»"i- CRETAN ■ PICTOGRAPHS AND "!l^,m(M:'',,y;:-^'fi- PRAE- PHOENICIAN • SCRIPT ..Vj^s,' n f^&~^. '€)) / % "S^iSSatOBUBiBawoi. ARTHUR J EVANS [-/ j;VANS (A. J.). Cretan Picto^iaphs and Frae-Pooenician Script. With an account of a Sepulchral Deposit at HagioB Onuphrioe nej,r Phaetos in its relation to Primitive Cretan and Aegean Culture. Roy. 8; and 139 illustrations in the text ; clo'h ; VERY SCARCE 1^^^ CRETAN PICTOGIUPHS PKAE-PHOENICIAN SCRIPT WITH AN ACCOUNT OF A SEPULCHRAL DEPOSIT AT IIAGIOS ONUPHRIOS NEAR PIIAESTOS IN ITS RELATION TO PRIMITIVE CRETAN AND AEGEAN CULTURE BY ARTHUR J. EVANS, M.A., F.S.A. KEETER OP xnE ASHMOLEAN MUSEUM AND HON. FELLOW OF BRASENOSE COLLEOF, OXFORIJ WITH A COLOURED PLATE, TABLES, AND 139 ILLUSTRATIONS IN THE TEXT fLontJon BERNARD QUARITCH, 15 PICCADILLY a. P. PUTNAM'S SONS, 27 WEST 23d STREET 1895 KiciiARD Clay and Sons, Limited, LONDON AND BUNGAY. SRLr URL NOTE. The first of these papers is reprinteJ, by permission, from the Hellenic Journal (Vol. xiv. Pt. 11. 1895) with some slight additions and corrections. The account of the Hagios Onuphrios deposit and its bearings on the prehistoric culture of the Aegean world is now issued for the first time. Already in 1893, on the occasion of my paper on 'A Mycenaean Treasure from Aegina ' I ventured to announce to the Hellenic Society that I had found what I believed to be a clue to the existence of a system of picture-writing in the Greek lands. The result of my explorations in Crete during the spring of 1894 was not only to confirm this discovery as regards the ' pictographic ' system but to add distinct evidence of the existence in the island at a very early period of a linear system of writing standing in a certain relation to the pictorial. A summary report of the results of my researches in Crete in the spring of 189-1 was sent by me to the Athenaeum from Candia on April 25 of last year, and appeared in that Journal on June 23. The Times of Aug. 29 published a further account of my Cretan discoveries, written by me at the request of the Editor, and I also read a paper on the subject in the Anthropological Section of the British Association, of which report.s appeared in the Academy and other papers. On that occasion I called attention for the first time to certain archaeological evidence connecting the Philistines with Mycenaean Crete. In the second paper of this book and the supplement are incorporated same further materials obtained by me during another Cretan journey undertaken this spring. SUMMAIIY OF CONTENTS. PlUMITIVK PKTOORAPIIS AND A I'EAE-PllOENlCIAN SCllIPT FROM Crete and the Peloponnese ... § I. — Cretan di.scoveiies ... § II. — The facetted stone.s with pictographic and linear symbols § III. — Evidences of a pictographic f-cript § IV. — Cla.ssification and comparison of the symbol.s ... § V. — The Mycenaean affinities of the Cretan pictographs § YI. — The earlier classes of Cretan seal-stones § VII. — The linear signs and their relation to the pictographic series ... The Sepulchral deposit of Hagios Onuphrios near Phaestos in its rel.\tion to the primitive cretan and aegean CULTURE Supplementary notes Index (270)- -(372) (270)- -(288) (288)- -(299) (300)- -(302) (302)- -(317) (317)- -(324) (324)- -(345) (346)- -(372) (105)- -(136) (137) (140) ILLUSTlUTIOiNS. PRIMITIVE PICTOGRAPIIS ETC. Fig, 1. — Signs oa vase-handle, Mycenae ... ... ... ... (273 Fig. 2. — Signs on amphora -handle, Mycenae ... ... ... ... (273 Fig. 3. — -Terracotta OS, GouU'is ... ... ... ... ... ... (278 Fig. 4. — Clay cup with incised characters, Goulas ... ... ... (278 Fig. ib. — Characters on Gouliis Clip ... ... ... ... ... (278 Fig. 5. — Vase with incised characters, Prodroinos Botzano ... ... (279 Fig. 56. — Chai-acters on vase from Prodromos Botzano ... ... (279 Fig. 6. — Bronze axe with incised character, Selakonos ... ... (280 Fig. 7. — Signs on bronze axe from Delphi ... ... ... ... (280 Fig. 8. — Engraved amethyst from Knosos ... ... ... ... (281 Fig. 9. — Signs on blocks of Mycenaean building, Knusos ... ... (282 Fig. 10. — Block at Phaestos with engraved signs ... ... ... (283 Figs, lltt, lib. — Engi-aved whorl from Phaestos ... ... .. (284 Fig. 12. — Button-seal with linear signs, Phaestos ... ... ... (285 Fig. 13, — Engraved button-seal, Messan'i .. ... ... ... (285 14. — Terracotta pendant from cave of Idaean Zeus ... ... (28ri Engraved disk-bead, Knosos ... ... ... ... ... (286 16. — Steatite pendant, Arvi ... ... ... ... ... (28G 17«, 17i.— Perforated steatite, Central Crete (286 18«, 18/a— Steatite relief, Messara (287 Fig. 19. — Inscribed steatite, Siphnos ... ... ... ... ... (287 Figs. 20rt, 206. — Types of prism-shaped seals ... ... ... ... (288 Fig. 21. — Convoluted seal-stone ... ... ... ... ... ... (288 Figs. 21— 31.— Three-sided seal-stones with pictographs ... (290—294) 2 Figs. 32 — 36. — Four-sided equilateral seal-stones with picto- graphs (294—297) 2. Fig. 38. — Convoluted seal-stone witli pictographs. Eastern Crete ... (298 Figs. 39 — 41. — Pictographic seal-stones of Mycenaean types... ... (299 Pictographic Symbols (302—315)33 Fi Fig. 15 Fig. Figs, Figs Fig. 42. — -Gem with sjjirals and palmettos, Couh'is Fig. 43. — Template symbol Fig. 44. — Template symbol with palmette Fig. 44. — Outline of palmette formed by template Figs. 46 — 47. — Diagrams illustrating use of template Fig. 48. — Design formed by template ... Fig. 49. — Egyptian scarabs. Twelfth Dynasty, and Early Cretan seal stones with designs derived from them ... Fig. 50. — Steatite seal-stone with spiral relief from Hagios Onuphrios deposit, Ph.ie.-tos (.328 (319 (320 (320 (320 (321 (322 (327 •.\flK 4 4 9 9 9 10 10 11 11 12 13 14 15 16 16 17 17 17 17 18 18 19 19 -25 -28 29 £0 -46 50 51 51 51 52 53 58 59 o vi ILLUSTRATIONS. PAGE Fig. 51. — Early compact type of triangular seal-stone, Class III. ... (331) 62 Figs. 52a, 526, 52c. — Types of primitive vases fi-om Cretan seal-stones (332) 63 Fig. 53. — Triangular seal-stoue bouglit at Smyrna ... ... ... (334) 65 Fig. 54. — Clay stamps from Early Italian deposit.*!, (a) Pollera Cave, Liguria, (b) Sanguineto Gave, Liguria, (c) Terramara of Montale (336) 67 Fig. 55. — Early seal-stone, grey^steatite, Praesos ... ... ... (337) 68 Fig. 56. — Early seal-stone, yellow steatite, spearman etc., Candia ... (338) 69 Fig. 57. — Early seal-stone, black steatite, Central Crete ... ... (338) 69 Fig. 58. — Early seal-stone, brown steatite, Crete ... ... ... (339) 70 Fig. 59. — Early seal-stone, lion-headed figure (Berlin)... ... .. (339) 70 Fig. 60. — Early seal-stone, dark steatite, two figures ikc, Central Crete (340) 71 Fig. 61. — Early seal-stone, black steatite, two-headed figure, Crete ... (340) 71 Fig. 62. — Early seal-stone, greyish-yellow steatite, camel ifec, Crete (341) Fig. 63. — Early seal-stone, yellow steatite, Crete ... ... ... (341) 72 Fig. 64. — Early seal-stone, yellow steatite, Crete, ostrich itc. ... (341) 72 Fig. 65. — Early seal-stone, brown steatite. Central Crete, cock etc. . (342) 73 Fig. 66. — Brown steatite, disk-bead, Kamares ... ... ... ... (342) 73 Fig. 67. — Green steatite, disk-bead, Crete, two figures ... (343) 74 Fig. 68. — Triangular bead-seal, Siteia ... ... ... ... ... (344) 75 Fig. 69. — Triangular bead-seal, three fish itc, Berlin ... ... ... (344) 75 Fig. 70. — Triangular bead-seal, green steatite. Twelfth Dynasty motive etc.. Central Crete (344) 75 Fig. 71. — Triangular bead-seal, yellow steatite. Twelfth Dynasty motive &c., Crete ... Fig. 72. — Triangular bead-seal, black steatite, bull's head itc, Candia Fig. 73. — Triangular bead-seal, Hagios Onuphrios deposit, Phaestos ... Fig. 74. — Black steatite seal with linear scrijit from Lower Egypt ... Fig. 75. — Signs on pot.sherds at Tell-el-He.sy compared with Aegean forms Fig. 76. — Inscription probably in Eteocretan language from I'raesos, in ai-chaic Greek letters ... Table I. — Cretan and Aegean linear characters compared with Aegean signs found in Egypt and Cypriote forms Table II. — Groups of linear symbols, from Ci'ete, Mycenae, and Siplinos Table III. — Pictographs and linear signs comp.ired with Cypriote and Semitic parallels ... THE IIAGIOS ONUPHRIOS DEPOSIT, ETC. Fig. 77. — White steatite scarab Fig. 78. — Steatite scarab. Twelfth Dynasty Fig. 79a, 79J, 79c. — Steatite bead-seal Fig. 80. — Steatite bead-seal with convoluted relief ... Fig. 81a, 81i, 81c.— Clay cylinder Fig. S16i«. — Dark steatite button-.seal with Twelfth Dynasty motive (344) 75 (345) 76 (345) 76 (347) 78 (351) 82 (355) 86 (349) 80 (35.3) 84 (365) 96 106 106 106 106 ... 107 107 ILLUSTRATIONS. vii TAOE Fig. 82a, 826. — Eagle-shaped seiil of greeu steatite 107 Fig. 83a, 836. — Eagle-shaped seal from llauiiin ... ... .. ... 107 Fig. 84a, 846.— Steatite cone seal 107 Fig. 85rt, 856. — Ivory cone seal ... . . ... ... ... ... 108 Fig. 86. — Ivory cone seal ... ... .. ... ... ... ... 108 Fig. 87. — Ivory cone seal ... ... ... ... ... ... ... 108 Fig. 88.— Steatite pendant 110 Fig. 89a-J.— Beads ... 109 Fig. 90. — Crystal pendant mounted with gold ... ... ... ... 110 Fig. 91— 94.— Gold ornaments 110 Fig. 95 — 96. — Bronze perforated objects plated with gold ... ... ... Ill Fig. 966i's.— Do. Steatite .'. Ill Fig. 97. — Bronze gold-plated object, perhaps hilt .. ... .... ... Ill Fig. 98. — Gold terminal ornament ... ... ... ... ... ...Ill Fig. 99. — Marble pendant in form of oeuochoe ... ... ... ... 112 Fig. 100. — Clay suspension vase with cover ... ... ... ... ... 112 Fig. 101. — Small clay suspension vase, Arvi ... ... ... ... ... 113 Fig. 102. — Cover of clay suspension vase ... ... ... ... ... 113 Fig. 103a, 1036.— Small .spouted vase 113 Fig. 104. — Small vase for suspension ... ... ... ... ... ... 113 Fig. 105. — Horned object of brown bucchero .. . ... ... ... ... 114 Fig. 106a. — Spouted vase painted yellow with terracotta stripes ... ... 114 Fig. 107. — Early painted pyxis ... 115 Fig. 108.— Early painted jar 115 Fig. 109. — Small limestone vessel, Phaestos ... ... ... ... ... 116 Fig. 110. — Variegated limestone vessel, Phaestos ... ... ... .. 117 Fig. 111. — Serpentine vessel on limestone pedestal of Fourth Dynasty date, Ghizeh Museum ... ... ... ... ... ... 118 Fig. 112. — Brown stone vase from Pinies near Elunta (Olous) ... ... 118 Fig. 113. — Grey steatite pot, Gouli'is ... ... ... ... ... . 120 Fig. 114.— Steatite vase, Arvi ' 120 Fig. 115. — Steatite cup, cist grave, Aivi ... ... ... ... ... 120 Fig, 116. — Steatite mug, cist-grave, Arvi ... ... ... ... ... 120 Fig. 117. — Steatite pot with cover, cist-grave, Arvi ... ... ... ... 121 Fig. 118. — Small limestone pot. Arvi ... ... ... ... ... 121 Fig. 119. — Banded limestone pot with cover, Arvi ... ... ... ... 121 Fig. 120.— Steatite lid, Twelfth Dynasty deposit, Kahun, Egypt 122 Fig. 121.— Steatite bowl, cave, Psychro, Crete ... 122 Fig. 122. — Limestone conglomerate pot, Chersonesos, Crete ... ... ... 123 Fig. 123.— Foliated steatite vase, Mi'lato 123 Fig. 124. — Marble ' idol,' Hagios Onuphrios deposit ... ... ... ... 125 Fig. 125. — Marble ' idol,' Hagios Onuphrios deposit ... ... ... ... 125 Fig. 126. — Marble ' idol,' Hagios Onuphrios deposit 125 Figs. 127a, 1276.— Small marble 'idol,' Phaestos 125 Fig. 128.— Marble 'idol,' Hagios Onuphrios deposit 125 Fig. 129.— Marble ' idol,' Hagios Onuphrios deposit 126 viii ILLUSTEATIONS. Fig. 130. — Marble 'idol' (female) Hagio-s Oniiphrios deposit Fig. 131. — Marljle 'idol' (female) Hagios Onuphrios deposit Fig. 132. — Head of marble ' idol,' Hagios Onuphrios depo.sit Fig. 133.— Marble ' idol ' (female) Siteia Fig. 13-1.— Marble 'idol; Siteia Fig. 135. — Marble ' idol,' cist-grave, Amorgos ... Fig. 136. — Stone mould from Selendj, Maeonia Figs. 137(1, 1376, 137e. — Lead figure and ornaments said to hav found near Candia Fig. 138. — Square ended bronze dagger, Hagios Onuphrios deposit . Fig. 139. — Double-pointed bronze spear, Hagios Onuphrios deposit . PI. I. [xii.]. — Design of Mycenaean ceiling reconstructed with the aid of the Goulas gem (Fig. 42) and the template symbol... been PAGE 126 126 126 128 128 129 133 134 135 136 At end PRIMITIVE PICTOGIiAPHS AND A PPiAE-PHOENICIAN SCRIPT, FROM CRETE AND THE PELOPONNESE. [270] PRIMITIVE PICT0GRAPH8 AND SCRIPT PRIMITIVE PICTOGRAPHS AND A PRAE-PHOENICIAN SCRIPT FROM CRETE AND THE PELOPONNESE. § I. — Cretan Discoveries. In the absence of abiding monuments the fact has too generally been loot sight of, that throughout what is now the civilized European area there must once have existed systems of picture-writing such as still survive among the more primitive races of mankind. To find such ' pictographs' in actual use — the term is used in its most comprehensive sense to cover carvings on rocks or other materials whether or not actually overlaid with colour — we must now go further afield. Traces of such may indeed be seen on the rude engrav- ings of some megalithic monuments like that of Gavr Innis, on the rock carvings of Denmark, or the mysterious figures known as the Maramglic wrought on a limestone cliff in the heart of the Maritime Alps, to which may be added others quite recently discovered in the same region. In Lapland, where designs of this character ornamented the troll-drums of the magicians till within a recent period, survivals of some of the traditional forms may still be found to the present day, engraved on the bowls of their reindeer-horn spoons. Of actual rock-paintings perfectly analogous to those of Cherokees or Zulus, I have myself observed an example — consisting of animals and swastika-like figures painted probably by early Slavonic hands on the face of a rock overhanging a sacred grotto in a fiord of the Bocclie (li Cattaro. But the perishable nature of the materials on which picture-writing, having for most part only a temporary value, was usually wrought has been fatal to the survival of primitive European pictographs on any large scale. If we had before us the articles of bark and liide and wood of early man in tiiis quarter of the globe or could still s(!e the tattoo marks on his skin we should have a very dift'erent idea of the part once played by picture-writing on European soil. As it is, it is right- tiiat the imagination sho\ild supply the deficiency of existing evidence. In the areas embraced by the older civilizations such as Egypt, Babylonia and China, a different kind of influence has been at work, by which the void caused by the disappearance of the more primitive materials may in a great measure be filled up. For there the early pictographic elements, such as we 2 FROM CRETK AND THE PELOPONNESK. [JTIJ still liiid tlicm uniong savage lacus, were, in tliu liands of priestly and official castes, developed into a more complicated and exact system of writing, by which however we are enabled in many cases to trace back the original forms of the object selected. The same development from the simple pictographic to the hieroglyphic or quasi-alphabetic stage might naturally have been expected to have taken place in more than one European area had it not been cut short by the invasion of tlie fully equipped Phoenician system of writing. Even as it is however, it must be allowed that there are strong a ^n-iori reasons for believing that in the Greek lands where civilization put forth its earliest blossoms on European soil, some such parallel evolution in the art of writing must have been in the course of working itself out. For we now know that in the South-Eastern part of our Continent there existed long before the days of direct Phoenician contact an inde- pendent form of culture which already as early as the first half of the second millennium before our era might be regarded as in many respects the equal contemporary of those of Egypt and Babylonia. In view of the extraordinary degree of artistic and mechanical development reached by the representatives of what is now conveniently known as the Mycenaean civilization — at least as early, approximately speaking, as the seventeenth century, B.C. — and the wide ramifications of their commerce, is it con- ceivable, it may be asked, that in the essential matter of writing they were so far behind their rivals on the Southern and Eastern shores of the Mediterranean ? There is moreover a further consideration which tends to make the absence of any system of writing among the Mycenaean peoples still more improbable. At the dawn of history Asia Minor, whether we regard the predominant elements of its population from the point of view of race or of culture, may be said to belong to Europe. Its area from the earliest times of which we have any record was largely in the occupation of the great Thraco-Phrygian race and its offshoots. Its prehistoric remains, as far as we know them from Cyprus to the Troad, fit on to those of a large archaeological area, the continuation of which may be traced over the island stepping- stones of the Aegean to the mainland of Greece, while in the other direction kindred forms extend along the Danubian system to reappear amongst the pile-dwellings of Switzerland and Carniola, the trrrc-inare of the Po valley and even in Ligurian caves. But it is on the Eastern borders of this wide field of primitive culture that recent researches have brought to light the principal seats of the higher form of early civilization conveniently known as Hittite. Living in the Syrian and Cappadocian regions in the immediate proximity of upper Mesopotamia, and almost in the highways as it were of old Chaldean culture, its representatives yet show independent characteristics and traditions, the sources of which seem to be drawn from the North or West. And of these one of the most noteworthy is the possession of an original system of hieroglyphic writing, the relics of which are scattered from the banks of the Orontes to the Western shores of Anatolia. At a later date ' B 2 [272] PRIMITIVE PICTOGRAPHS AND SCRIPT 3 again we fi ud the Greeks of Cyprus and the inhabitants of a large tract of Asia Minor in the possession of syllabic scripts altogether distinct from the Phoenician alphabet. When it is once realized how largely the early civilization of the Aegean Islands and even the mainland of Greece was evolved out of similar elements to those of Asia Minor, it must certainly seem surprising that on this side no system of writing belonging to prae-Phoenician times should as yet have been clearly ascertained. The geographical contiguity to Anatolia, and the early trade relations which can be shown to have existed between the Aegean Islands and the valley of the Nile would assuredly, it might be thought, have given an impulse to the higher development of whatever primitive form of picture-writing was already to be found amongst the inhabitants of this Mediterranean region. It is impossible indeed to suppose that this European population was so far below even the Red Indian stage of culture as not to have largely resorted to pictography as an aid to memory and communication. And — even if an existing system was not perfected under the influence of foreign example — the race which laid the arts of Egypt and Western Asia under such heavy contribution was at least capable of borrowing and adapting a system of writing. It is true that Schliemann's great discoveries at Mycenae produced nothing that could be safely interpreted as a form of script. The objects seen in the field of many of the ordinary Mycenaean gems — the so-called 'island-stones' — are simply inserted as the space left by the principal design suggests, and are primarily of a decorative character — and due to the horror vacui of primitive art. Nevertheless, especially when we see a part standing for a whole — as a branch for a tree or the head of an animal for the animal itself — it may be fairly said that many of these gems do bear the impress of people familiar with the expedients of primitive picture- writing, such as we find it still in so many parts of the world. The Icntoid and amygdaloid gems in question did not, as we now know, serve the purpose of seals, but were simply ornamental beads worn round the wrist or neck.* Like the oriental periapts, however, worn in the same manner at the present day, they may often have been intended to serve as amulets or talismans ; and both the principal type of the intaglio and the smaller or abbreviated forms introduced into the field may have possessed something beyond a mere artistic significance. Still more is this likely to have been implied in the case of the engraved designs on the besils of the gold rings from the Mycenaean graves which seem actually to have served the purpose of signets. It certainly is not unreasonable to suppose that in this case some of the smaller objects in the field may have had a conventional religious mean- ing, and that they were in fact ideographs taken from a recognized hiero- glyphic code. The bulls' heads and lions' scalps, the ears of corn and double ' See Tsountas, 'AvacfKo^ol toi^wv iv Mukt)- to this rule in case of some Cretan lentoidgems TOis. 'Ef. 'Apx- 1888, p. 175. There arc Iireaenting groups of symbolic figures, probibly, as will be seeu below, some cxceptious 4 FROM CRETE AND THE PELOPONNESE. [273] axe certainly suggest that we have here to deal with symbols of divinity, perliaps standing for the divinity itself, or ideas of cult and sacrifice, — the latter form of symbolism being well brought out by the gold ornaments representing oxes' heads with a double axe between the horns. In the same way, to take an example from the practice of modern savages, a drawing of eyes and beak stood among the Iroquois for the Thunder-Bird or a rayed liead for a Spirit among the Ojibwas. The whole of later Greek symbolism may in fact be regarded as a survival, maintained by religious conservatism, from a wide field of primitive pictography. The figure that stands as the personal badge of the names of individuals at times actually appears as the equivalent of the written form of the name, as when a monetary magistrate called Leon places a lion on his dies. The same symbolic script is frequent in the rendering of city names, one of the most interesting examples being found on a coin of Mesembria where the part of the civic legend signifying day is supplied by a simstika — the emblem of the midday sun.'^ The symbols on the Mycenaean seals are themselves of too isolated occurrence to be used straight away as examples of a hieroglyphic system — though there seem to me to be good reasons for supposing that some at least among them did fit on to such a system. But more recently one or two objects have been found at Mycenae itself and in Mycenaean deposits else- where which are calculated more effectually to shake some of the preconceived notions of archaeologists as to the non-existence in Greece of a prae-Phoenician system of writing. The most important of these are the handle of a stone vase apparently of a local material (Fig. 1) found at Mycenae, wliich has Fig. 1.— SIGN.S on Vase- Handle, Mycenae. four, or perhaps five, signs engraved upon it, and the handle of a clay amphoi'a from a chambered tomb in the lower town of Mycenae with three Fig. 2. — Signs on Amphora-Handle, Mycenae. characters (Fig. 2). Single signs have also been noticed on the handles of two amphoras of the same form as the last found in the Tholos tomb of 2 P. Gardner, Num. Chron. 1880, p. 59 ; Head, Hist. Num. 237. [274] PRIiMITIVE PICTOGRAPHS AND SCRIPT 5 Menidi,^ on a three-handled vase from Nauplia* and a stone pestle from Mycenae.'' Dr. Tsountas in describing these finds lays stress on their occurrence in two cases in groups of three and four respectively, and rea.son- ably asks whether we have not here to deal with some form of writing. Professor Petrie again has discovered a series of isolated symbols on what he considers to have been fragments of early Aegean pottery discovered by him at Gurob in a deposit which he assigns to the period of the Twelfth Dynasty, and again at Kahun amongst Eighteenth Dynasty I'elics.'' Notwith.standing these indications, however, the last writer on the Mycenaean and early Aegean culture, M. Perrot, sums up the evidence as follows : ' The first characteristic which attracts the historian's notice when he tries to define the prae-Homeric civilization is tliat it is a stranger to the use of writing. It knows neither the ideographic signs possessed by Egypt and Chaldaea nor the alphabet properly so called which Greece was afterwards to borrow from Phoenicia.' He admits indeed that some of the marks recently observed on the vase-handles bear resemblance to letters, either Greek or Cypriote, but observes that they do not seem to form words, and that they are perhaps uotliing more than the marks of the potter or the proprietor, or ignorant copies of Phoenician or Asianic characters. 'As at present advised,' he concludes, ' we can continue to affirm that for the whole of this period, nowhere, neither in the Peloponnese nor in Greece proper, no more on the buildings than on the thousand objects of luxury or domestic use that have come out of the tombs, has there anything been discovered which resembles any kind of writing.' ' The evidence which I am now able to bring forward will, I venture to think, conclusively demonstrate that as a matter of fact an elaborate system of writing did exist within the limits of the Mycenaean world, and moreover that two distinct phases of this art are traceable among its population. The one is pictographic in character like Egyptian hieroglyphics, the other linear and quasi-alphabetic, much resembling the Cypriote and Asianic sjdlabaries. In the course of a visit to Greece in the spring of 1893 I came across .some small three- and four-sided stones perforated along their axis, upon which had been engraved a series of remarkable symbols. Tlie symbols occurred in groups on the facets of the stones, and it struck me at once that they belonged to a hieroglyphic system. They were however quite distinct from •' Tsountas, yiTiKTivat p. 213. One has a sijfn marks (see below, p. 282) M. I'crrot had pievi- rescmblini; tlie Greek fF, thi! other, 4= the otisly admitted (»/). cit. 461) that tlie Cypriote Cyjiriote, pa, ha, or p/ia. signs may have had an Aegean extension 'dnriug * 'Af>Xaio\oyiKhv AeATior, 1892, p. 73. It a certain time.' Hut the subseiiuent passage on was discovered by Dr. Stnis in a tomb of the p. 985 retracts this admission as far as the My- I'roHoea. On (^acli liaiidle was engraved a sign cenacan period is concerned. Dr. Kcich(d sng- likc the Greek H but witli offshoots from the gests {Ifomcrisdic Waffen, p. 112) that the tn|i of the upriglit strokes. linear designs below the coniliatants on thi' silver •'■ nf)oi(TiKoT5s'Apx«"'^<'7". characters so remarkably alphabetic that they might well be taken to belong to much later times — Byzantine, for instance. But the evidence against this view must be regarded as decisive. The H and A are both found among the early marks observed by Professor Petrie on the Kahun pottery ; read another way the X is a Cypriote ve. On the upper side of this whorl (Fig. II«) is seen a rude engraving of a horned animal — probably a bull or ox — which is quite in the style of the animal representations of a series of very early Cretan intaglios.