Myths of Crete and Pre-Hellenic Europe, by Donald A. Mackenzie, [1917], at sacred-texts.com
Obsidian Finds in Troy--Early Shipping Traffic--Copper Age in Cyprus--Doubt about Crete--Transition from Stone to Bronze in Troy--Was Copper first worked in Egypt?--The Oldest Bronze Articles--Bronze manufactured in Crete--Probable Sources of Tin Supply--A Visit to Troy--Homeric Memories--The Nine Cities at Hissarlik--The First and Second Citadels of Troy--Hand-made and Wheel-made Pottery--Symbolic Decorations--Trojan Eye Symbol on Yorkshire Relic--The Mother--goddess--Treasure of Priam and a Cretan Hoard--Engravings of Ships with Sails--Cretan and Egyptian Jewellery--Silver Cup and Silver Bowls--Homeric References--Ægean Influence on Anatolian Coast--The Inland Hittite Power--Ethnics of Anatolia--Danubian Cultural Area--Troy's Connections with Thrace--Ancient Conflicts on Plain of Troy--Problem of the Jade Traffic--European Jade Objects not all imported--Crete and the European Trade Routes-- Distribution of the Developed Spiral.
THE influence of Ægean culture, which assumed its specific character in Crete, extended as far distant as Troad, that strip of north-western Anatolian coastland which came under the sway of the Trojans. "In the Early Minoan period 'Crete'", writes Mr. and Mrs. Hawes. 1 "was in contact with Egypt on the one hand and with Hissarlik (Troy) and the Cyclades on the other--pupil of the former, teacher of the latter." It is possible that Troy's earliest connection with Crete goes back to the Neolithic Period, for finds have been made in the stratum of the first city of flakes and small artifacts of obsidian. This highly-prized stone was probably carried over the sea from Melos rather than along an overland trade route from Sinai.
It would appear that there was a certain amount of regular shipping traffic on the Ægean Sea in Neolithic times. Crete, as we have seen, imported obsidian from Melos long before the introduction of metal working. The beginnings of the trade can be traced at Magasa, where the flakes were found to be associated with an extremely crude pottery of great antiquity, and it was well developed apparently during the later stage of Neolithic culture, to which the obsidian knives from Knossos are assigned. It is unlikely that Melos was uninhabited when obsidian was first worked there. Ultimately its people exchanged it for marble from Paros, which was utilized to shape rough amulets or figurines of the mother goddess. But, so far, except for the evidence afforded by these finds of obsidian, no other indications that the Cycladic islands were occupied during the Neolithic Age have been forthcoming. Stone weapons have, however, been found in southern Greece and on the large island of Euba. Some of these are so small that they seem to have been charms, or votive objects, rather than real weapons. The Ægean Neolithic folk were evidently a peaceful people, and it may be that island communities utilized wood freely for implements of daily use. Wooden hand ploughs and wooden bowls were used in the Scottish Hebrides until a comparatively recent date, and the Egyptian peasants carried staves to drive their herds, and found them sufficient for purposes of defence.
The early peoples who reached Crete probably came by way of the Cyclades, either from the Anatolian or Grecian coasts. Before they accomplished this feat, the art of navigation must have advanced considerably. If it is held, on the other hand, that they passed direct oversea from Cyprus or Libya, we must conclude that they were skilled mariners who possessed well-equipped vessels and
were quite capable of conducting a sea traffic from the very beginning. Perhaps when the Cretan inscriptions can be read some light will be thrown on this aspect of the problem.
Among the isles, Crete, with its long record of human activity, was ever prominent in promoting commercial intercourse, and as mercantile enterprise was the principal factor in its development, Troy was probably reached by its wind-bronzed and adventurous mariners, who, having familiarized themselves with the "swan ways" of the Cyclades, undertook the exploration of the eastern and western shores of the Ægean Sea, gaining knowledge of prominent landmarks like Mount Athos and the massive mountain ridge of Samothrace.
Traffic by the sea, as well as by the land routes, must have been greatly stimulated after the knowledge of how to work metals became widespread. Ships could then be constructed more stoutly and with greater celerity, and must consequently have increased in number. Pharaoh Sneferu's order for a new fleet of forty odd vessels to convey timber from Phnicia is an interesting example of the manner in which ambitious monarchs might strive for mercantile supremacy. No doubt it was in consequence of the growing competition that experienced seafarers made voyages of exploration and opened up new routes in all directions. Malta, as we have seen, received obsidian from Melos; it also imported jade, which probably came from Anatolia. Jade was carried as well to Sicily, and as the Cretans imported liparite from the Lipari islands, after they had established a connection with Egypt, it was probably by them that jade objects were distributed westward.
It is uncertain when Cyprus was first visited by the Cretan mariners. The Neolithic relics of that island are
SEA-TRADERS FROM CRETE,
From the Painting by John Duncan, A.R.S.A.
notably scanty, and some think it was not occupied prior to the age of metals, as it is devoid of Neolithic strata. No doubt the earliest Cypriotes, who settled in the eastern river valleys, came from the Syrian coast. Their pottery was hand-made, and ornamented with incised designs, and compares more closely to Anatolian than pre-Dynastic Egyptian or Cretan varieties. The island had its Copper Age, and towards the close of it wheel-made pottery was manufactured.
It is held by some authorities, including Myers and Hall, that copper was first worked in Cyprus. If such was the case, it is remarkable that the island has not yielded traces of early commercial connections with Crete and Egypt. "Up to the present," says Mosso, "there is no evidence that copper was worked in the Isle of Cyprus before it was used in Egypt and Crete. . . . The word Cyprus comes from the name of the plant κύπρο?<υ?>σ?/υ?> {Greek kúpros}, which is the henna (Lawsonia inermis), used for dyeing the nails red." 1 Cypriote copper blades are of later date than those found in Crete, and the earliest flat axe of copper is of Egyptian Neolithic form. 2
There can be no doubt that Cyprus had a Copper Age before the Age of Bronze. The same cannot be said with certainty, however, regarding Crete. Copper weapons have been found in tombs, but they are small and of votive character, and the larger ones, of which they were copies, were perhaps of bronze. The few copper dagger blades that have been unearthed are difficult to place, and the view has been urged that bronze is as old in Crete as copper. The island of Minos "shows",
[paragraph continues] Mrs. Hawes says, "the same phenomenon as Hissarlik, 1 the sudden appearance of bronze at a date not later than 2500 B.C. On the evidence at present available no Copper Age can be predicated for the island. . . . The natural conclusion is that Crete knew nothing of copper until it knew tin also and the superiority of the alloy. This knowledge must have come through the extension of trade relations., not by conquest, for no country shows more independence in its metal series than Crete." 2
Whence was the bronze obtained by the Cretans? Was it from Egypt or Anatolia? Both Crete and Troy were able soon after the dawn of their Bronze Ages to import silver, which during the Old Kingdom Period was rarer than gold in Egypt. The silver may have come from the same region as tin. One possible source of supplies of silver was Cilicia, where silver mines are still worked; the other was Spain, in which country evidence has been forthcoming of early commercial relations with Crete.
