By W. M. Ramsay
Language and Letters in North Galatia
IT has been shown that the Gaulish tribes, when they entered the land which took from them the name Galatia, found there a much more numerous population amid which they settled as a ruling aristocracy, and thus formed a distinct country and government, recognised by the surrounding governments as one of the powers among whom Asia Minor was divided.
At first the two sections, which composed the population of this new country, Galatia, spoke two separate languages. The aristocracy spoke a Celtic tongue. Of the populace, presumably some few could speak Greek, but Phrygian was the sole tongue generally known, and even those who knew Greek must also have spoken Phrygian. There seems to be no reasonable doubt on these points, though no actual evidence remains on the subject.
The problem is to determine what was the fate of these languages. It is certain that at last Greek came to be the one sole language used in Galatia; but the dates at which Celtic and Phrygian ceased to be spoken are unknown, and form the subject of the present investigation.
The subject has been briefly discussed by a distinguished French scholar and traveller, M. Georges Perrot. But he has not taken into account all the conditions of the problem, and subsequent exploration has added considerably to the scanty stock of evidence available to him. As his authority and arguments have convinced many recent scholars — though Mommsen unhesitatingly and decisively rejects them — it will be best to begin by briefly stating his reasons, and showing why they must be pronounced inadequate to support his conclusion, that before the time of Christ the Celtic language had ceased to be spoken in Galatia, and Greek had become the sole language of the country.
It will be observed that he leaves out of sight one factor. He does not take into consideration the Phrygian language. He speaks as if the struggle had been only between Greek and Celtic.
The omission is due to that singular prepossession in the minds of almost all scholars — except Mommsen — who have touched this subject: they all speak and reason as if Galatia had been inhabited by Gauls only. If occasionally some one, like Lightfoot, p. 9, refers to the Phrygian element in the population, he forthwith dismisses it again from his thought and his argument. Mommsen alone declares positively and emphatically that the Galatian people must be regarded as a mixed race, in which the tone and spirit was given by the Gaulish element.
Though it cannot be proved, yet we must regard it as probable, that the Celtic language became the common tongue of the mixed race. The impressionable Phrygian population, devoid of energy, yielding readily to the force of circumstances, accepted the language of the conquerors,1 just as of old that older race which had been conquered by the Phryges adopted the speech of their rulers. The Phrygians of Galatia, though far more numerous, contributed much less to the prominent characteristics of the mixed race: they gave their religion and their manual labour in some of the simpler and more fundamental arts of life.
Thus M. Perrot’s first assumption may be accepted as probably correct. In the century before Christ the battle of tongues in Galatia was between Celtic and Greek.
His next argument is founded on the supposed fact that the ancient Lydian and Phrygian languages had died out before the time of Strabo, about A.D. 19, so that “in the whole country from the Sangarios to the sea nothing but Greek was spoken”. That supposition is incorrect. Strabo, XIII 4, 17, is quoted as the authority; but Strabo’s words do not imply that. Strabo does not mention the Phrygian language: he says that the Lydian language had ceased to be spoken in Lydia and was used only in Cibyra, a city in the south-west corner of Phrygia, which contained a Lydian colony.
Epigraphic discovery has now proved that the Phrygian language was known in various parts of central and eastern Phrygia at least as late as the third century after Christ. Some of the Phrygian inscriptions of the Roman period were published before M. Perrot wrote, but had not yet been identified as Phrygian.2 Their number has now been much increased. One is bilingual, a Greek and Phrygian epitaph. Two are longer, untranslated documents. The rest contain only a concluding formula in Phrygian, while the body of the inscription is in Greek: the Phrygian formula is a curse on the violator of the grave, and there seems to have been an idea that this appeal to Divine power was more efficacious in the old religious speech. The formula varies so much as to show that it was expressed in a living language, and was not merely a repetition of an ancient hieratic form of words.
Moreover, the exceeding badness of the Greek in some inscriptions found in Phrygia proves that they were written by persons who were almost utterly ignorant of the language. They were composed by uneducated rustics, who had only a smattering of Greek, and who ordinarily spoke in another tongue.3
In fact, it is no longer a matter of doubt that the native languages of Phrygia, Pisidia, Lycia, Lycaonia, Cappadocia,4 etc., persisted in common use far longer than was believed. It was only in the cities that Greek was much used, while the rustic population continued to speak their own native languages.
