale. Bordreuil, P. & E. Gubel 1988 “Bulletin d'Antiquités Archéologiques du Levant Inédites ou Méconnues V.” 421 Syria 65.3: 437-456. Boyce, George K. 1937 Corpus of the Lararia of Pompeii. Rome: University of Michigan Press for the American Academy in Rome. 1942 “Significance of Serpents on Pompeian House Shrines.” AJA 46.1: 13-22. Brandt, E. 1968 Antike Gemmen in deutsche Sammlungen I: Griechische Gemmen von minoischer Zeit bis zum späten Hellenismus, Staatliche Münzsammlung Münschen, (Teil 1). Munich: Staatliche Münzsammlung. Brandt, E. & E. Schmidt 1970 Antike Gemmen in deutsche Sammlungen I: Italische Gemmen, etruskisch bis römisch-republikanisch, italische glaspasten vorkaiserzeitlich, Staatliche Münzsammlung Münschen, (Teil 2). Munich: Staatliche Münzsammlung. Brandt, E., W. Gerke, A. Krug & E. Schmidt 1972 Antike Gemmen in deutsche Sammlungen I: Gemmen und Glaspasten der römisch Kaiserzeit sowie Nachträge, Staatliche Münzsammlung Münschen, (Teil 3), Munich: Staatliche Münzsammlung. Bregstein, Linda 1996 “Sealing Practices in the Fifth century B.C. Murašû Archive from Nippur, Iraq.” In Archives et sceaux du monde hellénistique, edited by Marie- Françoise Boussac and Antonio Invernizzi, pp. 53-63. Bulletin de correspondence hellénique, Supplement 29. Athens: École française d’Athènes. Brown, Blanche R. 1957 Ptolemaic Painting and Mosaic and the Alexandrian Style. Cambridge, Mass: Archaeological Institute of America. Brown, Shelby 1992 “Perspectives on Phoenician Art.” The Biblical Archaeologist 55.1: 6-24. Bruneau, Philippe 1967 Exploration archélogique de Délos 26: Les lampes. Paris: Édition de Boccard. 1972 Exploration Archéologique de Délos, Faites par l’École Française d’Athènes, 29: Les Mosaïques. Paris: Éditions E. de Boccard. 422 1978 “Les cultes de l’établissement des poseidoniastes de Bérytos à Délos.” In Hommages à Maarten J. Vermaseren: Receuil d'études par les auteurs de la série études préliminaires aux religions orientales dans l’empire romain à Maarten J. Vermaseren à l’occasion de son soixantième anniversaire le 7 Avril 1978. Volume 1, edited by Margreet B. De Boer and T. A. Edridge, pp. 160-189. Leiden: E.J. Brill. Bulard, Marcel 1926 Explorations archéologiques de Délos faite par l’École Française d’Athenes 9: Déscription des revêtements peints à sujets religieux. Paris: Éditions de Boccard. Cahill West, Jane M. 1995 “Rosette Stamp Seal Impressions from Ancient Judah.” Israel Exploration Journal 45: 230-253. Callieri, Pierfrancesco 1997 Seals and Sealings from the North-West of the Indian Subcontinent and Afghanistan (4th Century BC- 11th Century AD): Local, Indian, Sassanian, Graeco-Persian, Sogdian, Roman. Naples: Instituti Universitario Orientale. Cameron, George G. 1948 Persepolis Treasury Tablets. Oriental Institute Publications 65. Chicago: University of Chicago Press. 1965 “New Tablets from the Persepolis Treasury.” Journal of Near Eastern Studies 24: 167-92. Carradice, Ian 1995 Greek Coins. Austin: University of Texas Press. Carter, Elizabeth, Frank Hole, Zainab Bahrani, Agnès Spycket, & Joan Aruz 1992 “Prehistoric Susa.” In The Royal City of Susa, edited by Prudence O. Harper, Joan Arruz & Françoise Tallon, pp. 25-46. New York: Metropolitan Museum of Art. Cassirer, Ernst 1979 Symbol, Myth, and Culture: Essays and Lectures of Ernst Cassirer, 1935- 1945. Edited by Donald Philip Verene. New Haven: Yale University Press. Chase, George Henry 1902 “The Shield Devices of the Greeks.” Harvard Studies in Classical Philology 13: 61-127. 423 Chapouthier, Fernand 1935 Les Dioscures aux service d’une déesse. Paris: Éditions de Boccard. Chittenden, Jacqueline 1947 “The Master of Animals.” Hesperia 16.2: 88-114. Ciasca, Antonia 1988 “Masks and Protomes.” In The Phoenicians, edited by S. Moscatti, pp.354-369. New York: Abberville Press. Cintas, Pierre 1946 Amulettes puniques. Tunis: Institut des hautes études de Tunis. Clark, R.T. Rundle 1959 Myth and Symbol in Ancient Egypt. London: Thames & Hudson. Collon, Dominique 1987 First Impressions: Cylinder Seals in the Ancient Near East. Chicago: University of Chicago Press. 1996 “A Hoard of Sealings from Ur.” In Archives et sceaux du monde hellénistique, edited by Marie-Françoise Boussac and Antonio Invernizzi, pp.65-84. Bulletin de correspondence hellénique, Supplement 29. Athens: École française d’Athènes. Colvin, Stephen 2004 “Names in Hellenistic and Roman Lycia.” In The Greco-Roman East: Politics, Culture Society, edited by Stephen Colvin, pp.44-85. Cambridge: Cambridge University Press. Conze, Alexander, O. Bertlet, A. Philippson, C. Schuchhardt, & Fr. Graber 1913 Altertumer von Pergamon I: Stadt und Landschaft. Berlin: G. Reimer. Cornell, Leslie 1997 “A Note on Molded Bowls.” In Tel Anafa II, i, edited by Sharon C. Herbert, pp. 407-416. Ann Arbor: Kelsey Museum. Courby, Fernand 1922 Les vases grecs à reliefs. Paris: Éditions de Boccard. Coulson, William D. E., Margaret S. Mook, James W. Rehard, & Virginia R. Grace 1997 “Stamped Amphora Handles from Tel Beersheba.” BASOR 306: 47-62. Courbin, Paul 424 1966 La céramique géometrique de l’Argolide. Paris: Éditions de Boccard. Crowfoot, G. M. 1957 “Megarian Bowls.” In Samaria-Sebaste III: The Objects, edited by J.W. Crowfoot, G.M. Crowfoot & Kathleen M. Kenyon, pp.272-281. London: The Palestine Exploration Fund. Culican, William 1976a “Phoenician Demons.” JNES 34.1: 21-24. 