Once the secret of how to work metals passed from centre to centre of Neolithic culture, the ingenuity expended for long Ages in the shaping of artifacts of flint, obsidian, and jade was directed into new and inspiring channels. Cretans, Trojans, Cilicians, and Cappadocians alike may have been stimulated to inaugurate a new era by foreign influences, but they did not remain as slavish imitators. The pupil not only strove to excel the teacher, but even to surpass him. As in our own day a new invention may be improved by a people who have borrowed it, so at the dawn of the Metal Age the borrowers
appear to have contributed towards the development of a discovery which was to revolutionize the ancient world. Ægean Bronze Age culture has distinctive features which establish its independent character. It was not of sporadic development. The indigenous influences which were manifested during the lengthy Neolithic Age were not cut off by the importation of metal, but were rather given opportunity to achieve freer and more brilliant growth in every sphere of human activity. That being so, we are confronted by an exceedingly difficult problem when we seek to discover whence either Crete or Troy imported bronze, or the copper and tin with which to manufacture it. The influences exercised by local cultures tend to conceal the sources from which borrowings were made.
Copper was known in Egypt in pre-Dynastic times. Indeed, some authorities hold that it first came into use in that country. "It was the custom of the proto-Egyptian women, and possibly at times of the men also," says Professor Elliot Smith, "to use the crude copper ore, malachite, as the ingredient of a face paint; and for long ages before the metal copper was known, this cosmetic had been an article of daily use. It is quite certain that such circumstances as these were the predisposing factors in the accidental discovery of the metal. For on some occasion a fragment of malachite, or the cosmetic paste prepared from it, dropped by chance into a charcoal fire, would have provided the bead of metallic copper and the germ of the idea that began to transform the world more than sixty centuries ago." At first copper was used for small ornaments and then to make needles, one end of a copper wire being bent down to form an "eye". In time, chisels and axes and other implements were manufactured in imitation of those of stone which were in use. "Every
stage in the history and evolution of the working of copper", he holds, "is represented in Egypt, and is preserved under circumstances that enable us to appreciate in some measure the motives which led the Egyptians on, step by step, to the full realization of the immensity of the power they had thus acquired." 1 Professor Elliot Smith follows Dr. Reisner in this connection. 2
Others hold that copper was first worked in Asia. Professor Myers, as we have indicated, favours Cyprus. 3 Mr. Hall, who supports the view that the knowledge of corn passed from Palestine to Egypt and Babylonia, thinks that the knowledge of metal may have come from the same quarter, Sinai, Syria and Cyprus being "the original focus of the distribution of copper over Europe and the Near East. Copper came gradually into use among the prehistoric Southern Egyptians towards the end of the predynastic age. And they must have obtained this knowledge of it from the Northerners." Mr. Hall adds: "Dr. Reisner considers the Egyptian evidence alone. and not in connection with that from the rest of the Levant". 4
It is also contended that the manufacture of bronze was not an Egyptian invention, and that Troy and Crete were probably in touch with the centre where copper was first hardened by tin and antimony. Mr. Hall suggests that this art "came from the Middle East, where tin is found, to Greece, as well as Babylonia and, eventually, to Egypt". 5 Babylonia, like Cyprus, had a long Copper Age.
No direct proof has yet been forthcoming, however, that Egypt imported its first bronze implements. The fact cannot be overlooked that the oldest bronze relics yet
found come from the Nile valley. No discovery has yet been made that bronze was manufactured elsewhere prior to 3000 B.C. A few objects of bronze have been found in First Dynasty tombs. Maspero gave Angelo Mosso a piece of metal plate from an Abydos tomb to analyse. The test showed "copper 96.00 and tin 3.75 per cent". 1 Another important relic is the famous "bronze rod of Medum", which belongs to the Third Dynasty period. It was found embedded in the fillings of a mastaba associated with the pyramid of King Sneferu. Pure copper was also used extensively throughout Egypt for the manufacture of weapons and implements from pre-Dynastic times till the Twelfth Dynasty. Iron was known at an early period, and is referred to in the Pyramid texts. It probably had a religious significance.
The Egyptians may have received their earliest supplies of copper from Sinai, which they visited to obtain turquoise in the Neolithic Age. 2 We know that expeditions were sent to work in the copper mines in that region at a later period (Third Dynasty). Whence was the tin obtained to harden the copper? A possible source of supply is North-western Arabia. That it could be found there is suggested by the Biblical reference to the spoils taken by Moses from the Midianites, which included "the gold and the silver, the brass, the iron, the tin and the lead". 3 Another possible source is Anatolia, where tin is said to exist. The raiders against whom Pharaoh Sneferu of the Third Dynasty waged war on the Delta frontier may have come down an ancient trade route, having ascertained that rich plunder could be obtained in Egypt. There is also tin in Italy as well as copper, but the earliest copper weapons found in that country are of
advanced Cretan type (Middle Minoan). Local forms which have been found are not of earlier date.
It may be that Egypt's scanty supplies of tin during the Old Kingdom Age came from more than one source. Mr. W. M. Muller sees on a Sixth Dynasty relief "Ægeans bearing tin into Egypt". If the figures referred to are Ægeans, they were certainly Cretans. It is of special interest to find in dealing with Egypt's early imports of metal that a socketed bronze hoe of the Sixth Dynasty resembles examples from Cyprus and South Russia which are preserved in the British Museum. This artifact may have come down the sea trade route by which sporadic supplies of tin and bronze were carried. The manufacture of bronze in Egypt never assumed great dimensions, on account of the difficulty experienced in obtaining tin, prior to the Twelfth Dynasty. Its early Metal Age was mainly a Copper one.
After the mariners of Crete began to bring home supplies of bronze, its traders no doubt did their utmost to acquire the secret of how to manufacture it. It may be that, like Solomon, who sent Hiram of Tyre annual supplies of wheat and oil in return for timber from Lebanon and skilled workers in metal, 1 a Cretan monarch made arrangements with an Egyptian or Anatolian Hiram to send him artisans who were skilled in the manufacture of bronze.
One of the places in Crete where bronze was cast was a headland on the Gulf of Mirabello about three miles east of Gournia. An ancient copper mine there is called by the peasants "Chrysocamino", which signifies "the oven of gold" or "the golden furnace". Describing it, Dr. Hazzidaki writes: "The seashore rises for above 100 metres, and here is the cave with so small an
entrance that one has to go down and creep in on hands and knees. The cave is 52 metres long, the roof is irregular in height, about 2 metres near the entrance, that is, 2 metres from it, and in the middle it reaches a height of 20 metres, and at the far end it is 12 metres high. The walls and roof are covered with stalactites, and the rock is calcareous. Great blocks of stone have fallen from above, especially at the far end of the cave." Small fragments of primitive pottery of uncertain date were found in the cave. and also pieces of Middle Minoan times.
Smelting operations were carried on near the entrance of the cave, as is indicated by a piece of crucible found by Dr. Hazzidaki. Inside, pieces of scoria were picked up. The copper appears to have been entirely worked out. 1 Specimens of rock taken from a cliff in the vicinity have yielded a small percentage of copper.
