A very striking proof of the independent development of Elamite culture prior to the Semitic conquest is now furnished by the texts inscribed in the so-called "proto-Elamite" system of writing.[29] The majority consist of small roughly-formed tablets of clay, and the signs upon them are either figures or ideographs for various objects. Though they have not been fully deciphered, it is clear that they are tablets of accounts and inventories. A very few of the signs, such as those for "tablet" and "total," resemble the corresponding Babylonian characters, but the great majority are entirely different and have been evolved on a system of their own. Lapidary forms of the characters have been found in inscriptions accompanying Semitic texts of Basha-Shushinak;[30] and, from the position of each upon the stone, it was inferred that the Semitic text was engraved first and the proto-Elamite section added to it. That they were contemporary additions seemed probable, and this has now been put beyond a doubt by the discovery at Susa of a stone statuette seated upon a throne, which was dedicated to a goddess by Basha-Shushinak.[31] On the front of the throne at each side of the seated figure is an inscription; that on the left side is in Semitic, and that on the right in proto-Elamite characters. The one is obviously a translation of the other, and their symmetrical arrangement leaves no doubt that they were inscribed at the same time.
It is therefore clear that at the time of Basha-Shushinak the two languages and scripts were sometimes employed side by side for votive inscriptions, while the clay tablets prove that the native script had not yet been superseded for the purposes of everyday life. The "proto-Elamite" characters present very few parallelisms to Babylonian signs, and those that do occur are clearly later accretions. Thus it would be natural enough to borrow the Babylonian sign for "tablet," at a time when the clay tablet itself found its way across the border; and, though the signs for "total" correspond, the Elamite figures differ and are based on a decimal, not on a sexigesimal system of numeration. It may therefore be inferred that the writing had no connection in its origin with that of the Sumerians, and was invented independently of the system employed during the earliest periods in Babylonia. It may have been merely a local form of writing and not in general use throughout the whole of Elam, but its existence makes it probable that the district in which Susa was situated was not subject to any strong influence from Babylonia in the age preceding the Semitic expansion. This inference is strengthened by a study of the seal-impressions upon many of the tablets;[32] the designs consist of figured representations of animals and composite monsters, and their treatment is totally different to that found on early Sumerian cylinders. In the total disappearance of its local script Cappadocia offers an interesting parallel to Elam. The Hittite hieroglyphs were obviously of purely native origin, but they did not survive the introduction of the clay tablet and of cuneiform characters.
The earlier strata of the mounds at Susa, which date from the prehistoric periods in the city's history, have proved to be in some confusion as revealed by the French excavations; but an explanation has recently been forthcoming of many of the discrepancies in level that have previously been noted.[33] It would seem that the northern and southern extremities of the Citadel Tell were the most ancient sites of habitation, and that from this cause two small hills were formed which persisted during the earlier periods of the city's history. In course of time the ground between them was occupied and was gradually filled in so that the earlier contour of the mound was lost. It thus happens that while remains of the Kassite period are found in the centre of the tell at a depth of from fifteen to twenty metres, they occur at the two extremities in strata not more than ten metres below the surface. Even so, the later of the two prehistoric strata at the extremities of the mound, representing an epoch anterior to that of the "proto-Elamite" inscriptions, contains only scattered objects, and it is still difficult to trace the gradual evolution of culture which took place in this and in the still earlier period. It should also be noted that the presence of a single stratum, enclosing remains of a purely Neolithic period, has not yet been established at Susa. There is little doubt, however, that such a stratum at one time existed, for stone axes, arrow-heads, knives and scrapers, representing a period of Neolithic culture, are found scattered at every level in the mound. It is thus possible that, in spite of the presence of metal in the same stratum, much of the earlier remains discovered at Susa, and particularly the earlier forms of painted pottery,[34] are to be assigned to a Neolithic settlement upon the site.