Thus, in place of the argument that, since Phrygian had been forgotten in Phrygia before A.D. 19, Celtic probably had been forgotten in Galatia, we must substitute the exact opposite. Since Phrygian was still spoken in Phrygia in the third century after Christ or later, Celtic might be expected to persist in Galatia at least as long, inasmuch as Galatia was distinctly less open to Hellenic influence than Phrygia, and the Galatian people had much stronger national pride than the Phrygians.
Again, it was argued that no Galatian inscriptions in the Celtic language remain, and therefore the Celtic language could not have been spoken in Galatia.
This argument would serve equally well to prove that Greek was spoken universally in Isauria, Lycaonia, Cappadocia, Pontus, etc. Strabo says that in Cibyra four languages were spoken in his time; yet not a trace of any tongue except Greek occurs in the inscriptions of Cibyra. Are we to conclude that Strabo was wrong, and that only Greek was known there?
In truth, that line of argument is founded on a misconception as to the facts of society in Asia Minor, and has no force. Fashion was powerful. It was thought rude, barbarous and uncultured to use any language but Greek. All persons that had even a smattering of Greek aired their knowledge of the educated speech. Moreover, it is highly probable that nobody who was ignorant of Greek was able to write: those who got any education at all learned Greek, and hardly anybody in Asia Minor wrote in any language except Greek. The thirty or forty late Phrygian inscriptions mentioned above are the only exception, and they have mostly a special character.
The dearth of Celtic inscriptions in Galatia only shows that Celtic was not the educated speech of the country — a fact which we know independently. Few inscriptions in Galatia are older than the second century after Christ; the epigraphic evidence tends to prove that the mass of the people were only beginning in that century to think of engraving epitaphs on the tombs of their dead.
As to the natural probabilities of the case, there is no doubt that the Roman influence was on the side of Greek. While Rome favoured the Galatic spirit in many respects, it never seems to have admitted the Celtic tongue in official matters. Greek, the language of education, found full official recognition, and Rome made no attempt to force Latin on the eastern Provinces; but it admitted no third language. Those who wished to make full use of the opportunities of the Empire must speak either Latin or Greek. All whose knowledge was confined to some other tongue were barbarians and outsiders. The civilisation that Rome sought to impress on the East was Graeco-Roman; and the constitution of the Roman Province would naturally exert a powerful influence in forcing a knowledge of Greek upon all that sought honours and official employment, if they did not know it beforehand.
Even under the kings Deiotaros and Amyntas, before the Province was constituted, Greek must have been much used in diplomacy and foreign affairs. Greek at that time filled a place like what French filled no long time ago in Europe, as the international and diplomatic tongue. But Greek was more than that: it was the speech of education and of all educated men (like Latin in the Middle Ages): it was the language in whose literature almost all scientific and artistic knowledge was locked up. No Galatian could play a part in the extra-Galatian world without Greek. There is no doubt that Cicero and Deiotaros5 conversed in Greek. Coins struck in Galatia bore Greek legends; coins with Celtic legends could never have found international currency in Asia Minor at that time, as any numismatist will testify. In many such ways Greek was a necessity in Galatia.
But those facts do not prove that the Celtic language was unknown: they prove nothing as regards the speech of the uneducated mass of the population, and they prove nothing about home and family intercourse. They only show that Greek must have been familiar to the few: they do not show that it was used by the many. The strong Celtic tinge in certain respects, which indubitably coloured the Galatian State, could hardly have maintained itself so long amid the just and even tenor of Roman imperial rule, without a national language to support it.
We have more than this general presumption to trust to. There is distinct evidence to prove that Celtic was still spoken during the second century in Galatia. Both Mommsen and Mitteis6 are fully convinced by the evidence on this point.
