From Asianic Elements in Greek Civilisation, The Gifford Lectures in the University of Edinburgh, 1915-1916, by Sir William M. Ramsay, D.C.L., LL.D., New Haven: Yale University Press, 1928; pp. 243-266.

Asianic Elements in Greek Civilisation.





IN an article in the Journal of Hellenic Studies, 1920, p. 200 f., on the four Ionian tribes which are widely known in the Ionian cities and colonies, and which are familiar to all as an early institution in Attica, I suggested and wrongly rejected an explanation, which during subsequent study of Iono-Anatolian words has gradually established itself as the most probable.

We are, of course, in this book dealing only with probabilities, not with certainties, except where Hesychius or some other trustworthy authority makes a definite assertion.1 An occasional passage of Herodotus or other author is useful, e.g. Euripides, Ion, in the last speech of the goddess Athenaia.

The four tribes, which are indubitably Asian in origin, are usually enumerated as Geleontes or Gedeontes, Aigikoreis, Argadeis or Ergadeis, and Hopletes. The authorities differ greatly about this old enumeration and the proper order. That there was this ancient division of the Attic people into four tribes (phylae) is a primary fact: we cannot penetrate farther back. Euripides ascribes it to the four sons of Ion, i.e. the Ionian settlement in Attica. Aristotle, in the Athenian Politeia, ch. xli., mentions this system as the first and earliest; he does not name 244 the tribes, but connects them with Ion and his comrades, who united the whole of Attica into one State.2

Under Theseus, many of whose exploits were performed outside Attica, and who was really in a sense a foreign hero, though he lived in tradition as the great founder of Attic nationality and unity, there was some attempt made to introduce the tripe system, but only of classes, not of tribes. This European system of three, Eupatridai, Geômoroi, and Dêmiourgoi, is rather an interruption than a real step in development,3 and its author Theseus was often said to be a son of Poseidon and Aethra, an alien princess of Troezen:4 it has the appearance of Pelopennesian influence, and it was not permanent.

SECTION I.  THE QUADRUPLE AND THE TRIPLE TRIBAL SYSTEM. — The institution of the four tribes in Attica, which is associated in tradition with Ion and his four sons (i.e. with the early Ionian settlement in that country),5 has all the appearance of Asiatic origin.

The four castes in Hinduism are a very ancient institution, but the castes have been much mixed during the wars of nations; and yet, amid the forces of conquest and the rule of conquerors over subject races, there remains the fundamental idea of difference in occupation, Priests, Warriors, Agriculturists, and Artisans (as Professor Eggeling asserts).6 Brahmins, Kshatriya, Vaisya, and Sudra are restricted to four several duties, though other modern scholars would open to the Vaisya a variety of occupations, for four occupations are too narrow a range as civilisation grows. So 245 Plato classes together hunters, shepherds, and agriculturists in one of his ideal classes.7 However much Plato was guided by philosophic principles, the idea of four tribes devoted to four occupations and modes of life was too deeply fixed in his mind to be disregarded; and yet in his time society had become during a long development far too complex to be confined within such restricted limits. In Ionia, in European Greece, on the Anatolian plateau, and in India we must suppose that there did exist once a social state which was adapted to the fourfold way of life, and that wars, racial differences, and the rule of tribe over tribe, all tended to fall ultimately under that primaeval system.

Among the Dorians, on the other hand, immigration and conquest triumphed, and the beginning of their political system lay in the victory of two united and allied tribes over native races and the gradual admission of the conquered to a pseudo-equality. The numbers in each tribe differed greatly, but the tripartite system prevailed. This is a European idea.

In Attica political changes were far more rapid and thoroughgoing than among the Spartans and other Dorian tribes. In Crete, where so many races found refuge that no united nationality could establish itself permanently,8 the Dorians were simply one race of a number of old and new settlers, who could not overcome their rivals (Chapter III.). The geographical barriers interposed by mountain, glen, and sea were insuperable. On the west Asiatic coast the tribal system proved also evanescent; and a certain tendency to unify different races and cities, and to transplant the people of one city to increase another, is strikingly apparent.

SECTION II.  THE FOUR HINDU CASTES. — It is a remarkable fact 246 that, whereas in India (See Section I.) the idea of divine supremacy fixed the priestly caste as the highest, and their conservatism prevented human development, the priests in the Ionian and Attic quadripartite tribal system are rarely mentioned as highest. The cultivator of the ground was, as I venture to think, the great moral figure of the world, alike in the Hittite sculpture at Ibriz, far to the east in Asia Minor, and in the Iono-Attic tribal system. The priest adores the peasant-god at Ibriz; but god is the working, toiling power, much superior in importance and in size to the priest.

In Asia the religious influence is always stronger: in Europe the political struggle of party with party (usually originating in difference of race) is the supreme fact. Only in the ideals of Plato (Critias and Timaeus) are the priests placed first. In Athens and in Rome tribes became merely part of a political machine for voting. The four city tribes in Rome were the humblest among the total, which grew up to be thirty-five; and in these city tribes, all freedmen were enrolled.