1976b “Some Phoenician Masks and Other Terracottas.” Berytus 34: 47-87. Cunningham, A. 1884 Coins of Alexander’s Successors in the East. Chicago: Argonaut Publishers. Reprinted 1969. Daux, Georges 1965 “Chroniques des fouilles 1964: Stratos.” Bulletin de correspondence hellénique 89: 760-761. Davidson, Gladys R. 1942 “A Hellenistic Deposit at Corinth.” Hesperia 11.2: 105-127. Dayagi-Mendels, Michal 2002 The Akhziv Cemeteries: The Ben-Dor Excavations 1941-1944. IAA Reports no. 15. Jerusalem: Israel Antiquities Authority. De Mecquenem, R. 1927 “Inventaire de cachets et cylindres, Suse 1925-1926.” Revue d’assyriologie et archéologie orientale 24: 13-21. Delaporte, L. 1923 Catalogue des cylindres, cachets et pierres gravées de style oriental II. Paris: Hatchette. Deonna, W. 1938 Exploration archéologique de Délos, volume 18: Le mobilier délien. Paris: Édition de Boccard. Douglas, Mary 1973 Natural Symbols: Explorations in Cosmology. London: Barrie & Jenkins. Dunand, Maurice & Raymond Duru 1962 Oumm el-‘Amed: Une ville de l’époque hellénistique aux échelles de Tyr. 2 Volumes. Paris: Librairie d’Amérique et d’Orient Adrien Maisoneuve. 425 Dürrbach, Félix, Pierre Roussel, Marcel Launey, André Plassart & Jacques Coupry 1926 Inscriptions de Délos, Volume 4: Decrests postéieurs à 166 av. J.-C. (nos 1497-1524). Paris: Académie des inscriptions et belles-lettres. Dusinberre, Elizabeth R. M. 2003 Aspects of Empire in Achaemenid Sardis. Cambridge: Cambridge University Press. 2005 Gordion Seals and Sealings: Individuals and Society. Philadelphia: University of Pennsylvania Museum of Archaeology and Anthropology. Edgar, M.C.C. 1907 “Notes from the Delta: Clay Sealings from Thmouis.” Annales du Service des Antiquités de l’Égypte 8: 154-157. Edwards, Charles, M. 1986 “Corinthian Moldmade Bowls: The 1926 Reservoir.” Hesperia 55.4: 318- 419. Edwards, Katharine M. 1933 Corinth, the Results of the Excavations Conducted by the American School for Classical Studies at Athens, Volume VI: Coins, 1896-1929. Cambridge, Mass: Harvard University Press for the American School of Classical Studies at Athens. Eldritch, L. G. 1918 “A third Century Etruscan Tomb.” AJA 22.3: 251-294. Elworthy, Frederick Thomas 1958 The Evil Eye. The Origins and Practices of Superstition. New York: The Julian Press. Emmanuel-Rebuffat, D. 1988 Corpus-Speculorum Etruscorum; France: Paris-Musée du Louvre, fascicule 1. Rome: “L’Erma” di Bretschneider. Erlich, Adi & Amos Kloner 2008 Maresha Excavations Final Report II: Hellenistic Terracotta Figurinesfrom the 1989-1996 Season. IAA Reports, no.35. Jerusalem: Israel Antiquities Authority. Farbridge, Maurice H. 1970 Studies in Biblical and Semitic Symbolism. New York: Ktav Publishing House. 426 Finkielszteijn, Gerald 2001 Chronologie détaillée et revise des eponyms amphoriques rhodiens de 270 à 108 environ, Premier bilan. BAR International Series #990. Oxford: Archaeopress. Firth, Raymond 1973 Symbols Public and Private. Ithaca: Cornell University Press. Forbes, Barbara Ann 1978 Catalogue of Engraved Gems in the Art Museum, Princeton University. PhD Dissertation. Berkeley: University of California. Forsythe Johnston, Paul 1985 Ship and Boat Models in Ancient Greece. Annapolis: Naval Institute Press. Foss, Clive 1975 “Greek Sling Bullets in Oxford.” Archaeological Reports 21 (1974-75): 40-44. Fondation pour le Lexicon iconographicum mythologiae classicae 1981-1999 Lexicon Iconographicum Mythologiae Classicae. Zurich: Artemis Verlag. Fossing, Poul 1929 The Thorvaldsen Museum: Catalogue of the Antique Engraved Gems and Cameos. Copenhage: Thorvaldsen Museum. Furtwängler, A. 1900 Die antiken Gemmen. Geschichte der Steinschneidekunst im klassischen Altertum 1-3. Amsterdam: Adolf M. Hakkert. 1965 Reprint. Gardner, Percy 1878 A Catalogue of the Greek Coins in the British Museum: The Seleucid Kings of Syria. London: the British Museum. 1963 A Catalogue of the Greek Coins in the British Museum, Volume 6: Peloponnesus. Bologna: Arnaldo Forni. 1963b A Catalogue of the Greek Coins in the British Museum, Volume 9: Thessaly to Aetolia. Bologna: Arnaldo Forni. Garrison, Mark B. & Margaret Cool Root 2001 Seals on the Persepolis Fortification Tablets, Volume I: Images of Heroic Encounter. Oriental Institute Publications 117, Chicago: University of Chicago Press. 427 Gerke, S., V. Scherf & P. Zazoff 1970 Antike Gemmen in deutsche Sammlungen III, Die Gemmensammlung im Herzog-Anton-Ulrich-Museum Braunschweig, Die Gemmensammlung in Arcäologischen Institut der Universität Göttingen, Die Gemmen- sammlung der Staatlischen Kunstsammlungen Kassel. Wiesbaden: Prestel Verlag. Gerson, Stephen N. 2001 “Fractional Coins of Judaea and Samaria in the Fourth Century BCE.” Near Eastern Archaeology 64. 