About the middle of the second century after Christ Pausanias7 speaks of a native, non-Greek language, actually spoken in Galatia: “the shrub which the lonians and the rest of the Greeks call kokkos, and which the Galatians above Phrygia call in their native tongue hus”. This native tongue can only be Celtic. It is not possible here to plead that Pausanius is speaking on the authority of some old book, and passing off borrowed information about the past as his own true knowledge about the present. A few pages before he mentions a fact which he had learned in that way regarding the cavalry of the Gaulish invaders, and there he puts it in a different way: “this organisation they called trimarkisia in their own tongue”.8 Moreover, his statement about the Galatians of Pessinus9 is couched in a form suggesting personal knowledge; and he had been in the sanctuary of Zeus at Ancyra, I 4, 5.
A trace pointing to the persistence of the Celtic language in Galatia about the middle of the second century after Christ, is found in Lucian.10 When the false prophet Alexander was in repute at Abonouteichos on the Pontic coast, persons came to visit him from the countries round, Bithynia, Galatia,11 and Thrace. Occasionally questions were propounded to him by barbarians in the Syrian or the Celtic language: in such cases he had to wait until he could find some visitor able to interpret the question to him, and occasionally a considerable interval elapsed between the propounding of the question and the issuing of the reply, if a translator was not readily found. It is not necessary to understand that all questions in Celtic had to wait long for an interpreter: it was probably easier to find an interpreter in Celtic than in Syriac. But even if it were sometimes the case that Celtic interpreters were difficult to find, that would only prove that some of the Galatic visitors could not speak Celtic, while others could. But that might happen naturally. Most of those who came from Galatia, especially at first, would be traders and travellers, classes of persons who must have picked up in a rough way a good deal of education. The language of trade was, beyond all question, Greek throughout those regions; and those who were engaged in trade (many, of course, hereditarily), would be likely to be the most thoroughly Hellenised of the Galatians. Thus, there might be cases when an interpreter of a Celtic question was not readily found among Galatian merchants at Abonouteichos.
Such seems the natural explanation. The propounders of questions in Syrian or Celtic are called “barbarians” by Lucian; but that does not prove them to have been from regions outside the Roman Empire. Any one who spoke any language but Greek (or Latin) was called by the Greeks a barbarian; so, e.g., the people of Malta are called by Luke, although Malta had belonged to Rome for about 270 years when Luke visited it. Probably some of the questions were propounded in barbarian tongues merely for the purpose of testing Alexander’s skill, for the tendency to test even that in which one believes lies deep in human nature. Hence we need not suppose that those who put questions in Celtic were all ignorant of Greek.
Again, in the fourth century the witness of Jerome is emphatic — the Galatians spoke the universal language of the East, Greek, but they also spoke a dialect slightly varying from that used in Gaul by the Treveri. This clear testimony by a man who had travelled in Galatia and among the Treveri cannot be twisted and perverted (as Lucian and Pausanias are by some writers). There is therefore only one method: when testimony is dead against you, you can always refuse to believe it. And so Jerome is set aside, without any reason given that can stand a moment’s investigation.
But the old plain and simple method of disbelieving all that contradicts one’s prepossessions is now becoming discredited as belonging to the Dark Age of modem scholarship. The one argument which used to be counted sufficient — that Jerome was a Christian, and that anything stated in a Christian work is suspicious — is now no longer implicitly accepted.
Mitteis pronounces no decision on this point: it is not necessary for his purpose. Mommsen accepts Jerome’s testimony, and justifies it by solid reasons; and the voice of healthy historical criticism will assuredly be on his side.
That the Galatian people was bilingual for centuries is an interesting, but well-ascertained fact. Compare the Welsh in modern times after many centuries of English rule.
Now, as to the date when Greek spread most among them, the evidence is far from satisfactory.
Almost the only evidence comes from the reception of Greek names in Galatia; Already in the third and second centuries Gauls with Greek names occur: Apatourios B.C. 223, Lysimachus 217, Paidopolites 180. At that time the Gauls were serving as mercenaries in various camps, and their leaders must have found it convenient to use Greek names. Probably Apatourios and Lysimachus had two names, Celtic and Greek, according to a widespread custom in districts where a smattering of Greek was spread: it was convenient to have a Greek name amid Greek surroundings, and a native name amid the surroundings of home. But no evidence exists, and in fact Galatia is almost the only country of that kind in which no explicit proof of the use of alternative or double names has been found (though in all probability they were used).