The primaeval division goes back to Ion, who unified the people of Attica, and classified them into four Phylae, twelve Phratriae or Trittyes, 360 families (γένη), and 10,800 individual members of families (γεννηταί). This represents a tradition that implies a settlement of the Old-Ionians in Attica. The scattered tribes of Attica were divided by mountain ridges and mutual dislike. Those early Ionian settlers instituted a classification of the entire people into four tribes (φυλαί), which were necessarily subdivided. The phratriai may have been fixed for a time at the number twelve; but the regulation numbers of families and individuals could not have been more than a brief and temporary arrangement. The numbers must have grown larger steadily, as the population increased, while emigration to colonies had not begun.

The division into phratrai is the ancient method, recorded by Homer in Iliad, iii. 362, but he necessarily uses the term nations (phyla), not the technical term tribes (phylae). The sole point in 247 which Theseus diverged from the old Asian and Ionian system was in adopting apparently the Dorian triple system of classes, Eupatridae, Geomoroi, Demiourgoi, instead of the Ionian and Asian quadruple system of tribes; but his Eupatridae seem to have embraced both priests and warriors, and to have been in possession of all civil and religious offices and to have been the exponents of the law (which was still uncodified and unwritten) and the declarers of religious right and usage. The other two classes had to accept this interpretation from the aristocratic class in whose possession lay the principles of interpretation.

SECTION III.  THE FOUR TRIBES OF ICONIUM. — The only place in the interior plateau of Asia Minor where the existence of four tribes can be proved with almost complete certainty is Iconium, as is shown in the inscription on Fig. 6. The restoration is largely conjectural, being founded on a poor copy by Hamilton; and the stone must already in his time have been in a very bad condition. Hamilton, indeed, was a geologist by profession and interest, not an epigraphist; but his copies of inscriptions are almost always good, because he was careful and thorough in everything, and sought for truth, not glory. I have often called him “the prince of travellers in Asia Minor,” and consider him, after many years’ experience, to deserve this title fully.

The inscription was arranged between four garlands representing the four tribes.9 The tribes, unfortunately, have changed names in compliment to the Roman Emperors. It was a common practice to inscribe within a garland the name of a tribe, as here; sometimes “the Senate” or “the Demos” is engraved in this way.


Pen and ink sketch of a copy of an epitaph showing the four garlands of the Iconian colony.

FIG. 6.

The four Garlands of the Iconian colony.

Singly from other inscriptions we learn the names of three, but all four were inscribed within four garlands. One is Augusta, probably representing an original Dias, for Augustus was worshipped as Zeus. Another is the tribe of Heracles, the toiling god, and the benefactor of men by subduing the earth for their use: it became Hadriana. A third was the tribe of Athena, having a second name beginning with P,10 but the rest of the name is now lost. The fourth has perished entirely, but a plausible conjecture would make it Hephaestias, the tribe of craftsmen: or perhaps it may have been called after Hermes, the messenger, who is so frequently associated with Zeus as his servant and as the speaker (see Acts xiv. 12). It 249 is vain to indulge in conjecture as to the tribes. The garlands and other ornaments had evidently been chiselled off before Hamilton saw them; an three or four of the letters of the inscription had become illegible.

SECTION IV.  ATTICA. — According to tradition the primitive population of Attica was autochthonous. Pollux states that the Attic tribes were always four, called at first Kekropis, Autochthon, Aktaia, and Paralia, in the time of the mythical Kekrops, first King of Attica, a land which was at that time called after his name Kekropia. Kekrops was sprung from the earth; he was an autochthon by birth, but his name is given to a different tribe. His body was in its upper part human, but in its lower part a dragon or serpent. He is sometimes said to have instituted marriage; i.e. he founded the social system of families in Attica. His wife was Agraulos, and his son was Erysichthon. In his reign, at the beginning of Attic history, occurred the contest between Athenaia and Poseidon for the lordship of the whole country (mentioned below in this chapter). Other primitive institutions of society in Attica were attributed to Kekrops, such as the abolition of sacrifices of blood (presumably human blood, for the sacrifice of animals was never abolished); he substituted cakes in the place of slain victims.11

With the substitution of pelanoi for human victims compare the sacrifice of Isaac commuted into the sacrifice of a ram, and the statement in an inscription of Cilicia Tracheiotis which my friend and old pupil, professor T. Callander, communicated to me, that part of the sacrifice to the dead in that barbarous land considered of “nine souls of men,” evidently a substituted offering for nine human victims, for the period is Roman, when no such human sacrifice can have been permitted: to the Romans it was mere superstition, which they regarded as worse than wrong.12


In other legends of Attica, Kekrops was closely connected with Erechtheus (on whom see below in this section) and with Athenaia, the Mother-Goddess of Athens.13 To his three daughters (the eldest, Agraulos, bearing the name, i.e. being another form, of her mother) Athenaia entrusted a box containing the child Erichthonios, which they were commanded not to open, but they disobeyed the command, were terrified at the sight of a serpent, the guardian genius of the land.

In the days of another later mythical king, Kranaos, the names of the four tribes were changed to Kranais, Atthis, Mesogaia, and Diakris. Then under the king or chief Erichthonios their names became Dias, Athenais, Poseidonias, and Hephaistias, and they were called in the time of Erechtheus (from the Sons of Ion, Yavan), Geleontes,14 Hopletes, Aigikoreis, and Argadeis.

