STOP Early Journal Content on JSTOR, Free to Anyone in the World This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in the world by JSTOR. Known as the Early Journal Content, this set of works include research articles, news, letters, and other writings published in more than 200 of the oldest leading academic journals. The works date from the mid-seventeenth to the early twentieth centuries. We encourage people to read and share the Early Journal Content openly and to tell others that this resource exists. People may post this content online or redistribute in any way for non-commercial purposes. Read more about Early Journal Content at http://about.jstor.org/participate-jstor/individuals/early- journal-content . JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people discover, use, and build upon a wide range of content through a powerful research and teaching platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please contact support@jstor.org. Classical Philology Vol. VI October, IQII No. 4 KELTS AND LIGURIANS By Richabd Wellington Husband In my paper entitled "Race Mixture in Early Rome" {Trans. Amer. Phil. Assoc. XL, 1910, pp. 63-81) I held in connection with the extant Ligurian inscriptions that "by far the greater part of what is called Ligurian is strictly Gallic, and what is not Gallic is not Indo-European. The language of the country of the Ligurians became largely Gallic, after the coming of Gallic tribes to Italy." It was also suggested that a distinction should be made among the Gallic tribes migrating to Italy, and that those tribes which preserved original ku in the form qu became located in the western and coast regions of Liguria, while those which changed original ku to p settled in Italy farther to the north and east. In the present paper I shall give what evidence is available to determine the time of these migrations, and the routes taken by the migrating tribes. By this means I hope to show that the statements and assumptions made in the earlier paper had a genuine foundation in evidence. The fullest and most systematic account we have of the coming of the Gauls into Italy is that of Livy (v. 34r-35. 3), whose story runs as follows: In the reign of Tarquinius Priscus at Rome, prior to 578 b.c, the Bituriges held sway over a third part of Gaul, called Celticum. Their king Ambigatus, seeing that the country could no longer support the increasing population, sent out two hordes to find new homes under the guidance of his two nephews Segovesus and Bellovesus. The latter set out for Italy, accompanied by large [Classical Philology VI, October, 1911] 385 386 Richard Wellington Husband numbers of the tribes of the Bituriges, Arverni, Senones, Aedui, Ambarri, Carnutes, and Aulerci. While delaying among the Tri- castini, in despair of being able to cross the Alps, they learned that some Phocaeans had just landed near the mouth of the Rhone, where they were attempting to found a city, later Massilia. In this the Phocaeans were opposed by a neighboring tribe of Ligurians, the Saluvii, who were overcome by the Phocaeans with the aid of the horde of wandering Gauls. Thereupon the Gauls crossed the Alps and made their way down the Duria through the Ligurian tribe of the Taurini. Not far from the river Ticinus they encountered the Etruscans, who were then in possession of the valley of the Po, and defeated them. Here they heard that the name of the district was something that to them sounded like Insubres, the name of a canton of the Aedui, and taking this as a good omen, they founded there the city of Mediolanium. Later came another migration, of Cenomani, who settled farther east. These came by the same route as the first swarm. They were followed closely by two Ligurian tribes, the Libui and the Saluvii. A third host, composed of Boii and Lingones, entered Italy by the Pennine range, therefore passing through the Ligurian tribe of the Salassi, turned east no doubt through the territory of the Lepontii, and finally settled to the south of the Po. Finally the Senones came, and settled still farther south in the country of the Umbrians, from the river Utis to the Aesis, along the Adriatic. (In giving this summary I have adopted the text of Weissenborn in his annotated edition.) It is important to note the location of these tribes in Gaul prior to their migration. (In general, see Desjardins La Geogr. de la Gaule Rom. II, 462-97 ; Kiepert Lehrb. d. alien Geogr. 