3: 106-121. Goldman, Hetty 1956 Excavations at Gözlü Kule, Tarsus I: The Hellenistic and Roman Periods. Princeton: Princeton University Press Grace, V. 1934 “Stamped Amphora Handles Found in 1931-1932.” Hesperia 3.3, American Excavations in the Athenian Agora: Fourth Report: 197-310. Gramatapol, Mihai 1974 Les pierres gravées du cabinet numismatique de l’Académie Roumaine. Brussels: Latomus, Revue d’Étude Latines. Green, J. R. 1995 “From Taranto to Alexandria.” In Trade, Contact, and the Movement of Peoples in the Eastern Mediterranean: Studies in Honour of J. Basil Hennessy, edited by Stephen Bourke and Jean-Paul Descoeurdres, pp.271- 274. Mediterranean Archaeology Supplement 3. Sydney: Mediterranean Archaeology. Gubel, Eric 1987 “‘Syro-Cypriot’ Cubical Stamps: The Phoenician Connection.” In Studia Phoenicia V, Phoenicia and The East Mediterranean in the First Millenium B.C.: Proceedings of the Conference Held in Leuven from the 14th to the 16th of November 1985, edited by Edward Lipiński, pp. 195- 224. Leuven: Uitgeverij Peeters. Guz-Silberstein, Bracha 1995 “The Typology of the Hellenistic Coarse Ware and Selected Loci of the Hellenistic and Roman Periods.” In Excavations at Dor, Final Report. Volume IB. Areas A and C: The Finds, edited by Ephraim Stern, pp.289- 434. Jerusalem: The Hebrew University of Jerusalem. Hackens, Tony & Ghislaine Moucharte, editors 428 1992 Numismatique et histoire économique phéniciennes et puniques: actes du colloque tenu à Louvain-la-Neuve, 13-16 mai 1987. Studia Phoenicia, volume 9. Louvain-la-Neuve: Université Catholique de Louvain. Hägg, Robin 1988 “The Swedish Institute at Athens. Report for the academic years 1984-85 and 1985-86.” Opuscula Atheniensia: Acta Instituti Atheniensis Regni Sueciae 17: 220-242. Hallock, Richard T. 1969 Persepolis Fortification Tablets. Oriental Institute Publications 92. Chicago: University of Chicago Press. Harper, Prudence O. & Pierre Amiet 1992 “The Mesopotamian Presence: Mesopotamian Monuments Found at Susa.” In The Royal City of Susa, edited by Prudence O. Harper, Joan Arruz & Françoise Tallon, pp. 159-182. New York: Metropolitan Museum of Art. Harper, Pudence O., Joan Aruz & Françoise Tallon, editors. 1992 The Royal City of Susa: Ancient Near Eastern Treasures in the Louvre. New York: The Metropolitan Museum of Art. Head, Barclay V. 1878 A Catalogue of the Greek Coins in the British Museum: Attica, Megaris, Aegina. London: the British Museum. 1901 A Catalogue of the Greek Coins in the British Museum: Lydia. London: the British Museum. 1906 A Catalogue of the Greek Coins in the British Museum, Volume 25: Phrygia. London: the British Museum. 1963 A Catalogue of the Greek Coins in the British Museum, Volume 2: Central Greece (Locris, Phocis, Boeotia and Euboea). Bologna: Arnaldo Forni. 1963b A Catalogue of the Greek Coins in the British Museum, Volume 3: Corinth, Colonies of Corinth, etc. Bologna: Arnaldo Forni. 1964 A Catalogue of the Greek Coins in the British Museum, Volume13: Catalogue of the Greek Coins of Ionia. Bologna: Arnaldo Forni. Head, Barclay V. & Percy Gardner 1963 A Catalogue of the Greek Coins in the British Museum, Volume 8:The Tauric Chersonese, Sarmatia, Dacia, Moesia, Thrace, etc. Bologna: 429 Arnaldo Forni. Head, Barclay V., Reginald Stuart Poole & Percy Gardner 1963 A Catalogue of the Greek Coins in the British Museum, Volume18: Sicily. Bologna: Arnaldo Forni. Herbert, Sharon C., editor 1997 Tel Anafa II, i: The Hellenistic and Roman Pottery. Ann Arbor: Kelsey Museum. Herbert, Sharon C. & Andrea M. Berlin 2003 “A New Administrative Centre for Persian and Hellenistic Galilee: Preliminary Report of the University of Michigan/ University of Minnesota Excavations at Kedesh.” BASOR 329: 13-59. Hildburgh, W. L. 1946 “Apotropaism in Greek Vase-Painting.” Folklore 57.4: 154-178. Hill, George Francis 1900 A Catalogue of the Greek Coins in the British Museum, Volume21: Lycaonia, Isauria and Cilicia. London: the British Museum. 1910 Catalogue of the Greek Coins of Phoenicia in the British Museum. London: the British Museum. 1964 A Catalogue of the Greek Coins in the British Museum, Volume11: The Greek Coins of Cyprus. Bologna: Arnaldo Forni. 1965 A Catalogue of the Greek Coins in the British Museum, Volume27: The Greek Coins of Palestine (Galilee, Samaria, Judaia). Bologna: Arnaldo Forni. Hogarth, David George 1908 Excavations at Ephesus: The Archaic Artemisia. London: the British Museum, London. Holloway, R. Ross 1986 “The Bulls in the “Tomb of the Bulls” at Tarquinia.” AJA 90.4: 447-452. Howland, Richard Hubbard, 1958 The Athenian Agora, Volume 4, Greek Lamps and their Survivals. The Princeton: American School of Classical Studies at Athens. Hurwitt, Jeffrey M. 430 1993 “Art, Poetry, and the Polis in the Age of Homer.” In From Pasture to the Polis: Art in the Age of Homer, edited by Susan Langdon, pp.14-42 Columbia: University of Missouri Press. Invernizzi, Antonio 1966 “Bullae from Seleucia.” Mesopotamia 3-4: 69-124. 1996 “Stelle e rosette tra le impronte di sigilli degli archive di Seleucia al Tigri.” In Alle soglie della classicità: Il Mediterraneo tra tradizione e innovazione. Studi in onore di Sabatino Moscati. Volume Secondo: Archeologia e arte, edited by Enrico Acquaro and Sabotino Moscati, pp.801-811. Rome: Istituti Editoriale e Poligrafice Internazionali. Invernizzi, Antonio, Ariela Bollati, Vito Messina & Paolo Mollo 2004 Seleucia al Tigri. Le impronte di sigillo dagli archivi, I-III. Alexandria: Edizioni dell Orso, Centro Ricerche Archeologiche e scavi per il Medio Oriente e l’Asia. Kaptan, Deniz 2002 The Daskyleion Bullae: Seal Images from the Western Achaemenid Empire. Leiden: Nederlands Instituut voor het Nabije Oosten. Kachatrian, Z’ores 1996 “The Archives of Sealings found at Artashat (Artaxata).” In Archives et sceaux du monde hellénistique, edited by Marie-Françoise Boussac and Antonio Invernizzi, pp.365-370. Bulletin de correspondence hellénique, Supplement 29. Athens: École française d’Athènes. Karageorghis, Vassos, 1996 “Anthropomorphic Clay Masks from Cyprus.” In Alle soglie della classicità: Il Mediterraneo tra tradizione e innovazione. Studi in onore di Sabatino Moscati. Volume Secondo: Archeologia e arte, edited by Enrico Acquaro and Sabotino Moscati, pp.813-821. Rome: Istituti Editoriale e Poligrafice Internazionali. 