[table 1]

These and other tales and names evidently refer to the primitive state of Attica and of Athens, and to the gradual settlement there of the Asian Sons of Ion (Yavan), “by whom the isles and lands of the sea were parcelled out and occupied” (Genesis x. 4). There were contests even among the gods for this almost sea-girt 251 land, separated from the rest of Greece by mountains (especially Parnes), mountains not now so difficult to cross as formerly when no roads had been made.

Those earliest four tribes in the time of Kekrops derive their names from the king himself, from the indigenous population, from the long promontory, and from the sea-shore. They are purely native and local; but the prominence assigned to the promontory and the sea-shore indicates that settlements by the Old-Ionian sailors were beginning in a barbarous land.

The four tribes of Kranaos hardly represent any development from those of Cecrops, but one tribe Atthis may stand for the country in general (like the older Autochthon), while the two last are purely local divisions. The reference to the Old-Ionians has been lost. Kranaos, in fact, represents almost an earlier stage than Kekrops, but little is related about him. Both kings are really native to the soil; both represent the primitive state of the land.

[table 1] 15

Erechtheus marks a more developed stage than either of the two previously mentioned. The land has a guardian spirit in the form of the household serpent, social life is beginning, and the gods have settled as patrons of the land.

SECTION V.  THE FOUR TRIBES OF ATTICA. — As to the number of the early Iono-Attic tribes all authorities are agreed, though 252 Pollux has substituted in one of his lists Kadeis for Argadeis. He seems to have found Gedeontes in his authority, but he can hardly be quoted in proof of this spelling, which rests mainly on Plutarch and Pollux.

Moreover, Pollux gives a still more primitive enumeration of the Athenian tribes, according to the names of their patron deities, as Dias, Athenais, Poseidonias, and Hephaestias. His two lists must be connected, and are of the same nature and number. The first list enumerates the four tribes as four religious groups protected by four special deities, whereas the other uses technical or tribal titles. Dias is the tribe of Zeus Geleon (originally Gedeon); Euripides connects the Aigikoreis with Athenaia and her Aigis; these two are respectively the toiling class which works the earth, and the class of priests, who are devoted to the service of Athenaia, the goddess of the united country of Attica.

The sequence of Poseidonias after Athenais is due to the famous incident of Athenian history which made Athenaia and Poseidon contend for the place of guardian deity of the county.

For what divine figure Poseidon may perhaps have been substituted remain uncertain, but the double name Poseidon-Erechtheus on the Acropolis of Athens weighs in favour of Erechtheus-Erysichthon16 as the guardian of the land; now there can hardly be any doubt that this guardian deity was the sacred serpent, who lives in the house and under the ground, brings good fortune and must never be killed,17 as with him dies the luck of the household. Poseidon, 253 for Poti-davan, is the lord-earth, though ordinarily the Earth was conceived of as the Mother. Kekrops himself was half serpent, half man, and he was first king of Attica.

Hephaistias is naturally the class of the skilled artisans, as we have seen; and it remains that Poseidonias is the tribe of the Hoplitai or Hoplitai warriors, who made their own bows and arrows in primitive time, and tested them, as they tested all their own implements of war: he was also the horse-god, and the horse was used in war.

Thus there emerges the explanation of the strife between Athenaia and Poseidon, which of the two should rank highest as guardians of the land and its fortunes. Athenaia created and appealed to her olive, the tree of peace and comfort:18 the olive produces the oil, the one great source of wealth in the plain of Athens: the olive is a very slow-growing tree; and the planter of the olive must look forward to a long period of peace before any result can accrue from his labour. Poseidon appealed to his horse, the animal of war, which is inevitably fatal to that distant comfort and the riches that accrue from growing the olive. The alternative was peace or war; and the aristocratic class was the horse-riding and horse-using class, as is apparent during the fifth century before Christ in Athenian history. The peasant used the ox: the noble used the horse. The peasant wanted peace: the noble was far from averse to war, though not all nobles actually desired it.

Plato placed hunters, agriculturists, and shepherds together as a single class (see Timaeus, 24, and Critias, 110). The hunters sought food, not sport. In Anatolia at the present day it is considered a crime to kill any animal except for food; just as it is wrong to throw away bread instead of eating it. Plato was guided by ancient ideas, and was not inventing novelties: his model is 254 often to be sought in Anatolia or farther east.19 In the Critias he states what he considers to be the true facts of Crete, but facts under a veil of phantasy. It is his pleasure to mention the warriors last, as lowest in his estimation; but in this he agrees with the official order in Attica (as below stated), and with the order given by Strabo.