444, 446) . The Bituri- ges were divided into two sections. One branch, called the Bituriges Vivisci, dwelt near the mouth of the Garumna (Pliny NH. iv. 108; Strabo iv, p. 190; Ptolemy Geogr. ii. 7, 8), and the other, the Bituriges Cubi, west of the upper part of the Liger (Pliny iv. 109; Strabo iv, p. 191 ; Ptolemy iii. 7, 13). As the other tribes moving toward Italy surrounded the Bituriges Cubi almost to the extent of a semicircle, it cannot be doubted that these are the Bituriges referred to by Livy. We have no information except from this passage in Livy that the Bituriges had ever been in control of the central part of Gaul, and Kelts and Ligurians 387 not the slightest evidence for the belief of D'Arbois de Jubainville (Les premiers habitants de VEurope II, 297-305) that they were in possession of a large continental Keltic empire. In the time of Caesar they were subject to the Aedui (BG. vii. 5. 2), but in the gen- eral revolt of the year 52 b.c. headed by the Arverni under Vercin- getorix, the Bituriges took a prominent part (BG. vii. 5. 7; 8. 6). Twenty of their cities were burned to prevent the advance of the Romans (BG. vii. 15. 1). Their greatest city, Avaricum, which is described as the most beautiful city in Gaul (BG. vii. 15. 4), was made the object of Caesar's attack (BG. vii. 14-28). It would appear, therefore, that notwithstanding their subjection to the Aedui they were still a vigorous and nourishing tribe. The Aulerci were divided into three pagi, the Aulerci Brannovices (Caes. BG. vii. 15. 2), the Aulerci Eburovices (BG. hi. 17. 3; vii. 75. 4), and the Aulerci Cenomani (BG. vii. 75. 3; Pliny NH. iv. 107; Ptol. Geogr. ii. 8. 9). Caesar uses the word Aulerci by itself four times, but he does not use the names of the pagi without prefixing the general name of the tribe. The Cenomani, the largest of the three, became known, especially in Italy, only by their cantonal name. Thus Livy probably means that in the earliest migration people of the two smaller cantons came to Italy, while later the third came in a movement by itself. Livy does not here mention the nationality of the Saluvii, but it is clear from Epitome LX that he believed them to be Gauls, Fulvius Flaccus primus Transalpinos Ligures domuit hello, missus in auxilium Massiliensibus adversus Saluvios Gallos, qui fines Massi- liensium populabantur. But Pliny (NH. iii. 47) includes them among the Ligurum celeberrimi ultra Alpes, and that the Romans considered the Saluvii to be Ligurians is shown by the triumphal inscription of Fulvius and Sextius, CIL. I, 460, DE:LIGURIB:VOCONTIEIS: SALVVIESQ. However, Strabo in three passages contrasts the Salu- vii with the Ligurians in such a way as to indicate that he regarded them as Gauls. Thus he says : i? e<£e£»j? irapaXia, yv exovaiv oi re Maao-aXtarai ical oi 2,d\ve$ l^expi Aiyvmv iirl ra irpbs 'lraXiav ftept) ical tov Ovappov irorapov (iv, p. 178) ; and oi Maao-aXicoTai .... irapeSocrav . . . . to 8e Tavpoevnov ical ttjv 'OXBiav ical 'AvtittoXiv ical Niiccuav ra> t5>v SaXucoK eOvei ical Tot? Aiv VTrepaXireiwv K.e\ra>v ) ttoXvv xpovov 7roXefirfcravTe<; ical tovtok ical tois Aiyvaiv (iv, p. 203). Strabo explains why he calls the Saluvii (or Salyes as the Greeks called them) Kelts, for he says: icaXovai Be tov<; 2a\iw ol fiev iraXaiol to>v 'EXXtfvcov Aiyvas ical ttjv ftdipav, rjv eyovcnv ol Macro-aXtwTai, Aiyvo-Tucrjv, ol 8' varepov KeXroXt'yi'a? opofid- £ovo-i, teal rrjv fie^pi 'Aoveviavos /cal tov 'PoBavov ireSidBa tovtok irpoave/iovaiv (iv, p. 203). Livy also contrasts them with the Ligurians, et P. Cornelius .... profectus ab urbe sexaginta longis navibus praeter oram Etruriae Ligurumque et inde Saluvium montis pervenit Massiliam (xxi. 26. 