Bronze was also cast in Gournia. This is proved "by the finding of scraps of bronze and slag, pure copper adhering to smelting vessels, a crucible pot for carrying a charge of metal, and by numerous stone moulds, into which the molten metal was run for making knives, nails, awls and chisels". Copper was used for the manufacture of bowls, jars, and other utensils, but "weapons were of bronze, containing as much as ten per cent alloy with copper". 2 Copper daggers with an extremely small percentage of tin have also been found.
But although copper could be found in Crete, the tin, as has been indicated, had to be imported. "By the beginning of the Bronze Age", 3 writes Dr. Mackenzie in this connection, "the valley of the Rhone must have played a dominant role of communication between the
great world of the Mediterranean and the north; by that time it was probably already the high continental trade route towards the tin mines of Britain." Angelo Mosso also favours the hypothesis that Crete's early supplies came from England. "We know the road", he says, "followed by the caravans bringing English tin through France to the mouth of the Rhone at the end of the Neolithic period, while no trace of any trade in tin has so far been discovered in the East." 1 Mosso's reference to the "East" applies to "the mountains of China where tin is found".
Mrs. Hawes, who favours a Nearer Eastern source, writes as follows: "When the Pumpelly expedition returned from Turkestan in 1904, one of the members brought potsherds indistinguishable at first sight from the brilliantly mottled ware found at Vasiliki during the same season. . . . The strong likeness between the two fabrics, of which the writer has personal knowledge from having handled them together, is more reasonably explained by intercourse than by accident. Moreover, Dr. Hubert Schmidt, who accompanied the expedition, reports that a neighbouring tumulus (near the large one in which the pottery was found) gave him a three-sided seal-stone of Middle Minoan type, engraved with Minoan designs--man, lion, steer, and griffin. How shall we explain those evidences of Ægean influence in Southern Turkestan? They must be brought in line with other proofs of contact."
This distinguished lady archæologist refutes Dr. Muller's view that the Ægeans who carried tin into Egypt obtained their supplies from a trade route that connected Central Germany with the sea coast. "The backwardness of Europe in learning to employ metal", she says, "is
undeniable." Hungary, like Cyprus, had a Copper Age before bronze became known. "We see", she writes, "that at C. 2500 B.C. Asia Minor shared with the Ægean the knowledge of bronze, whereas three centuries later Europe was still in the Stone Age. . . . As further explanation of the priority of bronze in Asia Minor, we may now suggest the probability that, long before tin was discovered in Europe, it was being brought overland through Asia Minor, and also by way of Transcaucasia and the Black Sea from distant Khorassan, Strabo's Drangiana, where its presence has been confirmed. Excavations at Elizabethpol in Transcaucasia have revealed a culture in early contact with the Ægean." 1 She thinks that carriers "not unlike the swift Scythians of Herodotus, frequented both the tin-producing region south-east of the Caspian and the copper region of the Danube at an early date". 2
Troy was a probable "clearing house" of the early tin and bronze trade. We should therefore visit it before dealing with Ægean commercial connections with Western Europe.
Our course is a north-eastern one across the island-strewn Ægean Sea. This way went the Homeric Achæans who fought for the possession of Helen, the heiress of the Spartan throne, and no doubt with desire also to expand their area of political influence in the interests of commerce. We cast anchor as we draw near the southern shore at the mouth of the Hellespont. Since the dawn of history myriads of vessels have passed beyond this point to navigate the narrow strait, the modern Dardanelles, that leads towards the Sea of Marmora and the great Black Sea beyond it.
The famous Troad lies before us. It is a country which does not make much appeal nowadays, but must have offered many attractions to early settlers. The valleys are suitable for agriculture; there is excellent herbage on the hillsides for flocks and herds, and an abundance of game among the mountains. During winter the south winds from the Mediterranean impart to it a milder climate than prevails in the Balkans, or the uplands of Phrygia, and the summer heat is tempered by the cool Etesian winds. Water is plentiful; there are numerous springs and generous rivers flowing from the mountains. Withal there is an abundance of timber, much good clay for brick-making, and an endless supply of limestone with which to erect dwellings and strong, high walls to protect citizens and their domesticated animals against the attacks of bears and lions and cunning wolves that prowl through the forests and up and down the green valleys, not to speak of human enemies.
We land at the mouth of the famous river Scamander, turning our backs on the unpicturesque tongue of European land known to the ancients as Chersonesus, and in our day as the peninsula of Gallipoli; we also take our eyes from the shouldering hills of the island of Imbros, behind which towers sublime Mount Saoce, the loftiest peak of Samothrace, on which the god Poseidon aforetime sat to watch the Homeric heroes performing mighty feats of arms.
Our steps are directed inland, and we proceed to cross the long and windy Plain of Troy, remembering
Old unhappy far-off things
And battles long ago.
Yonder towards the south-east, blue above the ridges of woody hills, is the Anatolian range of Mount Ida, which
forms a noble frontier of the Troad; there Paris was once a shepherd; thither, too, fled Æneas after Troy fell. To the west is the high coastland of the Ægean Sea, and eastward and north-eastward are broken groups of featureless mountains divided by pleasant valleys. Less than 4 miles in front of us we can distinguish a boat-shaped hillock, on the spur of a sloping hill, rising abruptly from the plain: that is famous Hissarlik, the site of the ruins of the various citadels of Troy.
The memoried plain is bordered on either side by the Rivers Simis and Scamander. There are marshes to avoid, as in Homer's time, but these are easily detected at their utmost limits by the clumps of long grasses and weeds, and of whispering tamarisks which also fringe the steep and crumbling river banks.
The Simis has shrunk to a few inches in depth, for it is now late summer; puffs of wind blow clay dust from its clay-caked and stone-strewn bed. Down a beautiful valley it flows westward, as if to cross the plain towards the Ægean Sea, until it curves round a ridge of hills and directs its course to the shore of the Hellespont. The more famous Scamander is about 2 feet deep and about 20 feet in breadth. When, however, the snows are melting on the Ida range it is exceedingly turbulent, and of such great volume that it carries down trees and boulders, and occasionally overflows its reedy banks to submerge the plain. The Simis similarly rages furiously at this period.
There is an interesting reference in the Iliad to the sudden rise of the rivers after a "cloud burst". When Achilles drove one part of the Trojan army into the city and another into the Scamander,
the plain he found
All flooded o'er, and, floating, armour fair,
And many a corpse of men in battle slain.
The Scamander was supposed to be increasing for the express purpose of resisting his advance. The roar of its spring flood resounds in the sonorous hexameters of Homer, but sinks to a spray-like hiss in an English translation.
Rearing high
His crested wave, to Simis thus he 1 cried:
"Dear brother, aid me with united force
This mortal's course to check; he, unrestrained,
Will royal Priam's city soon destroy.
Nor will the Trojans his assault endure.