The official and regular order in Attica, according to a good modern scholar,20 whose original authority I am unable to verify, was Geleontes, Argadeis, Aigikôreis, Hopletes. So far as I can see, the nearest approach to an authority is Boeckh’s commentary on C.I.G. ii. 3665, p. 928, where the order at Cyzicus is Geleontes, Argadeis, Aigikoreis, Hopletes, but Bôreis21 and Oinôpes are there interposed before Hoplêtes; being presumably two tribes meant to contain population (probably native) added to the Old-Ionians. The authority would require further study, which I assume the scholar has given. The division was based on difference of occupation and ways of life, as Strabo says. He and Pollux are agreed that there were stages in the history of the four tribes, as is Aristotle in his Athenian Politeia, ch. xli. Plutarch gives the impression that there existed at an earlier time four tribes who subsequently chose four different occupations; but this is an impossible inference from his authority — who, however, is good. We must imagine a primitive state of Attic society, in which there were four chief and fundamental occupations, agriculturist, artisans, priests, and warriors. Ion, having first, as Strabo says (viii. 7. 1, p. 383), divided the people of Attica into four tribes (phylae), thereafter classified them according to four ways of living. There can, however, be no doubt, judging according to natural probability 255 and the general evidence derived from many sources, that the four tribes and the four ways of living are identical. The four tribes bore the technical and Old-Ionian names as enumerated above; but Strabo designates them from their ways of living, agriculturists (γεωργοί), skilled workmen ([GREEK TEXT), doers of religious things ([GREEK TEXT), and guardians ([GREEK TEXT), agreeing with the order quoted above in this paragraph as official.

We take the agreement between Strabo and the official order, already mentioned, as the most trustworthy. Plato is philosophically inclined to place the priests first: they have to do with divine things and to interpret to men the will of the gods; and it is really difficult to see why an Asian system should not take the same view (Section I.). the fact that the Anatolians put the priests after the agriculturists and the artisans is illuminative regarding the ancient Anatolian and Old-Ionian ideas.

Otherwise Plato, Critias, 120, tends towards the tradition expressed by Strabo. He calls the Ergadeis or Argadeis the skilled workmen (demiourgoi), classes together with them all that gain their nurture from the ground, and at the end he places the fighting class, as the least useful in a well-ordered state, but prescribed by tradition and old custom, and as not wholly unnecessary even in a philosophic state. In the Timaeus, 24, he makes a clear distinction between the skilled workman (demiourgoi) and the class of shepherds and hunters and agriculturists. While Plato is not bound to an ancient Attic or Ionian custom, but speaks as a philosopher aiming at eternal truth, he was not wholly unmindful of the ancient Asian ways of life, and preserves the fourfold classification. He could not bring his philosophic mind to rank the agriculturists high; but he could not disregard the facts of life: the earth must be tilled and they that tilled the ground were more useful than mere warriors. There is every probability that he thought of Geleontes as priests, and Aigikoreis as herdsmen. Stephanus of Byzantium seems to consider the Aigikoreis as the first of the four tribes which came 256 down from Ion, the rest being in order Ergadeis, Geleontes, and Hoplêtai.

It is a disappointment to find that the order of Euripides agrees neither with the “official” order nor with Strabo, an excellent authority for everything connected with or derived from Asia Minor. Euripides, however, clearly considers that the Aigikoreis, wearing the goatskin of Athenaia, the goddess of the land, where therefore her priests. His enumeration is perhaps modified by metrical considerations, and by the desire to make Athenaia end her enumeration with her own priests, giving them prominence and emphasis. The last place in such a poetic list is emphatic,22 like the first; the middle two are slurred over, and are mentioned as less important, “then second are Hoplêtes and Argades.” He also connects them with the four sons of Ion, multiplying a popular and accepted tradition which could not be seriously violated in a drama acted before the people.

The name Geleon (Teleon in Euripides, Ion, 1579 f., and in Plutarch, is a mere error of form) is often placed first in our authorities. Strabo means the Geleontes or Gedeontes when he uses the Greek word γεωργοί,23 viz. those that work the earth to make it useful to mankind. Euripides and Herodotus, v. 66, made Geleon the eldest son of Ion, and first among the eponymous heroes of the four tribes; but Euripides has the subsequent order Hoplêtes, Argades, and Aigikorês,24 while Herodotus enumerates the sons of Ion as Geleon and Aigikoreus and Argadeus and Hoples.

Plutarch, Solon, 23, puts the Hoplitai, or warriors, first, by a very 257 natural error.25 Warriors come first in a conquering tribe. There is, however, no reason to think that the Ionians settled as conquerors in Attica, rather they came as civilisers of that barbarous country. Then he mentions the artisan class, Ergadeis, the agriculturists, Gedeontes, and the class devoted to pasture and breeding of sheep, Aigikoreis.

Stephanus of Byzantium puts the Aigikoreis first; but, like Plutarch, he understands them to be herdsmen, and after he had made this first error his order is disturbed and becomes untrustworthy: Argadeis, Geleontes, Hoplitai.

Plutarch depends on a good ancient tradition, which used the names Gedeontes and Ergadeis (true old forms, according to the view taken in this chapter).

The four tribes fell into disuse in Attica at a comparatively early date during the gradual development of the Athenian constitution. Partly though the prejudices of some authorities, partly through actual want of knowledge (for accuracy about past institutions was not much sought for), accounts are conflicting. Herodotus, v. 66, says that the names were derived from the four sons of Ion; and so also say Euripides and Aristotle. It is a notable coincidence that in Genesis x. 4 the sons of Yavan (the Semitic form of Ion) were four; but beyond the fourfold division here is no resemblance. The four sons of Yavan are four separate states or cites, and they can only be four divisions of the Ionian people, who settled in different regions; and one would expect to find that in each of the states (as in Ionian colonies generally) the division into four tribes existed.