^^ This figure is followed by a peculiar symbol and, what is extremely remarkable, on the lower side of the stone the same symbol recurs in immediate juxtaposition to what appears to be the bull's or ox's head reduced to a linear form.^^ The engraving of the upper and lower side of the stone seems to be by the same hand. The material itself, a greenish steatite, and the irregular form both occur moreover in the case of another inscribed stone from Siphnos to be described below, bearing letters showing a very marked affinity with Cypriote. Again, every other object from the deposit in which this inscribed whorl was found seems to be of very early fabric.^'-"^ The prima facie view of the characters on this curious stone might easily lead to the conclusion that it was of much later date. But the early, irregular form and material, the rude animal, the curious association of signs unknown to the later Greek alphabet, and the place of finding point to an antiquity corresponding with that of the other relics from the same sepulchral stratum. From the same deposit was obtained a button-like pendant of black '* I may specially cite a rudely triangular Compare too tlie animal ou Fig. 18«. steatite, with a horned animal in a very primi- '" Sec below, p. 364-366. tivo style, found with other early pendants in ""^ See below, ]i. \0i scqq. a grave of prac-Myceuacan date at Milato. 16 FROM CRETE AND THE PELOPONNESE. [285] steatite engraved with linear signs (Fig. 12) and a sketch of another object of the same class has been kindly placed at my disposal by Professor Halbherr. Tiie object in question (Fig. 18) is of a green stone somewhat more regularly formed tlian the other and shows on its face a K-like character, Fig. 12 (2 diams.) though it is uncertain which way up the sign should be set. It was found by Dr. Halbherr in a necropolis of the last Mycenaean period in Messara consisting of oven tombs, but the pendant itself may possibly belong to a still older stratum. Fig. 13. — Engraved 'Button-Seal,' Messaka (2 diams.).. iSLy attention has been further called by Dr. Hazzidaki to a perforated terracotta object, apparently also a kind of pendant (Fig. 14), witli an incised symbol consisting of a horizontal line with two cross-strokes, like the Cypriote prt turned on its side, from the cave of the Idaean Zeus. On a perfor- ated disk from the site of Knosos (Fig. 15) there occurred a sign like a Cypriote po. From one of a series of early cist-graves at Arvi (Arbi), on the South-East coast of the island, containing stone vessels and other relics of prae- Mycenaean date I obtained a green steatite pendant (Fig. 16) with two linear symbols, one on each side, curiously resembling an Alffand Gimcl. Fig. 17, from Central Crete, a perforated triangular steatite of irregular form, also shows on two of its faces curious linear signs. Fig. 18a and & is a dark brown steatite ornament from the Messara district, having on both sides of c [286] PRIMITIVE PICTOGRAPHS AND SCRIPT 17 Fig. 14. — TF.nnAcoTTA Pkxdaxt from CAvn nr Idaf.an Zeus. Fig. 15. — EyGRAvnn Dlsk-Bead, Knosos (2 diams.). F[i:. It).— Steatite Pendant, Ar.vi (2 diams.). Fro. ni.— rmu'oiiATEi) Steatite, Centuai. Crete (2 (li.ainfi.). 176. 18 FROM CRETE AND THE PELOPONNESE. [287] Fig. 18re. Fio. 18ft.— Steatite Relief, Mesr.a.iiX (2 diains.). Fig. 19.— Siphnos. C 2 [288] PRIMITIVE PICTOGRAPHS AND SCRIPT 19 it figures iu relief. On one side are what appear to be two primitive repre- sentations of animals, the style of one of which recalls the ox on the Praesos disk, while on the other face are two tortoises and an uncertain symbol grouped together like some of the pictographs on the triangular seals to be described below. To these Cretan examples I may add a pale gieen perforated steatite (Fig. 19) from Siphnos, in material somewhat resembling the Phaestos disk, one side of which is engraved with characters of curiously Cypriote aspect. II. — The Facetted Stones with Pictographic and Linear Symbols. As forming a group by themselves it has been found convenient to reserve the detailed examination of the facetted stones presenting picto- graphic symbols for a separate section, and at the same time to place with them the prism-shaped seals of the same type with more linear characters. Another form of bead-seal and two examples of lentoid gems with picto- graphic groups are also added. The facetted stones themselves are of three principal types, all of them perforated along their major axis. I. — Tiiree-sided or prism-shaped (Fig. 20 a and h). This type is divided into two varieties — one elongated (a) the other more globular (/3). Fig. 20a.— (2 dianis.). Fig. 206.— (2 diaius.). IF. — Four-sided equilateral. III. — Four-sided with two larger faces. IV. — With one engraved side, the upper part being ornamented with a convoluted relief (Fig 21). Fig. 21. — (2 Jiams.). 20 FROM CKETE AND THK PELOPONNESE. [289] This form may peiliaps be rogarded as a later devolopineut of an earlier type of Cretan bead, the upper part of which is carved into the shape of two Ncrita shells lying end to end with a common whorl, a specimen of which was found in the Phacstos deposit above referred to. The other stones, which are of ordinary Mycenaean forms including the lentoid type, are grouped with the above as Class V. The figures are taken from casts, so that, assuming that the originals were seals, this gives the right direction of the symbols. In some cases however it is not easy to decide which way up the impression should be shown, and the order in which the sides are arranged is for the most part arbitrary. When one side presents a single type of an evidently ideographic character it has been given the first place, and at times a boustrophedon arrangement seems to be traceable. In Fig. 23 for instance, the first side seems to run from right to left, the second from left to right, and the third again from right to left. The drawings were executed by Mr. F. Anderson with the guidance of magnified photographs from casts, and the stones are in all cases enlarged to two diameters. Effects due to the technique of the early gem-engraver's art, such as the constant tendency to develop globular excrescences, must be mentally deducted from the pictographs. Unless otherwise indicated, the stones or their impressions were obtained in Crete by the writer. [290] PKIMITIVE PICTOGRAPHS AND SCPvlPT 21 Glass I. Three-sided or Prism-shaped. /I. (Fig. 21). — Brown steatite. Crete. Uncertaiu locality. 22f(. h. (Fig. 22). —Green jasper. Pro\^ince of Siteia. Crete. FROM CRETE AND THE PELOPONNESE. [291] 23a. C. (Fig. 23). — White cornelian. Eastern Crete. ■Zia. 246. Fig. 24. 24(; D. (Fig. 24). — White cornelian. Eastern Crete. 25c. JS. (Fig. 25). — Crete. (Berlin Museum.) [302] PillMITIVE PIGTOGEAPHS AND SCPJPT 23 26c. (P^ li^', •26a. F. (Fig. 20). — Red coiucliau. Crete. Province of Sitcia. 27c. 0. (Fig. 27). — Brown steatite. Crete. Uncertain localitj-. Sides h and c contain wliat appear to be purely decorative designs. W>^/^ ^' 28c. //. (Fig. 28). — Steatite. Crete. Uncertain locality. 24 FI{OM CRETE AND THE PELOPONNESK. [293] 29r(. 29J. /. (Fig. 29). — Wliite steatite. Praesos. ft '■ J-' Fig. 30a. l^^^\tS Fig. oOJ. Fig. 30c. Fig. 30. -/". (Fig. SO). — Grey steatite. Kn6sos. (From a sketch.' [294] PKIMITIVE PIOTOGRAPHS AND SCRIPT 25 3U-. 31a. K. (Fig. 31). — This stone belongs to the more globular type, Class I. h, Crete^ (Berlin Museum.) Glass II. Four-sided Equilateral Stones. 32rf. 32c. 32i Fig. 32. 32a. A. (Fig. 32). — Red cornelian. Crete (Ashmolean Museum ; Mr. Greville Chester; wrongly labelled as ' from Sparta,' see p. 136). 26 FROM CRETE AND THE PELOPONNESE. [295) 33b. ^SiF^^' *s^ 336. 33c. P^Bu i^u^ ■i3d. Fig. 33. />'. (Fig. 33). — Crete. Province of Siteia. (PolytecLniou, Athens.) I2J61 PRIMITIVE PICTOGRAPHS AND SCRIPT 27 oi((. 346. :i^ 77T?' ^ 34f;. r^ ^MiM^ ^iS i.j^'j,;; 34rf. Fig. 34. C. (Fig. 34).— Crete. (Berlin Museum.) 28 FROM CRETE AND THE PELOPONNESE. I29< 35r(, 356. m4M^'^ 35c ^?; \v 35rf. Fin. 35. 1). (Fig 3.5). — Green jasper. Crete. Palaekastro, near site of Itanos, Sbff. 366 36c. E. (Fig. 36).— Steatite. Province of Siteia. (Polytechnion, Athens.) Sides a and c contain decorative designs. [298] PRIMITIVE PICTOGEAPHS AND SCRIPT 20 Class III. Four-sided Stones with Two Larger Faces. 37ffl. ■ill. Fig. 37. A. (Fig. 37). — Green steatite. Central Crete. This .stone properly belongs to an earlier class. Class IV. Stones with a Sinole Face: the Upper Part Convoluted. Fig 38. A. (Fig. 3S). — White cornelian. Eastern Crete. 30 FROM CRETE AND THE PELOPONNESE. [299] Glass V. Stones of Ordinary Mycenaean Type. Fig. 39. A. (Fig. 39). — From impression taken by Professor Sayce at Athens. Tliis form of gem was in use for the besils of rings in Mycenaean times. Fig. 40. B. (Fig. 40). — Brown steatite. Knosos. This and the following are ordinary types of perforated lentoid bead but of very early fabric. Fio. 41. C. (Fig. 41). — Black steatite. Messar^ district. [.iOO] PRIMITIVE PIC'TOGRAPHS AND SCRIPT 31 § III. — Evidences of a Pictographic Script. It is impossible to believe that the signs on these stones were simply idle figures carved at random. Had there not been an object in grouping several signs together it would have been far simpler for the designer to have chosen single figures or continuous ornament to fill the space at his disposal. As it is, single figures or continuous ornament are occasionally introduced on the vacant sides of stones where it w^as not necessary to cover the whole stone with symbolic characters ; and in the same way small ornamental forms are found in some cases filling, for decorative purposes, the spaces between the symbols. In Fig. 22 one side is purely decorative ; in Figs. 27 and 36, two sides, and such features as the small chevrons in the vacant spaces of Fig. 31f, or the network behind the designs on Figs. 33a and 34c and d, are obviously supplementary ornaments. But these extraneous features only bring out more clearly the fact that the signs themselves are introduced with a definite meaning, and are in fact a form of script. A method and intention in the choice and arrangement of the symbols is moreover percep- tible, quite incompatible with the view that they are mere meaningless ornaments. The signs themselves are chosen from a conventional field. Limited as is the number of stones that we have to draw from, it will be found that certain symbols are continually recurring as certain letters or syllables or words would recur in any form of writing. Thus the human eye appears four times and on as many different stones, the 'broad arrow' seven times, and another uncertain insti'ument (No. 16 of the list given in the succeeding section) as much as eleven times. The choice of symbols is evidently restricted by some practical consideration, and while some objects are of frequent occurrence, others equally obvious are conspicuous by their absence. But an engraver filling the space on the seals for merely decorative purposes would not thus have been trammelled in his selection. Two other characteristics of hieroglyptic script are also to be noted. The first is the frequent use of abbreviated symbols, such as the head for the whole animal, the flower or spray for the plant. The second is the appear- ance of gesture-language in graphic form — an invaluable resource of early pictography for the expression of ideas and emotions. Amongst such may bo noted the human figure with arms held down (Fig. 36/y), the crossed arms with open palms and thumbs turned back (Fig. 31i), and, closely allied to this, the bent single arm with open palm (Fig. 3.5rf). Such features, again, as the wolf's head vvitli protruding tongue — also found on Hittite nioiiu- ments — or the dove pluming its wing, have probably a significance beyond the mere indication of the animal or bird. The symbols occur almost exclusively in groups of fiom two to seven ; the most frequent however are of two or of three, which seems to show that the characters tlms appearing had a syllabic value. Certain fixed prin- 32 FROM CRETE AND THE PELOPONNESE. [301] ciples, also, are traceable in the arrangements of the symbols in the several groups. Some signs are almost exclusively found at the beginning or the end of a line. Tlie human (>ye appears thus three times out of four ; the instrument No. 16 of the list below occupies the extremity of the group in seven, or perhaps eight, cases where it occurs. The same two symbols more- over are seen on difterent stones in the same collocation. Thus the horns and four-rayed star occur in close proximity on the stones (Fig. 23& and 32i) from Crete. The instrument (No. 10) above referred to occurs five times on as many different stones iii collocation with the ' broad arrow.' The arrow- head, again, is twice placed beside the y-like sign No. 54 (Figs. 23i and 35a). In four cases where the bent leg makes its appearance (Figs. 22b, 2oa, 34'& and Supplement, p. 136), it is in immediate contiguity with a symbol that seems to stand for a door or gate. Such collocations in the small number of instances at our disposal are alone sufficient to exclude the supposition that the signs on these stones were engraved haphazard for decorative purposes. It further appears, when we come to file the several columns, as on the Babylonian principle they would follow one another in the impression of a seal, that in several cases a boustrophedon arrangement has been adopted which recalls that of early Greek writing. This is specially noticeable in Figs. 22, 23, 33. as well as in Fig. 34, where by the analogy of other Myce- naean gems from Crete representing ships the vessel must be taken as going in the direction in which the oars slope. It seems usual to begin from right to left. That these seals were designed to convey information regarding their owners in a primitive form of writing is clearly brought out by another phenomenon with which we have to deal. On Fig. S6d the place of the pictographic symbols is taken by linear characters which no one will deny represent actual letters, and which fit on in fact to an Aegean or Mycenaean syllabary the existence of which can be demonstrated from inde- pendent sources. This phenomenon must certainly be taken to throw a retrospective light on the hieroglyphic forms that leplace the letters on the bulk of these stones. It will be further shown in the course of this inquiry that a certain proportion of these pictographic signs reduced to linear forms actually live on in this Aegean syllabary. In a succeeding section i-'" attention will be called to a still earlier class of Cretan seal-stones presenting for the most part the same typical tri- angular form as those of Class I. already described. These more primitive stones, which cannot in fact be separated by any definite line of demarcation from the later scries, throw a valuable light on the original elements out of which the more formalized pictograpliic system finally grew. In some cases the same symbols are actually seen in a more primitive stage of development. But on this earlier class the more purely pictorial and ideographic elements "» See p. 324 scqq. The stones, Figs. 21 , .37, 39, 40, miglit rerliaps with greater propriety have been grouped with tliis earlier ."Series. [302] PRIMITIVE PI0T0C4RAPHS AND SCRIPT 33 are uaturally preponderaut auJ the personal relation in which the seals stand to their owners is clearly revealed. They seem indeed to be descriptive of his individual character as an owner of flocks and herds, a merchant, a huntsman or a warrior. These more naive delineations, of a ruder stage of culture, supply a welcome clue to the interpretation of such ideographic elements as survive in the more conventional forms with which we are at present dealing. Here too we may often see a reference to the avocation or profession of the owner of the sear and may venture to conclude that the more purely symbolic characters have a personal application. Thus for example Fig. 34, exhibiting at the beginning of one column a ship with two crescent moons above it, may be reasonably supposed to have been tbe signet of one who undertook long voyages. Fig. 2-i, with the pig and door, would have belonged to some one who owned herds of swine : in which case the two figures of the axe and kid on the other face may contain the elements of the owner's own name. The fish at the head of Fig. 33 may indicate a fisherman. The seal-stone represented in Fig. 23, with the adze and other implements — including one in which I have ventured to recognize the template of a decorative artist, — probably belonged to a member of a masons' guild. The harp on Fig. 31 suggests a musician. It is possible that the individual element of ownership, which on the earlier class is" brought out by the complete human figure, may be elsewhere indicated by the human eye alone, which is of frequent occurrence in these stones. § IV. — Classification and Comparison of the Symbols. In the following list I have included all the above signs that have any claim to be regarded as of hieroglyphic value, excluding the small obviously ornamental devices that are occasionally found filling in the space between the symbols, but including one or two like the S-shaped figures that may after all belong to the same decorative or supplemental category. It will be seen from the arrangement adopted that the symbols, where it is possible to recognize their meaning, fall into regular cla.^ses like the Hittitc or the Egyptian. The Human Body and its Parts. 1 y Fig. 3G/'. Ideograj)!! of a man standing alone, with his arms f«n held downwards, perhaps denoting ownership. It is followed /l ^y linear characters on another facet of the stone. Human "* *• figures in this position are frequent on Cypriote cylinders. A similar figure also occurs on a cone from Kamleh, near JafTa, in the Ashmolean Collection. 34 FROM CRETE AND THE PELOPONNESE. [503] Figs. 29«, 32r/, Uh ^^^ ^.^ ^.^ aud c, Sob ami c, and r^< (Q) © •^•'^- The eye appears (a) (/O 00 (,l) (r) tNvice in conjunction ^ ^ ^ ^ ^ ^ ' ^ ' with No. 16. As an indication of meaning we may compare Egyptian cC^^S^ '"^ '' ^^^'^ deter- minative of ' siglit,' ' watching,' &c. On tlie Hittite monuments the eye does not seem to be separately portrayed. On the inscriptions of Haniath and Jerabis (Wright, Fmp. of the Hittitcs, PI. I. H. 1, line 1, and PI. VIII. A., line 1) the upper part of a figure of a man is represented, with his finger apparently pointing to his eye. In the delineation of this symbol on the Cretan seal-stones, four distinct stages are perceptible : (1) the whole eye with the lashes all round; (2) the whole eye with the lashes fully drawn on two diagonal sides of tiie eye only, elsewhere only faintly indicated ; (3) what appears to be an abbreviated form of the latter type ; (4) the pupil and iris only, indicated by concentric circles. In one case (Fig. 3.5) this latter type occurs on the same stone as the complete eye in a place where it would have been impossible to insert the fall symbol. It is, however, difficult to distinguish this latter simplified form, con- sisting of concentric circles with or without a central dot, from what appears to be a solar symbol. (See below, No. 02.) Y Fig. 31?). Another ideograph taken from gesture-language. The sign may have indicated 'ten' or any multiple often : thus any great number. So far as the crossing of the arms goes, the symbol may be compared with the two confronted figures that occur twice on a Jerabis monument (Wright, oj). cit. PI. IX.). ^ % ^ig- ^''^'^- ^1*"^ ^ gesture-sign. The Egyptian open hand B i^ indicates a palm measure. The forepart of the arm witli open hand is seen on one of the Jerabis inscriptions (Wright, op. cit. PI. VIII. B. 1. 2). Compare, too, the hand and forearm sculptured on a rock at Itanos above an archaic Greek inscription (Com- paretti, Lcricii di G'6rfyna,'&c., p. 442, No. 206). e. Figs. 22b, 25a, 34i. The bent leg /^ in Egyptian = pat, ret, men, &c., as a determinative, is \A applied to actions of the leg, as 'marching' and -^^ ' ajiproaching,' and to agrarian measurements, as arura, ' an acre.' Among Hittite symbols only the lower part of the leg is found, apparently B2 [304] PRIMITIVE PICTOGRAPHS AND SCRIPT 35 booted. Cp. n Kolitolu Yaila. So far as style is concerned, the greatest resemblance ti/ is presented by a bent human leg seen in the field of a gem from the lower city of Mycenae (Tomb 10, 'E0. 'Xpx- 1SS8, PI. X. 9). ft Fig. 32rf. Possibly = a rump. Arms, Implements, and Instruments. Fig. 32& and cf. 41. Resembles an arm holding a curved instrument. As such it may be compared with the Egyptian determinative jF^ p. = a hand holding a club {ne')(t), applied to i ' forcible action. The forepart of the arm holding weapons or implements is common among Hittite symbols. S ^^ A Fig. 24?). The single axe occurs on early seal-stones in ^^^M the Ashmolcan Collection, from Smyrna and N. Syria. ^ I It is perhaps represented by a symbol on the Hittite I monument at Bulgar Madeu (Ramsay and Hogarth, Pre- hcUcnic Monuments of C'appadocia, PI. II. line 2, near middle). On an inscription from Jerabis (Wright, o^J. cit. PI. II. C. line 1, and A. 1. 4) the axe seems to occur in combination with another object. In Egypt the single axe is a sign of divinity. The present type of axe, however, is altogether non-Egyptian. I Fig. 87?'. Perhaps an early form of double axe-head. M 10 ^^ ^ ^^ Figs. 23?', 39. The double axe is a form altogether foreign to Egypt. As a Hittite hieroglyph it has been recently detected on an inscription, and it is seen re- peated in pairs on a Cypriote cylinder (Cesnola, Si. cit. PL IX. lines 7, 8). Long adzes are among the most typical forms of bronze implements found in Crete. They are found in Mycenaean deposits, and one in my possession from the Cave of Psychro is 1P35 inches in length. It is probable that the end of the wooden handle of the Cretan implement repre- sented above was shaped like the hind leg and hoof of an animal, as in the case of many Egyptian tools. 38 FROM CRETE AND THE PELOPONNESE. [307] 23 .,. f Fig. 34rf. Saw, shaped like the jaw of an animal, probably *^f^fj foimed of wood set with flint flakes. Compare the Egyptian '?^y^^I^^^^?\\ = saw. For a somewhat similar saw of wood set with flint teeth from Kahun, see Petrie, Illahun, Kaliun, and Gurob, PI. VII. Fig. 27. Houses and Household Utensils. 2i f , I Figs. 22/;, 24«, 2oa, 29b. Gate, door, or part of a fence. No. 2 in connexion with a pig. 25 ^J Figs. 30«, 32f, 36(Z. Perhaps variant of above, but cf. the Egyptian symbol for ' shutter ' ^^ 3 26 • Fig. 346. Gate or shutter. Ol 27 V \/Vr Fig. 32e. Fence. O' ^ 28 ^^ Fig. 39. This vase evidently represents a metal original closely resembling the Oriental ibrik, which serves an ewer for pouring and sprinkling water. Vessels of this shape form the principal type of a class of Mycenaean gems specially common in Eastern Crete (see below, p. 370), sometimes fitted with a conical cover like Persian ewers of the same kind. The curving spout recalls that of an Egyptian libation-vase — y X -^^'«^''' = ' libation,' 'sweet water' — but a simpler parallel is found in \ / the ordinary water- vessel (> C^ nnm = ' water.' It is probable that ^ the Cretan sign also stands ^^ — ' for ' water ' ; indeed, on the lentoid gems referred to, this vase and others closely akin, with high beaked spouts, are seen beside a plant or spray.i'"-' All this clearly indicates the purpose of watering. ^ 29 ^\_^ Eigs- 32(;, 31c. This form of vessel is of ceramic character, and the seal on which it occurs belongs to an early class. It corresponds with a primitive type ot high-beaked vases of very wide distribution, extending from Cyprus and the '"'' In the case of a closely allied form of vase Gouliis a vase of this kind is seen beside a with two handles the spray is seen inserted in plant, above which is a rayed disc indicating the mouth of the vessel. On a gem from the midday sun. [308] PRIMITIVE PIGTOGRAPHS AND SCRIPT 39 Troad to the Aegean Islands aud the mainland of Greece. They occur at Hissai-lik.and in the early cist-graves of Amorgos of prae-Mycenaean date, and I found part of the beaked spout of one of equally early fabric on the site of Praesos. Vases of this form are seen on the most primitive class of Cretan engraved gems, going back to the third millennium B.C. (see p. 332), and continue — taking at times a more metallic form — into the Mycenaean period. On two Vaphio gems ('E^. 'A^px- 18D0, PI. X. 35 36) a closely allied prochvus is seen in the hands of the mysterious beast- headed daemons of Mycenaean art, -who in one case are engaged in watering nurseling palm-trees. Another representation of the same form of vase occurs above two bulls in the field of a gem from Tomb 27 of the lower town of Mycenae ('E,^. 