added two fine examples of this language, which is written in an early alphabet akin to the Greek, and which has been thought to date back as far as 800 B.C., though this date is perhaps somewhat too early. He publishes a learned treatise by De Saussure on these texts, but the writer does not attempt to read them. The Phrygians were European Aryans, and the texts found in Phrygia may probably be theirs, though the name does not appear to be recognisable in any known case. There can be no real doubt, when the terminations and other indications of grammar are considered that this dialect is Aryan, and closely connected with-though also distinct from-Greek. In the new examples the name of the Ionians may perhaps be read, and they appear to be mortuary texts set up by the sons or other relatives of the person in whose honour they were carved.*

Intermediate between the period of conquest by Gyges and Croesus and the later age of Alexander, comes that of domination by the Medes and Persians, of which there are many known remains in Asia Minor. M. Chantre has published a number of Persian cuneiform tablets, which have been regarded (for various reasons) as forgeries,† but which may still prove to be genuine. The name Darius, and other well-known words, are clearly legible, and a fine seal from the same region ‡ has been published, which shows the same characters, with the well-known Persian name Mithradata. Why Dr. Sayce regarded this first as 'Hittite,' and then as a forgery, is not clear. The work in the design is far too delicate and original to be the production of a modern forger. But we do not depend on this evidence alone; for the coins of the later Perso-Greek age, when, after the great defeat of Persia at Issus, Greek population swarmed over into Asia, are found in Cappadocia, while further west the Lycian language, written in a distinctive alphabet, was early recognised by Grotefend and Rawlinson as connected

6

*No. 1 may be read, Vasthus the great man, by race an Ionian, akin to Zeus.' No. 2, To Otos, son of Vetes, who was born a citizen of the Ionian city Antioch in Asia, this his sons have inscribed.'

With one exception these tablets appear legible in Persian, though not translated in M. Chantre's volume. Three of them appear to be letters, in the name of 'Darius, King of Kings, son of Hystaspes,' appointing provincial governors; a fourth, also of official character, may prove to be from an Antiochus.

Proceedings of the Society of Biblical Archæology, November 2, 1897, p. 301.

*

with Persian, and it appears to represent the speech of the Medes who, under the elder Harpagus, took Xanthus in 545 B.C.,† the extant texts however being a century and a half later than this date. In Cappadocia the Medes appeared as early as 650 B.C., when King Pharnaspes is mentioned, who married Atossa, sister of Cambyses, the great-grandfather of Cyrus. Strabo mentions Persian gods (Vahman and Amendat and Anahita) in Cappadocia, and the later calendar of this region gives Persian month names. Probably the small bronzes found by M. Chantre, which represent the Indian humped ox, may belong to this age; and near Cæsarea (Mazaca) he collected a great number of coins-both silver and bronze-including, besides those of Antiochus Euergetes, Ormisdas and Tigranes, others of Ariarattes III. and Ariobarzanes I. Such Persian names of the pre-Roman period are very distinct in character from those of the earlier Kati, which recall no Aryan known names at all.

We may glance for a moment at the later Greek and Roman remains of the region, beyond which period M. Chantre's researches do not extend, though the beautiful early Ottoman buildings in Persian style, which Mr. Hogarth has described, show us at a much later period the Mongol Turk restored to power, yet borrowing his civilisation from the Aryans who had, in old days, learned all they knew from his true ancestors the Kati, and from the Phoenicians, long before they came into contact with Babylonia-the old home of both Kati and Phoenician culture. The art of Mycena and Egina was Asiatic, and the Greeks of the latter island city used the same weights which were used in Babylon. But when the current turned again east, and the generals of Alexander set up petty kingdoms in Asia Minor, it was Persian rather than Babylonian civilisation which they encountered. After Eumenes had reigned a few years in Cappadocia a dynasty of Persians, descended from the royal family, succeeded, and lasted from Ariarattes III. for more than two centuries, until the kingdom of Armenia was established by Tigranes, son-in-law of the great Mithridates of Pontus. Thus until Pompey's conquest in 65 B.C., Cappadocia, like Pontus, was more Persian than Greek in culture; and these kings quite possibly continued to use

* See the study of this language in the Journal of the Royal Asiatic Society,' October 1892.

Herod. i. 176.

Persian cuneiform, side by side with the Greek lettering of their coins; for in Babylon the old characters were still employed even after the Christian era. The seal above mentioned might be that of the celebrated Mithridates IV. of Pontus; * and some of the tablets found by M. Chantre may belong to this same period-the fourth and down to the first centuries B.C.

The most remarkable monument of this mingled GrecoPersian style is however found rather further east, at Nimrûd Dagh, above Samosata, near the Euphrates in Armenia. The huge statues built up of masonry, with well-carved heads the heads alone being six feet high-are Persian in character. The accompanying bas-reliefs are partly Persian, partly in ill-copied Greek style. The Antiochus of Commagene, who built this shrine shortly before Pompey's conquest, calls himself a 'lover of Greece,' and writes his long inscriptions in Greek. His family was an offshoot of the great house of Seleucus; but among the gods whom he invokes in Greek speech are Ahura-mazda and Mithra of Persia, whom he worshipped side by side with Zeus.