Haste to the rescue then, and from their source
Fill all thy stream, and all thy channels swell;
Rouse thy big waves, and roll a torrent down
Of logs and stones, to whelm this man of might." 2
We reach Hissarlik and ascend it to survey a maze of ruins. The fields around us were tilled and irrigated aforetime, when there were watchmen on the "topless towers" to give warning of the approach of raiders. These keen-eyed men could see far up the valleys; nor could vessels cross the Hellespont without their knowledge; and they had glimpses to the west, across the Scamander, of the Ægean Sea, which is but 3¼ miles distant, and were thus able to herald the approach of the galleys of Crete.
Before Schliemann began to excavate on this wonderful hillock, by cutting a deep broad trench through the various strata, it towered about 160 feet above the level of the plain; but when the earliest Neolithic people first chose it as a settlement, it was not much more than 50 feet high. Distinct traces survive of nine cities in all, the latest being the Troy of the Roman Age. Each city, after the first, had been erected on the levelled debris of
the previous one. So the hill, like a stooping giant, gathered from age to age an increasing burden for its great unwearied back.
Troy I was built in the Neolithic Age. Its deposit of from 12 to 14 feet indicates that it endured for many long centuries. Portions of its walls constructed of small stories, here and there in herring-bone pattern, were laid bare by Schliemann. As the foundations, in some parts, do not reach the bedrock, it is evident that the hillock was occupied for a considerable period before stone was utilized for building purposes. The earliest defensive works may have been ramparts of earth.
Hissarlik was apparently from the earliest period the citadel of the city which lay round it on the plain. Here dwelt, in a palace, the king and his family, and here also were stored the treasure and winter food-supply of the tribe. When enemies poured down the mountain passes, or across the Hellespont from Europe, the citadel became a shelter for women and children, and for flocks and herds. Inside its walls, too, the warriors found safe retreat when attacked by overwhelming numbers. The hill forts and brochs of Scotland appear to have served a similar purpose.
Within the area of Troy's Neolithic citadel traces survive of the stone foundations of houses and of certain erections usually referred to as "sheep-folds". Of special interest are the remains of pottery which have come to light. The fragments unearthed by Schliemann were of the hand-made variety, and these are numerous and varied enough to show that the Trojan ceramic art was developed locally and attained a comparatively high degree of excellence. Invariably the pottery is dark and decorated with geometric designs, the incisions being filled in with white chalk as in Crete and Egypt. A fine surface finish was effected by the use of the smoothing-stone.
Doubt has been expressed as to whether all the bronze implements which Schliemann associated with this early stratum really belong to it. Some of these may have fallen down the sides of his trench, and got mixed up with the relics of a deposit with which they had originally no connection. It appears certain, however, that the Neolithic city was in existence at the dawn of the Metal Age in Crete, for some of the bronze implements in question are unlike those found in later strata.
The second city was erected before 2500 B.C. Whether or not there was a fresh racial infusion we have) as yet, no means of knowing. It is significant to find in this connection that there are distinct traces of development from the Neolithic period, especially in the ceramic relics, a sure indication that a considerable portion of the old stock remained. For the first time the hillock was levelled, a process which no doubt obliterated much valuable evidence, and it then stood about 100 feet above the sea-level. Retaining stone walls, which sloped inward, were also erected, and those round the south-western and western sides of the eminence can still be traced.
This was the city which Schliemann believed to be Homer's Troy, because it contained a great amount of burnt debris. But in this he was mistaken. Shortly before he died, however, he found some Mycenæan potsherds which afforded a clue to the mystery and enabled Dr. Dörpfeld, the distinguished German archæologist, who conducted subsequent excavations, to locate Homer's city in the sixth stratum.
Dr. Dörpfeld has divided the history of the second stratum into three periods. These may be referred to as Troy II A, B, C. The citadel of Troy II A, was little more than a tribal fortress about 100 yards in diameter. There were two main entrance gates, one on the south-western
side and the other on the southern. The pottery which was manufactured resembled the hand-made variety of the Neolithic settlement, but the workmanship displayed was on the whole inferior. Apparently we meet here with the decadent period during which vessels of stone were being constructed with the use of copper drills.
In the Stratum II B the new pottery makes its appearance. The Egyptian potter's wheel had evidently reached Troy as well as Crete, while the enclosed baking-furnace also came into use. There can be no doubt, therefore, that a brisk trade was being conducted along the trade routes both by land and sea. Considerable progress was effected also in architectural work, brick as well as stone being largely used.
The evidence of Stratum II C shows that the citizens of Troy were progressing by leaps and bounds. Traces of destruction by fire of earlier buildings suggest that frequent conflicts were waged round the fortress, and it is possible, therefore, that the extensions and alterations which were effected from time to time were rendered necessary to maintain the prestige of the city in stirring and difficult times, when hordes of nomads were enabled by the acquisition of metal weapons to overrun large portions of territory.
It was during the period covered by the deposits of the second city of Troy that the great masses of Asiatic pastoral nomads pressed into Europe and conquered the more passive and more highly-cultured agriculturists of the Mediterranean race. As much is indicated by the burial remains of the Early Bronze Age in Europe, which show that a broad-headed people pressed westward, first along the uplands and then across the valleys, in increasing numbers, here adopting the funerary customs of their predecessors, and there introducing their own.
Troy continued to develop its own civilization, resisting, it would appear, for a long period the raids of plundering barbarians. That its wheel-made pottery was not imported is made evident by its distinctly local characteristics. The hand-made jars, with side projections, pierced for suspension, which were characteristic of Stratum II A, assumed more artistic character in Stratum II B, when the wheel came into use. Another link between earlier and later times is the "face urn". These interesting Trojan products indicate that the decoration of pottery may have had a mythological significance. Zigzag, St. Andrew's Cross, herring-bone, and V-shaped designs, as well as rippling lines, "trickle ornaments", and dots, may therefore have meant much to the people who believed that their food-supply was the gift of a deity, or group of deities, whose favours they constantly invoked by performing ceremonies and offering sacrifices. In the Odyssey the Phæacians toasted the deity when they drank together. King Alcinous, addressing his guests after Odysseus had partaken of his meal, spoke as follows:--
Pontonoüs! mingling wine, bear it around
To ev'ry guest in turn, that we may pour
To thunder-bearer Jove (Zeus) . . .
When, at length,
All had libation made, and were sufficed,
Departing to his house, each sought repose. 1
Food and drinking vessels may have been dedicated to deities as well as the potter's wheel, which, as has been indicated, was credited to the god Ptah in Egypt. The spirit of the god, or of one of his emissaries, may have been in the cup. It is of interest, therefore, to find that the lips of some of the Troy vessels are ornamented with
GENERAL VIEW OF "THE TREASURE OF PRIAM"
(From the photograph by Schliemann in "Atlas Trojanischer Alterthümer.)