'Apx- 1888, PI. X. 24). 50 31 1 Fig. 40. This symbol belongs to the same class as the above. Fig. 40. Possibly some kind of vessel. Maeine Subjects. Figs. 34a, 28a. Tlie first of these vessels is accompanied with two crescents, one on either side of the mast — perhaps a sign of time as applied to the duration of a voyage (see below, No. 65). One ship has seven oars visible, the other six. In form these vessels show a great resemblance to those which appear as the principal type on a class of Mycenaean lentoid gems, specimens of which are found in Crete. One of these in my possession shows fifteen oai's and a double rudder, and perhaps an upper row of oars. The double end of the first example — like an open beak — may recall the swan-headed ships of the confederate invaders of Egypt ' from the middle of the sea' in Rameses III.'s time as seen on the frescoes of Medinet Habou. In the present case, however, no yards are visible. 33 '^^^B^ l''o- •^•^"^- Apparently a tunny-fish : the hatched-work r^^^^^ behind may indicate a net. Fish as hieroglyphic symbols are common to Egypt and Chaldaea. It looks as if tunny-fisheries had existed off the Cretan coast in Mycenaean times. The well-known gem with a fislicrman in the British Museum {Gem Catalogue, 80, Pi. A) may refer to the same industry ; and tunny-fish occur on two more Cretan gems of Mycenaean date in the same collection. A fish of the same type occurs as a symbol on Cypriote cylinders (cf. Salaminia, PI. XIV. 48). 40 FROM CRETE AND THE PELOPONNESE. [309] I li-i W Fig. 39. Also apparently a fish. The head is more rounded than No. 33, but this may be due to rudeness of design. Fish of the same rude form are seen on Cypriote cylinders (of. Cesnola, Salammia, PI. XIV. 48). Animals and Birds. Fig. 33c. Head of he-goat. This symbol presents a re- markable similarity to the Hittite hieroglyph of the same object V— .^. the value of which from its occurrence on 3i- ^^^ 1 35 '^ ^■9 the bi- 'V-—' lingual seal of Tarkutimme (Tarkondemois) ■^ in Hittite and cuneiform characters is known to represent the syllables Tarrik or Tarku (Sayce, Trans. Soc. Bill. Arch. Vol. VII. Pt. II. (1881), p. 297, and Emp. of Hittitcs, p. 182; Theo. Pinches, lb. p. 220, and Trans. Soc. Bibl. Arch. March 3, 1885; and of Halevy Rev. Sdm. 1893, p. 55 seqq.). The element ' Tarrik,' again, in the name of this prince, seems to refer to the god Tark (cf. RanLsay and Hogarth, Pre- helleiiic Monuments of C'aj^padocia, p. 9 seqq.). The Egyptian goat's-head sion ^ir'^s" is of a different character. The neck is given as well as the head, ^""V^ and there is no beard. 36 ) Fig. 37a. Bull or Ox. The seal on which it occurs is of primitive type. ^ 37 ^ Fig. 24&. A doe or kid. f^ ^^ 4Jj vV. ^'§®- ^^^' ^^^- Apparently intended for deer-horns. V 39 i y Fig. 2()rt. Horned head of an uncertain animal, apparently an ox. V 40 ^^m^ Fig. 21«. This appears to be rather a &Mcranz««( or skull ▼ of a bull or ox, than the actual head of the animal. As an ornament of tlie reliefs of altars the iKcmnium occurs already in Mycenaean art. This appears from a lentoid gem in the British [310] PRIMITIVE PICTOGRAPHS AND SCRIPT 41 Museum, ou wliicli is seen an animal of the goat kind freshly slaughtered, with a dagger thrust into its shoulder, lying on an altar or sacrificial bench, the front of which is adorned with four lucrania much resembling the above. In this ease, to complete the parallel with later classical reliefs, fillets attached to the extremities of the horns are seen hanging down between the skulls. \ 41 \l Fig. oM. This symbol must be regarded as uncertain. It is placed here, however, as showing a great resemblance to the Hittite sign which has been interpreted as an elongated form of the M ass's head. (Palanga.) 42 V _ ■■■•; Fig. 37?'. Perhaps a variant of the above. 43 ^^^^^ Fig. 24«. Pig. A similar ideograph occurs on a three- sided stone of the earlier Cretan type presented to the Ashmolean Museum by Mr. J. L. Myres. 44 ^[^^ Figs. 23«, 325. Wolfs head with the tongue hanging out. This symbol shows a remarkable likeness to the Hittite j^^o^ (Jerabis, o}-). cit. PI. VIII. D. 1. 3, PI. IX. (?^ — -^ 1. 3), where again we find the same pro- truding tongue. Fig. 31«. Dove pluming its wing. Fig. 40. Perhaps variant form of above. Fig. 39. Bird standing. Birds in a .somewhat similar posi- tion occur among the Hittite symbols at Jerabis and Bulgar iMaden, and are frequent in Egyptian hieroglyphics. Fig. 26«. Apparently a bird's head. Heads of various kinds of birds arc common among Egyptian hieroglyphics. 42 FEOM CRETE AND THE PELOPONNESE. [3111 49 \ Fig. 32c. This symbol apparently consists of two birds heads turned in opposite directions. 50 ^^ Figs. 28c, 30a. Perhaps a conventionalized sea-horse. The uppermost symbol on Fig. 18?* (No. 76 below) may be a simplification of (") 1^'')' this. Compare V^--//^ on a ' Hittite ' seal-stone from Smyrna. A very similar form ^""^^ occurs on an early truncated cone from Tartus. 51 ^ On the steatite relief (Fig. 18b). Apparently a tortoise. Vegetable Forms. Fis. 34i. Fig. 25b. @®® Figs. 23&, 33rf, 35«, 35c. This may perhaps be regarded as an abbreviated form of one of the above, with possibly a differentiated meaning. The form is common to the Hittite monuments, occurring at Jerabis, ^ (Wright, op. cit. PI. VIII. B 1. 5) in a more floral, and also q£p (op. cit. U PI. XIX. 6) in a geometrical form ; while at Bulgar U Maden (Ramsay and "iiogVLxih, PrcheUenic Monuments of Cai^padocia, PI. II. 1. 3, beginning) it forms a purely linear sign q9q . The same, or a closely allied symbol, is also of 1 : seen on the lion Marash (Wright, op. cit. PI. XXVII. Ill, 1. 1). :jo ¥ Fijr. Tob. [312] PRIMITIVE PICTOGRAPHS AND SCRIPT 43 5G wl.^ -^i?- '"l'- Perhaps a lily. Tiiis form is more pictorial ^j^L, than the others. Compare the Hittite ^^ Hamath pi^ST (Wright, 02?. cii. Pi. IV. 11. 2 and 3). lh, 40. Tree symbol. On a Mycenaean lentoid gem, now in the Museum of the Syllogos at Candia, a votary is seen blowing a conch-shell before an altar, behind which is a sacred grove with trees in the same conventional style. ^ A similar degeneration of the sacred tree occurs iz on Cypriote cylinders. 59 J Fig. 28&, repeated. Spray or branch, and the same is seen duplicated on Fig. 29c. I Heavenly Bodies and Derivatives. 60 '^ "V Fig. 33c. Day-star, or sun, with eight revolving rays. ^1v 01 vly ^^o" """ ^^^^ '^y^ more revolving). Day-star, or sun, with twelve rays. Star-like symbols occur on Syrian and Asiauic seal-stones. 62 J^ -* Fig. 35/'. This symbol, with the tangential offshoots ^@) suggesting revolution, seems to fit on to No. 60 and to be ^*^*^ of solar import. For the concentric circles as a solar emblem compare the Egyptian (q) Sep = times (vices), and the circle with a central dot is also the Chinese symbol for sun. The eye symbol, No. 4, approaches this very closely. 44 FROM CEETE AND THE PELOPONNESE. [313] C3 A > ^ -v Fig. Sod. This form suggests a combiaatioii of solar and lunar symbols. C4 VV Fig. 32& and cf. 39. Star of four rays. This symbol is frequent on Cypriote cylinders. X u 65 / \ Two small crescent-moons are seen on either side of the mast of the vessel on Fig. 34rt. They perhaps indicate duration of time — months — as applied to the length of a voyage. Geographical or Topographical. 6G A k ^ M Figs. Sod, 2ob. Apparently variants of the ^^^^ ^H^^V same symbol which seems to represent a (a) (h)" widely distributed pictograph for mountains and valleys, and so country or laud. On the boss of Tarriktimme (Tarkondemos) M = country (Sayce, Trans. L'ihl. Arch. Vol. VII. Pt. II. (1887), p. 297 '^ scqq. ; and cf. Halevy, Fuv. Si- ■mitique, 1893, p. 55 scqq.). It is found again in Jerabis (Wright, o^j. cit. PI. IX. J. II. 1. 1) and apparently on the monument near Bulgar Maden (R. and H. FrchcUcnic Monuments, &c., PI. II. 1. 2) AA. The Egyptian r\_,^ '"d = mountain is applied in the same way as a determinative I ' for 'districts' and 'countries.' As snut = granary, it reappears, with one or two heaps of corn in the middle, in the simple sense of a ' plot of ground.' The Accadian symbol, again, signifying a plot of ground, exhibits a foim P\/] closely parallel to the above. And in this connexion a truly remarkable coincidence is observable between the pictographic symbolism of old Chaldaea and that of the Cretans of the Mycenaean period. The linear form of the Accadian Ut-tu f^^ shows a sun above the symbol of the ground with a plant growing l ^"^J out of it. But on specimens of Mycenaean gems observed by me in Eastern Crete, side by side with the vase for watering already reftrred to, are seen symbolic or conventional representations of the plant gro^Ying out of the ground, recalling the Accadian version almost totidcm li^tcis M on amygdaloid cornelian; Zero (near Praesos). r^te; on amygdaloid / V \ cornelian; GouKis. In another case the ewer t=l divides the two sym- bols !^ \\/\ o" ''^" almond-shaped stone of the same character; Girapetra. [314] PREMITIVE PICTOGRAPHS AND SCRIPT 45 Geometrical Figures. G7 \ / Figs. 23&, 23c, 25a, 25c, 8Sd, 345, 38. This sign may bo simply a supplementary figure. On Fig. 38 it is thrice repeated with the sign No. 16, and might, like the similar Egyptian sign N/, indicate multiplication. X 08 A ^ Figs. 34rt, 34d This may be an intercalated sign, perhaps of the nature of a break between words. K S2 («) (?') G9 f\ ^v Figs. 21«, 23c. Repeated in two directions on Fig. 23c. This, too, is possibly an ornamental insertion, but it may however be compared with the Egyptian ^ , a coil of thread, signifying ' to reel.' 70 /^N Fig. 24c. This may be the same as No. 69 with an additional ornamental flourish. Uncertain Symbols. 71 ^ t^ ji ll_ Figs. 31?*, 3.5c. The late Hittite sign occurs at Gurun (R. and H. op.cit. Pi. IV. 2, 1. 2), and perhaps in the inscription near Bulgar /f,) /^^ Maden (o^j. cit. PI. II. 1. 3). <^C^ 72 "^T" Fig. 27«. I 73 y-^N Fig. 25c. I''i". 25c. Somewhat fractured below. 46 FROM CRETE AND THE PELOPONNESE. [315] 75 ^^ Fig. Sid. A certain analogy is presented by the Hittite sign ^^T' \ Hamath (Wriglit, op. cit. PI. I. H. II. I. 2), X Jerabis (cqx cit. PL VIII. B. 1. ',), and on the 'Niobe' (Ed. GoUob. in oji. eif. PI. XXII.). 7G ^^^ ^ Fig. IS//. On tlie .steatite relief (Fig. 18i); possibly a conventionalized form of No. .50. 77 \ / Fig. 2Gn. 78 •"'^ Fig. 2ort. Perhaps a variant of No. G9. S 79 ^-\ Fig. 33c. This symbol presents a certain resemblance to the " ^ Hittite forms (^^ Hamath (Wright, oj^. cit. PI. I. 1. 1, PI. II. H. III. 1. 1, PI. IV. H. V. 1. 1) ; (f^ Jerabis (op. cit. Pi. VIII. J. I. A. 1. 3, B. 1. 2) ; (7~^ Bulgar Maden (R. and H. oj;. cit. PL II. 1. 3) ; (^~^ Gurun (o2-). cit. PL IV. 1). - 80 4^1 Fig. 22a. This recalls the Egyptian V = ' skein of thread,' r^\ the determinative for ' linen,' ' bind- U ing ' &(,_ Compare, ^^ too, the twisted cord O sen - ' to turn back,' and / te, the tied up bundle A = 'to bury.' On the Hittite ) 82 "•>^^ Fig. 32(;. This symbol, if rightly comj)leted, recalls the -■-•■^^^ Egyptian >! This is in faet an ordinary Jlycenaean of form, fjeni representing apparently a kind of base. [318] PRIMITIVE PICTOGRAPHS AND SCRIPT 49 Akropolis graves at Mycenae,-*' and it is a common ornament of tlie stamped glass plaques of tlie later Mycenaean interments. Anotlier example of this vegetable form may be seen on a low vase found by Professor Petrie in the ' Maket ' tomb at Kaliun, the approximate date of which is now fixed at about l-ioO B.C. by the new evidence supplied by the foundation deposits of Thothmes III. at Koptos.-^ A very similar type of leaf is also seen on a Mycenaean fragment from Tell-el-Amarna, belonging to the age of Akhenaten (Khuenaten) and the early part of the fourteenth century before our era. In a still more literal form, moreover, it appears executed in a brilliant blue on the fresco decoration of the Palace itself.^^" The leaf on the Goulas gem presents the distinguishing feature of being decorated with hatched lines ; -'^ and tliis peculiarity recurs in an example of the same motive upon one of the vases from the first shaft-grave at Mycenae, the ceramic contents of which, fitting on as they do to some of the types of Thera,-^ must be regarded as earlier rather than later than the Tell-el-Amarna fragments. On these grounds I would approximately refer the Goulds gem to the fifteenth century B.C. A more globular variety of the three-sided stones is also represented among Mycenaean gems. On one obtained by me from Central Crete the same leaf-shaped ornament occurs as that described above. On another from Malia, also a cornelian, engraved on two of its faces, are designs of a wild goat struck by an arrow, and of a flying eagle with two zigzag lines proceeding from it — possibly a Mycenaean ' thunder-bird.' An en- graved amethyst, again, of this type was found in the Vaphio tomb ; and here again we have an indication of date taking us to the middle of the second millennium B.C. and to the most flourishing period of Mycenaean art. The peculiar form of stone (Fig. 21) with the spirally fluted back. "" Schuchhardt, Schliemann's Excavaiions, Tliera class of vasos would alone be fatal to \>. 187, fifjs. 161 — 163. Mr. Potrie's former view, that the beginning of "' Mr. Petrie in his Egyptian Bctscs of Greek natural designs on Mycenaean pottery should History {Hell. Joiirn. xi. (1890), p. 273) and be brought down so low as this in date. But Mr. Ulahun, &c. , pp, 23, 24 had dated this tomb c. Cecil Torr, who in a letter to the Classical Review 1100 B.C., though he noted as a somewhat makes much of the inconsistency between the incongruous circumstance that the latest scarabs results obtained at Tell-el-Amarna and Mr. found belonged to Thothmes III. The new Petrie's former opinion as to the date of the comparisons supplied by foundation deposits of Maket tomb, will hardly be gvatilied to find Thothmes III. excavated by him at Koptos, that tlie chronological revision that has to be sucli as the ribbed beads, &c., of the same type made is in favour of a greater auticiuity. there found, have now led him however to '"^ Specimens of this design presented by revise his opinion, and to carry back the date Mr. Petrie are now in the Aslimoleau Museum, of the Maket tomb to the same time as these '" Petrie, TcU-cl- Amanui, PI. XXIX. and deposits. An examination of the Koptos relics, cf. p. 17. In this case however the leaf is more which I had the advantage of making in Mr. lanceolate. Petrie'.s company, leaves no do\ibt in my mind '-'■' This is notably the case with the vase that this conclusion must be regarded as final. which bears on its neck two breasts surrounded On other grounds, especially since the discovery with dots. Compare Schuchliardt, op. oil. fig. of the Tellfl-Amarna fragments, 1 had already 166, p. 189 and Dumont et Chaplain, Circ^- b(!cn led to infer that! 100 li.i;. was too late a date miquc ilc la Grecc 2>roprc. for the 'Maket' deposit. The existence of the 50 FROM CRETE AND THE PELOPONNESE. [:il9j which, as pointed out above, seems to originate from a twin Nerita shell type of an earlier period, also occurs among the Mycenaean gems of Crete. One of these, obtained in Candia, is engraved with a typical design of a sepia : another, found at GouU\s, has a combined spiral and vegetable motive of great interest (Fig 42). The leaves in this composition evidently belong Fig. 42.— Gem, GoulXs (2 diauis.). to the same water-plant as that seen on a painted ossuary in the form of a hut discovered by Professor Halbherr in a Mycenaean tholos tomb at Anoya Messaritika.^* The same palmette-like form however recurs in a still more literal guise, occupying the arched interior of the symbol No. 14 on the three-sided stone Fig. 23ff. And here an interesting combination suggests itself The observation has already been made above that the symbol No. 14 (see below. Fig. 48) which occurs on stones (Fig. 2ob, 3.5c) is the same as No. 15 (Fig. 44), minus the leaf and spirals. I had therefore at first looked upon this latter as a kind of decorative excrescence not essential to the symbol itself. But the symbol in its simplified form, with its arched space below and two curved incisions on the top, remained a puzzle. Judging by the analogy of other signs, it was probably some form of instrument or implement, and the suspicion did cross my mind that it might be connected with house-building and possibly the decoration of ceilings. But the Goulas gem places tins conjecture in quite a new light. The combination of triquetral curves and vegetable ornament that it presents, at once declares the design to be a part subtracted as it were from a more spacious ornamental surface. The divergent spirals, coupled with foliate or ^* P. Orsi, Urne Funchri Crclcsi, PI. I. Terrot, La Grice Primitive, p. S30, quotes with approval a theory of M. Houssay, a zoologist (which he had previously applied to a large cuttle-fish on a Mycenaean vase from Pitane in the Aeolid), that the ducks, fish and starlike objects seen between the branches of the plant upon the ossuary were supposed to have been generated by it, and that it is in fact the 'barnacle-tree' of folk-lore. For myself how- ever the plant simply represents a water-plant by the side of a stream, the ducks which follow next behind it are flying over the surface of the water, and the fish alone, in the third line, are actually in the water. In fact it is not difficult to trace in this design a reminiscence of a commonplace of Egyptian painted pavements and frescoes, in which river-plants with ducks flying over them or poising on their branches are seen besiilo a tank or stream containing fish. Only here the forms of the leaves are diflereut from those of the lotos or papyrus seen on the Egyptian models. E 2 [.■520] PRIMITIVE PICTOGRAPHS AND SCRIPT 51 floral forms, are tlie animating principle of a wliole series of large decorative compositions, of which the ceiling of Orchomenos is the most conspicuous example in Mycenaean art, but which are in fact the almost literal copies of Egyptian prototypes. Fig. 43. Fig. 44. Fig. 45. In view of these comparisons it occurred to me that the symbol connected above with the palmette — belonging ex hypotliesi to a form of Mycenaean ceiling — might have been a simple kind of stencilling plate known as a ' template,' such as is still in use among decorators, and that it was employed for a similar purpose by the artists whose business it was to adorn the palaces of the Mycenaean lords. I accordingly cut out a symmetrical model of the sign (Fig. 43), and made a practical test of its utility in the mechanical procedure necessary for producing such a design. The use of the incurved notches at the top of the figure became at once apparent. The symbol, first applied with the top of the arch uppermost so as to stand on a line ready ruled, gave the upper outline of the leaf, for which the , inner margin of the arch supplied the tracing. Now turning the figure upside down, and carefully adjusting its feet to the terminal points of the upper border of the tracing already made, it will be seen (Fig. 45) that the double curves fit into the lower opening of the arch, and give the two incurving lines required for the lower margin of the palmette (Fig. 44). The form of template suggested by the symbols fulfils the following conditions : — (I) It will be contained in a square, its height being equal to its width. (2) The opening at the base of the arch is equal in width to the space between the exterior horns of the summit. (3) The top of the arch forms a semicircle, the radius of which is equal to that of the curves of the notches at the top. I. — Now apply the template thus formed to a sloping line A B twice on each side of it, as shown in the diagram Fig. 4G, so that in all four positions one of its feet rests on the portion C D of the said line A II. II. — Apply the template sideways to the sloping lino A ]'>, as in diagram Fig. 47, and adjust the foot in each case to the lines E E' , mark the point of the extreme horns F F' and rule the two lines F G, II F, which are parallels. FROM CRETE AND THE PELOPONNESE. [321] Now complete the circular heads of the arches round the points F F , which form, in the case given, the centres of the circles thus drawn. Fig. 46. III. — Produce the parallels F G, E F and join the points F F'. Taking F F' as a, side, mark off as often as required the same distance on the produced parallels F G, H F, drawing at each such distance a fresh parallel to the line ^^:%;:^^ Flo. 47. F F\ and thus producing a series of rhombi. At each of these points repeat the small circles, and to complete the groundwork of this band of the design it is only necessary to draw the curving lines tangentially to them. [322] PRIMITIVE PICTOGRAPHS AND SCRIPT 53 H The first section 'of Fig. 48 shows the simple rhombus, the second the same with tangential lines straight and curving. In the third section on the line A B, and upon the base D E already obtained in Fig. 46, a palmette is formed by reversing the template as in Fig. 46c, and so on in the other rhombi. The curving stems and cress lines are then filled in as in the Goulas gem (Fig. 44), the result being that shown in PI. XII. Observation. — In order to fit the design into a square or rectangular surface, as in PI. XII., each new band of rhombi must be taken back to a start- ing point (Z), which must be at the same distance from a right or left margin as was F at the commencement. Fig. 48. The complete design as restored in Plate XII. by the aid of the template symbol may well have decorated the ceiling of a palace hall or princely sepulchral chamber in the great Mycenaean city where the gem was found which suggested this practical application of the pictograph. The typical combination of the volute and vegetable motive which it exhibits affords in turn a secure chronological standpoint. The design before us belongs to the same class as the ceiling of Orchomenos and the fragment of wall- painting from the palace at Tiryns,-^ and was, like them, undoubtedly executed under the immediate influence of the Egyptian style of ceiling decoration that came into vogue under the Eighteenth Dynasty, and the finest examples of which are to be seen in the Theban tombs. The colours on Plate XII. have in fact been suppUed from Egyptian analogy.^^* 26 Schliemann, Tinjns, PI. V. "» The tangential curves of this group of designs are in nearly all cases coloured yellow as if to imitate gold, :ind this rule also holds good in the case of the wall-painting in the Palace at Tiryns (Schliemann, Tiryns, PI. V.). The alternation of red and blue fields is also common in Egyptian ceilings of this class. I I am indebted to Mr. ,1. Tylor for some un- published examples of similar patterns from the 54 PROM CRETE AND THE PELOPONNESE. [323] It is probable that at the time when these gems were executed this decorative pattern combining the palmettos and returning curves or spirals was widely prevalent in Crete. The template symbol itself recurs on two seal-stones, in one case with palmette and spirals attached, and on the triangular seal, Fig. 22c, there is a combination of two palmettes and curving lines going in opposite directions, which may be regarded as a simplified version of the fuller motive, as seen in the Goulas gem. The volute form of the latter stone is, as already shown, characteristic of a class of Cretan gems with purely Mycenaean types, and the connexion that has been established between the design that it presents and Cretan jjicto- graphic symbols on the one hand, and the Egypto-Mycenaean ceiling decoration on the other, gives us a fresh basis for a chronological equation. The later pictographic class is once more brought into close relation with Mycenaean art, while the Egyptian parallels take us once more to the middle of the second millennium before our era for the approximate date of the seal-stones on which these suggestive forms occur. In examining the symbols on the Cretan seal-stones various other parallels with Mycenaean forms have already been pointed out. The single figures which occur, such as the young doe or kid in Fig. 24&, the dove pluming its wings on Fig. 31a, fit on both in style and execution to the Mycenaean class. The ship on Fig. 34<« and 28a is found again in all its typical lines on lentoid beads of Mycenaean fabric found in Crete. The double axe No. 10, the bent leg No. 5, the bucranium No. 40, all make their appearance as accessories of Mycenaean seals and gems from Pelo- ponnesian tombs. The forms of vases seen in Nos. 28 and 29 are elsewhere held in the hands of Mycenaean daemons, and are the distinguishing types of a whole series of lentoid and amygdaloid gems of Mycenaean character found in Eastern Crete, on the ethnographical importance of which more will be said later on. It is always possible, as already observed, that some of the smaller objects seen in the field of the typical Mycenaean gems beside the principal design may belong to the same pictographic class as the signs on the angular seal- stones. Such correspondences as those noted above certainly tend to add to this probability. But, bearing in mind the known tendency of the primitive artist to fill up the vacant places of the field with supplementary figures, it does not seem safe to assume that, because small figures identical with the pictographic forms occasionally found their way on to these more decorative objects, they are necessarily to be regarded as having in that position a hieroglyphic value. When however symbols of this character occur in groups, occupying the whole surface of field, the case assumes a different complexion, and it is with this phenomenon that -we have to deal in the class of early lentoid gems from Crete represented by Figs. 40 and 41. Of these ceilings of grottoes near Silsilis, of the of red and blue, enclosed by yellow tangential Eighteenth and Nineteenth Dynasties. One of curves, affords a close parallel to the Cretan these, a series of rhoiuboidal fields alternately design as restored in PI. XII. [324] PRIMITIVE PICTOGRAPHS AND SCRIPT 55 Fig. 40 exhibits a group of four distinct symbols and part of a fiftb, whicli has unfortunately been broken off. Fig. 41 again contains three signs apparently of the same hieroglyphic character, one of which — the arm holding a curved instrument — resembles the symbol on Fig. 326. These specimens belong apparently to the earlier class of lentoid beads and, like all those of this eaily class, which in Crete is especially well represented, are cut in soft stone, apparently steatite. One is from Knosos, the other from the Messara district of Central Crete, and with them may be grouped another similar lentoid bead from the latter region, with a figure which clearly represents an insular copy of the Egyjitian AnkJi. ^ VI. — The Earlier Classes of Cretan Seal-stones. The comparisons already accumulated sufficiently warrant us in refer- ring the most characteristic of the hieroglyphic stones to the great days of Mycenaean art. The connexion established is indeed from many points of view so intimate that it is impossible to avoid the conclusion that there existed within the regions dominated by the Mycenaean culture — in Crete certainly, perhaps in the Peloponnesus — a form of picture-writing of much the same general character as that in use throughout this same period in the ' Hittite ' countries of Asia Minor. But with these Mycenaean comparisons the last word has by no means been said on the origin and evolution of the hieroglyjjhic forms. There are distinct indications that the beginnings of this picture-writing go back to a far more remote period of Cretan story. Everything tends to show that they are in fact deeply rooted in the soil. The most typical forms of the stones themselves come, as will be seen, of an old indigenous stock. As we go farther back the signs become more pictorial, but they seem still to stand in a personal relation to their owners not to be found on merely decorative gems, and they serve essentially the same, purpose as elements of seals. Of the types described the four-sided equilateral prisms represented by Class II., all of which seem to belong to the Mycenaean period, correspond with an Egyptian form of seal-stone that was in vogue in the time of the Eighteenth Dynasty, and a good specimen of which in green jasper dating from the reign of Thothmcs II. (c. 1516 — 1503 B.C.) was found by Jlr. Petrie in the Makot Tomb at Kaliun. But the three-sided form seems to be a characteristically Cretan product and to go back in the island to a much more remote period. In the course of my journey through Central and Eastern Crete I came across a series of stones whicli, though of distinctly earlier fabric, showed the same typical triangular form as Class I. of the later hieroglyphic series. Some of tliese have the same elongated form, others resemble in shape the more globular variety, but they are larger, and unlike the others, always cut in steatite and never out of harder materials such as cornelian or jaaper. 