3). From these several passages Long draws the following reasonable conclusion: "This shows that the Ligurians of Gallia, or the country west of the Var, became known to the Roman by the name of Salyes. Strabo's remark that these Salyes, whom the early Greeks named Ligures, were called Celtoligyes by the later Greeks may explain how Livy or his epitomizer has called the Salyes both Ligurians ('Transalpinos Ligures,' Epit. 47) and Galli (Epit. 60). They were a mixed race of Galli and Ligures" (Smith Did. of Class. Geogr., s.v. " Salyes") . D'Arbois de Jubainville breaks away decidedly from the belief of the Romans, and certainly misinterprets Strabo, in holding that the original Ligurian inhabitants of this territory were conquered by the Gallic Salyes, and that the word Celto-Ligurian indicates a condition of Gallic supremacy on Ligurian ground (Les premiers habitants I, p. 373). If the early Greeks called the Salyes Ligurians, it must mean that at the time of the founding of Massilia there was such a tribe of Ligurians in the vicinity. But with the first southern movement of Gallic tribes, as the Allobroges, Volcae, and others, the earlier people were largely subdued and mingled with the newly arrived Gauls. However their former fame as the opponents of the attempt to found the Greek city of Massilia would justify Pliny in calling them celeberrimi, and in classifying them as Ligurians. Such a mistake could easily be made, for without doubt Latin was the only language used in this part of the Provincia in Pliny's day. But there is equally little doubt that prior to the use of Latin, the Keltic of the moving Gallic tribes had prevailed over the still earlier Ligurian Kelts and Ligurians 389 spoken there. The degree to which Gallia Narbonensis was roman- ized is shown by Strabo's statement in regard to the Cavares, that they had become Roman in language and life, and some of them even in their form of government (iv, p. 186; cf. Budinsky Die Aus- breitung der lot. Sprache, pp. 102 ff.). It is noteworthy that all the inscriptions found in the province are written in Latin, except eight, which, although using the Greek alphabet, are in a Gallic dialect. No Ligurian inscriptions have yet been found in this section. Fur- ther, it is significant that Varro calls the Massilians trilingual, hos (Massilienses) Varro trilingues esse ait, quod et Graece loquantur et Latine et Gallice (Isidor. Orig. xv. 1. 63). They did not speak Ligu- rian, and therefore their neighbors, the Salyes, must have spoken Keltic and not Ligurian. This is the first clear instance of my con- tention that the Keltic of the moving Gallic tribes largely superseded the Ligurian of southern Gaul and northern Italy. How the Boii happened to be concerned in this migration is by no means clear. The first record we have of their appearance in Gaul is in connection with their participation in the movement of the Helvetii in 58 b.c. Apparently they had but recently appeared in Noricum when they received the invitation to go along with the Helvetii (Caes. BG. i. 5. 4). Caesar says that they had lived trans Rhenum, but this is so indefinite that we must fall back upon the statement of Tacitus (Germ. 28): Igitur inter Hercyniam sylvam Rhenumque et Moenum amnes Helvetii, ulteriora Boii, Gallica utraque gens, tenuere. Manet adhuc Boihemi nomen, signatque loci veterem memoriam, quamvis mutatis cultoribus. It can scarcely be that this home in Bohemia is that referred to by Caesar in the words trans Rhenum, for they must at that time have lived even west of the Rhine. They seem to have been very nomadic in their tendencies, for they participated in almost all the various Gallic movements for several cen- turies. Niese (Zeitschrift fur deutsches Altertum, NF. 30, p. 149) very plausibly suggests that in the war against the Cisalpine Gauls after the close of the Second Punic War it was the Boii who suffered most heavily from the revenge of the Romans. They were driven from their homes, and many of them forced out of the country. They took refuge among their former allies, the Taurisci, and finally after a further defeat at the hands of the Dacians, settled in Bohemia, 390 Richard Wellington Husband where they remained until driven out by the Marcomani. In the uncertainty surrounding their early location, it is very unsafe to build up a theory from the brief statement of Caesar, as Mullenhoff does, that the whole movement of the Gauls started from the vicinity of the Rhine rather than from central Gaul (Deutsche Altertumskunde 2, 265-69). These are the only tribes requiring special mention, but there are some facts in connection with the relation existing among