The Romans in Western Asia generally, accepted the Greek civilisation which preceded them. The language of literature and of trade alike was Greek, and the number of Latin inscriptions is very small as compared with the Greek official, religious, or funerary texts of the Roman age in Asia Minor or in Syria. Roman architecture equally borrows from Greek in these regions, and the origin of the division of the Roman Empire is found in the essentially Greek character which the Asiatic provinces maintained. But the various elements of population were, as we have seen, numerous, and the old worship of Ma continued in Roman times. Strabo describes her sanctuary at Comana in Cappadocia, and the institution of the temple girls,' which was probably of Mongol origin-a peculiar morality common among Turanian races and still existing in Japan, of which we read not only in Babylonia and Phoenicia, but also in Lydia † and Etruria.

6

As time went on new elements of race appeared among the early Christians (of whom a few texts are found in Phrygia, besides many that are doubtful) and the Jews, of whom also inscriptions in Greek have been discovered by

About 240-190 B.C. The text reads probably Mithradata Gtrtu, 'Mithridates the Fourth.'

Herod. xii. 13.

Ramsay. The inscription of Asbolus (Yashub-el), who left money for 'burning the papoi on the customary day,' belongs to about the third century A.D. The name is Jewish, and Asbolus belonged to a guild of dyers-a common Jewish trade. The papoi' were perhaps woollen' objects burnt at the tomb-a custom still preserved by Jews in the East; and the supposed allusion in this and in other texts of this region to Christian societies appears to be unfounded, though there is no doubt of the very early spread of Christianity in Cappadocia, Phrygia, and Pontus.

The Semitic race existed, as we have seen, in Cappadocia by 2000 B.C. In the fifteenth century B.C. the Princes of Elishah (or Cilicia) wrote to Egypt in a Semitic dialect. The coins of Tarsus at a later period bear Phoenician texts, and the same race is believed by Renan and others to have still existed in this south part of Asia Minor in the days of the Apostles. This, it may be noted, casts a remarkable light on a passage in the New Testament* which has not so far been explained. At Lystra, north-west of Tarsus, the natives, who took the Apostles for gods, called Barnabas Dios,' and Paul Hermes,' in the speech of Lycaonia.' They apparently misunderstood the Apostles' names, and rendered the first Bar-Nebo, son of the god Nebo,' while Paul became Bul-the name of a Phoenician deity which frequently occurs on Palmyrene texts of the same century. The population of these regions-Semitic, Aryan, and Mongol-Jews, Greeks, Persians, and aborigines, was as mixed in Paul's time as it is now, when the Aryan Greeks and Armenians live among Jews, Turks, Tartars, and Mongol Lazis from the Caucasus, as described by M. Chantre.

The latest ruins which this explorer describes belong to the Roman age, including the fine baths of Saravena-probably of Justinian's time-and the earlier temples, palaces, and theatres of Comana. It may be noted in passing that some of the smaller buildings, which he calls temples, are clearly family tombs, such as are found also in Syria with texts showing them to have been erected during the life of the owner.† M. Chantre himself found funerary texts in the examples which he describes, one erected by the senate and people of Cæsarea Hieropolis (the older Mazaca), to Marcus Ulpius Valerius Cornelius, and another for himself and his wife Euphemia, by a certain Eutyches. Both

*Acts xiv. 12.

+ See Edinburgh Review,' No. 371, January 1895, p. 216.

of these are in Greek, though the name in the first case is Roman.

Our author's conclusions from his evidence are sensible, if not as definite as the translations of the Kati texts allow. He supposes the Hittite' monuments to be older than Jensen and others have thought, and to belong to an age earlier than that of the Egyptian conquest of Syria (about 1600 B.C.). He concludes that this race came from the East, and even suggests that Sargon of Ur was their leader. This, of course, is speculative, as we do not know exactly how far Sargon I. pushed his conquests, though he is said to have reached the Mediterranean. The later Babylonians believed him to have lived about 3800 B.C., and he probably belonged to the Akkadian race. By about 2000 B.C. he was deified, and is mentioned as a god on Semitic seals, and bricks, and bas-reliefs of that age. M. Chantre quotes the opinion of Col. Conder that this race was of Mongol stock, and attributes to Dr. Sayce the discovery that they were not Semitic, which had, however, been shown as early as 1866 by Chabas. He considers the peculiar type of their features to be still observable in Armenia, which is only natural as, in addition to the Turkish peasantry, he himself speaks of the Lazis and other Tartars whom he met. The Armenians themselves have probably a strong infusion of the old Mongol blood in their veins, which accounts for their using the Turkish instead of the Armenian language. There is much evidence on this question which he does not mention, and the symbolism of the monuments, the physical type and dress of the figures on bas-reliefs and seals, the language, and the history, alike clearly connect the aboriginal civilisation of Cappadocia with that of the Kassites and Akkadians. The evidence is constantly increasing, and even since the publication of Col. Conder's recent work on the subject* several new texts have been published, including a fine cylinder signet of very Babylonian design but with a very clear Hittite inscription. It represents (as do others) the sacrifice of a king by a stream, while a pillar has been erected behind the worshipper. Another bas-relief, from near Malatiya in Armenia, again gives a picture of sacrifice by two persons, with a text in three lines. Thus, including the new discoveries of M. Chantre, we possess some forty texts of various lengths, and upwards of fifty seals, with Hittite inscriptions. They occur as far east as Nineveh, and as far west as the

The Hittites and their Language. London: Blackwood, 1898.