2000 Ancient Art from Cyprus: The Cesnola Collection in the Metropolitan Museum of Art. New York: The Metropolitan Museum of Art. Keel, Othmar & Christoph Uelinger 1998 Gods, Godesses, and Images of God in Ancient Israel. Translated by Thomas H. Trapp. Minneapolis: Fortress Press. Klose, D.O.A. 1984 “Nikopolis und Doliche: Neue Tonsiegel aus dem ἀρχειον des syrischen 431 Nikopolis.” Jahrbuch für Numismatik und Geldgeschichte 34: 63-76. Kroll, John H. & Alan S. Walker 1993 The Athenian Agora, Volume 26: The Greek Coins. Princeton: The American School of Classical Studies at Athens. Kushnir-Stein, Alla 2002 “New Hellenistic Lead Weights from Palestine and Phoenicia.” IEJ 52: 225-30. Kyrieleis, H. 1996 “Ptolemäische Poträts auf Siegelabdrücken aus Nea Paphos (Zypern).” In Archives et sceaux du monde hellénistique, edited by Marie-Françoise Boussac and Antonio Invernizzi, pp.315-320. Bulletin de correspondence hellénique, Supplement 29. Athens: École française d’Athènes. Lang, Mabel & Margaret Crosby 1964 The Athenian Agora, Volume 10: Weights, Measures and Tokens. Princeton: The American School of Classical Studies at Athens. Langdon, Susan 1993 From Pasture to Polis: Art in the Age of Homer. Columbia: University of Missouri Press, Columbia. Lawrence, Arnold Walter 1957 Greek Architecture. Baltimore: Penguin Books. Leith, Mary Joan Winn 1997 The Wadi Dalieyh Seal Impressions. Discoveries in the Judaean Desert, 24. Oxford: Clarendon Press. Lenger, Marie-Thérèse 1957 “Timbres amphoriques trouvés à Argos.” BCH 81.1: 160-180. Lipinski, Edward 1995 “Le ‘caducée’.” In Actes du IIIe congrès international des études phéniciennes et puniques. Tunis, 11-16 novembre 1991. volume 2, edited by Muhammad Fantar and Mansour Ghaki, pp.203-209. Tunis: Institut National du Patrimoine. Maaskant-Kleibrink, Marianne 1971 “Cachets de terre de Doliche?” Bulletin antieke Beschaving 46: 23-63. 1978 Catalogue of the Engraved Gems in the Royal Coin Cabinet, the Hague: The Greek, Etruscan and Roman Collections. Weisbaden: Franz Steiner 432 Verlag. Maddoli, Gianfranco 1964 “Le Cretule del Nomophylakion di Cirene.” Annuario della Scuola Archeologica di Atene e delle Missioni Italiane in Oriente, Volume XLI- XLII, New Series XXV-XXVI: 39-146. Manoukian, Hasmik 1996 “Cachets d’Artachate.” In Archives et sceaux du monde hellénistique, edited by Marie-Françoise Boussac and Antonio Invernizzi, pp.371-373. Bulletin de correspondence hellénique, Supplement 29. Athens: École française d’Athènes. Markle, Minor M. 1999 “A Shield Monument from Veria and the Chronology of Macedonian Shield Types.” Hesperia 68.2: 219-254. McDowell, Robert Harbold 1935 Stamped and Inscribed Objects from Seleucia on the Tigris. Ann Arbor: University of Michigan Press. McEwan, G. J. P. 1982 Texts from Hellenistic Babylonia in the Ashmolean Museum with Notes on the Seal Impressions by the Late Briggs Buchanan, Oxford. Editions of Cuneiform Texts 9, Oxford: Clarendon Press. McGowan, Elizabeth P. 1995 “Tomb marker and Turning Post: Funerary Columns in the Archaic Period.” AJA 99.4: 615-632. Meijer, Fik & Onno von Nijf 1992 Trade, Transport and Society in the Ancient World: A Sourcebook. London: Routledge. Merritt, Lucy Shoe 1982 “Some Ionic Architectural Fragments from the Athenian Agora.” In Studies in Athenian Architecture, Sculpture and Topography. Presented to Homer A. Thompson, edited by Homer A. Thompson, pp. 82-92+ 204-205. Hesperia Supplements, Volume 20. Princeton: American School of Classical Studies at Athens. Meshorer, Ya’akov & Shraga Qedar 1999 Samarian Coinage. Jerusalem: The Israel Numismatic Society. Michaelidou-Nicolaou, I. 433 1979 “Inscribed Clay Sealings from the Archaeion of Paphos.” In Actes du VIIe congrès d’épigraphie grècque et latine: Constantza 9-15 septembre 1977, edited by D. M. Pippidi, pp.413-416. Paris: Institut d’Archéologie de Bucarest. Mildernberg, Leo 1992 “The Philisto-Arabian coins, a preview: preliminary studies of the local coinage in the fifth Persian satrapy. Part 3.” In Numismatique et histoire économique phéniciennes et puniques: actes du colloque tenu à Louvain- la-Neuve, 13-16 mai 1987, edited by Tony Hackens and Ghislaine Moucharte, pp.33-40. Studia Phoenicia, volume 9. Louvain-la-Neuve: Université Catholique de Louvain. Milne, J. G. 1906 “Clay Sealings from the Fayum.” JHS 26: 32-45. 1916 “Ptolemaic Seal Impressions.” JHS 36: 87-101. Milner, N. P. 2004 “Ancient Inscriptions and Monuments from the Territory of Oinoanda.” Anatolian Studies 54: 47-77. Minns, Elis H. 1913 Scythians and Greeks: A Survey of Ancient History and Archaeology on the North Coast of the Euxine from the Danube to the Caucasus. Cambridge: Cambridge University Press. Mlynarczyk, Jolanta 1997 Alexandrian and Alexandrian-Influenced Mould-Made Lamps of the Hellenistic Period. BAR International Series #677. Oxford: Archaeopress. Moore, Mary B. 1997 The Athenian Agora, Volume 30, Attic Red-Figured and White-Ground Pottery. Princeton: The American School of Classical Studies at Athens. Moore, Mary B., Mary Zelia Pease Philippides, & Deitrich von Bothmer 1986 The Athenian Agora, Volume 26, Attic Black-Figured Pottery. Princeton: The American School of Classical Studies at Athens. Mørholm, Otto 1991 Early Hellenistic Coinage: From the Accession of Alexander to the Peace of Apamea (336-188 B.C.) Cambridge: Cambridge University Press. Moscati, Sabatino 1988 “Stone Reliefs.” In The Phoenicians, edited by S. Moscatti, pp.300-303. 