The sons of Yavan (Ion) had a natural tendency towards this fourfold division; but in different settlements the four tribes would 258 naturally acquire a different character. What is certain is that the four Attic tribes or divisions must be connected with the very ancient settlement of the Ionians in Attica, when they imposed their institutions on an uncivilised country.

As being of Asian origin the four sons of Ion in Attic story, viz. the four tribes descended from him, represent four occupations and modes of life. So Strabo says, and he is probably our best authority. Plutarch26 agrees in a hesitating way, omitting priests altogether, and misunderstanding the term Aigikoreis. The four were priests, soldiers, artisans, and agriculturists; of these the agriculturists are really the most important, though they seem the most humble. They are also, especially in a simple state of society, necessarily the most numerous. During the time when this Attic classification was a real political fact there was no such thing as popular voting, or counting of individual votes and numbers. In any public deliberation the popular feeling made itself apparent in other ways, as, e.g., by applause or by crowding towards a speaker who expressed a widespread sentiment. Otherwise the government lay in the hands of chiefs or kings, with leading men or elders as a sort in the hands of chiefs or kings, with leading men or elders as a sort of consultative body. The names of the Attic tribes appear in two forms, Gedeontes and Geleontes.27 It seems to me now most probable that the transition from “d” to “l” was a stage in the development of the Ionian and the Anatolian languages (as it was in Latin). Gedeones, then, must be the older form, and it has all the appearance of being connected with Da or Gda,28 the earth. Gedeontes are Gadavantes (with interposed vowel between two consonants), the men of the earth, the toiling peasants, who support the State. Their 259 god is Zeus Geleon (for Gedeon), who enjoyed an ancient and obscure cult in Athens. The “Peasant God” was the lord of the state, the god that cleared the land, and worked and made the earth useful to man, a sort of toiling Heracles, whose “labours” were almost all for the benefit of man; compare e.g., the Lernaean Hydra (or marsh) and the Lion of Nemea, etc. which conceal truth under a guise of fancy and myth. With regard to the “Peasant God,” I may refer to the chapter on his character in Anatolia as “the great moral figure” of the world, in Luke the Physician and other Studies in Religious History. He is dressed as a peasant, in contrast to the gorgeous robes of the priest who adores him, but he wears the cap of power, in the famous Hittite sculpture at Ibriz.

There is considerable variation in the text of Pollux. He gives the tribes as being always in early times, before and after Ion, four (according to the text of Bekker, which is here followed). They were changed in name by different kings, Kekrops, Kranaos, and Erechtheus (who called them after the four sons of Ion), and finally for political reasons they became ten by the reforms of Alcmaeon and Cleisthenes, when the Pythian oracle selected names which were ancient.29

The Argadeis30 were the skilled workers, the artisans, and their name is connected with the Greek word ἔργον, applied to the arts of the household practised by women, and to the constructive arts practised by men. A woman is praised both by Homer and in the late Phrygian dirges (Chapter IX.) for her beauty and her skilled household management (ἔργα). Their deity was Hephaestus, the god of all skilful things and labour-saving devices.31 The Anatolian 260 village Manarga may perhaps retain a memory of the name, as the Erga or Arga of Mannes — Men. Probably “arga” was the true Anatolian form, not “erga.”

In many Anatolian inscriptions of the Roman period the term “skilled workman” is used almost as a title of distinction, and is added to a man’s name according to certain important trades, e.g. tekton, carpenter and house-builder32 (these two occupations being usually united, for a house needed both wood and stone in its walls, as in Asia Minor at the present day), chalkeus, worker in bronze or copper, and so on.

The meaning of the term “tekton” has a bearing on the often-discussed question whether Jesus, like his father Joseph, was a “carpenter” in our sense, or a “mason.” The truth is that Joseph was both, and his “son” in the usual fashion was brought up to the same trade as his father; but their implements (ὅπλα) were far from being as accurate as at the present day. There can be no question what was the meaning of architect (ἀρχιτέκτων); why then of tekton? A man who could do only one of the two trades was of little use at that time. The architekton planned the entire house, wood and stone.

Aigikoreis are, as Euripides says, the priests that wear the goat’s skin (aigis). They were the goat-priests of Chapter VII., the Attabokaoi33 of Pessinus, the priests that taught the culture of the goat in a 261 country where goat-breeding was of prime importance, as it is in the great plains that stretch south of Pessinus, the main central plateau of Asia Minor. From the misunderstanding of the term Aigikoreis spring many of the difficulties that we encounter in trying to reconcile ancient authorities. Much confusion was caused to the ancients by mistaking Aigikoreis as goat-herds instead of “goat-priests,” and the mistake has seriously influenced the lists transmitted by them.

The Hoplêtes must be the warriors, and must denote a special caste. Hoplôn is a common personal name in Pisidia, amid whose mountains old Anatolian customs and names lingered longest, and who were soldiers and brigands and robbers, all fighting men, hated and feared by the men of the fertile plains, the cultivators of the soil. The word ὅπλον was in origin Anatolian; it was used both in a general sense as an implement, and in a special sense as a weapon of war.