Fortunately for the study of the early ceramics of Elam, we have not to depend solely on the rather inconclusive data which the excavations at Susa have as yet furnished. Digging has also been carried out at a group of mounds, situated about ninety-three miles to the west of Susa, which form a striking feature on the caravan route to Kermanshah. The central and most important of the mounds is known as the Tepe Mussian, and its name is often employed as a general designation for the group. The excavations conducted there in the winter of 1902-3 have brought to light a series of painted wares, ranging in date from a purely Neolithic period to an age in which metal was already beginning to appear.[35] This wealth of material is valuable for comparison with the very similar pottery from Susa, and has furnished additional data for determining the cultural connections of the earlier inhabitants of the country. The designs upon the finer classes of painted ware, both at Susa and Mussian, are not only geometric in character, but include vegetable and animal forms. Some of the latter have been held to bear a certain likeness to designs which occur upon the later pottery of the predynastic age in Egypt, and it is mainly on the strength of such points of resemblance that M. de Morgan would trace a connection between the early cultures of the two countries.[36]
But quite apart from objections based on the great difference of technique, the absence of any pottery similar to the Egyptian in Babylonia and Northern Syria renders it difficult to accept the suggestion; and it is in other quarters that we may possibly recognize traces of a similar culture to that of the earlier age in Elam. The resemblance between the more geometric designs upon the Elamite pottery and that discovered at Kara-Uyuk in Cappadocia has been pointed out by Professor Sayce;[37] and Mr. Hall has recently compared them in detail with very similar potsherds discovered by the Pumpelly Expedition at Anau in Russian Turkestan,[38] and by Professor Garstang[39] at Sakjegeuzi in Syria.[40] It should be noted that, so far as Elam is concerned, the resemblance applies only to one class of the designs upon the early painted pottery, and does not include the animal and a majority of the vegetable motives. It is sufficiently striking, however, to point the direction in which we may look for further light upon the problem. Future excavations at Susa itself and on sites in Asia Minor will doubtless show how far we may press the suggested theory of an early cultural connection.
While such suggestions are still in a nebulous state, it would be rash to dogmatize on the relation of these prehistoric peoples to the Elamites of history. A study of the designs upon the Elamite potsherds makes it clear, however, that there was no sudden break between the cultures of the two periods. For many of the animal motives of a more conventionalized character are obviously derived from the peculiarly Elamite forms of composite monsters, which are reproduced in the seal-impressions upon "proto-Elamite" tablets.[41] Moreover, it is stated that among the decorative motives on pot-sherds recently discovered in the lowest stratum at Susa are a number of representations of a purely religious character.[42] It is possible that these will prove to be the ancestors of some of the sacred emblems which, after being developed on Elamite soil, reached Babylonia during the Kassite period.[43] How far Babylonia participated in the prehistoric culture of Elam it is difficult to say, since no Neolithic settlement has yet been identified in Sumer or Akkad. Moreover, the early Sumerian pottery discovered at Tello, which dates from an age when a knowledge of metal was already well advanced, does not appear to have resembled the prehistoric wares of Elam, either in composition or in design. It should be noted, however, that terra-cotta female figurines, of the well-known Babylonian type, occur in Elam and at Anau[44]; and it is possible that in Babylonia they were relics of a prehistoric culture. On sites in the alluvial portion of the country it is probable that few Neolithic remains have been preserved.[45] But it should be noted that fragments of painted pottery have been found at Kuyunjik, which bear a striking resemblance to the early Syro-Cappadocian ware;[46] and these may well belong to a Neolithic settlement upon the site of Nineveh.[47] It is thus possible that the prehistoric culture, which had its seat in Elam, will be found to have extended to Southern Assyria also, and to non-alluvial sites on the borders of the Babylonian plain.