This use of Greek names, beginning so early, taken in conjunction with intermarriages, might have been expected to have spread very widely in the second and first centuries. But, as we saw on p. 66, the tendency to adopt Greek ways was checked, and a strong reaction of the Gaulish spirit occurred in the second century. The anti- Hellenic tendency was strengthened by the Mithridatic Wars (in which Hellenism rallied to the Oriental king against Rome and the Galatian tribes), and by the subsequent Romanisation of Galatia under Deiotaros. The almost exclusive use of Celtic names in the ruling families, B.C. 90-40, proves that the national feeling was still strong against Hellenisation. Many names are known in the three tetrarchic dynasties, and almost all are Celtic. There is, however, one notable exception.
Amyntas bears a Greek, especially a Macedonian name. At this time the great Galatic families seem to have used Gaulish names almost exclusively.12 Was Amyntas, then, a Greek?13 This is highly improbable, because it would have been difficult for a Greek to govern the Galatian aristocracy, and Augustus was too politic to offend a strong national feeling. Moreover, Dion Cassius calls him Amyntas the Galatian.14
Now, it is probable that Amyntas did not belong to one of the great ruling families. He had been secretary to Deiotaros, and his selection for that office implies that he had not merely natural ability, but also considerable education; and the educated classes always tended to use Greek names. Very probably Amyntas had a Celtic name also; but in his relations with his South Galatian subjects and with foreign nations he would use the name which marked him as of the educated class.
Similarly, of the four envoys sent by Deiotaros to Rome in B.C. 45 three bear Greek names;15 it is, however, not certain that all were Gauls; the king might have found some convenient tools among the Greeks. His physician, Pheidippos, was of course a Greek,
M. Perrot, in a lucid survey of the evidence, fixes on the year A.D. 10 as about the decisive turn in the tide of naming.16 Henceforward Celtic names are exceptional, and Greek or Latin names are customary. On this quite correct result two remarks are to be made.
In the first place, the disuse of Celtic names was not so complete as it is said by some writers to have been. In Ancyra, the centre of Galatian civilisation, they might be expected to disappear most rapidly; but even there we find in M. Perrot’s inscriptions of the second century the following names, certainly or probably Celtic:17 133 Epona, 123 [Kau?]aros, Borianus, Mamus, Barbillus, An[. . .]natus; and in a rural district, 151 Masclus.18 In the only rustic part of Galatia where inscriptions have been found in appreciable number, the following Celtic names occur (all probably second century A.D. or later): Vastex, Barbollas, Meliginna, Zmerton, Leitognaos, Dobedon. A short inscription of Laodiceia Combusta19 (third or fourth century), with the names Kat[t]oios and Droumamaris, probably shows a Celtic family in that Lycaonian or Galatic city. These specimens out of a larger number known will suffice: they are taken from the first two sources that suggested themselves.
Secondly, it is hardly correct to say as some do, that native names lingered far longer than the native languages in Asia Minor. That is true where a language dies out in presence of the speech of a more energetic section of the population (as Phrygian did in Galatia): in such cases, as M. Perrot says, on sait que les noms propres survivent en général aux noms communs, qu’ils restent comme le dernier vestige d’une langue sortie de l’usage. This rule is perhaps true in a sense in Asia Minor, but it is far from expressing the whole truth. It is also true, and a more vital point in the present question, that proper names began to be disused, and Greek names came into wide use, centuries before the native language disappeared. The very persons who inscribed Phrygian formulae on their graves20 bore Greek, not Phrygian names.
The disappearance of names not Greek or Roman in Asia Minor is too large a topic for our pages: it is only part of a much wider subject. The fact is that at this period and throughout the Empire, the old national names were everywhere discouraged by the prevailing tone of society, which was Graeco-Roman in the East, and Roman in the West. It was generally esteemed barbarous, rustic, the mark of a mere clown, to bear a native name: as the comic poet of an older time said: “It is a shame for a woman to have a Phrygian name”.21 The aristocratic feeling of the old Gaulish families made them cling for a time to the hereditary names; but the fashionable tone was too strong for them.