434 New York: Abberville Press. Moscati, Sabatino, editor. 1988 The Phoenicians. New York: Abberville Press. Moss, Leonard W. & Stephen C. Cappanari 1976 “1. The Mediterranean. Mal’occhio, Ayin ha ra, Occulus fascinus, Judenblich: The Evil Eye Hovers Above.” In The Evil Eye, edited by Clarence Maloney, pp.1-15. New York: Columbia University Press. Murray, M. A. 1908 “Ptolemaic Clay Sealings.” Zeitschrift für ägyptische Sprache und Altertumskunde 44: 62-70. Muscarella, Oscar White, Annie Caubet & Françoise Tallon 1992 “Susa in the Achaemenid Period.” In The Royal City of Susa, edited by Prudence O. Harper, Joan Arruz & Françoise Tallon, pp. 215-252. New York: Metropolitan Museum of Art. Negro Ponzi, M. M., A. Invernizzi & G. Gullini 1971 “Fourth Preliminary Report of the Excavations at Seleucia and Ctesiphon.” Mesopotamia 5-6: 9-104. Newell, Edward T. 1977 The Coinage of the Western Seleucid Mints: From Seleucus I Antiochus III. New York: American Numismatic Society. 1978 The Coinage of the Eastern Seleucid Mints: From Seleucus I to Antiochus III. New York: American Numismatic Society. Nicolaou, Kyriakos 1978 “Oriental Divinities Represented on the Clay Sealings of Paphos, Cyprus.” In Hommages à Maarten J. Vermaseren: Receuil d'études par les auteurs de la série études préliminaires aux religions orientales dans l’empire romain à Maarten J. Vermaseren à l’occasion de son soixantième anniversaire le 7 avril 1978. Volume 1, edited by Margreet B. De Boer and T. A. Edridge, pp. 849-853. Leiden: E.J. Brill. Oelsner, J. 1996 “Siegelung und Archivierung von Dokumenten im hellenistischen Babylonien.” In Archives et sceaux du monde hellénistique, edited by Marie-Françoise Boussac and Antonio Invernizzi, pp. 101-112. Bulletin de correspondence hellénique, Supplement 29. Athens: École française d’Athènes. 435 Onians, John 1979 Art and Thought in the Hellenistic Age: The Greek World View, 350-50 BCE. London: Thames and Hudson. Ornan, Tallay 1995 “The Transition from Figured to Non-Figured Representation in First Millennium Mesopotamian Glyptic.” In Seals and Sealings in the Ancient Near East: Proceedings of the Symposium Held on September 2, 1993, Jerusalem, Israel, edited by Joan Goodneck Westenholtz, pp.39-56. Jerusalem, Bible Lands Museum, Jerusalem. Pantos, P.A. 1985 Τα σφραγισµατα της Αιτολικης Καλλιpiολεος. Athens. 1996 “Poträtsiegel in Kallipolis: Einige methodologische Bermeerkungen.” In Archives et sceaux du monde hellénistique, edited by Marie-Françoise Boussac and Antonio Invernizzi, pp. 185-194. Bulletin de correspondence hellénique, Supplement 29. Athens: École française d’Athènes. Paraire, D.J. 1980 “Les Figurines.” In Tell Keisan (1971-1976): une cite phénicienne en Galilée, edited by Jacques Briend and Jean-Baptiste Humbert, pp.331-352. Paris: J. Gabalda. Payne, Humphrey 1931 Necrocorinthia: A study of Corinthian Art in the Archaic period. Oxford: Clarendon Press. Perdrizet, Paul 1904 “Steles Peintes de Sidon.” RA 4.3: 234-244. Picard, Colette G. 1966 “Sacra punica: Étude sur les masques et rasoirs de Carthage.” Karthago 13: 7-115. Pisano, Giovanna 1988 “Jewellery.” In The Phoenicians, edited by S. Moscatti, pp.570-574. New York: Abberville Press. Plant, Richard 1979 Greek Coin Types and Their Identification. London: Seaby Publications Ltd. Plantzos, Dimitris 436 1996 “Female Portrait Types from the Edfu Hoard of Clay Seal Impressions.” In Archives et sceaux du monde hellénistique, edited by Marie-Françoise Boussac and Antonio Invernizzi, pp. 307-313. Bulletin de correspondence hellénique, Supplement 29. Athens: École française d’Athènes. 1999 Hellenistic Engraved Gems. Oxford: Clarendon Press. Platt, Verity 2006 “Making an Impression: Replication and the Ontology of the Graeco- Roman Seal Stone.” Art History 29.2: 233-257. Platz-Horster, G., M. Schulter & P. Zazoff 1975 Antike Gemmen in deutsche Sammlungen IV, Hannover, Kestner-Museum, Hamburg Museum für Kunst und Gewerge. Wiesbaden: Franz Steiner. Poole, Reginald Stuart 1883 A Catalogue of the Greek Coins in the British Museum: The Ptolemies: Kings of Egypt. London: the British Museum. 1963a A Catalogue of the Greek Coins in the British Museum, Volume7: The Ptolemies, Kings of Egypt. Bologna: Arnaldo Forni. 1963b A Catalogue of the Greek Coins in the British Museum, Volume5: Italy. Bologna: Arnaldo Forni. 1964 A Catalogue of the Greek Coins in the British Museum, Volume10: Alexandria and the Nomes. Bologna: Arnaldo Forni. Preka-Alexandris, Kalliopi 1989 “Seal Impressions from Titani, a Hellenistic Metropolis of Thesprotia.” In Technology and Analysis of Ancient Gemstones, Ravello, European University Centre for Cultural Heritage, November 13-16, 1987, edited by Tony Hackens, pp.163-172. Revue du Groupe Européen d'Etudes pour les Techniques Physiques, Chimiques, Biologiques et Mathématiques Appliquées à l'Archéologie, Volume 23. Rixensart: PACT. Price, Simon & Emily Kearns, editors. 