SECTION VI.  THE TRIBES AND THEIR SUBDIVISIONS IN EPHESUS. — In Ephesus, which had been subjected to many changes of constitution, as it was successively Greek or Lydian, or reconstructed, there were, according to the editor of the British Museum Inscriptions, Part III. p. 68 f., six tribes (phylai): (1) Epheseis (the original native population whom the Attic colonists found in the land); (2) Sebaste (“which owes its name and perhaps its origin to Augustus”); (3) Têioi (new colonists brought in from Teos to compensate for great losses sustained by Ephesus in a war); (4) Karenaioi (of similar character, from a town in Mysia); (5) Euonymoi (early Attic colonists); (6) Bembinaioi, or Bembeinaioi.

Little can be learned from his classification, dating from the Roman period, about the real state of early Ephesus. Possibly the tribe Sebaste was the tribe of Zeus, from the usual identification of Augustus with that god. The spelling and forms are often of late Roman period and are incorrect.


More important is the subdivision of the tribes into groups called thousands (chiliastyes). Among tribe 1, Epheseis, there are known Argadeis, Bôreis (as at Cyzicus), Lebedioi (presumably a group of settlers from Lebedoes, not numerous enough to constitute a tribe), Oinôpes (as at Cyzicus); in tribe 2, Sebaste, there are known Labandêoi (perhaps either settlers from Caria, or worshippers of the god of the double axe, labranda), Sieis, and another ending in -mêoi; in tribe 3, Têioi, Eurypom-[pou ?], Echeptolemeus, Hegetoreioi, [Gel ?]eontêoi, and at least one other unknown.

In tribe 4, Karênaioi, there are Althaimeneus, probably named from a legendary hero of Crete, who fled to Rhodes and was there worshipped (Diodorus Sic. v. 59); according to Strabo (x. 479, 481, and xiv. 653), he had come from Argos to Crete, and thence to Rhodes. There was a tribe Althaimenis at Camiros in Rhodes (Inscr. Brit. Mus. ii., cccliii.: see also Pausanias vii. 2. 3). There are also “thousands” named Echyreos, Peios, Simôneos, and Chelôneos. Peios is local, named after Mount Pion.

In tribe 5, Euonymi, where old names might be expected, there are known only Glaukêos and Poly[klê ?]os.

In tribe 6, Bembine, Bembinaioi, Bembineis, there are known only two “thousands,” Aigôteos and pelasgêos: the latter name suggests an ancient division, and the former an early goat-worship.”

Boeckh, on C.I.G. 2855, conjectures (without solid ground of any kind to stand on, but with mere probability) that the four old Ionian and Attic tribes once existed at Miletus, Teos, and Cyzicus. This may be so. Wiegand’s excavations at Didyma may yield proof; but at present no mere guess is of any value.

The constitution and the tribes of Ephesus are unusually well known among all the Ionian peoples, because an account of them was given by Ephoros the historian about 400-350 B.C., which is summarised by Stephanus of Byzantium34 (unfortunately very 263 much abbreviated), and because the constitution as fixed during the Roman imperial time is attested by numerous inscriptions, which have been skilfully published by the late Bishop Hicks (Greek Inscr. of Brit. Mus. Pt. III. p. 69 ff.). The city was subjected to numerous vicissitudes. New citizens were introduced and new tribes had to be made for them. The city was conquered, or rather taken over peaceably, by the Lydian king Alyattes, about 580 B.C., and more violently by the Macedonian king Lysimachus; and it was radically changed in constitution by both. Ephesus was, doubtless, first a Lydian village, the property of the goddess, then a free Ionian city, then a Lydian town under the guardianship of the goddess, then a Hellenistic city, then Roman. The number of tribes grew to six in the Roman time. The old Ionian names disappeared; but the Argadeis and perhaps the Gedeontes35 remained as chiliastyes or “thousands” in one or other of the tribes. It is possible that there were five chiliastyes in each tribe, making thirty in all; but this is quite uncertain: others of the old names may have survived; but the list of the chiliastyes is or seems to be very defective.

The Bembinaioi was the last tribe in later time; but the order may have changed in various ways. Bembinaioi is perhaps a reduplicated form of Bennaioi: compare Salouda and Salsalouda in the Tchal-Ova, mere variations, also Pasa and Paspasa in Cappadocia.36 Bishop Hicks, however, thinks that an Argive village Bembina, close to Nemea, sent colonists to Ephesus after the return of the Heracleidae, and that these formed a special tribe; but that seems in my judgement unlikely. Still, it is highly probable that the Athenians under Androclus were accompanied by many of the 264 old Peloponnesians, especially from the Argolid territory, fleeing from the Dorian invasion, the so-called “Return of the Heracleidae.”


Roman period.

1.  Epheseis

1.  Sebaste

1.  Têioi

1.  Karênaioi

1.  Euônymoi

1.  Bembinaioi

Time of Ephoros.