In the dearth of inscriptions — itself a proof of illiteracy — authorities for Galatian names are so few that the argument resting on them is feeble; but so far as it goes it is that the early Roman period was the time when Celtic names passed out of fashion; and the change heralded a marked increase in the use of the Greek language.
As to any literary interests in Galatia, not a sign is quoted earlier than the fourth century. Galatia like Cappadocia is a blank in literature; and those are the two countries in which fewest cities (in the strict Greek sense) existed.22
The evidence is overwhelming. About A.D. 50 Galatia was essentially un-Hellenic.23 Roman ideas were there super-induced directly on a Galatian system, which had passed through no intermediate stage of transformation to the Hellenic type. It was only through the gradual slow spread under Roman rule of a uniform Graeco-Roman civilisation over the East that Galatia began during the second century after Christ to assume a veneer of Hellenism in its later form.
Road-building in North Galatia seems to have begun under Vespasian, when Galatia was united to Cappadocia as a frontier and military Province. The only Roman colony was probably founded by Domitian. It was during the first century one of the least civilised corners of the Empire, remote, difficult of access, with little trade, lying apart from the world, with a strongly marked character of its own. As Mommsen with his unerring historic instinct long ago recognised, it had become a Celtic island amid the waves of the Oriental races, and remained so in its internal organisation even in the Roman Imperial period.24 . . . In spite of their sojourn of several hundred years in Asia Minor, a deep gulf still separated these Occidentals from the Asiatics (among whom the Greeks of Asia Minor must for some purposes25 be counted). The strong mutual dislike that kept the Asiatic Greeks and the Galatians apart is evident from the time of Mithridates onwards: at that time Galatians and Romans faced and conquered the Graeco-Asiatic reaction.
The dislike of the Asiatic for the northern barbarians may be paralleled at the present day by the hatred of the Turkish inhabitants of the same country for the Circassian immigrants, who resemble in many respects the picture that is drawn for us of the Gauls, free, proud, rapacious, unruly, a terror to their more peaceful and submissive neighbours. Every traveller in Asia Minor, who has come to know anything of the feelings and life of the people even in the most superficial way, learns that the Mohammedan Turk hates the Mohammedan Circassians far more than he dislikes his Christian neighbours; and his hatred is rooted in fear. So the Gauls were hated in ancient Asia Minor.
This hatred lasted late; and one observes its effects, in the fourth century, in the jealousy and contempt expressed for the Galatians by the Cappadocians. Thus Basil, Epist. 207, I, speaks with marked innuendo of Sabellius the Lydian and Marcellus the Galatian. Gregory of Nyssa, Epist. 20, mentions that the garden Vanota, where he writes, was called by a Galatian name, but deserved a name more in accordance with its beauty than a mere Galatian word. And the heretic Eunomios complained, as of an insult, that Basil had called him a Galatian, whereas he was a Cappadocian of Oltiseris.26
In view of these facts every one who considers how closely the writings of Paul and the other Apostles (so far as we know) keep to actual life, how vivid and realistic are their pictures of the Churches which they address — every such scholar must expect that, in a letter written by Paul to a group of North Galatian churches, there should be found touches which bring before us the special character and position of these churches. He must expect that the address would throw light on, and receive illustration from, the peculiar position of the Galatians, so distinct and apart from the type and tone of all the surrounding races, whether Greek or Anatolian.
This expectation is not realised. On the contrary, there are only three points in the Epistle that have ever been alleged as signs of Gallic character.
One is the stock joke, that the Galatian Christians changed their form of belief, and the French are a fickle people. It is surprising that such a sane and clear-headed scholar as Lightfoot should have repeated this from his predecessors. In truth, he was here misled by his own historic instinct: he felt that, if the North Galatian theory was true, there must be traces of Celtic character in the Epistle, and as he would not abandon the theory he must find the traces.
The sufficient and only reply is to quote Luther’s arguments that the Galatians must have been a Germanic race, because the Germans are fickle. As a matter of fact, Paul nowhere calls the Galatians fickle, or implies that their change of faith was caused by fickleness: see p. 255.