The topmost row shows the Golden Diadems, Fillet, Ear-rings, and small Jewels. Second row--Silver "Talents" and of Silver and Gold. The Silver Vases and curious Plate of Copper. Third Row--Silver Vases and curious Plate of Copper. Fourth row--Weapons and Helmet--crests of Copper or Bronze. On floor--Vessel, Caldron, and Shield (all copper).
circles enclosing dots. One characteristic fragment shows two circles with a straight line drawn down between them. It is obvious that the potter desired to represent a face with staring eyes. Schliemann believed that the face was intended for that of an owl, and constantly made reference to "owl-headed" vases. Another fragment, however, shows clearly that the crude artistic efforts were directed towards the representation of the human face. No attempt was made to indicate the nose line, but the eyes were fairly well shaped, and above these the eyebrows were drawn also. In other examples the eyebrows and nose were shaped like a bird in flight, the eyes being represented by perforated circles, while a straight line represented the mouth.
This tendency towards realism is found to be less pronounced, however, as the vessels become of more complicated and finer construction. The arched eyebrow, the eyes and ears, yield to purely decorative tendencies, and become symbols, as do also the dots, rings, and cones representing female breasts; the swastika on the lower part of the body is evidently a fertility symbol. This process of developing symbols from natural objects can be traced even in the Palæolithic Age. It does not follow, however, that the change robbed the ornaments entirely of their religious and magical character, difficult as it may be to discover where a symbol is divested of significance and a purely artistic motive begins.
The Trojan method of representing the human face, with the bird-wing-shaped nose and eyebrows and the eye dots, is paralleled by similar designs on objects from the Greek islands. Interesting examples of the same artistic motive have been found in the East Riding of Yorkshire. In a trench surrounding a burial cairn on Folkton Wold were discovered chalk drums associated
with unburnt burials. These are ornamented with spirals, St. Andrew's Cross, and other characteristic Ægean designs, and also with the eyebrows and eye symbols. As the latter appear on standing-stones of the Marne and Gard valleys in France, and on early Bronze Age vessels in Spain, it may be that the chalk drums are interesting survivals of racial or cultural influence which reached these islands across the English Channel by way of Spain. 1
The second stratum of Troy is remarkable for its treasure hoards. Schliemann found no fewer than seventeen of these. The most famous is the "royal treasure", or, as he called it, "the treasure of Priam", which, with the assistance of his wife., he concealed during the workmen's dinner-hour. The objects were of rich and varied character. In a silver jar had been stored two great diadems of elaborate construction, which were worn by females of high rank. One is composed of four rows of small heart-shaped leaves of gold connected with fine wire, and is fringed with a row of larger pendants suggesting the human form. On either side are tails, terminating with larger pendants in a bunch. This diadem is about the breadth of the forehead, and when clasped round the head the hair was bunched above it, while the tails fell downwards and lay on the shoulders. Elaborate ear-rings were also worn, as well as rich necklaces made of small gold rings strung together, and bracelets of twisted gold. Some of the ear-rings are of spiral design. The spiral is also associated with the rosette to ornament elaborate gold hairpins and broad bracelets. A small gold eagle-shaped ornament is of special interest, as it indicates the sanctity with which that bird was invested in this region.
Included in the hoard are several bars of silver, which
may, as Schliemann suggested, have been used for money. A silver dagger was no doubt a royal weapon used on occasions of great ceremony. Like the bronze daggers it was pierced so as to hold the rivet with which it was attached to the handle. One dagger handle is carved in ivory and is reminiscent of Palæolithic Magdalenian Art, for it is shaped to represent a crouched animal. A bronze handle of similar design has been discovered in Etruria, and is now in the Kestner Museum at Hanover.
Among the objects in lead, special reference should be made to a figurine of the mother-goddess. It is of somewhat conventional design, like the terra-cotta figurines found in Cyprus, Mesopotamia, and Greece, and those of marble and other stone in the Cycladic islands. The face is stern, with a hard drooping mouth, and the eyes stare cold and angrily. Long curls dangle down from the ears; the neck is exaggerated and crossed with symbolic markings, and the hands are clasped across the breast. The female characteristics are pronounced, and on the lower part of the body the swastika, or hooked cross, is depicted on a V-shaped projection surrounded by round bosses. The legs are merely suggested, and may have been used as a handle, or as a spike to be thrust into the soil of a holy mound. Votive figurines found at Anau in Turkestan, and those also from Sumeria, were attached to nails, or terminated like nails, so as apparently to be driven into sacred shrines, for the same reason as the visitors to sacred wells drop pins into them, or attach rags to overhanging trees. Prayer-nailing still obtains in the East.
It may be remarked here that the third, fourth, and fifth citadels of Troy, which cover a period between about 2000 B.C. and 1500 B.C., are of no great account. The city shrank in importance after the occurrence of a great disaster which is indicated by the fire-swept remains of
Stratum II C. The sixth, or Homeric Troy, will be referred to in a subsequent chapter.
Since Schliemann's day, attempts have been made to relegate the "treasure of Priam" to a comparatively late period, one nearer Troy VI than Troy II A. Indeed, it has been asserted that this rich hoard fell down the trench from the sixth city stratum. But although Schliemann sometimes nodded, like Homer, his location of the treasure can no longer be disputed. In 1908, Mr. Seager, the American archæologist, discovered a similar hoard on the island of Mochlos, which lies about two hundred yards off the north-eastern coast of Crete in the picturesque Gulf of Mirabello. For some 4500 years the treasure had reposed in a necropolis of the Early Minoan Period, happily secure from the attentions of generations of tomb robbers. The island is barren and without a water supply, and was consequently never suspected of containing anything of value. At one time it may have been part of a peninsula which sheltered a natural harbour much frequented by the earliest mariners.
The hoard included gold diadems, rings, pendants, hairpins, and fine chains, "as beautifully wrought", Sir Arthur Evans has remarked, "as the best Alexandrian fabrics of the beginning of our era". 1 There were no spiral designs as at Troy, but wonderful artificial leaves and flowers. Of special interest are the gold bands "with engraved repoussé eyes for the protective blind-folding of the dead". These, Sir Arthur suggests, were "the distant anticipations of the gold masks of the Mycenæ graves". Bead necklaces were probably charms. Associated with these articles were miniature stone vases of local material. Some were of Early Egyptian form, and all were of exquisite workmanship.
GROUP OF JEWELS FROM THE HOARD DISCOVERED IN THE ISLAND OF MOCHLOS
(See page 238)
An engraving on a ring in this hoard depicts a ship with a sail and a full equipment of oars. Troy may have been visited by the men who crossed the seas in vessels of this kind. Traces of Cretan commerce have been forthcoming at Hissarlik, and Trojan artifacts have been found in Crete. In 1909 discovery was made at Phæstos of a fragment of pottery which resembles fragments of the same date (Early Minoan II) found in the second city of Troy. Relics of Cretan connections with Troy have also been found at Vasiliki and other eastern sites.
Crete's reputation for metal-working was widespread among the ancients, but no one dreamed, before Mr. Seager made his important discovery, it was of such great antiquity. The remarkable technique displayed shows that the craft had a long history. It no doubt owed something to Egypt, if, indeed, it was not established on the island by Egyptian traders. "Of the jewelry worn by the Pharaoh and his nobles, in the Old Kingdom," writes Professor Breasted, "almost nothing has survived, but the reliefs in the tomb chapels often depict the goldsmith at his work) and his descendants in the Middle Kingdom have left works which show that the taste and cunning of the first dynasty had developed without cessation in the Old Kingdom." 1 The Cretan ornaments have distinct local characteristics. Like the painters and potters, the goldsmiths showed a distinct feeling for nature, as in their leaf and flower designs; one notable ornament is the Cretan equal-limbed cross. Of special interest, too, is a clover-leaf ornament-an anticipation of the Irish devotion to the shamrock.