them all which should be understood. With the exception of the Libui, the Saluvii, and the Boii, they dwelt in the central part of Gaul, in that third section of the country inhabited by those qui ipsorum lingua Celtae, nostra Galli appellantur. More particularly, they represented all the principal tribes between the Sequana and the Liger rivers, and two powerful tribes just to the west of the Liger. In the revolt of Vercingetorix, the Arverni, Bituriges, Senones, Aulerci, Aulerci-Cenomani, and Mandubii were closely allied (Caes. BG. vii. 4 ff.). The Aedui and Ambarri remained faithful to Caesar, although after some hesitation. In fact, from the reading of Caesar one gains the impression that these two tribes were of different mold from their neighbors. They were always at odds with the Sequani and the Arverni. Thus Diviciacus the Aeduan once told Caesar Galliae tohus factiones esse duas; harum alterius principatum tenere Aeduos, alterius Arvernos (BG. i. 31. 3). Diodorus claims to have unusually accurate information, for he tells us something which, he says, is unknown to many, that the people living in the southern part of Gaul, both those toward the Alps and those toward the Pyre- nees, were called KeXroi, while those living farther north (adopting the generally accepted reading aptcrov for votop of the MSS) and toward the Hercynian forest were called TaXdrai, but that the Romans incorrectly grouped all of them together under the common name of FaXdrai (v. 32. 1). This distinction has been felt by the majority of modern writers to have a strong foundation in fact. (A short history of the discussion to 1885 is given by Czoernig Die alten Vblker Oberitaliens, pp. 150 ff.) In recent years the discussion of the relationship of the tribes in central Gaul has had its basis in the interpretation of their linguistic remains. (The best treatments of these are Rhys Celtae and Galli, Kelts and Ligurians 391 1905; Rhys The Celtic Inscriptions of France and Italy, 1906 ; Nichol- son Keltic Researches, 1904, pp. 117-67.) Our knowledge of their language is of the meager and unsatisfactory kind, usual among the European peoples who did not early come under Greek and Roman influence. We possess a fairly large number of names of persons and places, and a few words cited by Roman writers. In addition there are the charms in Marcellus of Bordeaux, the Coligny Calendar, the Rom Defixiones, and forty shorter inscriptions, eight of which are suspected. While this is not the place for an extended examina- tion of the language of these inscriptions, it is worth while to mention that the eight suspected inscriptions bear some remarkable resem- blances to the inscriptions from Pisaurum in CIL. I, 167 ff., which make it a certainty that these eight are Keltic, and that those from Pisaurum are largely Keltic, with a slight mixture with Latin. We possess inscriptions from the majority of the tribes under consideration, and, apart from the Aedui, the Ambarri, and possibly the Lingones, linguistic evidence links all of them closely together. With them are also to be grouped the Sequani, probably the Parisii, the Allobroges, the Caturiges, the Volcae, and other tribes to the south. A close scrutiny of the linguistic remains shows that there was a decided dialectic cleft in central Gaul. The fundamental differences depend upon the treatment of original p and original ku. Adopting the names given by Diodorus it is seen that the Celtae retained original p much later than did the Galli, e.g., in the tribal names Parisii and Pictones, together with several Sequanian and Biturigan words. In all other Keltic dialects it was lost early, leaving only a possible trace in the form h in some dialects. In the treatment of original ku the Celtae correspond to the Goidelic branch in Scotland and Ireland by preserving it in the form qu or c, e.g., Sequani, Sequana, quimon, qutios, equos, etc. In the remainder of the continental Keltic groups this sound early became p, e.g., epos, Petrucorii. Thus between the Sequani on the east and the Arvernj and Bituriges on the west, a group of Galli, composed of Aedui, Ambarri, and Lingones, had pushed their way south, the advance guard of the invading Galli. The earliest Kelts who inhabited Gaul were those who have been