2003 The Oxford Dictionary of Myth and Religion. Oxford: Oxford University Press. Rahmani, L.Y. 1984 “Hellenistic Brazier Fragments from Israel.” IEJ 34: 224-231. Rappaport, Uriel 1992 “Phoenicia and Galilee: Economy, Territory and Political Relations.” In 437 Numismatique et histoire économique phéniciennes et puniques: actes du colloque tenu à Louvain-la-Neuve, 13-16 mai 1987, edited by Tony Hackens and Ghislaine Moucharte, pp.261-268. Studia Phoenicia, volume 9. Louvain-la-Neuve: Université Catholique de Louvain. Reifenberg, Adolph 1947 Ancient Jewish Coins. Jerusalem: Rubin Mass. Richter, Gisela M. A. 1968 Catalogue of the Engraved Gems of the Greeks and the Etruscans. New York: Phaidon. Rice, E. E. 1983 The Grand Procession of Ptolemy Philadelphus. Oxford: Oxford University Press. Rickman Fitch, Cleo & Norma Wynick Goldman 1994 Cosa: The lamps. Memoirs of the American Academy in Rome, Volume 39. Ann Arbor: University of Michigan Press for the American Academy in Rome. Roaf, Michael 1983 “Sculptures and Sculptors at Persepolis.” Iran 21: 1-164. Roebuck, Carl 1951 Corinth, Results of Excavations Conducted by the American School of Classical Studies at Athens, Volume XIV: The Asklepieion and Lerna. Princeton: The American School of Classical Studies at Athens. Roes, Anne 1933 Greek Geometric Art: Its Symbolism and Its Origin. London: Oxford University Press. 1935 “New Light on the Grylli.” JHS 55.2: 232-235. Rosenthal-Heginbottom, Renate 1995 “Imported Hellenistic and Roman Pottery.” In Excavations at Dor, Final Report. Volume IB. Areas A and C: The Finds, edited by Ephraim Stern, pp.183-250. Jerusalem: The Hebrew University of Jerusalem. Rostovtzeff, M. 1932 “Seleucid Babylonia: Bullae and Seals of Clay with Greek Inscriptions.” Yale Classical Studies 3: 3-114. Rotroff, Susan I. 438 1982 The Athenian Agora, Volume 22, Hellenistic Pottery: Athenian and Imported Moldmade Bowls. Princeton: The American School of Classical Studies at Athens. 1991 “Attic West Slope Vase Painting.” Hesperia 60.1: 59-102. 1997 The Athenian Agora, Volume 27, Hellenistic Pottery: Athenian and Imported Wheemade Tableware and Related Material. Princeton: The American School of Classical Studies at Athens. Rubensohn, Otto 1907 Elephantine-Papyri. Berlin: Weidmannsche Buchhandlung. Şahin, Mustafa 2001 “Hellenistic Braziers in the British Museum: Trade Contacts Between Ancient Mediterranean Cities.” Anatolian Studies 51: 91-132. Salinas, A. 1883 “Selinunte” Notizie degli scavi antichita communicate all R. Academia dei Lincei: 287-314. Schaefer, Joerg 1968 Hellenistische Keramik aus Pergamon. Berlin: De Gruyter. Schwabe, Calvin, Joyce Adams & Carleton T. Hodge 1982 “Egyptian Beliefs about the Bull’s Spine: An Anatomical Origin for Ankh.” Anthropological Linguistics 4.4: 445-479. Sellers, O. R. 1962 “Coins of the 1960 Excavation at Shechem.” The Biblical Archaeologist 25.3: 87-96. Seltman, Charles 1933 Greek Coins: A history of metallic currency and coinage down to the fall of the Hellenistic kingdoms. London: Methuen & Co. Ltd. 2nd edition, 1953. Seyrig, H. 1940 “Cachets d’archives publiques de quelques villes de la Syrie romaine.” Mélanges de l’Université Saint Joseph 23: 85-107. Sherwin-White, Susan & Amélie Kuhrt 1993 From Samarkand to Sardis: A new approach to the Seleucid Empire. Berkeley: University of California Press. 439 Smith, Tyler Jo 1997 “Votive Reliefs from Balboura and Its Environs.” Anatolian Studies 47: 3- 49. Speleer, L. 1917 Catalogue des intailles et empreintes orientales des Musées Royaux du Cinquantenaire. Brussels: Vromant & Co. Spier, Jeffrey 1992 Ancient Gems and Finger Rings: Catalogue of the Collections, the J. Paul Getty Museum. Malibu: The J. Paul Getty Museum. Spivey, Nigel 1997 Etruscan Art. London: Thames & Hudson. Stager, Jennifer M. S. 2005 “’Let No One wonder at the Image’: A Phoenician Funerary Stele at Athens.” Hesperia 74.3: 427-449. Stampoulidis, N. 1992 Τα σφραγισµατα της ∆ηλου 2: Ο ερωτικοσ κυκλος. Paris: Dépositaire de Boccard. Stern, Ephraim 1995 “Clay Figurines and Popular Cult Objects of the Persian Period.” In Excavations at Dor, Final Report. Volume IB. Areas A and C: The Finds, edited by Ephraim Stern, pp.435-459. Jerusalem: The Hebrew University of Jerusalem. 2000 Dor, Ruler of the Seas: Nineteen Years of Excavations at the Israelite- Phoenician Harbor Town on the Carmel Coast. Jerusalem: Israel Exploration Society. Stern, Ephraim, editor 1995 Excavations at Dor, Final Report. Volume IB. Areas A and C: The Finds. Jerusalem: The Hebrew University of Jerusalem. Sussman, Varda 2006 “A Unique Oil Lamp and Bowl of the Hellenistic Period.” IEJ 56: 39-50. Sütterlin, Christa 1989 “Universals in Apotropaic Symbolism: A Behavioral and Comparative Approach to Some Medieval Sculpture.” Leonardo 22.1: 65-74. 440 Tang, Brigit 2005 Delos, Carthage, Ampurias: The housing of three Mediterranean trading centres. Rome: “L’Erma” di Bretschneider. Thompson, Dorothy Burr 1963 “Three Centuries of Hellenistic Terracottas, III: The Late Third Century BCE.” Hesperia 32.3: 301-317. 1965 “Three Centuries of Hellenistic Terracottas, VII: The Early First Century BCE.” Hesperia 34.1: 34-71. 1966 “Three Centuries of Hellenistic Terracottas, VII: The Early First Century BCE.” Hesperia 35.1: 1-19. Thompson, Homer A. 1953 “Athens and the Hellenistic Princes.” Proceedings of the American Philological Society 97.3: 254-261. Trendal, A. D. 1978 “Three Apulian Kraters in Berlin.” Jahrbuch der berliner Museen 12: 153- 190. 