56 FROM CRETE AND THE PELOPONNESE. [325] One or two of these ecarlier types (Figs. 21, 30) have been inserted in the series of hieroglyphic seal-stones already given, as presenting symbols of essentially the same class though at times in a more primitive form and associated with more purely ideographic figures. It would not have been difficult, as will be seen from the contents of the present section, to have added others, and in truth no real lines of demarcation can be laid down between the earlier and the later group. These primitive types show a close correspondence in their designs with certain other classes of early engraved stones found in the island. Amongst these may be mentioned flat disks perforated along their axes and engraved on both faces, button-like stones, and others of truncated pyramidal and sub-conical forms, bored horizontally near the apex. For the dating of this early grouj) most valuable evidence is supplied by the deposit, already referred to, found at Hagios Onuphrios, near the site of Phaestos, and now preserved in the little Museum of the Syllogos or Literary Society of Candia (Heraklio). This depiosit, which contains nothing that can safely be brought down to Mycenaean times proper, is of a homo- geneous character, and seems to me to be of capital importance in the history of early Aegean art. Although exact details of the excavation are wanting,^** it is certain that it represents the remains of early sepulture, dating from the same period as the primitive cemeteries of Amorgos and presenting a series of objects in many respects strikingly similar to those from the Amorgan cists.-""' Here are the same rude marble idols and vessels, high- spouted clay vases and rude pots with perforated covers, as well as the first beginnings of painted ware, with red, white, and violet strijjes on the plain surface of the clay. Here is the square-ended triangular-bladed dagger of the Amorgan graves, the fluted jewelry, but of gold instead of silver; here are the same steatite pendants and spirally ornamented seals. In a word the Phaestos deposit covers precisely the same period as the earlier elements of the Amorgos cemeteries — a period which may be roughly defined as intermediate between the first prehistoric stratum of Troy and the early remains of Thera.-"'^ As a matter of fact a two-handled jar with red and white streaks on the blackish-brown ground which must be regarded as one of the latest objects in the Phaestos group approaches in technique some of the earliest ceramic specimens from Thera. These considerations would alone bo sufficient to afford a rough chrono- -" Professor Halbherr lias obligingly collected bones and skulls, but no regular tomb was noted, for me on the spot the following particulars of The whole deposit occupied a space of about the find, that are all that are now obtainable. four square metres. The hill of H. Onuphrios where the objects -»^ For the early cist-graves of Amorgos see were found rises opposite the double Akropolis especially F. Diimmler, MiUhcilungcn von den of Phaestos about a quarter of a mile to the GriecMschcn hiscln (Ath. Mitth. 1886, p. 15 North of the ancient city. The find-spot itself seqq. and 209 scqq.). The contents of some of was on the southern slope of the hill just above the Amorgan tombs, obtained by me in 1893, are the Khans on the Dibaki road and near the now in the Ashmolean Museum, aqueduct of a mill. Tho deposit was accideu- -^^ For the Hagios Onuphrios deposit see tally discovered in 1887 at a small distance p. 104 scqq, beneath the surface. The objects lay iu a heap of [326] PRIMITIVE PICTOGEAPHS AND SCRIPT 57 logical guide. The Thera vases may be justly regarded as the earliest examples of the Mycenaean class, which already by the middle of the second millennium B.C. had attained its apogee. On archaeological grounds there- fore it would certainly be unsafe to bring down the earliest of the painted vases found beneath the volcanic strata at Santorin and Therasia later than the eighteenth century before our era. On the other hand, the first pre- historic city of Troy must be carried back to a far more remote period. The recent excavations of Dr. Dorpfeld have now made it abundantly clear that the Sixth City on the site of Hissarlik belongs to the great age of Mycenae, or roughly-speaking 1500 B.c.^' But between this and the once miscalled ' Homeric ' City of the second stratum, an interval, estimated by Dr. Dorpfeld in round numbers at 500 years, must be allowed for the intervening settlements, and beyond this again lies the whole duration of the Second City, the beginnings of which go back at a moderate estimate to 2500 B.C. The earliest and most primitive stratum is thus in Dr. Dorpfeld's opinion carried back to the close of the fourth millennium before our era. But the Phaestos deposit contains direct chronological indications of a kind hitherto unique amidst primitive Aegean finds. Amongst the relics found there occurred in fact a series of Egyptian scarabs belonging to the Twelfth Dynasty and the immediately succeeding period. And happily in this case we have to deal not with cartouches containing names which might possibly have been revived at later periods of Egyptian history, but with a peculiar class of ornament and material that form the distinguishing characteristics of the Egyptian scarabs of Twelfth Dynasty date, and which, though partly maintained during the succeeding Dynasty, give way in later work to other decorative fashions. The amethyst scarabs with a plain face — intended to be covered with a gold plate, — characteristic of this period of Egyptian art, are represented among the Phaestos relics by an example, on which — probably by an indigenous hand, — three circles have subsequently been engraved. A more important specimen however is a steatite scarab engraved below, with a spiral ornament peculiar to this period, to which also in all probability belongs a white steatite bead with a vegetable motive and a scarab with a hieroglyphic inscription. Nor must this occurrence of Twelfth Dynasty scarabs be considered at all exceptional in Crete. From the Messara district I acquired another of the same class, with a returning spiral ornament of a typical kind ; while another scarab found in the same region, with an S-shaped scroll and a cowry-like back, apparently represents an indigenous imitation of a form that came into vogue during the Hyksos period."* '■^ For the chronology arrived at by Dr. Diirp- istic form with cartouches represeuting blun- feld, .see especially Troja : 1893, pp. 61 and dcred copies of the name of Ra-sehoteb-ab of 86, 87, the Thirteenth Dynasty, who reigned about '■'* This is Professor Petrie's opinion. In his 2510 is c. It is natural to refer these blundered nislory of Egypt (vol. \. )t. 208, Fig. 110) are imitations of this cartouche to the succeeding engraved two ' cowroids ' of the same character- Hyksos Period and with them this ' cowroid ' B8 Prom CHETE and the PELOtONNESE. [32?] The Twelfth Dynasty of Egypt is placed by the most recent chrono- logical researches-*' between the approximate dates 2778 and 2565 B.C. The succeeding Thirteenth Dynasty, which partly preserved the same style comes down on the same reckoning to about 2098 B.c.^*'' With the guide afforded by the presence of these Egyptian relics on the one hand and the approximation to the earliest ceramic types of Thera on the other, we may roughly take the period 2500 — 1800 B.C. as the time-limits of the Phaestos deposit, which no doubt consisted of successive interments. The generally 'Amorgan' facies of the whole group of objects found quite squares with this result and at the same time prevents us from bringing down the central period of the deposit too near the date of the more developed ceramic style found in Santorin and Therasia. But among the EGYPTIAN SCARABS XIIth DYNASTY EARLY CRETAN SEAL-STONES Fig. 49. engraved stones found here, together with specimens of other types described above, occurred a typical example of an elongated, three- cornered seal-stone of the earlier class (see below. Fig. 73), having upon it designs of a decorative rather than hieroglyphic character. Upon a button-like ornament of steatite from the same deposit were engraved three characters of the linear class (Fig. 12) ; and the remarkable inscribed whorl (Fig. 11), referred to above (p. 284), was found in association Avith the other relics on the same spot. form. A parallel to this shell-like type is found in the twin Kcritci bead of the Phaestos deposit, already rofurred to on p. 289. -'Pa Petrie op. cil. p. 147. =8t> Op. cU. p. 204, [328] PRIMITIVE PICTOGEAPHS AND SCRIPT 59 The influence of the decorative motives of Twelfth Dynasty scarabs is percejatible upon other early Cretan seal-stones, both of the three-sided and button-like classes. This will be clearly seen by a comparison of the designs of the three scarabs given in Fig. 49«, h, c, with motives taken from the faces of primitive stone ' buttons ' and triangular seal-stones of early fabric (Fig. Add-h). It will be seen that the I'ower part of the ornament on d has been ' crossed,' as it were, by the ' broad arrow ' symbol which occurs on another facet of the same stone. This and g are triangular stones of the same type as that represented in Fig. 19a, but of earlier technique than the conventionally pictographic class. The central design on d reproduces the principal motive of the scarab above it, and the two signs on / are simply incomplete and rude transcriptions of the very characteristic scrolls on c.^^ The buttons c and g were obtained by me from the Messar^ district, and the other of a closely similar type (k), which is unfortunately broken, is from the Phaestos deposit. It is not too much to say that this Fig. 50 (c>ril.-ir(;(>il 2 diams.) taking over of the decorative designs of Twelfth Dynasty scarabs on to these early Cretan stones is of capital importance in the history of European art. In the examples already given will be found simple examples of the borrowing at this early period — c. 2500 B.C. — of the returning spiral motive which was afterwards to play such an important part, not in the Aegean countries only, but in the North and West. On the Twelfth Dynasty scarabs this motive, as is well known to Egyptologists, was developed to an extraordinary degree, the whole field being often entirely occupied by divergent spirals to the exclusion of all other elements. These purely spiral types, like the other Twelfth Dynasty motives already noticed, were also copied by the native Cretan engravers. A good instance of this will be seen on another button-like steatite of quatrefoil shape (Fig. 50) from the same Phaestos deposit, exhibiting a series of four divergent spirals. •" This parallel was kindly supplied me by Mr. Petiie. 60 FROM fTtETE AND TUK PELOPONNESE. [:r29] From Crete, where we find these Aegeau forms in actual juxtaposition with tlieir Egyptian prototypes, we can trace them to the early cemeteries of Amorgos, presenting the same funeral inventory as that of Phaestos, and here and in other Aegean islands like Melos can see them taking before our eyes more elaborate developments.^"" Reinforced a thousand years later by renewed intimacy of contact between the Aegean peoples and the Egypt of Ameuophis III., the same system was to regain a fresh vitality as the principal motive of the Mycenaean goldsmith's work. But thougli this later influence reacted on Mycenaean art, as can be seen by the Orchomenos ceiling, the root of its spiral decoration is to be found in the earlier ' Aegean ' system engrafted long before, in the days of the Twelfth Dynasty. The earliest gold-work as seen in the Akropolis Tombs is the translation into metal of ' Aegean ' stone decoration. The spiral design on the Stele of Grave V is little more than a multiplication of that on the Phaestian seal. In the wake of early commerce the same spiraliform motives were to spread still further afield to the Danubian basin, and thence in turn by the valley of the Elbe to the Amber Coast of the North Sea, there to supply the Scandinavian Bronze Age population with their leading decorative designs. Adopted by the Celtic tribes in the Central European area, they took at a somewhat later date a westerly turn, reached Britain with the invading Belgae, and finally survived in Irish art. The high importance of these Cretan finds is that they at last supply the missing link in this long chain, and demonstrate the historical connexion between the earliest Euro- pean forms of this spiral motive and the decorative designs of the Twelfth Dynasty Egyptian scarabs.-^'' And it is worthy of remark that in Egypt itself, so far as it is possible to gather from the data at our disposal, this returning spiral system, which can be traced back to the Fourth Dynasty, is throughout the earlier stages of its evolution restricted to scarabs.-"'^ The -'•■' Compare especially the steatite button-seal that Egypt was the place where this motive from Kuphonisi between Naxos and Amorgos, first originated, considers that it first reached F. Diimmler {Jth. MiUh. 1886. Bcilaye 1. 1.) : the Greeks by Phoenician mediation in the the green marble box from Amorgos {0;j. cit. fifteenth cent. B.C.— a view which the Cretan Boilagcl. Fig. A) and the stone 'pyxis' in the and Aegean finds must certainly modify. He form of a hut from Melos ( Perrot et Chipiez, considers that it readied Central and Northern La Grece Primitive -p. 910, Fig. 461). Europe througli mercantile intercourse due to -"'' In the Hellenic Jmirnal, Vol. xiii. p. 221, the amber trade, and apparently favours the I had already ventured to point out that the view that it came to those regiona directly from early spiral work of the Mycenaean jewels fitted Egypt. But the early spread of these spiral on to that of the earlier stone ornaments of the motives among the Aegean popidations affords Aegean islands and the spiral decoration of these the most natural explanation of its first ap- in turn to the simple spiral system that attained pcarance in the Danubian regions. As its apogee in Egypt under the Twelftli and noticed below, it seems certain that the in- Thirteeuth Dynasties. But the 'missing fluence of this Aegean spiral system h.ad link' to comiilete the Egyptian connexion bcgnn to leave its mark on Central and was not then in my hands. Dr. Naue, in Northern European art in prae-Mycenacan his recent work. Die Bronzezeit in Obcrhayern times. (Munich 1801, pp. 21;", 216), while recognizing -»'= Professor Pctrie'.s observation. [330] PRIMITIVE PICTOGRAPHS AND SCRIPT 61 primitive Aegean imitations are also in the same way confined to stonework, and were only at a later date transferred to metal and other materials. The ^yhole weight of the archaeological evidence is thus dead against the generally received theory that the spiral ornament, as it appears on Mycenaean art, originated in metal-work,^"^' though its later application to this and other materials naturally reacted on its subsequent development. It may be regarded as certain that the early Aegean spiral system born of this very ancient Egyptian contact was beginning to spread in a Northern direction at a date anterior to the great days of Mycenae. At Lengyel in Hungary and at Butmir in Bosnia the spiral decoration appears already on pottery of late neolithic date, and some Hungarian clay stamps with a quadruple spiral design might be taken to be the direct copies of the Cretan steatite seal-stone represented above.^"^ Nor are there wanting indications that the Aegean spiral system was leaving its impress on Italian pottery before the days of Mycenaean contact. On the present occasion it has been impossible to do more than call attention to the far-reaching importance of this decorative result of the early contact between the Aegean islanders and the Nile Valley in the third millennium before our era. Of that early contact I was able in the course of my Cretan explorations to collect other interesting evidence in the shape of a series of primitive stone vessels of strikingly Egyptian types. In particular, I acquired a stone pot with a cover identical with those found by Professor Petrie in a Twelfth Dynasty deposit at Kahun. It was found beside a skeleton in an early cist-grave at Arvi, on the South-Eastern coast of Crete, in company with other stone vessels, some of a more indigenous character, and a clay suspension vase, very like one from the Phaestos deposit. The Twelfth Dynasty parallels above instituted are of special value to our present inquiry from the corroboration that they afford to the chrono- "M I am informed by Professor Petrie tliat his Dynasty accumulations. In Cyprus similar researches on this class of scarab lead to this vessels are found in graves anterinr, though not conclusion. An illustrative series of these, long anterior, to the period of Mycenaean including one of Tat-ka-ra of the Fourth influence. Milchhijfer, who like other.s derived Dynasty, has been published by Dr. Naue the Mycenaean spiral decoration from wire-woik (Die Bronzrzdl in Ohcrhayern, p. 145) from designs (Die Aiifimye dcr Kuiust, p. 16 seqrj.), impressions supplied by Mr. Petrie. It would saw a corroboration of this theory in the gold appear, however, that at least as early as the jewelry from the 'Treasures' of Hissarlik Thirteenth Dynasty this spiral decoration was (Schliemann, liios, p. 453 scrjq.). But many bcinnin" to spread in Egypt to other objects objects from those 'Treasures' do not by any besides scarabs. There is in the Aslimolean moans belong to tlio remote period to which Collection a black-ware vase from Kgypt of a they were originally referred by Dr. Schliemann. style characteristic of Twelfth and Thirteenth Their whole fades shows that they are not far Dynasty deposits (cf. Petrie, A'a/«Hi, ffitroft rt7(rf removed in date from Mycenaean times and llaicara, p. 25 and PI. XXVII. figs. 199-202) belong to the Sixth rather than the Second City. which has a punctuated returning spiral orna- -»= See especially the Hungarian clay seals mcnt running round the upper part of its body. rein-esented in the Coviptc llcndu du Cont/ris Specimens of similar ware, tbougli without the Prihistorique, Budapest 1S78, PI. LXX. Fig. 14 spiral decoration, were found at Khataneh by and cf. Fig. 13. The Sshapcd design so fre- M. Naville in company with Thirteenth Dynasty iiueul on the Cretan sealstones is also reprc- scayabs, in graves deep down below Eighteenth sented on Fig. 12 of the Same serjps, 62 FROM CEETK AND TIFK PELOPONNESE. [331] logical evidence suggested by the Pliaestos dejoosit. In tlie one case we have actual association with Egyptian relics belonging to the first half of the third millennium before our era ; in the other case we have, — what is even more significant, — unquestionable imitation of the same. Both lines of evidence enable us to refer to this early period some of the more archaic of the three-sided seal-stones and certain types of engraved stone ' buttons.' But the evidence of the influence of Twelfth Dynasty decorative motives on this group of early Cretan seal-stones, while itself supplying a landmark of extreme antiquity, enables us to carry back to a still earlier date a yet more primitive class of stones still untouched by this Egyptian influence. Our chief standpoint for this chronological result is supplied by the three-sided stones which of all the forms exhibiting the symbolic figures may be described as the most characteristic. Setting aside for the moment the most globular variety presenting purely Mycenaean designs, these triangular stones may be divided into the following classes : — Class I. — Elongated triangular stones presenting groups of symbols or ornaments enclosed in an oval groove somewhat resembling an Egyptian 'cartouche.' Seen at their extremities the central perforation of the stone is surrounded by a triangular groove {sec Fig. 20h, p. 288). The seals of this class are generally of harder materials, such ascornelian, jasper or chalcedony. They present the hieroglyphs in their most conventional form. The materials and some of the designs show that they belong to the Mycenaean Period proper. This class has already been dealt with in Section II. Class II. — Elongated triangular stones of the same shape as the other with or without the oval groove or cartouche, but of more primitive execu- tion, and of softer material, such as steatite. Both hieroglyphic and linear symbols already occur on some of these, but there is a greater frequency of single designs on the sides, and of purely decorative motives, in some cases derived from Twelfth Dynasty scarabs. Class III. — Triangular stones of shorter and more compact form (Fig. 51), with or without ' cartouche.' Like Class II. they arc of soft materials, such Fio. 51. as steatite. S-shaped designs occasionally occur on these, which may possibly be due to Egyptian suggestion, but more elaborate attempts to copy Twelftli Dynasty motives are as yet rare. Human figures, birds and animals, or parts of such, vases and other objects occur, occasionally grouped, and representa- [332] PRIMITIVE PICTOGRAPHS AND SCRIPT 63 tions of men in various attitudes and employments, but no linear symbols are. found. The designs are more pictorial and less conventionalized than in the other groups. This Class seems to overlap Class II., but on the whole is distinctly earlier in style. The subjects represented show a remarkable parallelism with those on certain perforated disk-like stones found in the island. Some of them are very rude and apparently go back beyond the period of Twelfth Dynasty Egyptian influence. The existence of this most primitive class of triangular seal-stones is of special importance to our subject as showing the indigenous character of the material out of which the later hieroglyphic script was evolved. Many of the subjects, such as the vases, the heads of animals, the birds, branches and Fig. 52i. horn-like figures, are esseniially the same as those that we find conven- tionalized and grouped together on the later series. Amongst the ceramic forms we may even see traces of the earlier stages out of which the more advanced types, such as the beaked ocnochoe of Mycenaean times, were evolved. These beaked vases take in fact, on some of the stones, the same simple ' askos '-like shapes — betraying their origin from skin vessels — that are characteristic of the earlier strata of Hissarlik and of the most primitive cist-tombs of Amorgos. Others, again, are 'suspension' vases with round bottoms of equally primitive character, and arc actually seen hanging from poles. This independent evidence would alone suffice to carry back the early seal-stones of this class to the third millennium before our era. The ceramic forms that thcv portray, Fig. 52, a, h and r for c.