At the time the articles in the Mochlos hoard were manufactured, there must have been many wealthy men in Crete. Those whose ships visited Troy and Spain
were probably the possessors of articles of silver as well as gold. But none of these have been discovered. Perhaps some of the Early Minoan silver artifacts were so highly prized that they were kept as heirlooms. Dr. Xanthondides found two silver daggers in a tomb at Kumasa, near Gortyna, while excavating tombs of the Early Minoan III Period. They were ribbed and of triangular shape, like other daggers of bronze. Associated with these metal objects were steatite "libation vases", a rough marble figure of the mother-goddess, three miniature vases with lids on a reel-shaped stand, and an earthenware vessel of teapot shape with geometric ornamentation. Sir Arthur Evans discovered several silver bowls of the Middle Minoan Age at Knossos. Among the finds of the American archæologists at Gournia is a shapely silver cup with handles, from a house tomb, which recalls Homer's reference to "a silver cup, the work of the Sidonians". 1 It is, however, of much greater antiquity than anything which can be credited to the Phnicians. Perhaps it was won by the individual in whose grave it lay for displaying skill as a boxer. A double silver cup was awarded to the Homeric athlete Epeius, who "knocked out" Euryalus at the funeral games that followed the burning of Patroclos. 2 Joseph, 3 who was so greatly honoured by the Pharaoh, was the possessor of a silver cup, and must therefore have been wealthy as well as influential.
The Cretans may have received their supplies of silver from Troy, where, as is shown by the articles made from that metal in "Priam's treasure", it was abundant enough.
Some hold that this silver came from Spain, and their theory will be dealt with later in this chapter. Others favour the view that the Trojans and Cretans imported it
from Lydia or Cilicia. It is possible that silver was obtainable by the island mariners at primitive commercial centres at or near Miletus, Ephesus, or Pitane. But of this there is no direct proof. The remarkable fact has to be given recognition in this connection that no traces of early Ægean trade have been found at any of these points. 1 Even the islands of Samos, Chios, and Mitylene have failed to yield any indications of commercial connections with Crete and the Cyclades during the Early Bronze Age. "Except for their north-western corner", writes Mr. Hogarth, "the Asiatic coasts of the Ægean lay, until very late, outside the culture-area associated with the name of that sea. But if 'tis true, 'tis strange! Why did the Cretan and other Ægean sea-rovers, whether pirates or merchants, or both, fail to settle on these particular coasts and isles? They had pushed their wares into Hissarlik, and had filled all the opposite shores of Europe with a culture much higher and more vigorous than any which has left a contemporary trace in Anatolia." Mr. Hogarth believes that "there must have been some strong continental power dominating all the west-central coast of Asia Minor from an inland capital. It must have been a non-maritime power, careless about developing its coast lands, but careful to keep others away from them." This power was the Hittite--the confederation of peoples controlled by the Hatti, the "white Syrians" of Greek tradition, whose ancient capital was situated at Boghaz'köi. It is possible that the early Ægean influences
which permeated Anatolia were introduced through the medium of Troy.
Troy appears to have existed during the Late Stone and Early Bronze Ages as the capital of an independent state. Its earliest settlers were probably of the Mediterranean race, and congeners of the Neolithic folk of Thrace and the Danube area, who had pressed northward through Syria and round the southern Anatolian coast, or by way of the "Cilician gates", to the western shores of the Ægean Sea, afterwards crossing into Europe. This racial movement, which radiated also throughout the agricultural valleys of Anatolia, appears to have taken place before the broad-headed Hatti, who were a pastoral people, became the dominant race. It may be also that there survived among the mountains descendants of the ancient Palæolithic races. The Etruscans, for instance, whose racial affinities are obscure, are believed to have come from Anatolia.
The Danubian cultural area was of wide extent. It included part of southern Russia and part of southwestern Austria, the whole of Thrace and Macedonia, and a portion of Thessaly. At several centres a high form of Neolithic culture was developed. "There is reason to believe", writes Mr. Hogarth in this connection, "that some population, racially kin to that which developed the Ægean culture, was present on the Anatolian coasts from early times, and also that there had been very early passage of influences, and perhaps of peoples, from Balkanic Europe to Asia Minor. Not only has the earliest sub-Neolithic stratum at Hissarlik produced pottery and weapons closely resembling those of Neolithic Danubian graves, but at two other places where sub-Neolithic settlements have been explored in north-west Asia Minor, Danubian analogies are even more certainly to be
remarked. Those places are Boz Eyuk in Central Phrygia, and Yortan in Mysia. The vases of the latter site, where there is a cemetery of the earliest Bronze Age, show close analogies with Cypriote forms, and suggest that the earliest migrants from Europe spread sporadically far down through the peninsula to the Levant." 1
Like Anatolia, the Danubian area was a melting-pot of races. In addition to the Armenoids of Hatti type who invariably clung to an upland habitat, but also fused in localities with the Mediterranean peoples, the fair northern peoples pressed southward to absorb the local culture and fuse with the earliest settlers. The ethnic friction which resulted caused periodic migrations of displaced peoples. There was, therefore, much crossing and re-crossing of the Hellespont and the Bosphorus in the Late Neolithic and Early Bronze Age Periods.
Troy, by reason of its situation, must have been ever a meeting ground of various ethnic elements. Many desperate conflicts, no doubt, were waged on its windy plain long ages before the Homeric era. There were rich spoils besides in its citadel to attract the invader. It lies at the end of the northern trade route which runs through Anatolia towards Mesopotamia, and must ever have been a "market-place" for traders, who could exchange there their far-carried commodities for the products of Thrace and the Ægean.
Various axes of green and white jade, which Schliemann found in the stratum of the first city, may be relics of an ancient trading connection with the east, as the knives and arrow-heads of obsidian appear to be of a connection with the Cyclades.