The words quoted from Ephoros, about 400-350 B.C., are very important in their bearing. Androclus, son of Codrus, king of Athens, was said to have led an Attic migration to Asia Minor, and to have founded the Greek city of Ephesus; but he made war on Priene disastrously; and the Ephesians lost so many men, that the survivors rebelled against the sons of Androclus. Colonists and helpers came to Ephesus from Teos and Karina. These were enrolled in two new tribes; the original inhabitants, whom the new colonists had found there, were called Ephesioi, and the two tribes of introduced settlers were called Têioi and Karinaioi, while the Attic founders of the Greek Ephesus were called Euônymoi from the name of an Attic deme.37 It is implied in this account that the old native population of Ephesus must have constituted a tribe. They could not be expelled; there is no precedent for thinking that they were massacred; at that time the older population were always accepted and incorporated in the new city. In Ephesus, especially, the older population was strong in the protection of the native goddess; and to have degraded them in any way would have 265 offended the divine power that ruled the land. They were therefore made the first tribe, and the chief priest of the goddess by hereditary right, as sprung from the Lydian king Kandaules, was a conspicuous and splendid figure among them (see Chapter XII. B).

It is evident in history that Ephesus was the least thoroughly hellenised and the most truly Lydian of all the Ionian cities; and this is indicated by the primary position of the ancient Lydian population among the five tribes about 400 B.C., and in the Roman period. They held the priesthood; the temple was at the Lydian end of the city: the temple, the Christian church of St. John Theologos, and the old Seljuk mosque are all close together.38 At this point the immense power of the goddess and her priests was concentrated. When Alyattes gave his daughter in marriage to Melas, a wealthy citizen of Ephesus, it is probable that he chose one of the Lydian element and tribe; for this marriage would strengthen that element.

it is highly probable that there was by no means a harmony of feeling between the Greek part of the city and the Lydian. Hipponax, an Ionian Greek, seems to have made a specially scurrilous attack on the chief priest (see p. 180). The priest is said to have been called by the Persian title Megabuzes; and it is possible that a Persian title was given to him under Persian domination; but it seems not improbable that buzes or bozes was an Anatolian word meaning the representation of the god: see my Cities and Bishoprics of Phrygia, Pt. I. p. 152 f., No. 52, where it is shown that Bazis-Bozis meant “the estate of the god” (both at Tyana and on the Lydo-Phrygian frontier.)39

It would not be possible here to study closely the “thousands” of Ephesus; they imply Carian elements as well as Ionian and 266 Greek and Lydian. One was called Labandêos,40 where the disappearance of the “r” after “b” is an Anatolian peculiarity.41 Aigôteos points to an early form of goat-breeding according to the rules prescribed and enforced by religion. The goddess probably had her earliest home among the mountains south of Ephesus, and she was then a goat-goddess as well as a bee-goddess (see Chapter VII.). Hicks, however, connects Aigôteos with an ancient Arcadian town Aigus, and the “Thousand” Pelasgêos with Arcadian Pelasgians who accompanied the Attic emigrants. Perhaps he is right.

There was a local element among the “thousands,” as, e.g., Peios took its name from Mount Peiôn;42 and probably other examples might be given with fuller knowledge of the “thousands” and of the parts or divisions of Ephesus.

Two of the “thousands” in the tribe Epheseis, Bôreis and Oinôpes, bear names that are markedly Anatolian. They also occur at Cyzicus along with the four old Ionian tribes. The only town in Upper Anatolia where four tribes are known to have existed is Iconium, that ancient city where the old legend of the flood and the mourning of Nannakos was localised. They are given as four, but three only are known, as shown in Fig. 6, taken from Journal of Hellenic Studies, 1918, p. 183. According to Boeckh on C.I.G. 2855 and 3064, whom Hicks follows, there were originally the four old Ionian tribes at Miletus and Teos. Two were added at Miletus.

NOTE (to p. 106). — In Beowulf (p. 104) the hP>



 1  The Etymologicum Magnum is a small collection from old scholia, and is far less useful. Pollux and Suidas are too general, but contain useful information: they try to do too much. Diodorus, except incidentally, confines himself to Roman, Josephus to Jewish antiquities.

 2  ωνος  καὶ  τῶν  μετ’  αὐτοῦ  συνοικησάντων : these words place it beyond doubt that an immigrant Ionian people formed Attica into the first stage of its growth as a single State, and grouped the people into four tribes.

 3  It is not mentioned by Aristotle, Polit.

 4  Or he was half legitimate son of the Attic king Aegeus and Aethra.

 5  With Ernst Curtius, I regard an early settlement in European Greece, mainly in Attica, as one of the great early events in Greek history.

  6  Encycl. Britannica, ed. xi. vol. 13, p. 502 f.

  7  See below in this chapter, Section I.

  8  Although, yielding to the authority of Lang, Leaf, and Myers, I have accepted their translation, “Dorians with waving plumes,” I incline personally to the belief that “triple-tribal Dorians” is the true translation and the true European idea, even in Crete.

  9  τὰ  τέσσαρα  στέμματα  τῆς  κολωνίας. Iconium was made a Roman colony by Hadrian; but no Latin or Romanists were sent to it. The title was an honour. See J.H.S., 1918, p. 183; Boeckh, 3995 b; but Boeckh’s publication, based on Hamilton, is unintelligible. My restoration is conjectural, but seems guaranteed by the words.