\anii)le, correspond 64 FROM CRETE AND THE PELOPONNESE. [333] with the round-bottomed types that precede the earliest class of Aegean painted pottery, such as that of Thera or from tiie Kamares cave in Crete itself. It will thus be seen that the most typical forms of seals on which the hieroglyphic characters occur, as well as the prototypes of the hieroglyphics themselves, go back on Cretan soil to a very remote period. The earliest class seems, indeed, to iiave received its characteristic stamp already before the days of that intimate contact with Twelfth Dynasty Egypt which has left its impress on some of the later decorative designs. The evidence collected by Professor Petrie, at Kahun, tends to show that already by the time of Usertesen II., c. 2681 — 26G0 B.C., Aegean foreigners were settled in Egypt. If, therefore, the beginnings of the Twelfth Dynasty Egyptian influences perceptible on the Cretan intaglios date approximately from that epoch, this still earlier class on which this influence is as yet non-apparent may well go back to the early part of the third millenium before our era. It stands to reason indeed that the indigenous European culture repre- sented by the primitive Cretan population must have reached a comparatively advanced stage before it could have placed itself in the direct contact with the higher Egyptian civilization. Nor was it with Egypt only that the sea- faring enterprise of the Cretan islanders was already at this early date opening up communication — whether predatory or commercial, it might be hard to say. A remarkable piece of evidence is supplied by a seal-stone of the earliest class (Fig. 62), which certainly seems to point to a connexion with the Syrian coast. On one side of this stone is the unmistakable figure of a camel in the act of kneeling, the knees of its fore-legs however being bent in the wrong direction, as if drawn by one wdio had but a distant knowledge of the animal. An interesting pendant to this evidence of Oriental intrusion is supplied by a triangular stone, in every respect resembling the early Cretan type, brought back by the late Mr. Greville Cliester from the North coast of Syria, and now in the Ashmoleau Museum at Oxford. The facets are, in this case, surrounded by the oval groove or cartouche which apparently belongs to the more advanced specimens of the primitive series, but both from its compact form and the rude style of the engraving the stone in question must be referred to the same general period as those groujied above under Class III., and can hardly be brought down later than the api^ro-ximate date 2000 B.C. Other independent evidence points to the same early intercourse with Northern Syria. Certain seals in the form of a truncated or obtuse-ended cone occur in Crete, some of which seem also to have been derived at the same early date from this Oriental source. In the Phaestos deposit, above referred to, three of these, and apparently a fragment of a fourth, were found, and it is to be noted as a significant feature that one of these and the fragment were made of ivory. This imported material might in itself warrant the suspicion that this class of seal, which in Crete seems to be of exceptional occurrence, was of foreign origin. As a matter of fact, in Northern Syria, where this must be regarded as a typical form, due no F [.VH] PRIMITIVE PICTOGRAPHS AND SCRIPT 65 doubt to Babylonian influence, these sub-coaical seals are frequently formed of ivory. Seals of this type do not seem to be at home in the intervening Anatolian region, though they are occasionally found there, and their appearance per saUum on Cretan soil must be reasonably construed as evidence of an early maritime connexion between the Aegean island and the North Syrian coast. The Hagios Onuphrios find indeed affords a still more irrefragable ]iYooi of this contact in a green steatite seal, the ujjper part of which represents a seated eagle. An exactly similar type from the Hauran is to be seen in the Ashmolean Collection. Are we therefore to believe that Crete in the third millennium before our era was occupied by a sea-faring race — perhaps Semitic — from the Syrian coast ? Such a supposition might explain some of the phenomena with which we have to deal, but in any case it must be allowed that there is a distinctly local character about many of these early Cretan stones. The primi- tive seal-stones of the triangular form described arc, as we have seen, at home in Crete. That their range may have e.xtendeil to other parts of the Aegean is possible, and an example of a somewhat later type procured at Smyrna by Mr. Grcville Chester (Fig. 53) and now in the Ashmolean Collection rather b'6a. points to some such diffusion, Smyrna being a well-known gathering point of Aegean finds. On the other hand these stones do not seem to be found on the mainland of Asia Minor. Certain three-sided stones of a peculiar ' gabled-shaped ' class are indeed widely diffused in Cilicia and Cappadocia, but they are as a rule much larger and seem to have no immediate conni'xion with the Cretan form."** The occurrence of a single example of a seal-stone iileutical both in shape and technicjue with the most typical Cretan forms on the North Syrian coast is as yet an isolated phenomenon in that region, whereas in Crete itself this form is clearly indigenous and of wide distribution. Wc liave here therefore in all probability to dc:\] witli an object brougjit to the 3" Ic the ca.se of tliese stones only one side, which is Larger than tlic others, is engraved, the other two beinj{ set at .-m obtuse angle and forming a slojiing iiack like a gable. 'Gable- bliai:ed ' may therefore bo a convenient term to apply to this well-marked East-Anatolian class, wliich bears no obvious resemblance to the e(|uilateral stones with which wc are concerned. It may yet have a coninmn origin. 66 FROM CRETE AND THE PELOPONNESE. [335] Syrian coast from Crete by the same maritime agencies that in tlie contrary direction brouglit Syrian forms to the Aegean island. Tiic materials that my recent researches have enabled me to put together point clearly to the conclusion that tlie early engraved stones of Crete are in the main of an indigenous and non-Asiatic character. At the outset indeed we are confronted by a negative phenomenon wiiich brings tliis archaeological result into strong relief. The influence, namely, of Babylonian cylinders is altogether non-aj^parent. At Melos and Amorgos in deposits of the same age as the early Cretan seal-stones cylinders of native work are found in which the Clialdaean form is at times associated with a decoration whicli appears to be derived from the Egyptian spiral motives already referred to. On the mainland of Asia Minor again early indigenous imitations of Babylonian cylinders are also widely diffused. In Cyprus they are predominant, and they are very characteristic of the finds along the coast of Syria. It is evident then that a people settling in Crete from that side would have imported this type of seal, and we should expect to be confronted with the same prevalence of the cylindrical type as in Cyprus. But, as has been already observed, this characteristically Asiatic type is at any rate so rare in Crete as to be hitlierto unknown among the insular finds. This noteworthy fact seems to exclude the supposition that Crete was occupied by colonists from the Syrian coast at any time during the long period when Syria itself was dominated by Babylonian culture. We must therefore suppose that if such an occujDation took place it was at any rate at an extremely remote period. The parallelism between certain Syrian types and those of Crete is certain. There is moreover a great deal besides in the figures and style of engraving of many of the Cretan stones which strongly recalls other primitive stones found on the easternmost Mediterranean coasts. The early Cretan relics may indeed be said to belong to the same East Mediterranean province of early glyptic design as many similar objects from Syria and Palestine. But, after duly recognizing these undoubted affinities which can to a great extent be explained by the as.similating influences of early commerce, it must nevertheless be allowed that the most characteristic of the early types of Cretan seal-stones are true native products. They are in fact in situ geographically. If in the one direction they seem to find parallels per saltum on the coasts of Syria and Canaan, in another they fit on to the early engraved stones of Cilicia and the more western part of Anatolia, and they are equally linked on the other side with primitive types of the Aegean islands and the Greek mainland. Some early forms of seal-stones found in Crete have a much wider diffusion, extending not only to the neighbouring tracts of Asia Minor and the Aegean islands, but still further afield to the West. The button-like stones for example have a very extensive range in Greece and the Levant, they are found in Cyrene and even appear as imported foreign forms in the Nile valley. These stone buttons may eventually prove to have quite an excep- tional interest in the history of Aegean art, as the direct progenitors of the lentoid beads so much affected by the Mycenaean engravers. The most F 2 [338] PRIMITIVE PICTOGRAPHS AND SCRIPT 67 primitive types of the Mycenaean lentoid gems exhibit somewhat conical backs, which may be regarded as a modification of the perforated hump of the typical buttons. The 'buttons' themselves in their original form go back to a much earlier period than the Mycenaean proper, for, as has been shown above, it is upon their decorations that the influence of the Twelfth Dynasty scarab motives is peculiarly apparent.^^ But these button-like ornaments themselves, with their protuberant perforated backs, what are they but the reproduction in soft stone of proto- types of pinched-up clay ? A clay seal of an incurving cylindrical form, but, unlike the Asiatic cylinders, having incised devices at top and bottom and sicie perforations, was found in the early deposit of Hagios Onuphrios near Phaestos already refeiTed to. Aud the almost exact reproductions of some of the stone buttons in clay actually occur in the Italian terrcmarc and in the Ligurian cave deposits of the neolithic and seneolithic periods (see Fig. di a — c). The clay ' stamp ' from the tcrramara of Montale in the Modenese, Fiu. 5i.— Clay Stami'S fhom Early Italian Deposits (leduceil to about I liuear). a. Pollera Cave, Fiuale, Liguria (in the Morclli Collection at Genoa). h. Caverua del Sanguineto, Finale, Ligiu'ia. (Cf. A. Issel, Nolo paldnologichc sulla coUezione del FiG. 62. — Greyish Yellow Steatite (Crete) where the owner is associated on another side witli the head of a long- horned ram, a not infrequent feature on these early seals. 63a. 6Sb. 63c. Fig. 63.— Yellow Steatite (Crete). On Fig. 64 the ram's head is seen again associated with a bird and scorpion, the latter a favourite symbol on early Asianic and Syrian seal- stones. 64n, 64i. 64c, Fig. 64. — Yellow Steatite (Crete). It seems probable that the long-necked stout-legged bird engraved on this stone is intended for an ostrich, in which case we have another interesting- iudication of Southern commerce. The intimate contact already at this [342; PRIMITIVE PICTOGRAPHY AND SCRIPT early date existing with Egypt makes it not improbable that the trade-route by which ostriches' eggs — and no doubt their plumes as well — found their way to Mycenae had its origin in the Aegean enterprise of the third millen- nium B.C. In Fig. 6-5, an S-shaped design, similar to that noticed above, is asso- ciated on the remaining sides of the stone with two pairs of pictorial symbols, in one case two ibexes' heads, in the other apj)arently a cock and an uncertain object. This is the earliest evidence of the cock, — the original home of which is traditionally sougiit in Persia, — on European soil. 65«. 65b. Fig. 65.— Bnow.N- Si'eatite (Central Ceete). 65c. A commercial purpose is occasionally indicated by a number of incised dots or pellets which occur beside the figures on these primitive stones, and which in all cases seem to belong to a duodecimal system. In Fig. 37 of the pictographic seals already represented, which might so far as style is con- cerned liave been included in this earlier group, there are seen on one face 66«. 66i. Fio. 66. — Brown SrEATiTE Disic.F.i ah (Kajiaiie.s, Ci'.ete). twelve pellets and on two of the imrrowor sides of the stone two groups of three. On a remarkable engraved disk, Fig. GO, obtained by mo at Kamnros on the Southern slope of Ida, also of early date, a standing figure clad in a long tunic appears with four dots on either side of him. On the other side in the spaces FROM CRETE AND TTTR PELOPONNESE. [3.t3] between the various figures are three dots. On an ivory cone, again, from the I'haestos deposit four similar pellets appear, two on each side ol' a rude figure of an eagle. This early duodecimal system is found again on an interesting series of engraved stones, one n seal of curiously Cilician or ' Ilittite ' type found at Palaeokastro near Baia, opposite the island of Elaphonisi on the Laconian coast, containing a graduated series of similar groups of pellets, first twelve arranged in three rows of four, two seals with six on each, and other small perforated cubes which see'm to have stood for units. The stone Fig. GG is of great interest as affording one of the earliest examples of a group of pictorial symbols. Round the goat which forms the principal type on one side are three smaller figures — one api^arently representing the upper part of an ai'cher in the act of shooting, another a human eye, and below the goat an uncertain object. In certain cases the figures on these early engraved stones seem to have a reference to some episode in personal or family history. On the green steatite disk Fig. G7, the other face of which is occupied by two goats, a branch, and other objects, we see what, owing to the naiveness of the art, may either be interpreted as a comic or a tragic scene. A figure in a long tunic, behind which is a high-spouted vase, is represented attacking and apparently overthrowing a naked figure seated on a stool. 67«. 67J. Fio. 67.— Green Steatite Di.sk-Bead (Crete). Various designs in the primitive series recur in a more conventionalized form in the later class of Cretan seal-stones. On Fig. G8, found near Siteia, are already seen two symbols like the 'broad arrow' of the later hieroglyphic series, and the goat and the skin buckets slung on the pole again make their appearance. On Fig. G9, what seems to be a ruder veision of the same symbol is seen in front of an animal or perhaps a centaur. Then follow on the remaining sides three spearmen and perhaps a dog. The Twelfth Dynasty influence, as already remarked, is very perceptible [344] PRIMITIVE PICTOGRAPHS AND SCRIPT 75 G8a. 68i. Fig. 68.— Steatite (Found nf.au Siteia, Crete) Fig. 69.— (Crete, Berlin Mfseum). 70a. 706. 70c. Fig. 70.— Green Steatite (Central Crete). 71(1. 71/'. 71c. Fig. 71.— Vkllou- STEATrrr. (Crete). 76 Fl^OM CRETE AND THE PELOPONNESE. [345] on some of these eai-Iy scal-stoncs. Tho oii < u X ;'c) ^¥ O /»/ z ^/! Zf A V /<0 SA J S/ CD AtO [350] PKIMITIVE PICTOGRAPHY AND SCRPIT 81 on the southein steeps of Mount Ida, immediately above a Mycenaean nekro- polis, two of the bee-hive tombs of which I had occasion to visit and in which Professor HalbheiT has now excavated an intact Mycenaean tomb. The above cave was excavated by Dr. Hazzidaki, the President of the Syllogos or Literary Society at Candia, and the objects found are now exhibited in the little Museum of that Society. ^"''' My own observations of these have led me to the conclusion that the ceramic class here represented, though of archaic aspect, may slightly overlap the more purely Mycenaean pottery in the island. A spray on one specimen resembles a design on a Mycenaean pot from the prehistoric Palace at Knosos ; a fish on another recalls similar forms on the painted hut-urns from Cretan tholos-tombs, and a barbaric head and arm finds a close parallel in a painted fragment from tomb 25 of the lower town of Mycenae. Nos. 1, 6, 7, and 14 and No. 13 of Professor Petrie's Plate of Aegean pottery show, so far as their shape is concerned, a greater affinity with this Cretan class than with any hitherto known ceramic group, and the analogy certainly suggests an early Mycenaean date of some of the Kahun sherds. Both the Kamares pots and those from Kahun find, on the whole, their best comparisons with some early types from Tiryns (Schlicmann, PI. xxiv. c. xxvirf. and xxviifZ.). It may be confidently stated that during the Aegean period, which roughly corresponds with that of the Twelfth and Thirteenth Dynasties, and for which the name ' Period of Amorgos ' has been here suggested — no such finish of ceramic fabric either in form, glaze or colour as either the vases of Kamares or the fragments from Kahun had yet been achieved. If then these vessels were imported into Egypt at that early date they could not have come from the Aegean islands and still less from the mainland of Greece or from Italy. But while this, presumably the latest class of pottery found in the Kahun rubbish-heaps, is for the most part of early Mycenaean date, there seems no good reason for doubting Mr. Petrie's conclusion that the ruder pottery from the same deposit exhibiting the incised characters of non-Egyptian forms may go back in part- at least to the days of the Twelfth Dynasty. Isolated appearances will not mislead the archaeologist as to the general character of tiie deposits with which he is dealing, and all their associations point to the time of the Twelfth Dynasty as the chief period of their formation.^^ At Gurob again certain of the signs occurred under circumstances which seem to involve the same early date, while others were found on sherds which from their character and the position in which they lay belonged as clearly '"•'' A paper on the Kainaros pottery was read by elude a later date than that of the Twelfth Mr. J. L. Myres iti the Anthropological Section Dynasty. Yet these .signs belong to the .same of the British Association in 1893. It is to he class as the otliers, and occnr on pottery of the hoped that this ini]icjrtant study may shortly same rude fabric which occurs, together with >:ee the light in a fuller form. 1 believe that some r.f the marks, in foundation deposits of njy own conclusion.s as to the date of the ]iot- Usertesen II., and which, in Mr. Petrie's tery agree with those of Mr. Myrcs. ojiinion (Ktthntt, Gurvh, and Ifinrara, p. 43), '■''' The special circumstances under which ' c:annot bo mistalien for that of any Nubseiiucut the signs numbered HI, 21, 12.1, 120 in Mr. age.' I'etric'b list were found, seem altogether to e.\- 82 FROM CRETE AND THE PELOPONNESE. [351] to tlie (lays of the Eighteenth Dynasty and to the most llouiishing period of Mycenaean culture. So far as the early date of many of these signs is concerned, their extraordinary correspondence with those on the Cretan stones must be regarded as a striking corroboration of Mr. Petrie's views. Another close parallel to these linear characters and at the same time another proof of their early date has been supplied by the discovery of similar marks on potsherds discovered by Mr. Bliss in the earliest strata 1 (X 1 t hi z < 0( 1 1- Y % R Fig. 75.— Signs on ToTsnEiiDS at Tell-el-Hesy CoMrAUED with Aegean Foejis. (Cities I. and Sub. 1) at Tell-el-Hesy, which on a variety of evidence are referred by him to a date anterior to 1.500 B.C.^^* The examples given above (Fig. 75) will show that there is something more than a general resemblance '■"'^ ScE F. J. Bliss, A Mound of Many Cities, nr Tell-cl-IIcjiy Excavated, jip. 21, 23, 25, 28, 29, 30, 33, and 42. These marks on potsherds are described as found exclusively, with llie exception of No. 21, in the earliest strata. No. 21 is the last on the list below. G 2 [352] PRIMITIVE PICTOGRAPHS ANP SCRIPT ^3 between these marks and the Aegean signs. By including those of Kahun and Gurob the number of parallels may be appreciably increased.^* The correspondence of forms in the case of several of the characters found at Kahun and Gurob with those of the Aegean series is in several cases of such a nature as to exclude the supposition of a merely fortuitous resemblance. Few, I imagine, will believe that such a sign as No. 22 was about the same time evolved independently at Gurob, Knosos and Mycenae. The same holds good of several of the Cypriote letters. But the Cypriote comparisons are specially valuable since the possibility cannot be excluded that they supply a clue to the actual phonetic value of sonic of the Aegean characters. On Table II. I have put together various examples of the Aegean characters which occur in groups of two or more. They are from the follow- ing sources : — 1. Vase, Prodromos Botzano (p. 279). 2. Cup, Goulks (p. 278). 3. Ametliyst, Knosos (p. 2S1). 4. Seal, Knosos (p. 293). Signs on two sides, but unfortunately much worn. 5. Seal from Siteia Province (p. 297). On another side ideograph of a man. 6. Block of Mycenaean building, Knosos (p. 282). 7. Do. 8. Seal, Praesos (p. 293). Signs on two sides, two sprays as pictograph No. 59 on the third. 9. Amphora-handle, Mycenae (p. 273). 10. Handle of stone vessel, Mycenae (p. 273). 11. Buiton-seal, Phaestos (p. 285). 12. Perforated steatite, Siphnos (p. 287). To these must be added the Phaestos whorl. Fig. 11/'. The parallels supplied by the Cypriote syllabary suggest the following attempt to transliterate some of these groups : — 1. II -Ic- lo. 2. Il-pa-lo. 3. Kv ■ sa • ja ' Iv. 4. E-le. It remains however uncertain whether tlie characters should be read from "* Where so much still remains to be dis- railic Anieiaiio iieai' Fiuiilmariiui in Ligiiria. covereJ, it is worth while contcni|ilating at In connexion with the linear forms I cannot least the jiossibility that these early signs had help referring to certain sigus on early pottery also a Western and European extension. In the from the lake-dwellings of Paladru, near Voiron case of the purely pictograjjliic class, the in the Isere, some of which are remarkably parallel supplied by the Maravii/lic in the suggestive of Aegean parallels. For the pottery JIaritiine Al|i3 has already been cited, to which see Chautre, I'alafiUcs du Lac de I'aladru, may now be added another .similar group of Allium, I'l. X. Figs 1-5 and 7. sculptured signs more recently discovered by 84 FROM CRETE AND THE PELOPONNESE. [353] left to right or from right to left, neither is it clear where the inscription on the Siphnos stone which presents the largest number of parallels with the Cypriote should begin. Beginning with the sign which as tlie drawing stands is the topmost on the right, continuing with the lowest and then proceeding TABLE II. CROUPS OF LINEAR SYMBOLS ^ -^ Ml\t / lo D boustrophedon, the inscription as transliterated by Cypriote letters might road : — Si ■ mo' 11 ' no • se • to. The indications however are too slight to base upon them any too definite conclusions. So far as they go it must be admitted that the phone- tic equivalents suggested by the Cypriote parallels seem to belong to a [354] PRIMITIVE PICTOGEAPHS AND SCPJPT 85 language other than Greek. That we have to deal with a syllabary seems to be clear from the small number of characters contained in the several groups. The close correspondence of this whole series of signs with the Cypriote has already been sufficiently demonstrated. But the very fact that the Cypriote syllabary seems to have been derived from this earlier Aegean and ' Mycenaean ' script, or perhaps .some parallel Asianic branch, reacts against the Hellenic character of the original. For the Cypriote characters were never originally framed for Greek use. The Greek of the Cypriote inscriptions always seems to be clothed in a foreign dress ill-iittiug at the best. There is indeed the strongest presumption for believing that in Crete at least the race amongst whom the earlier Aegean characters were originally rife was of non-Hellenic stock. It was clearly recognized by the Greeks themselves that the original inhabitants of Crete weie ' barbarian ' or un- Greek. Herodotos, who brings the Lykians as well as the Kaunians of Karia from Crete, expressly says that the whole of Crete was once occupied by ' barbarians.' ^'" But the most authentic evidence of this non-Hellenic origin is the name of Eteokretes or ' true Cretans ' applied by the Dorian colonists of the island to the representatives of the indigenous stock, who long continued to live on in the fastnesses of Ida and Dikta. It would even appear that the language of these Cretan aborigines maintained itself in the extreme East of the island to the borders of the historic period. The evidence of this is supplied by an inscription recently found among the ruins of Praesos^" and now preserved in the Mu.seum of the Syllogos at Candia. This inscription, though written in archaic Greek characters, is composed in a non-Greek language, in this respect recalling the two Lemuian in- scriptions, from which however it differs in epigraphy and apparently in language. The following facsimile is from a photograph kindly made for rne by Professor Halbherr. The Praesian stone contains letter-forms in some respects diverging from those of the archaic Greek inscriptions of the island, and in the types of iuta and j9t that are there presented as well as in the early use of S shows a greater approach to Phoenician models. In the concluding letters which form the word Anait there seems indeed to be a direct reference to the Semitic Anat or Anaitis, 'the Persian Artemis,' whose image appears on one of the shields found in the cave of the Idaean Zeus.