The second is that among the sins against which Paul warns his Galatian correspondents are “drunkenness and revellings,” “strife and vainglory,” and that he charges them with niggardliness in giving alms: it is said that these are characteristic vices of the Celtic character. They are only too characteristic of most nations and most Churches. On their nature in Galatia, see p. 450 ff, 458 f.
The third is that the Celtic people were superstitious and “given over to ritual observances,” and Deiotaros was characterised by “extravagant devotion to augury: the Gauls in Galatia would find the external rites of the worship of Cybele attractive from their analogy to their own Druidic ritual,” though “the mystic element in the Phrygian worship awoke no corresponding echo in the Gaul”. Hence, it is argued, the Galatians were likely to fly from Pauline to Judaistic Christianity.
One can only marvel at this pedantic analysis of Galatian character. It is hardly worth while to point out that the best authorities consider Druidism a very late fact in Gallic history, and that scholars who study Galatia observe that not a trace of Druidic religion can be discovered there. The superstition of the Galatians amounts to this, that they had adopted the religion of Asia Minor !
The truth is that, though North Galatia had a peculiar and strongly marked character, not the slightest reference to its special character can be found in the Epistle. Yet the Epistle is full of references to the circumstances and everyday surroundings of the persons addressed — full even to a degree beyond Paul’s custom.
Note. — It may be here added that, in the article Galatia in Hastings’ Dictionary of the Bible, I have gone too far in admitting Hellenic influence in North Galatia, being overanxious not to colour favourably to my own theory an account which ought to be strictly impartial. But in that article the term “Graecised city,” applied to Ancyra, is intended to indicate “Greek-speaking,” and not “Hellenised”.
 But perhaps on the southern frontier near Kinna Phrygian was still spoken in the Roman time: one example of the Phrygian formula (see below) occurs there, Journ. of Hell. Stud., 1899, p. 119, no. 117.
 See Phrygian Inscr. of the Roman Period in Zeitschrift f. Vergleich. Sprachforsch., 1887, p. 381 ff. Literature of the subject quoted by Anderson, Journ. of Hell. Stud., 1899 (second half).
 Cities and Bish. of Phrygia, I, p. 131.
 See Mommsen, Röm. Gesch., V, pp. 92, 315. On Pisidian, the present writer’s Inscriptions en Langue Pisidienne in Revue des Univ. du Midi, 1895, p. 353 ff. On Lycaonian, Acts XIV 11.
 See p. 92.
 Mommsen, Röm. Gesch., V, p. 314; Mitteis, Reichsrecht, etc., p. 24.
 X 36, I: Frazer’s translation is quoted: his note endorses this obvious interpretation.
 X 19, II.
 Quoted on p. 85.
 Alexander Pseudomantis, 51. The attempt to explain away this evidence, Revue Celtique, I, p. 179 ff, is a failure.
 Pontic intercourse, see p. 143.
 Kastor is an exception (yet Holder gives Castoriacum as a Celtic city).
 Van Gelder, p. 200, thinks he was a Greek.
 Dion, L 13, 8, Ἀμύντας ὁ Γαλάτης. Compare Plutarch, Amat. 22, τῷ Γαλάτῃ simply, when speaking of the Tetrarch Sinorix.
 (Van Gelder, p. 200, says that all the names are Greek; but Blesamios is obviously Celtic.)
 Perrot, de Gal. Prov. Rom., p. 78, 89 f.
 Evidence in Holder passim.
 Anderson in Journal of Hellenic Studies, 1899, p. 81 ff.
 Athen. Mittheil, 1888, p. 266.
 See p. 149 f.
 See p. 30.
 See p. 135 f; Strabo, p. 537, says there were only two cities in. Cappadocia.
 On the talk about evidence to the contrary, see p. 173.
 Colonia Julia Augusta Felix Germa was not founded by Augustus (see Mommsen’s commentary on his colonies in Monumentum Ancyr., p. 120): Domitian named it after his beloved Julia Augusta, see Revue Numismatique, 1894, p. 170.
 In certain ways, of course, Greeks are Occidental as contrasted with Asiatics.
 Greg. Nyss. contra Eunomium, pp. 259, 281.