It would seem that the influence of Sumerian culture during the historic period first began to be felt beyond the limits of Babylonia at the time of the Semitic expansion. The conquest of Syria by Shar-Gani-sharri undoubtedly had important results upon the spread of Babylonian culture. The record, which has been interpreted to mean that he went still further westward and crossed the Mediterranean to Cyprus, is now proved to have been due to the misunderstanding of a later scribe.[48] It is true that some seals have been found in Cyprus, which furnish evidence of Babylonian influence in the island, but they belong to a period considerably later than that of the Akkadian empire. Of these, the one said to have been found in the treasury of the temple at Curium by General di Cesnola refers to the deified Narâm-Sin,[49] but the style of its composition and its technique definitely prove that it is of Syro-Cappadocian workmanship, and does not date from a much earlier period than that of the First Dynasty of Babylon. The most cursory comparison of the seal with the clay-sealings of Narâm-Sin's period, which have been found at Tello,[50] will convince any one of this fact. The other, which was found in an early bronze age deposit at Agia Paraskevi with its original gold mounting, may be definitely dated in the period of the First Babylonian Dynasty,[51] and Nudubtum, its original owner, who styles himself a servant of the god Martu (Amurru), may well have been of Syrian or West Semitic origin. Beyond such isolated cylinders, there is, however, no trace of early Babylonian influence in Cyprus.[52] This is hardly compatible with the suggested Semitic occupation during Shar-Gani-sharri's reign; there may well have been a comparatively early trade connection with the island, but nothing more.
Yet the supposed conquest of Cyprus by Shar-Gani-sharri has led to the wildest comparisons between Aegean and Babylonian art. Not content with leaving him in Cyprus, Professor Winckler has dreamed of still further maritime expeditions on his part to Rhodes, Crete, and even to the mainland of Greece itself.[53] There is no warrant for such imaginings, and the archaeologist must be content to follow and not outrun his evidence. Babylonian influence would naturally be stronger in Cyprus than in Crete, but with neither have we evidence of strong or direct contact. There are, however, certain features of Aegean culture which may be traced to a Babylonian source, though some of the suggested comparisons are hardly convincing. The houses at Fâra, for instance, are supplied with a very elaborate system of drainage, and drains and culverts have been found in the pre-Sargonic stratum at Nippur, at Surghul, and at most early Sumerian sites where excavations have been carried out. These have been compared with the system of drainage and sanitation at Knossos.[54] It is true that no other parallel to the Cretan system can be cited in antiquity, but, as a matter of fact, the two systems are not very like, and in any case it would be difficult to trace a path by which so early a connection could have taken place. It has indeed been suggested that both Babylonia and Crete may have inherited elements of some prehistoric culture common to the eastern world, and that what looks like an instance of influence may really be one of common origin.[55] But, as in the case of a few parallels between early Egyptian and Elamite culture, it is far more probable that such isolated points of resemblance are merely due to coincidence.
A far more probable suggestion is that the clay tablet and stilus reached Crete from Babylonia.[56] Previous to its introduction the Minoan hieroglyphs, or pictographs, had been merely engraved on seal-stones, but with the adoption of the new material for writing they were employed for lists, inventories and the like, and these forms became more linear.[57] The fact that the cuneiform system of writing was not introduced along with the tablet, as happened in Anatolia, is sufficient proof that the connection between Babylonia and Crete was indirect. It was doubtless by way of Anatolia that the clay tablet travelled to Crete,[58] for the discoveries at Kara-Uyuk prove that, before the age of Hammurabi, both tablet and cuneiform writing had penetrated westward beyond the Taurus.[59] Through its introduction into Crete the Babylonian tablet may probably be regarded as the direct ancestor of the wax tablet and stilus of the Greeks and Romans.[60]
Unlike the clay tablet, the cylinder-seal never became a characteristic of the Aegean cultural area, where the seal continued to be of the stamp or button-form. A cylinder-seal has indeed been found in a larnax-burial at Palaikastro, on the east coast of Crete; and it is a true cylinder, perforated from end to end, and was intended to be rolled and not stamped upon the clay.[61] The designs upon it are purely Minoan, but the arrangement of the figures, which is quite un-Egyptian in character, is similar to that of the Mesopotamian cylinder.[62] In spite of the rarity of the type among Cretan seals, this single example from Palaikastro is suggestive of Babylonian influence, through the Syro-Cappadocian channel by which doubtless the clay tablet reached Crete.