^^'' That at the period when the Praesian inscription was written the indigenous element in the island may have been still largely under Phoenician influence is probable enough, but the inscription itself does not seem to be Semitic. We may fairly conclude that the language liere found represents that of the Eteocretans of whom, as we know, Praesos was a principal stronghold, and it is reasonable to suppose that this was the originrd language of the ^"■' i. 173 riji' 7a() K/)19T7)v (Ix"" ''^ iroXaifii' V(il. iii.), ]). 'I.')! scqq, iraaav $dp0apoi. '■'■'•' F. IliUlihen' (! P. Oi'.si, Anticluli'i dclV 2'' Coiiiiiarctti, ic Icmii di Gor/yna c Ic altre Antra di Xaix Jdi:o, \\ 100 sfqq., and Atlas hcriziuni arcaiche Crelcsi, 1893 (Mon, Ant. PI. II.; and cf. Conijiaietti, luc. cit. ji. 452. 86 PROM CRETE AND THE PELOPONNESE. [355] early script with which we are now dealing. But the materials for com- parison are as yet too imperfect on either side to admit of satisfactory results. In Roman letters the inscription seems to read as follows:— ■*" f / / N K A L M I T K / OS I BARXE I AllfO ARK/APSET I MEG^ ARKRKOKLES | GEP /ASEPGNANAIT The original is written boustrophedon, the first, third, and fifth lines running from right to left. The AI in the last line are in ligature. It is possible that in the earlier period during which the indigenous Cretan script, both pictographic and linear, seems to have taken its origin the sole or preponderating element is the island may have been the ' Eteo- cretan.' It is certain however that at the time when the Homeric poems were composed Crete contained representatives of several other races. The polyglot character of the island is indeed clearly brought out by the locus classicus in the Odyssey.'"" The Greek element both Dorian and Acliaian is already at home there and seems indeed to have been already of old standing in at least the central district of the island. But if, at any rate towards the close ot the Mycenaean period, there was already a Greek population in Crete, it becomes probable that the mysterious ■*" I liave foUowcil Comparetti's suggestions loc, cit. *"'■ xix. 1. 172 scciq. [356] PRIMITIVE PICTOGRAPHS AND SCRIPT 87 characters with which we are dealing may also have been used by meu of Greek speech. And from the fact that in Cyprus a similar script, in its origin apparently non-Hellenic, was in use amongst the Greek-speaking inhabitants it becomes in itself not unlikely that the same phenomenon may have occurred in Crete and the Peloponnese where a similar script was in use in much earlier times. The Greeks of Cyprus spoke a dialect approaching to Arcadian — may they not have taken over with their language a form of writing once in use in the more Western area from which they may be sup- posed to liave migrated ? In view of these possibilities it is worth while examining the grounds of the presumption that the Greek settlement in Crete goes back to Mycenaean times. In the lines of the Odyssey referred to, whicii belong to one of the earliest passages preserved to us, Crete is spoken of as the home of several races speaking a variety of tongues, Achaeans and Dorians, Pelasgiaus, Eteo- kretes and Kydonians : — • K.p7]Trj T(? yaV eari jxeaw iv\ oXvoiri itovtw, KOkKr] Kcu Trieipa, Trepippvro';- ev B' dvOpwrroi TToXXo!, aTreipeaioi, Kat, euvijKovra TrdXi/e?. aWyj 8' aWu)V yXoyaaa fj,ep,iyp,ein]- ev fiev Ayatol eV B' 'Ereo/cpjjTe? fi€yaXi]Top€<;, hv Be KuSiuz'e? Aw/jfee? re Tpi^diKe}<;, where, as has been shown by Hoeck,'" there is a distinct reference to the specially Dorian *i' time division of nine years or ninety-nine months, — the double Olympiad, — at the end of which 'long year' Minos according to the tradition used to return to the cave of Zeus to receive fresh instruction and repeat what he had learned before.*^ But Minos himself is not Dorian, and the mythical genealogist is content with making the son of the Dorian leader Teutamos, who came from Thessaly to Crete, adopt the children of the ■" Krcta, i. p. lid srqq. From the later usage fVrc'wpos spiitor in allgemci never Hedentung witli reference to the election of the Spartan angewaudt seyn, mag selhst .schon Homer sieh Eiihors Hoeek infers that the Dorian kings re- tliescs Ausdrucks nicht mit jener bestimmten quired a fre.'ih religious sanction for their sovc- Riicksicht hedient haben : so lag doch der reignty every nine years, so that they could he tiefste grand dcr Hedeutsamkeit diescr Nenu- said to reign 'nine years.' Ho concludes; zahl in jener alten .lahreshestimmung.' ' Dicss ist unstrcitig der tiefcre Sinn welchcr '"■^ Dodwell, dc Cyd. p. Z\Q scqq. dem homcrischen Wvws ^ccewpos $a(ti\tvf "• I'lato, vi. p. ],')8. Cf. Schol. ad Od. untcrliegt. Mag nun immerhin das Wort xix. ITS. 88 FIJOM CRETE AND TIFE PFXOPONNESE. [357] Cretan Zeus — Minos, Rliadamanthys and Sarpudon.''-' Accordintr to this version we have a Dorian settlement in Crete from tlio Thessalian Doris, the later Hestiaeotis, under a leader with a Pelasgian name, going back to prao-Minoan times. It is to bo observed that this Thessalian connexion fits in with the account of the Odyssey which couples ' divine ' Pelasgians and Achaeans with the Dorians in Crete, and with the fact that a son of Minos beais the name Deukalion. According to the native Eteokretan tradition of the Praesians preserved by Herodotos,'*-' the Greek settlement in Crete had beo-un before the Trojan war, as a consequence of the depopulation of Crete caused by the disastrous Western expedition that followed the death of Minos. The Chronicle of Eusebios goes so far as to fix the year 14in B.C. as tlio date when the Dorian, Achaean and Pelasgian settlers who had set forth from the country about the Thes.salian Olympus landed in Crete. It will be seen however that though both the native Eteocretan tradition as preserved by the Praesians and the Greek records of tlie Thessalian expedition assign a great antiquity to the first Dorian settlements in Crete, they are in some respects at variance. The Praesian version speaks vawuely of a first settlement of Greeks and other foreigners in Crete at the time when a large part of it was left uninhabited owing to the wholesale Western exodus that followed the death of Minos. It then refers to a second depopu- lation of the island, consequent on the expedition against Troy, followed by a second colonization, which might fit in with the Dorian occupation of the Peloponnese. The Greek account on the other hand plants Dorians Achaeans and Pelasgians in Crete two generations before Minos, who becomes the adopted son of King Astorios the son of the Dorian leader. ''-'=' Diorl. iv. 60. In other MSS. of Dioduios tliat the Praesians and oU Kyjoniaus were of the name of the Dorian leader (son of Doros) the same stock, on the otiior hand it does not appears as Teltamos. Andron, in Stepli. Byz. necessarily mean that Minoau Crete was then in s. r. Awpioi/, gives the same version of the Dorian other hands. It is, rather, a patriotic way of invasion from Thessaly in prae-Minnan times, accounting for the disappearance of tlie Eteo- where the name appears, probably erroneously, kretan pojiulation from tlie later Dorian area as Teksaphos. Teutamos, as Hoeck notes by the fact that their Western expedition had (AVcta, ii. 1, 24, note 6), recurs in Pelasgian left the land tenantlcss, for any one who ehose genealogies ; cf. Hoiner, //. ii. 843. to occupy it. The argument, in fact, runs as " Her. vii. 171 €s Se tjjx KpTjTijc ff-nnaiee'iaav, follows. The greater part of Crete is occupied is A67ou(Ti OpoiVioi, taoiKiCeaeai &\\ovs t6 av by foreigncr.s. The-^ie foreigners came in when epdirous Kui fxaAiara "E\^^p'as, TpiVi) Si yefefi the original native occupants had cone else- /ifTa Mifccv T(AivT-liiTai>7ayei'(a0aiTaTpuima.. . . where on a Western expedition whence they It is reasonable to bring ep-nfiweuaav into con- never returned. But we Praesians as well as nexion with the failure of the great Cretan ex- the Polichnites near Kydoni.a, represent the old pedition to avenge the death of Minus and the inhabitants of the land. Therefore neither we Cretan settleanent of lapygia described in the nor they took part in the Western expedition, preceding chapter. The direct reference by The survival of the indigenous element in the Herodotus to Praesian, i.e. Eteokretan, tradition Kydoiiian district in the extreme West of Crete in c. 171 gives a special importance to his supplies a presumption that the Doric coloni- stateraent in c. 170 that the Praesians .and in- zation of the i.sland did not come by way of habitants of Polichn.a, that is the old Kydo- Peloponnese. All traditions point to Central— nians, alone among the Cretans did not take ' MinGan '—Crete as the region where Hellenism part in the Sicilian expedition. It seems on first took root. the one hand to show a recognition of the fact [358] PRIMITIVE PICTOGRAPHS AND SCRIPT 89 But both traditions are at one in regarding the Dorian occupation of Crete as the result of peaceful settlement rather than of a war of extermination. The account of the ' adoption ' of Minos by the son of the Dorian chief, after the settlers had seen a second generation grow up on Cretan soil, certainly points to a gradual and bloodless amalgamation of the Hellenic and indigenous elements. It has been necessary to recall these traditions of the great antiquity of the first Dorian settlement in Crete, since the prevailing tendency is to regard that settlement as a secondary result of the Dorian occupation of the Pelo- ponnese. That the conquest of the Peloponnese may have brought with it a new flow of Dorian migration to Crete is likely enough. The earlier settle- ments may well leave room for the later attributed to Pollis and Delphos of Amyklae, or for that of Althaemenes from Megara or Argos. The native tradition as represented by the Praesians distinctly points to a fresh Hellenic settlement in the period that succeeded the Trojan war. But to regard the traditions of the early Dorian settlement from Thessaly as given by Andron and Strabo as simply fabricated from an erroneous interpretation of the Homeric passage seems quite unwarrantable. The Homeric collocation of Dorians Achaeans and Pelasgians points itself to Thessaly ; the name of Deukalion, applied already in the Iliad to a son of Minos, points in the same direction, and a mere comparison of many of the local names of Crete with Thessalian forms is sufficient to prove an early connexion with that region.'** Both tradition, then, and nomenclature favour the view that Greeks and ' Pelasgians ' from Thessaly may have settled in Crete at a date far anterior to that of the Dorian conquest of Peloi)onnesos, and it follows that among those who used the curious Cretan script of Mycenaean and earlier times there may well have been men of Hellenic speech. The archaeological evidence points the same way. Although on the present occasion it is impossible to go into the evidence in detail I may say that my own researches into the prehistoric antiquities of Crete have brought home to me the impression of their great homogeneity. From Kissamos and Kydonia in the extreme West to Praesos and Itanos in the extreme East the same characteristic forms are perpetually recurring. The same type of Mycenaean culture, with certain nuances of its own, is common to the whole island. The same rude terracotta images occur throughout, and, as far as o.ur evidence reaches, the funereal rite of enclosing the bones of the dead in painted hut-urns enclosed in tholoi, at times excavated out of the rock, was as widely diffused. Diversity of race may have eventually led to some local differentiation. It looks as if tlie later class of seal-stones with pictographic ** E.g. irtrm'rt, the ancient ii;uiu; (or Clortyiia lie compared willi Trikka. Tlii'rr, was also a accordinf! to .Stepli. liyz. («.«'.), doHyn itself ('rctan Marinciia, accoriliiig to some accounts compaiinj; witli GyrlOii in I'errhaebia (liechte!, foiuuled by Magnetes from Thessaly (I'arthen. cited by linsolt, Or. G'cscli. 1'^, !530, note) ; End. c. 5). Tliesc parallels extend to Mace- Pimentos, Vhalanna (cf. too I'lialasarna), and donia ; compare for instance Olons and Olyn- /teM are also found both in Crete and Thessaly. thos, lUerapytna and I'yilna and the rivev- Tritla, an old name for Knosiis, may pos.sibly names Axos and Axios. 90 FROM CRETE AND THE PEL(3P0NNESE. [359] symbols were the special product of the surviving representatives of the aboriginal race in the East of the island, while on the Southern slopes of Ida, — to judge by the relics found in Kamarcs grotto, — pottery of archaic fabric continued to be produced in early Mycenaean times. Regarding them as a whole however, a great family likeness is perceptible in Cretan remains of this early period ; and, together with the general homogeneity, a remarkable continuity is observable. From about 900 B.C. onwards, to judge from the bronzes of the cave of Zeus, there was a strong A.ssyrianizing influence, due no doubt to Phoenician contact; but the archaeological break which at Mycenae itself and in the Greek mainland generally is perceptible in the centuries immediately preceding the days of the miscalled 'Archaic' Greek art or, as we should now call it, the Greek art of the ' Early Renaissance,' is in Crete conspicuous by its absence. We have here what may be called late Mycenaean crossed by Oriental influences but still essentially continuous, a phenomenon which repeats itself in an almost identical aspect at Argos and in the Argive relics found at Kameiros. The break caused on the Greek mainland by the intrusion of a geometrical style of art fitting on to that of the Danubian valley and the Hallstatt culture of Central Europe is reasonably connected with a tide of invasion from the North, of which the Dorian invasion of the Pelopounese represents the southernmost wave. But the Dorian invaders who are supposed to have been hurried on to Crete by the same migrating impulse — where have they left their mark on Cretan antiqui- ties? Certain geometrical elements came in no doubt, fibulae are found identical with those of the Dipylon or the Boeotian cemeteries, but the evolution of Cretan art is still in the main continuous. That there was at this period a fresh Dorian colonization of parts of Crete is probable : but the new comers were merged in the body of Dorian inhabitants already long settled in the island, and received from them the artistic traditions that they had themselves handed down from Mycenaean times. And in architecture at least, let it be remembered, it was the Dorian element that was to represent the true Mycenaean tradition. Another piece of archaeological evidence completely disposes of any difficulty that might be felt as to a colonization of Crete from such a com- jjaratively distant quarter as Thessaly in Mycenaean times. Mycenaean culture was early planted in the Thessalian coastlands, as appears from the tombs of that period discovered on the headland opposite Volo, the ancient lolkos.^'"' But, among the vases found in these Thessalian tomb.s, is a peculiar class of one-handled pots displaying water-plants with arrowlike or cordiform leaves and waved lines below, apparently indicative of water. A vase of the same form but with a different ornamentation was found in Akropolis Grave No. III. at Mycenae,^" but in the Maket tomb at Kahun, now shown by Mr. Petrie to belong to Thothmes III.'s time,^^ there was deposited a *^ See Wolters, ilykcnisclie Vasen aus dem occurs on a vase from Grave I. and another niirdlichcn Gricchenland, Allien. MiUh. xiv. from Grave VI., as well as on a glnss paste (1889) p. 262 scqq. ornament from Grave III. ■"'• A leaf ornament of tlie same character *' See above, p. 318. [360] PRIMITIVE PICTOGRAPHS AND SCRIPT 91 pot not only of the same shape as the ThessaUan examples but with an identical design. lolkos and the Nile Valley were thus either in direct commercial connexion, or at any rate supplied from the same source, as early as the fifteenth century B.C., and it cannot be doubted that Crete, lying between the two, formed an important link in the chain. The vegetable motive described is indeed a characteristic feature on Cretan gems of the Mycenaean period ^'^ and will doubtless be eventually found to have played an important part in Cretan ceramics. The arcliaeological evidence makes it well-nigh certain that there was a direct, intercourse between Crete and the famous Thessalian port at the period when, according to tradition, the first Dorian colonists along with Achaeans and Pelasgians found their way to the island from that veiy quarter. Tliere are therefore good grounds for supposing that the Greek colonization in Crete goes back well into the period during which the primitive forms of script with which we are dealing were in general use in the island. As a matter of fact the later epigraphic monuments of more than one of the Dorian cities of Crete actually exhibit what appear to be survivals of some of the characters belonging to the prae-Phoeuician script with which we are now dealing. Professor Halbherr has made to me the valuable suggestion that some of the characters brought to light by the present investigation had influenced the forms of certain letter's that occur in the most archaic Greek inscriptions found in the island, while in other cases they seem actually to have survived as marks of division. Thus at Lyttos there is seen a form of O consistino- of two concentric circles, with or without a central dot,''" identical with the symbol No. Id of the pictographic series or 28 of the linear. At Eleutherna ^^ and Oaxos ^^ there is found a form of Van -Y which suggests a differentiation from the Phoenician Van under the influence of the linear character No. 20 £ On the other hand the double axe symbol X occurs both at Gortyna ^^ and Lyttos ^"^ as a mark of division.-'^* But in considering the possibility that this early script may have been made use of by men of Greek speech we cannot restrict our survey to Crete alone. The indications that we possess, at any rate in the case of the linear characters, point to a much wider diifusion, Mycenaean in its most compre- hensive sense. The early script that we find in Crete extends, as we have seen, to the Peloponnese, but quite apart from this phenomenon there is abundant evidence to show that the Mycenaean culture in the two areas, at least in its earlier stages, was singularly uniform in aspect. On this occa- sion it is impossible to cuter into details, but it may be sufficient to say that the engraved Mycenaean gems found in Crete show a remarkable correspond- ence with those from Mycen;ie itself, the Vaphio tomb and other Peloponne- ■"' See .ibovc, p. !i23. is used to si'pai'ate two very (lifTerpnt clauses. ''» Comparetti, Lcygi di Oorlyna, A-c, p. 201. "■' Op. cit. \t. 434, Iiisct. No. 203, 1. 7. In '■" Op. oil. p. 418, Inscr. 194, 1. G. this case the sif;u is written hnrizontally in- 5' Op. cit. p. 402, liiscr. 187, 1. 3. stead of vertii'ally. "" Op. cit. p. 117, col. ix. 1. 43. In the note ■'* At Coiinth the same sign is used for E, in it is spoken of as 'un segno insigiiificante.' It I'amphylia for E, &2 ti'ROM CRETE AND THE fKI.OPONNKSK [361] siaii sites. Tlie art of tlio Vapliio gold vasus Huds itsclCaii absolute counter- part oil a fragment of a stone vessel presenting similar reliefs obtained by me on the site of Kiiosos. The cult-scenes on the gold rings find their nearest pendant on a Cretan example. A bronze figure of the same early type as that found at Tiryns, and another from Mycenae, lias Intely been dis- covered in a cave near Sybrita. Tn short, whichever way we look, we sec Mycenaean art in Crete as it now begins to emerge before us displaying the same typical form that it bears in Peloponnesos. And few will be found to doubt that, whatever may have been the nationality of the dominant race in whose hands both in Crete and Peloponnesos this art first took its character- istic shape, in Pclopoimesos at any rate it was taken over by Greek-speaking tribes. The close relation with Crete into which the royal house of Mycenae is brought in the Iliad and in Greek tradition generally ^^ becomes in this connexion of special interest. Atreus himself or his son Pleisthenes marries AiJrope the granddaughter of Minos, who in turn becomes the mother of Agamemnon and Menelaos. Idomcneus, the uncle of these, is the sjuest of the Argive princes — notably of Menelaos — and connected with them in the affairs of peace as well as war.^® According to local sagas Agamemnon himself founded the Cretan Mykenae^'^ and other cities of the island. There are besides this a considerable number of local names common to Crete and the Peloponnese,''^ but some at least of these may be due to the later wave of Dorian migration from Laconia and the Argolid. The early connexion between Crete and other parts of the Greek mainland, notably with Attica and Boeotia, is borne out by the same evidence of tradition and nomenclature. In the case of Boeotia indeed it is tempting to see in the peculiar form of the E a trace of the influence of the linear or pictographic symbol resembling a four-barred gate. Incomplete then as our evidence still is, it tends to show that the uso of early script with which we are dealing may have been shared both on the mainland and in Crete itself by men of Greek speech. The data at our disposal seem to warrant the conclusion that the diffusion of this early system of writing was in fact conterminous with that of the Mycenaean form of culture. The pictographic class of seal-stones seems to have been princi- pally at home in Crete. But the linear script had evidently a very wide range. In Crete itself the linear characters occur on a greater viiriuty of materials than the more pictorial forms. In the Peloponne.se they are found not only at Mycenae itself but at Nauplia, they reappear at Menidi and at Siphnos, and in Egypt they are found on the early potsherds of Kahun and Gurob. On the early whorls of Hissarlik we already see traces of similar signs.-'*''' In Cyprus we find a closely allied system, which had also diffused itself along the coastlands of Asia Minor, survivino- into classical times. It further appears that very similar signs had invaded the coast of 5= See espuciiilly Hoeck, Krela, ii. p. 397, ^^ E.g. Jnojkla, Therapnae, Pharac, Boiat, "'. yciO. [366] PRIMITIVE PICTOGRAPHS AND SCRIPT 97 appears ia conjunction with a linear symbol ty/- On the other side of the same stone the head of the same animal like an y upside down is followed by the same symbol £})/. The A is thus brought into direct connexion with the bull or ox. On a seal-stone again (Fig. 26a) we find a pictorial representation of an ox's head accompanied by other symbols, while on the pendant from Arbi (Fig. 16) occurs what seems to be a linear form of the same, standing by itself It cannot be doubted that the A symbol of the Cretan series is derived from an ox's head. If we turn to Phoenicia we find the same symbol with a record of its joictographic original in its name Akf=an ox. According to De Kouge's theory however, which still holds the field, we are asked to believe that in Phoenicia the symbol, notwithstanding its name, was derived from the hieratic representation of an eagle. In Crete we see the double axe linearized into a symbol X, like a closed X or two crossed Z's. From the occurrence of this symbol as equivalent to Z on the Sabaean inscriptions there seem to be good reasons for believing that the original Semitic form of Zayhi was of this shape, and Zayin is generally translated ' weapons,' which would find its natural explanation in the pictograph of the double axe. But the received derivation is from the hieratic sketch of a flying duck. The Cretan pictograph for a tree is reduced to the same form as the Phoenician SameJch = a post, the origin of which is by De Rough's theory traced to the hieratic degradation of an Egyptian chair-back. The two Cretan pictographs which may stand for a gate, fence, or shutter and the accompanying linear forms are practically identical with the Semitic C/(t'a Smith, Historical Gcor/raphy of the Holy Land, nach altdgi/plischcn DcnlcmiiUrn, 1893, p. 389. ion FROM CRKTE AND THE PELOPONNESE. [3G9] part in tlic invasions of Egyjit under Ramses III. and Merenptali, and wlioso name when brought into connexion with that of tlie Cretans curiously recalls the Bioi, lleXaa-yoi,"* so early settled in the island side by side witii 'Ereo- Kpi]Te<;, Achaians — probably the Akayvasof the same Egyptian monuments — and Dorians. Another member of this group of Aegean and West Anatolian peoples whose maritime enterprise was now a terror to Egypt and its border- lands was the Takkara, — ex hypoihesi Teucri, the eponymus of whose race wliether he appears at Salamis or Troy is doubly connected with Crete. These people are brought into close connexion with the Pulasati and Danonas (presumably Danai) in the expeditions against Egypt, and from an interesting notice in the Golenischeff Papyrus ^^ it appears that Dore or Dor on the coast of Canaan was already by about 1100 B.C. known as a city of the Takkaras. In Greek legend this city was founded by ' Doros the son of Poseidon'^'' and its inhabitants were known as Aeopiei>;P The names are certainly suggestive, and in days when lonians and probably Achaians were already mentioned in Egyptian records a trace of a Dorian element on these shores hardly need surprise us. That among the various elements from the Aegean coastlands who took part in the Philistine Confederation men of Greek stock may already have found a place as early as the twelfth or eleventh century B.C. can no longer at least be regarded as an improbable hypothesis. It is perhaps not without some actual warrant in fact that in the Septuagint version of Isaiah '^ the Philistines themselves are translated by "EWyve^;. Hebrew tradition is unanimous in bringing the Philistines from the ' Isle of Kaphtor.' ' Island ' here may simply mean distant coasts such as those of the Aegean in general, but the alternative form of Cherethim applied to the same people certainly indicates that, in so far as it stands for an island, Kaphtor should be ajiplied to Crete rather than Cyprus. This consideration lends an additional interest to the suggestion that Kaphtor may be connected with Kefto, whence came the people who of all those represented on Egyptian monuments most clearly show Mycenaean characteristics. Their costume, their peaked shoes and leggings, the dressing of the hair, the characteristic vessels they are represented as bearing to Thothmes III., show the closest parallels with Mycenaean forms. This parallelism, as shown by the Pelo- ponnesian remains such as the wall-paintings of Mycenae, the shape and ornament of the gold cups and vases and notably the figures on the Vaphio cups, has already been pointed out.'