10  Perhaps Π[ολιάδοςθηνᾶς, according to Weigand. This I do not like; but it is possibly correct.

11  Pelanoi were offering to the gods, mixed of meal, honey, and oil.

12  ἐννέα  ψυχὰς  ἀνθρωπίνας.

13  The cult of Athenaia-Mother was practised, but hardly anything is known about it, as it was overwhelmed by the idea of the Virgin Athenaia.

14  Teleon for Geleon is used by Euripides, Ion, and by Pollux, but is certainly an error, showing how uncertain tradition and text were.

15  Correction of Canter, adopted in the text of Bekker, which I follow.

16  Though separated in mythological record, these two names seem to me to denote the same heroic personality, like Perseus Bellerophontes, who slew the monster and led the Ionian colonisation of the south coast of Asia Minor. These first colonists became known to the Semites as the four sons of Yavan.

17  There are practically no venomous serpents in Greece or Anatolia. I have never heard or known of any person that had died from the bite of a serpent, though in one or two cases serpents have been pointed out whose bite was said to be dangerous or even fatal; but this I regard as due to popular superstition. Erechtheus, Erichthonios, Erysichthon, all mean Saviour of the Land.

18  I may venture to refer to my study of the olive as a factor in civilisation, Pauline and other Studies in Religious History.

19  That Plato was strongly influenced by Oriental ideas is admitted by all. The half philosophic, half religious myth or narrative with which his Republic ends, is the narrative of Er, son of Armenius.

20  See Teopffer in Pauly-Wissowa, Real Encyclopaedie, who quotes as his authority Meier, De gentil. Att. 4.

21  On Bôreis see p. 266.

22  Only Aigikoreis at the end and Geleontes at the beginning are mentioned with some description of each: the middle two are mere names.

23  Strabo viii. 7. 1, p. 383.

24  The enumeration is put by Euripides in the mouth of Athenaia herself, who appends her own Aigikoreis last, in order to finish the list, “named from my aigis the Aigikoreis will constitute one phylon,” see Ion, 1579 f.

25  The process of “civilising” a barbarous people often begins in all such countries by fighting and conquering. There is, however, the general impression that the old Greek regularly began as a peaceful trader; “and on the beach undid his corded bales”; but we have no proper trustworthy information.

26  Solon, 23.

27  Teleon in Euripides, Ion, and in Stephanus is a mere error (see Pollux viii. 109); but it shows that there was some doubt as to the former. Plutarch, following some good but unspecified authority, has Gedeontes.

28  Gda for Da is probably older and Anatolian, not original Ionian, but received from the Ashkenaz, the people of the inner country.

29  Kühn rightly altered τρία to τέως in the list of Kekrops. Canter, with equal justice, changed Kadeis to Argadeis in the list of Erechtheus. The list of Kekrops contains four names. The sons of Ion under Erechtheus were four, and Argadês or Ergades was the last. Each of the four tribes were divided into three parts.

30  Ergadeis is the form used by Plutarch.

31  Compare, for example, Iliad, i. 607-608, xx. 12.

32  In the great earthquake at Smyrna in 1880, the houses, being built of stone, with a framework of wood embedded in the walls, suffered little; whereas at Scio in 1881 the houses, being mostly built of stone alone, tumbled like a set of card houses. I saw the next day in Scio some large, lofty, modern houses standing hardly shaken, while the poor, old-fashioned houses were a heap of fallen stones. Even in houses built mostly of sun-dried mud bricks in Anatolia there is a good deal of woodwork in those of the better class.

33  Attago is the common form; but Buckler points out to me that the reading Kubeke in Hipponax (which the editor declines to emend) is a good parallel: Kubeke — Kubebe, like Attago — Attabo. There are other Anatolian instances of the variation, Mobolla now Mougla. I have often found some difficulty in distinguishing the Anatolian pronunciation of consonants.

34  Stephanus, s.v. Benna. Unfortunately Stephanus, in condensing the account which he took from Ephoros, makes it rather obscure in some points; but the meaning can be elicited with fair certainty. Probably Stephanus did not really understand the system very well; but he does his best to be accurate, and succeeds quite well.

35  According to a conjecture of bishop Hicks: [Ged or Gel]eontes.

36  See quotations in Historical Geography of Asia Minor (index).

37  A good discussion of these incidents by Hicks (as quoted above on p. 261 f.). The old book by E. Curtius, Beiträge zur Gesch. Und Topographie, written from personal local study, is worthy of much more attention than it has received from those who have not seen the district. Other authorities are given by Hisks, loc. cit.

38  They are all embraced in one photograph published in my Letters to the Seven Churches, Pl. III., facing p. 216.

39  Compare the Sclavonian word bogu, god. The late Byzantine name Phobabindi Hierapolis is for Boga.

40  The termination -ηος is probably a bad form of -ιος or -ειος : many similar cases occur.

41  Labraundis was a tribe at Mylasa in Caria (Hicks).

42  Hicks prefers to connect it with the Emperor Antoninus Pius.