1984 “Medea at Eleusis on a Volute Krater by the Darius Painter.” Record of the Art Museum, Princeton University 43.1: 5-17. Tsakirgis, Barbara 1989 “The decorated Pavements of Morgantina I: The Mosaics.” AJA 93.3: 395- 416. Van Buren, E. Douglas 1939 “The Rosette in Mesopotamian Art.” Zeitschrift für Assyriologie und vorderasiatische Archäologie 45: 99-107. Vandorpe, Katelijn 1996 “Seals on the Papyri of Greco-Roman and Byzantine Egypt.” In Archives et sceaux du monde hellénistique, edited by Marie-Françoise Boussac and Antonio Invernizzi, pp. 231-291. Bulletin de correspondence hellénique, Supplement 29. Athens: École française d’Athènes. Venedikov, Ivan 1977 “The Archaeological Wealth of Ancient Thrace.” The Metropolitan Museum of Art Bulletin, New Series 35.1: 7-71. Venit, Marjorie Susan 2002 Monumental Tombs of Ancient Alexandria: The Theater of the Dead. 441 Cambridge: Cambridge University Press. Verdiani, Carlo 1945 “Original Hellenistic Paintings in a Thracian Tomb.” AJA 49.4: 402-415. Vermeule, Cornelius 1966 “Greek and Roman Gems.” Boston Museum Bulletin 64.335: 18-35. Vollenweider, Marie-Louise 1967 Musée d’Art et d’Histoire de Genève, catalogue raisonné des sceaux, cylindres, intailles et camées, volume I. Geneva: Musée d’Art et d’Histoire Genève. 1979 Musée d’Art et d’Histoire de Genève, catalogue raisonné des sceaux, cylindres, intailles et camées, volume II: Les portraits, les masques de théatre, les symboles politiques, une contribution à l’histoire des civilisations hellénistque et romaine. Mainz: Philip von Zabern. Waage, Frederick O., George W. Elderkin & Richard Stillwell 1948 Antioch on the Orontes, Volume IV, Part I. Princeton: Princeton University Press. Waites, Margaret C. 1920 “The Nature of the Lares and Their Representation in Roman Art.” AJA 24.3: 241-261. Wallenfels, Ronald 1990 Sealed Cuneiform Tablets from Hellenistic Uruk: An Iconographic and Prosopographic analysis of the Private Business Documents. New York: Columbia University Dissertation. 1994 Uruk: Hellenistic Seal Impressions in the Yale Babylonian Collection, I. Cuneiform Tablets. Mainz: Philipp von Zabern. 1996 “Private Seals and Sealing Practices at Hellenistic Uruk.” In Archives et sceaux du monde hellénistique, edited by Marie-Françoise Boussac and Antonio Invernizzi, pp. 113-129. Bulletin de correspondence hellénique, Supplement 29. Athens: École française d’Athènes. Walters, H. B. 1926 Catalogue of the Engraved Gems and Cameos, Greek Etruscan and Roman in the British Museum. London: the British Museum. Warner Slane, Kathleen 1997 “The Fine Wares.” In Tel Anafa II, i., edited by Sharon C. Herbert, 442 pp.247-407. Ann Arbor: Kelsey Museum. Webb, Pamela A. 1996 Hellenistic Architectural Sculpture: Figural Motifs in Western Anatolia and the Aegean Islands. Madison: University of Wisconsin Press. Webster, T. B. L. 1967 Monuments Illustrating Tragedy and Satyr Play (Second Edition with Appendix). London: Institute of Classical Studies. 1995 Monuments Illustrating New Comedy. Edited by J. R. Green and A. Seeberg. Third Edition. London: Institute of Classical studies. Weiss, P. 1992 Neue Tonsiegel von .” Chiron: Mitteilungen der Komission für alte Geschichte und Epigraphik des Deutschen Archaeologischen Instituts 22: 171-193. Westgate, R. C. 2000 “Space and Decoration of Hellenistic Houses.” The Annual of the British School at Athens 95: 391-426. Wiegand, Theodore T., Hans Schrader, Herman Winnefeld, Robert Zahn, G. Krumer, & Wilhelm Wilberg 1904 Priene: Ergebnisse des Ausgrabungen und Unterschungen in den Jahren 1895-1898. Berlin: Rheimer. Woodward, A.M & Margaret B. Hobling 1925 “Excavations at Sparta.” Annual of the British School at Athens 26: 116- 310. Wroth, Warwick 1963 A Catalogue of the Greek Coins in the British Museum, Volume 4: Crete and the Aegean Islands. Bologna: Arnaldo Forni. 1964a A Catalogue of the Greek Coins in the British Museum, Volume 12: Galatia, Cappadocia, and Syria. Bologna: Arnaldo Forni. 1964b A Catalogue of the Greek Coins in the British Museum, Volume15: Pontus, Paphlagonia, Bithynia & the Kingdom of Pontus. Bologna: Arnaldo Forni,. 1964c A Catalogue of the Greek Coins in the British Museum, Volume16: Troas, Aeolis and Lesbos. Bologna: Arnaldo Forni. 443 1964d A Catalogue of the Greek Coins in the British Museum, Volume14: Parthia. Bologna: Arnaldo Forni. Zeitlin, Froma I. 2009 Under the Sign of the Shield: Semiotics and Aeschylus’ Seven Against Thebes. Second Edition. Lanham: Lexington Books. Zoppi, C. 1996 “Le cretule di Selinunte.” In Archives et sceaux du monde hellénistique, edited by Marie-Françoise Boussac and Antonio Invernizzi, pp.327-340. Bulletin de correspondence hellénique, Supplement 29. Athens: École française d’Athènes. Zwierlein-Diehl, E. 1969 Antike Gemmen in deutsche Sammlungen II, Staatliche Museen Preußischer Kulturbesitz Berlin. Munich: Antiken-abteilung. Online Sources: American Numismatic Society Online collection Catalogue: http://numismatics.org/collection/accnum/list Ancient Sources: Appian, Syriaca. Athenaeus, Deipnosophistai. Herodotus, The Histories. Homer, The Illiad. Homer, The Odyssey. Ovid, The Fasti, Loeb Classical Library. Sir James George Frazer trans. Harvard University Press. Cambridge. Plato, Timaeus. Pliny, Natural History. Loeb Classical Library. D.E. Eichholtz trans. Harvard University Press, Cambridge, 1971. Strabo, The Geography. The Greek Anthology. Loeb Classical library. W.R. Patton trans. Harvard University Press, Cambridge, 1979. 