^ The identification of the Kefti with the '* Chabas, who transliterates ' Pulasati ' view that they are Pelasgiaus. But he accciits as 'Pelestas,' had already identified them the identification of the Shardiu, Turshas, with the Pelasgians in his Antiquiti his- Akayvas, and Jevanas, with Sardinians, Tyrseni, toriquc. So too Kenan (Histoire g6n6ralc Achaians, and lonians. dcs lanqwcs simUiqucs, V, p. 53) : ' line hy- "^ W. Max Miiller, Asicn und Europa nacU pothese trfes vraisemblable, adoptee par les meil- alldgyptischai Denhnalcrn, p. 388. leurs exegetes et ethnographes, fait venir les "'' Steph. Byz. s.v. Aiipos. Philistins de Crfete. Le nom seul de Plishti... ''' Steph. Byz. I.e. navaavla^ Si eV tJ ttjs rappelle celui des Pelasges. ' This view also naTpiSos aitrov kticx^l Awpiers avToiis Ka\u commends itself to Maspero {Hist. Anc. des "' C. ix. 12. peiiples d'Orient, p. 312). AV. Max Miiller '' This comparison, first instituted by Puch- {op. cit. p. 368), while admitting the possibility stein, has been further brought out by Stein- that the Pulasati are Philistines, rejects the HoxS, Archdologischcr Anzeigcr,\i^2,]).\2 scqq [370] PRIMITIVE PICTOGRAPHS AND SCRIPT 101 Phoenicians lias been further shown to rest on a confusion of Ptolemaic times.^" The ruddy hue of the Kefti chiefs in the Theban paintings, — which seems to be the Egyptian way of rendering the rosy European cheeks.^^ — as well as their dress and facial type are clearly non-Semitic. Isolated resemblances such as those presented by the bronze figure from Latakia, the Syrian Laodicea, now in the Louvre,*^ or by the details of some Hittite or early Cilician reliefs cannot weigh against the much greater conformity with Mycenaean types, and, to the Peloponnesian examples already cited, my own researches now enable me to add a striking array of Cretan parallels. Here it may be sufficient to say that throughout Eastern and Central Crete the commonest types of Mycenaean gems show as their princi- pal designs a series of vessels evidently representing originals in the precious metals, some with beaked spouts, some with S-shaped double handles and slender bases which reproduce several of the most characteristic types of the vessels offered by the Kefti chiefs to Thothmes III. on the Theban tombs. The men of the Vaphio cups, who present such a striking resemblance to the Kefti tributaries as seen in the walls of the Rekhmara tomb, recur with the same flowing locks on a fragment of a stone vessel from Knosos. It is true that, if on the one hand the Kefto folk are brought into connexion with the people ' of the islands of the sea,' ^^ on the other hand they are found in the company of Hittites and of men of Kadesh and Tunej) (Daphne) and the Upper Rutenu of Inner Palestine. But if, as there is good reason for believing, the carrying trade of the East Mediterranean was at this time largely in Mycenaean hands, these associations and perhaps the tribute of silver and copper — it may be from Cilicia and Cyprus — that the Kefti bore in addition to their artistic vases would be accounted for without difficulty. The matter will appear even simpler if we may accept the view that the name of Kefto is to be identified with that of the Caphtor ^■' whose inhabi- tants included both the Aegean islands and the coast of Canaan in their 8" In the Canopus Decree ' Kefti ' is trans- numerous engraved stones found tliere, like lated ^oiriKr), which led Ebers and otlier Egyjit- others recently brought back by Mr. D. G. olofists to accept the identification of the Kefti Hogarth from Ain-Tab in Commagene, are with Phoenicians. W. Max Miiller however of Hittite and non-Mycenaean character. {Asicn uiul Europa nach allayyptischcn Dcn/c- *' Op. cit. p. 351. mdlem, p. 337) has shown how valueless the *- Longperier, Musee Napoleon, 21 ; Perrot et Ptolemaic tradition was in such matters. Chipicz, Ph&iiicie, d-c, 429, 430. From the place in which the name appears — ^•' In the Rekhmara inscription, after Naliariu and Heta — in early Egyptian ^* Tomb of Men-Kheper-ra-seneb, Mission lists, he himself conchules that it represents arclUologiquc fraM;aisc an Cairc, 5, 11, and cf. Cilicia. Stcindorlf, who also [op. cit. p. 15) W. Max Miiller, op. cit. p. 347, and SteindorlV, rejects the identification with Phoenicia, is loc. cit. led to seek the Kefti in the Gulf of Issos *° Ebers' suggestion that Caphtor = ' Kafl- or Cyprus. Hut, as noticed above, the rcrc' or Great Kefto (which he assumed on the archaeological evidence does not favour cither strength of the Canopus decree to be Phoenicia) Oilicia or Cyprus. Cyprus, as we know, was is rejected by W. Max Midler (np. cit. p. 390), touched by Mycenaean culture in comparatively who however expresses the opinion that the late times, but it w.as never, certainly, a centre name Kefto has nevertheless a real connexion of its propagation. The early Mycenaean .spiral witli Caphtor: '1st der Name Keftu (the or- work, .such as is seen on the Kefti vases, is foreign thograjdiy approved by him, p. 337) auszu- to Cypriote remains. On the Cicilian mainland sjirechcn so ist allerdings der Auklang muhr als Mycenaean traces altogether fail us. The zufiillig.' 102 FROM CRETE AND THE PELOPONNESE. [371] field of activity. The later confusion of their land with Phoenicia in the Canopus Decree is in this connexion not without its significance. In considering the question of possible Philistine influence on the origin of the Semitic script it must always be borne in mind that the actual colonization of Palestine is only a comparatively late episode in a connexion which goes back to far earlier times. The parallels supplied by the more primitive class of Cretan seal-stones abundantly show that there was a lively intercourse between the Aegean island and the Easternmost Mediter- ranean coast as early as the third millennium before our era. Aegean enter- prise, according to Mr. Petrie's researches, penetrated at an equally early date into Egypt, and of this again we have now the counter-proof in the Twelfth Dynasty Egyptian relics found in Cretan interments. Whether or not a ' proto-Semitic ' element may have existed in Crete itself and other parts of the Aegean world from very early times is a question beyond our present scope. Should this prove to have been the case it might simplify some problems that are at present enigmatic. There certainly seems to be a deep- lying community of early tradition between Crete and the Semitic world older than can be accounted for by Phoenician agencies of post-Mycenaean times. A river-name like lardanos, Minos the Cretan Moses, Diktynna in some respects so closely akin to Derke and Atargatis, the evidence supplied by Mycenaean relics of the early cult of Astarte, are only a few of a series of suggestive indications. There are Thraco-Phrygian elements no doubt which must be set off against these, but the possibility that the later coloni- zation of Canaan by the Philistines and their allies was in part at least a return wave of Europeanized Semites cannot be altogether ignored. Conjecture apart, however, the evidence accumulated by the present inquiry may be fairly taken to establish certain fixed points in the early archaeology of Crete and the Aegean lands. Proofs have been given of the existence of a pictographic system of writing which in Eastern Crete at any rate survived into Mycenaean times, but the earlier stages of which, on the evidence of Cretan seal-stones, may be traced far back into the third millennium before our era. The pictographic system of Crete is itself of independent growth and, though perhaps modified by Egyptian influences, is not a mere copy of Egyptian forms. In the Aegean world it occupies the same position as is occupied by the ' Hittite ' hieroglyphs in Asia Minor or Northern Syria, and it must in all probability be regarded as a sister system with distinct points of affinity and perhaps shading off into the other by intermediate phases. The pictorial forms are intimately connected with a system of linear signs which also goes back to a high antiquity, but which in certain cases at least may be referred with some confidence to a pictographic origin. These linear signs are of wide Aegean range, they fit on to the syllabaries of Anatolia and Cyprus and show besides many striking points of affinity with Semitic letters. They are found in Egypt at an early date in the wake of Aegean influences and seem to have been the common property of the Mycenaean civilization. In all this we have an interesting corroboration of an ancient Cretan tradition recorded by Diodoros. According to the Cretan version the [372] PRIMITIVE PICTOGRAPHS AND SCRIPT. 103 received accoiint of the invention of letters by the Phoenicians was only partially true. The Phoenicians had not invented written characters but had simply ' changed their shapes.'*" In other words they had not done more than improve on an existing system, — which is precisely what the evidence now before us seems to suggest. We may further infer from the Cretan contention recorded by Diodoros that the Cretans themselves claimed to have been in possession of a system of writing before the introduction of the Phoenician alphabet. The present discovery on Cretan soil of both a pictographic and a linear script dating from times anterior to any known Phoenician contact tluis affords an interesting corroboration of this little regarded record of an ancient writer. But the evidence of the Cretan seal-stones to which these remarkable results are mainly due does not end here. In many other ways they throw a new and welcome light on the early culture of the Hellenic world. The implements and instruments of Crete in Mycenaean times are here before us. The elements are present for the reconstruction in one case at least of a great decorative design. The pursuits of the possessors of the seals are clearly indicated, the ships that they sailed in, the primitive lyres to which they sang, the domestic animals that they tended, the game that they hunted, the duodecimal numeration that they employed. On the earlier seals we are able to trace the beginnings of this Aegean culture to an age much more remote than the great days of Mycenae. We see before us the prototypes of more than one of the characteristic forms of Mycenaean times. Here are its familiar vases in an earlier stage of development, its decorative beads approaching more and more the primitive clay button, its butterflies and polyps and even its mystei'ious lion-headed beings. Above all we find abundant proofs of a close contact with the Egypt of the Twelfth Dynasty, and of the taking over of the spiral system that characterizes the scarab decoration of that period. We can thus, as already pointed out, trace to its transported germ the origin of that spiral system wliicli was afterwards to play such an important part not in Mycenaean art alone but in that of a vast European zone. On the other side we find at this same early period, which may be roughly characterized as the middle of the third millennium before our era, accumulated j^roof of a close connexion with the Easternmost Mediterranean shores. The camel, perhaps the ostrich, was already familiar to the Cretan merchants and the ivory seals of Canaan were hung from their wrists. Already at that remote period Crete was performing her allotted part as the stepping-stone of Continents. ^' Diod. Lib. v. c. 74. (paal {sc. ot Kp^res) holleniqiies en Crete est venue bien lieureuse- Tous *oiViKas oiiK ^1 &PXVS (vpiiv &K\a tovs iiient confirnicr ces donnees des ancicns, qui, ■TiTTovsTuv-ypafi.ti.iTwvij.ei-a9iivatii.6vov. M. J. on le voit, en savaiout bien plus sur les temps P. Six kindly reminded nio of tliis passage. prdliolleuiqucs qu'on ne lecroit commuudmeut.' He adds 'la decouverte des liiuroglyplies pre- THE SEPULCHRAL DEPOSIT OF HAGIOS ONUPIIRIOS NEAR PHAESTOS IN ITS RELATION TO PRIMITIVE CRETAN AND AEGEAN CULTURE. 105 THE HAGIOS ONUPHRIOS DEPOSIT. THE SEPULCHRAL DEPOSIT OF HAGTOS ONUPHRIOS NEAR PHAESTOS IN ITS RELATION TO PRIMITIVE CRETAN AND AEGEAN CULTURE. The Phaestos deposit so frequently referred to in the preceding pages ^ is of such unique importance in the early archaeology of Crete and the Aegean shores that a more detailed account of some of the objects found there will not be out of place. The objects were found in a heap of human bones and skulls at a spot on the hill of Hagios Onuphrios, which rises about a quarter of a mile to the North of the double Akropolis of Phaestos. The find-spot, as already noticed, was on the Southern slope of the hill just above the Khans on the Dibaki road, and near the aqueduct of a mill. The dejjosit itself belongs to the period of Aegean culture so well illustrated by the early cemetei'ies of Amorgos, and to which the epithet ' Amorgan' may perhaps be conveniently applied. It represents a series of interments probably covering a considerable space of time, but the latest objects found, such as the painted vases (Figs. IOC — 108) below, are still prae-Mycenaean in their character, though showing some approximation to the earliest ceramic style of Thera. The deposit is in fact a j^art of what was evidently a prehistoric Necropolis of Phaestos. Among the seal-stones found in this deposit a three-sided steatite of the early type has been already engraved on p. 34.5 (Fig. 78), and the influence of a Twelfth Dynasty Egyptian motive has been traced in part of the design. Another quatrefoil seal is given in Fig. 50 (p. 328) showing a fully-developed spiral motive fitting on to the early Egyptian class. Several Egyptian scarabs referred by experts to the same period were discovered. Amongst these is a characteristic type of amethyst, though this, like the white steatite (Fig. 77), is ornamented with plain circles. A more elaborate decoration into which the spiral largely enters is seen in Fig. 78. Fig. 79 is a steatite bead-seal, perhaps suggested by a form of shell, and is somewhat analogous to the class described as 'cowroids,' while Fig. 80, with a similar leaf ornament, is carved above into a convoluted relief which has been compared to two nerita shells with a common whorl.^ The clay cylinder (Fig. 81) with side perforations is remarkable. Unlike the Babylonian cylinders, it is engraved only at top and bottom. The design above may perhaps be intoipretcd by the light of a better example of a similar design on an early seal from the Heraeon at Argos'* as a man standing before a large shield approaching the typical Mycenaean form. ' See p. 14 (283) ami 56 (.32i>) scqq. imlebted to Dr. WaUlstciii, Director of the ex- 2 Seep. 20(289). favatioiisof tlie AnjuricaiiSeliool ;it tlu' llcrai-'on. ' For the sight of an impression of lliis I am r'ltlMITIVK CRETAN AND AEGEAN CULTURE lOG Fig. 7". — White Steatitr Scar ah (2 iliams. ). Fill. 78. — Ste.\tite Scauai! (2 diams. ). 79n' :> 796. 19c. Fig. 79. — Ste.\tite (2 diams.). 80«. 80J. Fig. so.— SiE.n'irE (2 diams.). 107 THE HAGIOS ONUPHRIOS DEPOSIT. Jl '§:§'; <|4^' Slff. 81i. 81c. Fig. si.— Terracott.\ (Xatural Size). 82f(. S2i. Fig. 82. — Steatite (2 Jiams.). 83«, h. Fig. 83. — Steatite, H auras. 81(6i«) a. Fig. 81 (i/s).— Dark Stiutite (2 iliams.). 81 (A«) h. Sl((. Fk;. 81.— Steatite (2 diams.). S'li. PIUMFTIVE CllETAN AND AEGEAN CULTURE. 108 Of the button-seals, one with what look like rude linear characters has been engraved on p. KJ (285), Fig. 12. A broken specimen of another (Fig. 81 his) exhibits part of a curved design, the Twelfth Dynasty origin of which has been illustrated above.* The most remarkable for form is the eagle-shaped seal (Fig. 82) of green steatite, engraved below with a very rude figure of a goat. Tiiis seal, as already pointed out/' finds a close parallel in a Syrian 85«. Fig. 85.— Ivory (2 diams. example from the Hauran, which is here illustrated (Fig. 83 ", I) for the sake of comparison. The Hauran seal is engraved with curious characters. Equally Oriental in their connexions are a class of conical and sub-conical seals, represented in the Phaestos deposit by four examples. Fig. 84 a, b is of steatite, and the interlaced design appears to be of Egyptian derivation. The other three are of ivoiy, a material which, as already ' \ 86«. 806. Fig. 86.— Ivoev (2 diams.). ''"'- --" ...v-j.-TO 87f'. 876. Fig. 87. — Ivoky (2 diams). noticed (p. 64 (333)), is not unfrequently used for seals of the same form from the coast of Syria. Fig. 8.5 appears to represent a rude figure of an eagle with two pellets on either side. The geometrical designs on Figs. 80, 87 recur on the Maeonian mould and a curious leaden object to be de- scribed below.^* Among other small stone objects are a yellow steatite •' See p. 58 (327), Fig. Wi. P. 65 (334). 5^ Sec i>. 132, 133. 109 THE HAGIOS ONUPHRIOS DEPOSIT. mo CE O , 'vjjT Fk;. S!» (Xatuial Size).— (/, Marl pie and Steatite ; //. Dark Steatite ; c. Roek- crystal ; d. Dark Steatite ; c. Limestone ; /. Steatite ; (/. Ivory ; /(. Dark Steatite ; i., j. Colli. pendant which recalls a similar class of objects from the early cist-graves of Amorgos (Fig. 88), and some cylindrical heads of marble and steatite. Other types of beads from this deposit of various materials, steatite, rock-crystal, variegated limestone, and gold, are given in Fig. 89 a — j. It is of course conceiv- able, considering the circumstances of the Hagios Onuphrius find, that some one or other of these smaller objects like the beads may have got down from a higher level. But wherever it is possible to judge of the date of relics contained in this deposit their prae-Mycenaean character is well marked and there is every probability that these minuter objects belong to the same period. Among the jewels are a quartz crystal worn as a pendant with a gold mounting (Fig. 90), a small spiral band of gold (Fig. 91), and the small gold objects represented in Figs. 92, 93, the application of which is indeterminate. The granu- lated work of Fig. 94', which may have served as the end of a pin, is interesting from its occur- rence on some of the jewelry found at Hissarlik.'' It recurs on Mycenaean gold-work. In some cases a bronze core has been plated over v^ith gold. This method is followed in the case of three curious objects, two of which are engraved in Figs. 95, 96, and again of Fig. 97, perliaps part of a hilt, and three perfected knobs (Fig. 98). The fluting of the thin gold plate with which this latter is covered somewhat recalls the spirally fluted silver beads of an Amorgos grave. The objects represented in Figs. 95, 96, of which another (96 lis) occurred wholly formed of steatite, may perhaps be compared with a perforated instru- ment of diorito found in the burnt city at Hissarlik.'^ Among the jewelry may b? perhaps classed a miniature figure of a beaked vase carefully wrought in variegated limestone (Fig. 99), and another, less well shaped, of black steatite. These objects seem to have been pendants serving as amulets. As an ornamental appendage vases of similar form adorn the top of a gold pin^ found in one of the three smaller treasures of Hissarlik. " Cr. especially tlie gold earring, Sclilicmann, lUos, \K -ISO, Fig. 840. ' lUvD, Fig. Sii", \k -HO. * Schlii'niaiiii, Fig. 850. \K 488, Fig. 83J, and cf. PRIMITIVE CRETAN AND AEGEAN CULTURE. Fic. 88.— Yellow Stea- tite (Natural Size). Fio. 90.— Ci;y.st.4L with Golu MoUMTlSG (Natural Size). f^^ Fig. 91.— Gold (Natural Size). Fio. 92.— Gold (Slightly Eularged). Fio. 93.— Gold (Slightly Enlai'ged). Fio. 94.— Gold (Natural Size). Ill THE HACxIOS ONUPHRIOS DEPOSIT. 95({. 956. Fig. 95.— Bronze, Gold-Pl.ited (Natural Size). 95c. 96a. 966. Fig. 96. — BiioxzE, GoLD-PL.iXED (Xatiiral Size). Fig. 96 bis. — Ste.vtite. (Natural Size). Fig. 97. — Biionze, Gold-Plateo (Natural Size). Fig. 98. — Bkoxze, Giii,i)-ri,.\TKi) (Natural Size). I'KLMITIVK CItKTAN AND AE(!EAX CUJ/rURE. 112 Small gold ornaiiionts of tliu same ocnoclioe-like shape have come to light in Mycenaean sepulchral deposits at Mycenae itself, Menidi, Dimeni, and again at Arne;" in these cases however the spout is horizontal, whereas in the Trojan and Phaestian examples it slopes upwards. The religious signifi- cance of this type of vessel is shown by their appearance in the hands of the mysterious daemons of Mycenaean times. pX f_j Fic;. 99.— A'akiegated Limestone (Slightly Eulai-ged). Early types of pottery from the Hagios Onuphrios find are represented in Figs. 100, 102, 103. Fig. 100 is of a dark blackish brown colour with per- forated handles for suspension and a cover with four additional handles. The cover is almost identical with one found in the First City of Troy,"'' and the whole type of vessel, with the perforated, ear-like handles, answers to those of the earliest strata of Hissarlik. Fig. 101, from an early cist- grave at Arvi (see below), is added for the sake of comparison. It too is of the same dark bucchero, but of finer fabric, and it greatly resembles a vase from the earliest settlement at Tiryns.'"^ Fig. 102 is a cover of another hanging pot of the same bucchero ware. Fig. 103 is a small spouted vessel of the same dark paste, and Fig. 104 is a small reddish brown " In the Central Museum at Athens. "' Sehlieniann, Trojniiischc AUcrtlnimcr, Atlas, Taf. 21, 11 ni. llios, \\ 215, No. 27 (upside down). "'' ScMiemauu, Tinjns, p. 58, No. 1. 11:3 THE l£A(;io8 0NUPIIRI08 DEPOSIT. vessel with four hamUes, two of tlicm for suspension with double vertical perforations. Tlic iloublc-horued object of brown bucchero (Fig. 105) strongly recalls the horned handles {ansc cornuic) so characteristic of the Fig. 101.— Aiivi (5 diaiu.), Italian icrrcmarc. It is doubtless an attachment of a pot and from tlie two lioles below seems to have been intended to resemble an o.x's head. Its nearest parallel is the fragment of a vase representing a horned head of the same kind found in the early cemetery of Agia Paraskeve in Cyprus. 7 Fiii. 102. — PiiAEsTuS (i Jiaiii.). l''ii.. 103.— I'liAi.MM.- (Sligliily Ucaucfd). Flc. 101.— rii.\E.vTu> (Sliyluly Kciliiccd). Of a more a_0L0 J'S G - (icy 1 lie ^'("/ IS a gtuyish black Fio. 108.— P.viNTKU V.\sii iSlialiily rcduorid ; Vi cent, higli). PRIMITIVE CRETAN AND AEGEAN CULTURE. IIG with terracotta-coloured stripes. In form and coloiirini,' it shows some affinity to the later class of vases in the Amorgos cemeteries. Fig. 107 is a kind of pyxis with white bands on a terracotta ground. Originally, no doubt, it was provided with a cover. The jar represented in Fig. 108 shows still greater advance in th(! art of colouring pottery ; it bears red and white stripes on a greyish black ground, and its tints agree with those of some of the vessels found in the Kamares cave, on the Southern side of Mt. Ida.i" Indeed, the Kamares pottery must be brought into very direct relation Avith Phaestos, within whose territory it probably lay. On the other hand, the style of colouring shows a distinct approach to that of the earliest vases from Thera and Therasia. The design on this jar, and to a certain degree its shape, seems to me to stand in a direct relation to a very beautiful type of stone vase (B'ig. 123) which was in vogue in prehistoric Crete. Firi. 109.— Phaestos ; Limestone (Natural Size). Stone vases play a very important part in the early remains of Crete, as also in the contemporary deposits of Naxos, Amorgos, and other early Aegean sites. A small limestone vase from the Hagios Ouuphrios deposit is given in Fig. 109, together with its lid of the same material. Except that its cover is not provided with a knob at top, this vase bears a close resemblance to one (Fig. IID) obtained by me from a prehistoric cemetery at Arvi, the ancient Arbi, on the South-Eastern coast of the island. The variegated limestone vase, Fig. 110, now in the Ashmolean Museum, was found, like Fig. 100, near Phaestos, and probably belongs to the same early necrojiolis. These stone vases form such a characteristic feature in early Cretan deposits, and seem to afford in certain cases such a definite chronological clue, that a fuller account of those that I was able to meet with in the course of my recent explorations may not be amiss. They differ from the stone vessels 1" See above, pp. 79 (348), 81 (350) and note. 117 THE HAGIOS ONUPHRIOS DEPOSIT. of the 'Amorgan Period,' sucli as arc usually found in tlic prae-Mycenaean deposits of the Aegean islands, in two particulars. The stone-ware .such as is discovered for instance in the early tombs of Naxos and Amorgos is generally of white, apparently Parian, marble. The Cretan vessels are of far more varied materials. The other respect in which they differ from the kindred Aegean class is that they show a much greater conformity with certain types of Egyptian ^^ and, possibly too, of Libyan stone vessels.^' Fii!. 110.— riiAESTOs ; Vauiegated Limestone (J diaiii.). Massive pots of serpentine and diorite supported by pedestals of limestone or baked clay, forming incense altars, appear in Egyptian tombs from the time of the Fourth Dynasty (Fig. Ill), and several of these dating from tiie Fourth to the Sixth Dynasties are preserved in the Glzeh Museum. They are often provided with j)erforated horizontal handles, and the rim at top is generally broad and flat. A dark stone vessel (Fig. 112) found at Pinies above Ehinta, the ancient Olous, bears, as will be seen by a comparison with Fig. Ill, a very close resemblance to this archaic Egyptian class. A plainer type (Fig. 113) without handles was procured by me at Goulus, where larger vessels formed of a kind of conglomerate may still be seen beside some of the ancient cisterns Some of these were noticed by Spratt.