When the jade objects were first found they caused a flutter in archæological circles. It was pointed out that
scrapers and other articles made of jade had been found associated with the Swiss lake-dwellings, and at Neolithic sites in Brittany and in Ireland, as well as elsewhere throughout Europe. The belief obtained generally that these jade artifacts were imported into Europe from the borders of China, and Professor Fischer expressed the wish "that before the end of his life the fortune might be allotted to him of finding out what people brought them to Europe". 1 Professor Max Müller believed that the jade-carrying immigrants were the Aryans. "If", he wrote, "the Aryan settlers could carry with them into Europe so ponderous a tool as their language, without chipping or clipping a single facet, there is nothing so very surprising in their having carried along, and carefully preserved from generation to generation, so handy and so valuable an instrument as a scraper or a knife, made of a substance which is aere perennius." 2
It is not now believed, however, that all the jade objects found in Europe came from "a common far-distant home in the Kuen Luen Mountains". Since Müller connected his Aryans with jade, the two species of it, nephrite (jade proper) and jadeite, have been found in different parts of Europe. Nephrite has been discovered in Silesia, Austria, and North Germany, and it is believed to exist in Sweden, while jadeite, or a similar rock, was found not long since among the Alps. It is probable, therefore, that the Swiss and other scrapers were chipped from pebbles of jade picked up by the European Neolithic people. The quantity and quality of the Hissarlik axes, however, suggest an eastern source of supply, and it may be that these and the Maltese polished axe pendants of jade are genuine relics of primitive commerce. As the latter were charms, it would appear that the magical qualities
of jade were given recognition at a remote period. Among the Greeks it was the "kidney stone", and among the Spaniards, who imported it from Mexico, the "colic stone". Various rare stones were believed by the ancient peoples to have curative qualities. Instances could be cited of the possession, by representatives of ancient families at the present day, of stone charms of this kind that have long been treasured as heirlooms.
Although archæologists are less inclined nowadays than they were a generation ago to believe in the existence of Neolithic trade-routes which extended from the borders of China to Brittany, or to connect certain races with relics of similar character found in widely separated districts, there can be little doubt regarding the existence of commercial relations between different cultural areas. The introduction of metal appears to have done much to stimulate international trade. In the Early Bronze Age the influence of the Ægean, which may have "inspired every stage of culture" at Hissarlik, as Mr. Hogarth suggests, appears to have penetrated Thrace. Evidence has been forthcoming that two main trade-routes crossed Germany, one from the head of the Adriatic, and the other from the lower Danube valley. It has been suggested that some of the amber found in Crete came down these trade routes from the Baltic. 1 France was similarly crossed by the Rhone valley trade-route, down which, in time, tin from Cornwall was carried. That the Cretans were the earliest seafarers to come into direct touch with these routes is suggested by various interesting links of evidence. The most remarkable are the Egyptian glass beads found in South Germany, and the Egyptian blue-glaze beads taken from ancient graves on Salisbury Plain, which will be dealt
with in a later chapter, as they are connected with the Late Minoan Period.
Certain Continental archæologists incline to the belief that not only Crete but even Egypt was in direct touch with Western Europe at an extremely remote period. Summarizing their views, Angelo Mosso writes: "The vases found at Amerejo in Spain have the characteristic form of the Egyptian vases of the close of the Neolithic Age. The resemblance of the Egyptian idols with those of Crete and the Continent is an established fact; the burial sites are similar; the flat copper axes of Egypt cannot be distinguished from those of the Continent; the evolution of art in Southern France and in Spain went on during the Neolithic Age, and we know that navigation was general on the Mediterranean in the times preceding the introduction of copper-all these data give good reason to suppose that the pre-Dynastic Egyptians had relations with the west which enabled them to procure cassiterite, which when mixed with copper rendered it harder. . . . We hope", he adds, "that new discoveries may throw light on the relations of Egypt with England." 1
There can be little doubt that the Cretan mariners sailed westward as far as the coast of Spain, although the precise period at which they first undertook voyages in this direction may remain uncertain. Spain could supply silver, copper, and other metals. The brothers Siret 2 are of opinion that this country was the source of the earliest supplies of silver, the metal having been taken from the silver-bearing veins before the discovery was made how to extract it from lead as described by Pliny. 3
[paragraph continues] Mosso favours the view that the silver articles found in Crete were made from silver carried from Spain by the early mariners who sailed westward to fetch tin from the Cassiterides Islands. He makes no reference to the Cilician mines. 1
It is difficult to fix the movements of the early traders in chronological order. We cannot therefore ascertain from the archæological evidence available when the Cretans came into touch with the western Iberians, with whom they apparently shared a culture of common origin. Prior to the Bronze Age a comparatively high civilization was developing in southern Spain. The votive figures found in this region resemble those of Cyprus, Hissarlik, Crete, and the Cyclades; even the sacral horns were given recognition. Spanish Early Bronze Age artifacts also show close resemblances to Ægean forms, and the brothers Siret found in several places in Spain goblets similar to those taken from Early Minoan strata in Crete, and others from the tombs of Abydos in Egypt. These vessels were associated with flat copper axes and copper knives with silver rivets, as well as stone and bone implements. Tin appears to have been less plentiful at the period to which these finds belong than silver. It may have come from the Balearic Islands, Brittany, or England-the first named being the most probable source.
At Marseilles, where Greek merchants established themselves in later times, the visits of the Cretans must have stimulated trade along the Rhone valley route, which became gradually suffused with Ægean influences. The trade-route from the head of the Adriatic, leading towards the Brenner Pass, was similarly affected. Sicily and Italy have yielded suggestive evidence of early contact with Crete. Daggers and flat axes of Cretan shape
have been found in Italian tombs of the early metal age. Sardinia appears to have been visited also; it has yielded, among other things, specimens of characteristic Ægean axe adzes, which have also been found at Troy.
One of the most interesting links between Ægean, Trojan., Danubian, and Western European cultures is the spiral decoration, which appears to have been introduced along the trade-routes.
"The developed spiral", writes Mr. Hall, "appears suddenly in Egyptian art on seals and (rarely) in painting, at the beginning of the Twelfth Dynasty, 1 or shortly before," that is, "at the end of the Third Early Minoan, or beginning of the First Middle Minoan Period in Crete." 2 It appears to have been introduced into Egypt from Crete, for it occurs on objects of Early Minoan II and III date. There are spirals on the Trojan gold pins of "Priam's treasure". Mr. Hall favours the view of Much, the German archæologist, that "the spiral originated in metal wirework". He thinks it may have been "an invention of early gold workers in Lydia that reached Troy, was in the Cyclades translated into stone carving, in Crete transferred to pottery and to the designs of button seals, and as a seal design came to Egypt, where it was promptly adopted as the characteristic decoration of the new form of seal that had as suddenly become popular in the Nile land, the scarab". 3
The spiral ornament travelled along the trade-routes through Europe. Rings made of silver wire twisted in a spiral have been found by the brothers Siret in Spanish tombs which have yielded the goblets of Cretan form, already referred to. In the Danubian cultural area the spiral occurs on pottery of the early metal age. Following
DECORATIVE MOTIFS AND SYMBOLS
Figs. 1 to 8. Minoan and Celtic patterns compared. The treatment in different areas of motifs, which were probably of common origin, is of special interest. Numbers 7 and 8 are identical. fig. 9. The equal-limbed Cretan cross. Fig. 10, The swashtika symbol--cross with arms bent. Figs. 11, and 12. Celtic knot developed from swashtika by connecting points of bent arms by, curves--single treatment (point to point) in ii and double treatment with swashtika reversed (inner curves corner to elbow and outer curves point to point) in 12. Figs. 13 to 17. Religious Symbols, perhaps connected with belief in weapon spirits; 13, Shield and crossed arrows of Egypto-Libyan goddess Neith; 14, Mycenæan 8-form shield as symbol; 5, Cretan deity on seal; 16, Scoto-Celtic "spectacle" symbol shown upright as on standing stone; 17, Scoto-Celtic "crescent and arrow" symbol.
the road along the Moldau and the Elbe, it reached the shores of Jutland, and ultimately passed into Scandinavia. It reached England either along the Rhone or Danube valley routes. Reference has been made to the Yorkshire chalk drums on which it was inscribed. The New Grange stones are decorated with it, and early Scottish sculptured stones show local adaptations of the design. Eastward from the Danubian area it penetrated as far as Koban in Russian Armenia, between the Caspian and Black Seas, where it occurs on objects taken from a prehistoric cemetery in which Babylonian influence is also in evidence.