445 Appendix: The Kedesh Symbol Artist There are several seals within the corpus of the Kedesh symbol sealings that show marked similarities of execution which indicate that they might be the product of a single craftsman. The seals number nine in total, representing a set of twelve sealings. They are concentrated in categories such as masks and composites, though there are also examples in the ear, lyre and flower categories. The seals consist of MAS5, MAS14, BUC4, COM1, COM3-4, EAR3, FLO5-FLO6 and LYR3 (see figure App.1 below). The similarities of execution that link these particular seals are most readily apparent in the Mask and Composite categories where faces are present. The faces are shown either straight frontally or in profile with the majority being frontal and with there being no examples of a three-quarter view. The forehead area of each face is protuberant but short in height. The brow is heavy but is undifferentiated from the forehead. The eyes, if they are present, which they are not in MAS 14, are round dots, usually small, and placed about 2/3 to 3/4 up the height of the face. The noses are long and straight, sometimes looking like plain bars in the middle of the face, as in COM1. In the cases, such as COM3 and 4, where the face is in profile, the noses are long triangles with the outer edges having been done linearly with a cutting wheel. The lips, where present, are done as two short lines with a cutting wheel. The cheeks are broad and flattish, flowing seamlessly into a U-shaped jaw, or half-U in the case of the profile examples. The scene outside the face proper are usually done with distinct sweeping lines, as in the case of COM1, and areas of rounded flat volumes, such as with the balls over MAS 5. The five non-face examples (BUC4, EAR3, FLO5-6 and LYR3) have very 446 similar treatments of line and volumes. Namely, volumes are rounded but flat while the lines are broad and distinct. FLO6, for instance, consists of a series of flat sub-circular petals around a flat, circular middle. The seal EAR3 also depicts the human ear in a similar broad, flat volume with curved exterior outline. The bucranium BUC4 has the volumes of the bull’s head and bird’s body also depicted as broad, flattish volumes with gently curving outlines and the horns and ears are undifferentiated broad lines. Meanwhile, LYR3 has the body of the lyre and the upright done as a curved but flattish volume and the strings of the instrument are each broad and emphasised. Put together, there is a consistency of execution within this group which points to the presence of a singe artist in their fabrication. The identity of this putative artist is unknown and not recoverable, but there are certain clues within his work that may shed some light on his character. The first and most important thing to note is that this artist, unlike the Kedesh archive users in general, stands apart from the glyptic developments of the larger world, especially in regards to the Aegean region. Instead, his subject matter can be idiosyncratic since it includes otherwise rarely attested motifs such as the lone example of the Phoenician mask and the composite eagle. His faces eschew the ¾ view and exaggerated expressions popular in the New Comedy slave masks, opting instead for a frontal presentation and a solemn countenance. His depiction of a lyre is structurally very different from the lyres generally shown in Greek art, or even the other lyres in the Kedesh symbol corpus. Indeed, the depiction of LYR3 does not even suggest that the artist knew what a traditional Greek lyre looked like. The artist, however, does occasionally produce images such as the composite hippocamp that illustrate knowledge of specific points in the larger glyptic 447 world. As such, the idiosyncrasies of his compositions may be the result of personal preference rather than limited contact. Furthermore, the style of execution itself seems to draw more from older Near- Eastern styles than from contemporary Greek ones. Specifically, the treatment of volumes, with their broad flattish surfaces and their curving outlines, is reminiscent of the treatment seen in the Persepolis Fortification and Modelled styles, which respectively possessed similar flat volumes and curving outlines.670 As such, it may be that the artist was active early on, either in or shortly after the Persian period. At the same time, the presence of the composite hippocamps, otherwise only attested in the Late-Hellenistic archive at Delos, in his corpus would seem to make this unlikely. Rather it appears that the artist lives in the 2nd century BCE, and is consciously hearkening back to an older period and style. The fact that this artist is so well represented in the sealings of Kedesh itself, with twelve examples in the symbol corpus alone, is also noteworthy. Indeed, much of the composite category appears to be by his hand. As such, it would appear that the people using his seals interacted with a fair frequency with the archive of Kedesh. It is therefore likely that the Kedesh Symbol Artist was himself someone fairly local, either working in and around Kedesh itself or someplace fairly nearby so that several of his customers would be interacting with the Kedesh archive, but not necessarily repeatedly as only COM4 and MAS5 have multiple sealings. All of this evidence suggests that the artist was also probably not Greek himself, since his work was so outside the normal trends of contemporary Greek glyptics, 670 Garrison & Cool Root, 2001, pp. 16 & 18. favouring instead a style and subject matter with Near Eastern roots. Indeed he may himself have been a local Phoenician, since h mask at Kedesh. App.1: The Seals of the Kedesh Symbol Artist 448 e produced the lone example of a Phoenician