^^ Another prolific find-spot of this early Cretan stone-ware is Arvi, the site of the ancient Arbi, on the South-Eastern coast, where there existed an early cemetery of the same period as the sepulchral deposit of Piiaestos. Figs. 11+, 11-5, IIG, 117, and the clay suspensioTi vase (Fig. 101), were described as iiaving been found here on either si FiCr. 114.— Cist Giiave, Aiivi ; Steatite (?, diaiii. Fii:. 115. — Cist Ukave, Ai:vi ; Steatite (^ diam.). \*/ Fifi. 116.— Cist Grave, Ar.vi ; Steatite (J diaiu. V iL'l THE HAGTOS ONUPHRTOS DEPOSIT. 3% * 2^k ■■' Fio. 117.— Cist GiLwr:, Ar.vi ; Stf..\titf. (1 iliam.). Fin. lis. — Anvi; LiMF.SToyr, {%, ili.iiii.). Fig. 119.— Aiivi ; liANnr.n Limi>tcixe (1 iliam.), PRIMITIVE CRETAN AND AEGEAN CULTURE. 12: known, into Mycenaean times, but these display a still more advanced technique, and spiral and volute flutings and figures in relief. Fragments of this Mvcenaean class have been found at Knosos. Fig. 120. — SiEATrrE Lid, Twelfth Uvxasty Deposit, Kaiiun, Eotpt (-3 iliam.). The vases are for the most part of steatite, in many cases approaching that made use of for the early Cretan gems and seal-stones. Where this was Fig. 121.— Ca\e, Psychuo ; Steatite (§ diam.). procured is as yet a mystery, but from its constant employment it seems probable that it exists in great masses in the island. The material of Fig. 123 THE HAGIOS ONUPHRTOS DEPOSIT. 112 is less easily definable ; its lianler texture aud the porphyritic crystals that it contains are both notowortliy. The banded limestone of Fig. 110 seems to be foreign to Crete. Fin. 122.— C'liEr.soN'ftsos ; Limestonk C'oxglojier.vte (s diam.). It is evident with regard to the technique that in most cases the vases have been turned, and it looks as if in Crete the use of the wheel for stone vases may have preceded its application to clay. It is a significant fact that the clay vessels found with the stone pots (Figs. 114, ll.i, 110, Fii.. 12:1. — .Mii.Ai'o; SrE.MTrn (;; ili.-iin.). 117) from Arvi and tliat from Milato were hand-made, though in the former case at any rate the stone vases from tlie same grave had the appearance of having been turned. Tlie sfnne vases were furtlior finislicd by scraping (ir chiscllinfr. PRIMITIVE OI'.KTAN AND AE(iI':AN CLTLTUKE. liil I mil iiiilobtud to Mr. J. L. Myros lor the following detailed notes on the material and fabric of the above stone vases. Fit;, ll'i. ISlack matrix, willi dull brown ijatulics, wliicli api.i'ar to b(^ .slightly softfr, ami iivimovoiis wuU-ilcliiicd lon<< pmiiliyritic crystals of yellowish wliitu colour. Harder than the; otlicr vuswi but scratched without dilficulty with the knife, and slightly absorbent to the touch in spite of the coiniaratively high polish. (Pinies, Elunta.) Fi". 113. Steatite, greyish white with indistinct lighter and softer streaks. Sjilit by heat on one side. P.artly turned, partly .'craped. (Gonliis.) Fig 114. Steatite ; blackish brown, with wavy structure and white patches which are softer ; turned on tho lathe. (Avvi.) • Fig. 115. Steatite ; greyish with indistinct light coloured patches rather harder but less dense than Fi"s. 114, 116. The lid with this is of steatite ; black compact ground with distinct white patches which are softer. It is partly turned, partly scraped or ground. (Arvi.) Fig. 116. Steatite ; black with irregular white patches whicli are softer. Turned on the lathe and finished outside with a knife or a chisel. (Arvi.) Fig. 117. Steatite; black compact ground with rather rarer white patches. Partly turned, partly rubbed. (Arvi.) Figs. 111-117 seem to he of the .same origin, though their ipialities vary slightly. [They are from the same group of early tombs at Arvi.] Fig. 118. Limestone ; compact, greyish : mottled with white calcspar ; not unlike specimens of common cretaceous limestone of the Levant. Turned. (Arvi.) Fig. 119. Limestone ; compact and slightly argillaceous, finely banded in grey and cream wliite. (Arvi.) This material is not known in Crete, nor likely to occur there. The steatite lid with this did not originally belong to it. Its material much resembles that of Fig. 113. It is partly turned, partly rulibed. Fig. 120. Steatite ; lilackish brown with reddish patches which ai'c softer but less 'weathered' than Figs. 114-117. Partly turned, partly rubbed, and with a good surface polish. (From Kahun, Eg.yi't-) Fig. 121. Steatite; compact, brownish and very soft, much damaged. (Psychro Cave.) Fig. 122. Limestone breccia ; white, pink and yellow, red matrix. Not known in Crete. (Chcrsonesos.) Fig. 123. Steatite ; greyish, with indistinct lighter and softer patches and streaks ; very like Fig. 113. No traces of turning ; cut and .scraped. (Milato. ) Figs. 113 and 121 and the lid of Fig. 118 very closely resemble one another, and might very well be from the same mass. Among the stone objects discovered at Hagios Onuphrios the marble 'idols,' of which sketches are given below in Figs. 124 — 132 ^^, are of essentially the same class as those found in Amorgos and otlier Greek islands. The series here brought to linht bears witness to a degree of evolution of form which seems to indicate the lapse of a considerable period of time. Figs. 124 — 12G may be compared with the simplest Trojan examples. In Fig. 131, on the other hand, we have a more advanced type, in which the arms are clearly drawn laid across the body just below the breasts, and the breasts themselves are in full relief. In Figs. 127 — 130 we have vaiious intermediate types. Fig. 134 is a head alone with two perforations in the neck for attachment. To these Phaestian specimens I am able to add two other Cretan examples said to have been found in the Province of Siteia (Figs. 133, 134). ^' The small image represented in Fig. 127 is probably belongs to the same deposit, and in now in the Ashmolean Museum. It was ac- any case illustrates the same period of local 'luired by me at Candia and was stated by its culture, possessor to have been found at Phaestos. It 125 THE llAGiUS U^■L■i'lll;iU.S DEPOSIT. Fig. 1i!4 (.Sli-litly Enlarged). Fio. 12(3 (Natural Size). 127(1 Fil;. 125 (Xatural Size). Fk;. 127 (Xatuial Size). Fl(i. 12S {XaUival Size), pKiisir nVK CUlvTAX AND Al^iEAN CLLTlIlK l:iO Fig. 129 (Natural Size). Yw. 132 (SligUtly Eulargea). Fii:. 130 (I liuwi). Fig. 131 {| Huc-ai). 127 THE ll.\(;[()S OXUPHRIUS ])EP()SIT. Like the others, they iire ut' Parian marble, so tliat the material at least must have been imported. An exhaustive discussion of the various questions raised by these curious figures would require a separate dissertation. Here it must be sufficient to remark that the theory according to which we have simply to deal with degenerate copies from Chaldaean prototypes representing Istar or the Mother Goddess does not accord with the evidence at present before us. The simpler forms of these Aegean figures are so lacking in detail as to afford no definite points of comparison with the Asiatic types in ([uestion. On the other hand, if we turn to the West and North, we find a whole series of early images of clay, stone, and other materials which certainly seem to fit on to these Aegean forms. From the remains of the early settlements of Troy, we know that simple forms of this class of figures occur indifferently in marble, clay, and bone. Alabaster and clay figures of the same class are scattered through the Thracian lands and beyond the Danube as far afield as Rou- manian^ and the valley of the Maros in Transylvania. But beyond the Carpathians again there appears another parallel class of primitive figures which must perhaps be regarded as the most characteristic product of a vast Neolithic Province including a large part of Poland, East Prussia, and Western Russia. Stalagmite figures of this kind have been found in the Polish Caves,-" on the East-Prussian coast they recur in amber,"^ and a bone figure of the same kind has been found by Inostranzeff in the remains of a Neolithic station on the shore of Lake Ladoga.-- These Northern figures do not exhibit indeed any marked indications of sex. On one amber example, however, the body is marked by an imperfect triangular outline which resembles the representation of the vulva on some of the Trojan or Aegean types. The double perforation at the junction of the neck and body which characterizes some of the Baltic and Russian examples recalls the perforations on the neck of the 'idols' from Phaestos and Si tela represented in Figs. 133, 13-1, and that of the head by itself in Fig. 132.-^ The holes and grooves on .some of the Baltic forms suggest attachment to other objects, and a marble figure from an Amorgan cist (Fig. 13-5) shows lines upon it which seem to indicate the same application.-''' It would be unwiise to insist too much on these resemblances in detail, but taken in connexion with the appearance of this parallel class of Northern figures they can hardly be without some significance. '" Priiiiitivu clay statuettos have been found beig, 1882). at Cucuteiii in Roum.ania {Anlii/iia, 1890, Taf. "- Tisclilcjr, up. cil. 118 (30) Fig. 10. V. 2). C'f. Bulletin dr la S'ocictv d'Aniliropologic, -^ On a marble idol from Amorgos (Aslimolean 1889, p. 582, and S. Reinacli, Anlliropuhyic, Coll.) the thighs are bored. 1894, ]>. 293. -"' Now in the Ashmolcau Museum. 'I'lii.s ''" Ossowski iv. Zbiur Wiadomosci do Aiitrn- figure shows signs of jiainting on the right .side [lulogii Krajowej. (Third report on Tolisli of the neck, apparently representing a pendant Caves, (ip. cit. (1881) I. vi. ]>p. 28-.'>l, I'l. iii.- loek of hair. This is the same figure as that V.) JIateriaux, 1882, pp. 1 -Gl, I'l, I. II. sketched by Dr. Wolters (Ath. MiU/i. 1891 Tischler, Sle.inzcil in Ostpnnsscn, pp. 96, 97. p. -19 Fig. 3) and is from Grave D described by '-' Tischler, StcinseU in Ostprcusscn, p. 2."i (g), Dr. Diimmlcr {Alh. Mitlh. 188C, p. 15 sc/'y.). Figs. 6, 7 {Schri/Un d. phji.-ocl: Ocs., Kiinigs- PRIMITIVE CRETAN AND AEGEAN CULTURE. 12« Fiii. 133.— SiTEiA (§ liiiuar\ Fig. 134.— SiTEiA (5 linear). 129 THE HAGIOS ONUPHRIOS DEPOSIT. The Aegean aud Northern group together occupy in fact a continuous zone roughly divided by the Carpathians. It woukl even appear that this zone had a Libyan extension. A chxy female figui'e acquired by Mr. Petrie ^* from Abusir, near Sameineh, in the middle of the Delta, must clearly be regarded as a somewhat developed offshoot from the same primitive family. The lower part of this 'idol' resembles the Greek island figures, but the side-lock on the head gives this foreign relic from Egyptian soil a typically Libyan aspect. Small images in a squatting posture were also found by Mr. Petrie in his recently discovered settlement of an unknown, probably Fio. 13.'). —Cist Grave, Amuugo.s (Natiiiul Size). Libyan, race on the Upper Nile. These recall a class of squatting, obese female figures that may be traced from Thrace to Attica and the Peloponnese and which in turn curiously resemble the so called ' Cabiri ' found in the prehistoric building known as Hagiar Kim in Malta,-'"' — a structure the ^* See Illahun, Kalmn and Guroh, PI. xix. 43, and p. 19. It might also bo an interesting question how far the Egyptian wooden dolls ending above in a rectangular stump, to which the bcadstrings are tied to represent hair, m.ay go back to [iriniitive types of ' idols ' resembling the Aegean. The jiarnllelism is sometimes striking. -^^ Caruana Itcport on the. rhocnlcinn Jkc. ytnliijuilici of MiiUa, pp. 30, 31 and jihoto- graph : Furse, Prehistoric Conflrc.i.i, Norwich, p. 412 and I'l. ii. I'errot ot Chipiez, T. iii. Phdiiicic, &c., p. 305, Figs. 230, 231, where tl;o Ilagiar Kim itself, this counterpart of Talyots and chambered barrows, reappears as a ' Phoe- nician Temple.' I'KIMITIVK CHKTAN ANI> AKIJKAN CLMynKK. l.iO affinities of wliicli point in almost ciiual degreos to North Africa and tlic Iberian West. Proceeding westwards we find tlic X"li'''f lines on the bodies of some Trojan figures'-''' recurring in a more decorative form on the female clay ' idol ' from the Laibach pile-settlement, and on another primitive image from the prehistoric Station at Butmir in Bosnia.-'''' Not to speak of some general re- semblances presented by certain rude clay images found in the Swiss Lake dwellings and Italian terrcmare, recent discoveries on the Tyrrhene and still more westerly Mediterranean shores have supplied parallels of a very inter- esting kind. In the Finale Caves on the old Ligurian coast have been found clay figures -'' of ' Aeneolithic ' date approaching those of His.sarlik and perhaps atfording the nearest prototypes to Mycenaean forms. In Spain again the Brothers Siret have found figures of schist and bone in Neolithic and early Bronze Age deposits'-"^ which seem to stand in a direct relationship to the Aegean ' idols.' A whale-bone figure was found with Neolithic relics at Skara in Orkney, and images with the attributes of maternity strongly emphasized wore already known to the Eurojiean pojDulation of the Reindeer Period.^''^ In view of these widely ramifying and deeply-rooted European con- nexions'-""-' it seems in the highest degree unsafe to as.sume that the earliest Aegean ' idols ' of the present class are nothing more than degenerate copies of early Chaldacau images. Rather it is reasonable to suppose that the wide- spread primitive custom may have had a more easterly extension as far as the valley of the Euphrates. From the frequent association of these images with interments alike in the Aegean islands, the Ligurian caves, and South-Eastern Spain, we are led to infer that they had some connexion with ideas relating to the Nether World. This view is in fact supported by parallels from remote ^^ E.g. Sohliemaun, Troy, p. 331, Fig. 193. ="'' For the ' Venus of Brassempouy , ' and •'"^ Radinisky und Hoernes, Dio Ncolilhische other ivory figurines from the same snh-Pyren- Stalion von llutmir bci Sarajevo in Bosnicn, acan grotto, see AntkrojMlofjic, T. vi. (1S95), Taf. ii. 2. p. 141 sqq. '^'' Sec A. Issel, Ligwia Gcologicn c Prchis- "''= Since this was written 1 liave liad an op- torica Tav. xxviii. Figs. 11 and 14, and my portuuity of perusing M. Salomon Rcinach's ' Prehistoric Interments of the Balzi-Rossi Caves articles entitled ' La Sculpture en Europe avant and their relation to the Cave-Burials of the les influences greco-romaines ' (Anthrojioloijic, Fiualese' (.-htWw. Inst. Journ., 1893 p. 306 and 1894, 15-34, 173-186, 288-305; 1895, 18-39, note), where attention is called to the fact that 293-311). In these M. Reinach, like myself, one of these primitive image:i is painted and lays stress on the parallelism presented by the belongs to the Ligurian class of Neolithic painted Trojan and Aegean forms of primitive images pottery. (Cf. G. B. Amcrano, Dei Fast colorati with those of Spain, the Dauubiau regions, c dipinli delle Caverne di Finale.) These vases and the Amber Coast of the Baltic. He also are the Ligurian counterparts of tlic prae- maintains that these European forms were Mycenaean class of Sikel (or Sikan) painted evolved from the rudest and simplest prototypes ware. A somewhat more advanced painted rattier than that they degenerated from higher fignre was found in a neolithic 'cave' tomb, models: 'Etant donue un pilier on unc tablette Villafrati near Palermo. quadrangiilairc, on pouvait d'al.iord, pour sug- -■■ Premiers Ages de Metal dans le Sud-Est de gerer I'idee de la forme liumaine, amincir cer- VEsparjnc, pp. 32, 57, 257, and Plates vi. and taines parties, telle (pie la taille et le cou, de ix. ; La Jin de fepoijue neolithiijue en Espagnc, manierc a faire saillir les epaules et la tete ' ju Anthropologic, 1892, pp. 387, 399. {op. eit. 1894, p. 291). 131 THE H AGIOS ONUPHRIOS DEPOSIT, parts of the globe. In Japan there is actually a historic record of the sub- stitution in place of the slaves and retainers who sacrificed themselves to their deceased lord of small figures deposited in his grave. The Ushabtis or ' Kepondants' of the Egyptian tombs are said to have had a similar object, and it was their function to act as substitutes for the dead person when his turn came to work in the fields below. In ancient Mexican graves terracotta heads are of frequent occurrence, which seem to have been attached to figures of more perishable material ^^ and to have represented the wives or slaves of the departed. It is probable that the Phaestos head with two holes in the neck was also made for attachment to a body of some less lasting substance. What we have to deal with then is an 'Aegean' version of a primitive funereal custom which both in the Western Mediterranean basin and in the Danubiaii and Baltic lands seems to have had a wide European extension over a continuous area. In tlie Egyptian Ushabtis we have perhaps a conven- tionalized type of what may have originally been a southern offshoot of one archaic family. The occurrence of clay ' idols ' of the same general character in Cypriote ' Copper Age ' cemeteries like that of Alambra shows an Asiatic extension of the same custom, and we are thus led to the curious nude figures seen on early Chaldaean cylinders.-" These nude images have in this case been plausibly connected with the legend of the Goddess Istar, who, in order to pass through the seven gates of ' tlie Unchanging Land without return ' — there to procure the Waters of Life for her 'wounded Thammuz' — was forced to strip herself one by one of her robes and jewels till she went in at last mother-naked. It is certain that a direct piece of evidence connects them with her double — the Goddess Sala. Although no similar Chaldaean images of clay and stone are known of this early date, the absolute correspondence in type presented by Asiatic clay figures of a much later period permits us to suppose that more archaic examples will some day be brought to light. The connexion with the visit of the Mother Goddess to the abode of Death is just such a mythic outgrowth from the primitive custom of burying the naked image of a wife or mother with the departed as might have been expected. It was also inevitable — admitting such a mythic superstructure — that the Eastern family of such funereal images should afterwards undergo a religious transformation and be identified with or assimilated to Istar or some one or other of her Asiatic equivalents. The Syrian influence, resulting in. a more sensuous type of female image, with the organs of maternity strongly emphasized, undoubtedly spread through Anatolia, and early left its mark on the clay figures of the Cypriote graves. It may however be laid down as an absolute rule that the earlier the image the less trace there is of any such Asiatic '^ E. B. Tylor, Anahuac, p. 229. on one cylinder this ami a mule male figure that ^" See, for examples, Mcnant, Glypliqxtc often occurs with it are iJentilied hy the inscrip- On'cnfafci. p. 172, pp. 173-17.') ; Figs. 110-116. tion with Sala and Kamanu, in many respects Menant regarded the connexion with Istar as in reduplicate forms of Istar and Tammuz. these cases 'not proven,' Nikolsky, however Hamauu is the Syrian Rimmon, (Hei'. Arch. 1891, ii. p. 41), has now shown that- Primitive Cretan and aegean culturk. i.-52 iuflucnce. The figures in the earliest deposits of Hissarlik — those of tlie First City — are absolutely primitive and the most removed from all suggestion of these supposed Chaldaean prototypes. The existence of a continuous group of primitive 'idols' on European soil going back to Neolithic times, and extending from Crete to the shores of Lake Ladoga in one direction and to the Pillars of Hercules in the other, must in the absence of very direct evidence to the contrary be regarded as an independent phenomenon. So far as existing evidence goes, at the time when the Istar model first reached the Aegean shores their inhabitants were already in the age of metals, and it appears as an intrusive form beside the more primitive idols which they had handed down from Neolithic times. A leaden female image found by Dr. Schliemann in the second city of Hissarlik 2" clearly betrays its oriental parentage. The swastika engraved on the vulva is also evidently a stamp of godhead. This figure in turn finds its parallel in one of a pair of male and female divinities that appear on a serpentine mould, now in the Louvre, found at Seleudj, East of Thyatira in the ancient Maeonia^^ (Fig. 135), and these figures, as M. Salomon Reinach has shown,^^ take us back again to another Asiatic mould of the same material in which a God and Goddess are once more represented side by side. In this latter example the God with the horned headpiece evidently stands for a form of Bel, while his female companion, though in this case her lower limbs are draped after the flounced Babylonian fashion,^^ bears on her head a curious 'rayed half circle, which sufficiently betrays her identity. It is, in fact, the upper part of that speci.al variety of the radiate disk which in Chaldean symbolism indicates the star of Istar. The Western influence of the Babylonian type would find a curious illustration if we might accept the genuineness of a lead figure said to have been found with another lead object exhibiting cruciform ornaments near Candia. These objects were obtained in 1889 by Mr. Greville Chester, and are now in the Ashmolean Collection (Fig. 13G). But both the figure and the ornaments are almost line for line identical with the female divinity and two of the engraved objects that appear on the Selendj mould. It almost looks as if they had been actually cast in this individual mould, and if their claim to antiquity is to be allowed it would result tliat these leaden objects were imported into Crete from Maeonia in prehistoric times. The figure has the appearance of great age, but it is possible that some Levantine dealer may have profited by the existence of the mould to cast some lead figures from it. The fact that the square ornament is broken off at the same point ™ Ilios, p. 337, Fig. 226. The Babylonian produced), and cf. Penot et Cliipiez v. p. 300 parentage of this figure was clearly pointed out Fig. 209. by Sayce in his Preface to Schliemann's Troja, '- Op. cit. p. 46. pp. xviii., xix. He identifies the image with ^'' The draped lower limbs bring us nearer to the Trojan Ate and 'Athi, the Great Goddess of Mycenaean types. Compare especially the im- Carchemish. He thus traces the type to pressed glass figures of a female divinity from Chaldaea through Hittite mediation. Tomb II. of the lower city, Jlycenae. Tsountas, '" S. UeinSiCh, Esquisses ArchMogiques {ISSS), 'Ai-arrKoi^o! Ta'., on Mycenaean animal worship, [339 ».], 70, [340 n.\ 71. Cretans deny that Phoenicians invented letters, [372], 103. Curium, Mycenaean signs at, 137. Ci/liu(ler, clay, from H. Oniiphrios, 105. d/priote syUaharij, signs of, compared with ''Aegean' linear forms, [348], 79, [349], 80, [352-354], 83-85. compared with iSemitic letters, [366], 97, [367], 98. Cjjretie, button-seals found in, [335], 66. Dijrpfeld, Dr., on the Mycenaean city of Troy, [326], 57. on chronology of Trojan strata, [326], 57. Dare, pictograph, [310], 41. Diiiumler, Dr. F , on early cist-graves of Amorgos, [325], 56. on seal from Kuphonisi, &c. [329 «.], 60. Durirlecimal si/stem, evidences of, on early seal-stones, [342], 73, [343], 74. E(j)ipt'tan stone vases compared with Cretan, 117 seqq. Elcutheriia, form of Van at, [360], 91. Erii.v, Phoenician letters on walls of, [282], 13. Eteolretes, Cretan aborigines, their sur- vival in east of the island, [.354], 85. inscription in language of, found at Praesos, [354], 85, [355], 86. Eijc, pictograph, [303], 34. Excliel, LXX. version of, translates 'Chere- thites' by Kp^Tcs, [368], 99. Dagf/er, pictograph, [305], 36. bronze, from H. Onuphrios deposit, 1.35. Dawhias, probably Danai, [369], 100. connected with Takkaras (Teueri) and Pulasati, [369], 100. invade Egypt under Thothmes III., [368], 99. Delphi, bronze axe from, with symbol, [280], 11. De Hoiuic, his theory on origin of the alphabet, [366], 97. identifies Pulasati and Pelasgians, [369 «.], 100. Detilcalion, a son of Minos, [358], 89. Dildi/nna, relation of to Atargatis, [371], 102. Di'odiiros gives Cretan tradition as to inven- tion of alphaliet, [372 and it.], 103. Dolls, Egyptian, compared with primitive 'idol.s,' 129 «. Doi,r, pictograph, [307], 38, 137. Dnre or Dor, on Canaanite coast, city of Takkaras, [.•569], 100. in (Jrcek legend founded by Diiros, [3()'.i], 100. its inhabitants AaipuU, [369], 100. Dorians, pussiblc settlement of, on Canaanite coast, [369], 100. in Crete, mention of, in Odyssey, [356], 87. ' long year ' of, ai>i>lied to Minos, [3.')6 and ».], 87. setllenn'nt of, from Tliessalian Doris, [357], 88, [358], 89. Minos adopted by tlie son of thiMr leader, [.357], 88, [358], 89. Fahrifius, Dr., on Tjrehistoric building at JvnCsos, [281], 12. ' Fence, pictograpli, [307], 38. Finalniarina, Liguria, primitive 'idols' from, 130. painted ' idol ' from, nearest prototype of Mycenaean, 130. pictographic sculptures near, [352 )(.], 83.'' Fish, pictograph, [308, 309], 39, 40. Furtwiingler, Dr., supplies impressions of pictographic seals, [275], 6. ■yaXoTTfTpar or ' milkstones ' worn by Cretan women, [276], 7. G-'^ «■;• ^m ^m:^ 'i.'i^- V .- "*■- -'■'':'- ■■*■ ' A tr %■ . ■,,-- .' /V. Ml *.'''■•■ \' *-'.'' ►■.■ v--' .'.\ V".,. ■-.-■■' ^i? yr-r^^-V^'.;^^ v^-,'r-^, ■:#A^.^^t^ ^^v'-V'.^^i' ^■K . ■jt« ^^^f , , , ' ^- n^) fSlftSi ^^■.■^^s ■''\s''' ■'■■.'■.' ',i. !}.'. .',, ■ ;.'T*, ' "'- ',0 \ '■*"'. ^:^,.:v^•^V' V:.^, -;'■'<; ^'^?^:^i ■Bir- i'-t' "' ', '■*■,(.*!, I , -.^ V i - !■ ■ •, ; t- U?^#-' .gS-:v^; ri^r: :i»^ L*-^^^: ■m 'iMf'-i^ •;*. m^'ti^ ■J&ii^ KSy >;?■! ^A*-*^;: f«f, f*-;