The earliest connection between Crete and northern Europe is indicated by the finds of Baltic amber in Early Minoan strata. It probably had a religious significance. Amber was carried down the Elbe and Moldau route as well as through the Rhone valley to the shores of the Mediterranean, and across to England, Scotland, and Ireland. It is believed that this trade was flourishing along the Elbe route before 2000 B.C.
The manner in which early commerce was conducted between the peoples of northern and southern Europe is indicated by Herodotus, who refers to offerings sent to Delos by the Hyperboreans. "They" (the Delians), he wrote, "declare that certain offerings, packed in wheaten straw, were brought from the country of the Hyperboreans into Scythia, and that the Scythians received them and passed them on to their neighbours upon the west, who continued to pass them on until they reached the Adriatic. From hence they were sent southward, and when they came to Greece, were received first of all by the Dodonæans. Thence they descended to the Maliac Gulf, from which they were carried from city to city, till they came at length to Carystus. The Carystians took them
over to Tenos, without stopping at Andros; and the Tenians brought them finally to Delos." 1
Reference has been made to the engraving of a ship on a ring from the Mochlos hoard. It is shown sailing from the shrine of the mother-goddess, who evidently protected seamen as well as landsmen. A similar ship, carrying two sails as well as oars, was depicted on a seal stone of steatite, which also belongs to the Early Minoan Period. Two crescent moons above the mast seem to indicate that the voyage was to extend over a couple of months.
Other seal engravings show vessels with one, two, or even three masts. Some have complex riggings and well-braced yards. A seal from Mirabello shows a one-masted vessel with a square sail. 2 An ivory model of a ship found by Sir Arthur Evans in a tomb at Knossos has a hatch over its hold to protect the cargo. Terra-cotta and alabaster models were discovered at Aghia Triadha, near Phæstos, by the Italian archæologists. A terra-cotta model from Palaikastro belongs to the Early Minoan Age.
"The modern vessels of the Cretan fishermen, and especially those of the fishers for sponges from the Isle of Kalimnos, differ little", writes Angelo Mosso, "from the ships of antiquity." 3 Occasionally Maltese boats are found to have the Horus eye on the prow, like the ancient Egyptian boats of the dead found in tombs. Beside the eye a flag is sometimes painted. There were ensigns on the prows of pre-Dynastic Nilotic vessels. Neolithic ships carved on rocks in Upper Egypt had sails and oars like the Cretan vessels, which they resemble in shape. Maltese boats retain the high prows of the prehistoric ships, and
[paragraph continues] Italian cargo boats have oar helms similar to those of the Egyptian river vessels. Seafarers have ever been intensely conservative. Some of the curious superstitions that still prevail among them may be as old as the pre-Dynastic pottery of Egypt and the maritime seal stones of Crete. Early Minoan sailors may have whistled to conjure the wind spirit, like our own fisherfolks, as they steered between the rocky isles of the Ægean Sea, or struck out boldly now westward to Sicily, and anon eastward towards Cyprus and the Syrian coast.
216:1 Crete, the Forerunner of Greece, p. 19.
219:1 Dawn of Mediterranean Civilization, pp. 299 et seq.
219:2 Those who favour the Cypriote origin of copper-working urge that the earliest Egyptian copper artifacts are copies of those of Cyprus. It can be shown, on the other hand, that some of the Egyptian copper artifacts are copies of Neolithic forms.
220:1 Schliemann was wrong in asserting that Hissarlik (Troy) had a Copper Age.
220:2 Gournia, Mrs. Hawes and Others, p. 33. (American Exploration Society, Philadelphia, 1908.)
222:1 The Ancient Egyptians, pp. 3 et seq.
222:2 Prehistoric Cemeteries of Naga-ed-Der, Vol. I, p. 134.
222:3 Science Progress, 1896, p. 347.
222:4 The Ancient History of the Near East, pp. 89 et seq. (1913.)
222:5 Ibid., p. 33.
223:1 Dawn of Mediterranean Civilization, p. 57.
223:2 Ibid., p. 59.
223:3 Numbers, xxxi, 22.
224:1 I Kings, v, 1-12, and vii, 14. et seq.
225:1 The Dawn of Mediterranean Civilization, pp. 289-91.
225:2 Crete the Forerunner of Greece, pp. 289-91.
225:3 C. 2800 B.C.
226:1 The Dawn of Mediterranean Civilization, pp. 62-3.
227:1 Mrs. Hawes refers in this connection to E. Rössler, Zeits. f. Ethnol., XXXVII, 1905, pp. 114 et seq.
227:2 Gournia, p. 33.
230:1 Scamander.
230:2 Iliad, Book XXI (Derby's translation), 340 et seq.
234:1 Book VII (Cowper's translation).
236:1 British Museum Bronze Age Guide, pp. 89-91.
238:1 Times, 27th August, 1908.
239:1 A History Of Egypt. p. 94
240:1 Odyssey, IV, 618.
240:2 Iliad, XXIII, 741 et seq.
240:3 Genesis, xliv. 2.
241:1 Mrs. Hawes suggests that "the objects given in exchange by the Cretans for European products were of as inferior and ephemeral character as those with which modern traders dupe the native; hence the phenomenon noted by Burrows (The Discoveries in Crete, p. 190) that genuine Ægean articles are absent from districts where Ægean influence is undeniable" (Gournia, p. 10). Asia Minor may have received chiefly supplies of wine and food-stuffs. Pharaoh Meneptah of the XIX Dynasty sent shiploads of grain to the Hittites in time of famine (A History of Egypt, Professor Breasted, p. 465).
243:1 Ionia and the East, p. 58.
244:1 Schliemann's Ilios, p. 242.
244:2 Letter to Times, Dec. 18th, 1879.
245:1 Much of the Cretan amber is evidently from the Adriatic.
246:1 The Dawn of Mediterranean Civilization, p. 62.
246:2 Les premiers ages du metal, H. & L. Siret, p. 227.
246:3 Nat. History, XXXIII, 31.
247:1 Dawn of Mediterranean Civilization, pp. 372-3.
248:1 C. 2000 B.C.
248:2 Or Middle Minoan II, according to Hawes.
248:3 The Journal of Egyptian Archæology, Part II, pp. 115, 116.
250:1 Herodotus, IV, 33.
250:2 Probably "white sails and twisted ropes of ox-hide" (Odyssey, II 425-6).
250:3 Dawn of Mediterranean Civilization, p. 280.