called Celtae. They seem to have moved westward from the region 392 Richard Wellington Husband of the Danube to the Rhine, both sides of which they occupied for some time. Thence they moved into Gaul and covered the whole of the territory north of the Garumna. These were the Q-Kelts. related to the Goidels of Ireland, who retained original ku unchanged. The date of their arrival in Gaul is still much in dispute. Some schol- ars, particularly in France, do not admit that the Kelts reached Gaul until shortly before the year 400 b.c. (Bertrand La Gaule avant les Gaulois, p. 254; Sophus Miiller Urgeschichte Europas, pp. 54 and 154 ff.). D'Arbois (op. cit. I, pp. 262, 338) thinks they reached Gaul in the seventh century, while Salomon Reinach ("L'Etain Celtique" in L' Anthropologic, 1892, pp. 275 ff.) and Siret ("Les Cassiterides " in L' Anthropologic, 1908, pp. 129 ff.) date their arrival before the time of Homer. The difference of opinion is due to the fact that while there is an almost total lack of archaeological evidence for a Keltic culture in Gaul prior to 400 B.C., the conclusions drawn from historical documents are strongly in support of their earlier coming. The Celtae were followed westward by the P-Kelts, those who changed original ku to p. ' These are the Galli, related to the Brythons of the British Islands (Rhys The Welsh People, pp. 5 and 11). When this second mass entered Gaul we have at present no means of know- ing exactly. Their coming dislodged many of the Celtae, and drove some of them southward, while others took refuge in the dis- tant British Islands. This is usually regarded as the cause of the passing of the Goidels to England (Oman England before the Norman Conquest, pp. 9-10, 16; Rhys Report of the British Association for the Advancement of Science, 1900, p. 896; Holmes Ancient Britain and the Invasions of Julius Caesar, p. 455). Following the crowding-in of the Galli several southward move- ments, chiefly of Celtae, took place, going along three distinct lines. It is hard to trace these chronologically, but it seems that the earliest led ultimately to Spain, and established a number of Celtae there, where they mingled with the Iberians and formed with them the Celtiberians. Some of them were prevented from entering Spain by the Aquitani, a division of the Iberians, and were forced to remain between the Rhone and the Cevennes, e.g., the Volcae and the Helvii (Strabo iv. 1. 1; Livy xxi. 26). That a second attempt to enter Kelts and Ligubians 393 Spain was frustrated and turned west seems evident from the pres- ence of a section of the Bituriges at the mouth of the Garumna. The third movement led against the Ligurians to the south and east of the Rhone. Here we meet the Keltic names Cavares, Tricastini, Graioceli, Ceutrones, and Caturiges, whom Pliny calls Insubres (iii. 17. 125) . It may have been in this movement also that the Allob- roges crossed the Rhone. The invasion of Spain is usually placed somewhat subsequent to 600 B.C., but no definite date can be assigned. The earliest reference to their presence there is by Herodotus, who about 445 speaks of their being in the southwest corner of Spain (iv. 49; cf. ii. 33; Avienus Ora Mar. 205; Justin. 44. 4). But there seems to be no evidence to support the statement of Siret that an invasion of Spain by Kelts in the eleventh century put an end to the Phoenician empire there. It has been shown above that Livy expressly states that the Gauls were near the mouth of the Rhone at the time of the founding of Massilia, and that the Phocaeans invited these Gauls to help them drive back the threatening Ligurians, that is, the Saluvii. This would place the final southern movement of the Gauls not far from the year 600. That this is a not unreasonable conclusion is clear from a statement made by Hecataeus, writing about 500 b.c: MaaaaXia, 7ro'\(<»«aeW (Fr. 22, FHG. I, p. 2). When, therefore, Hecataeus wrote, it was still felt that the country near Massilia could be called Ligu- rian, but that the Gauls had encroached upon Ligurian territory until they were now not far from the city itself. However, according to our best authorities, the horde that helped the Phocaeans turned east and eventually found a home in northern Italy. Considering their location in Gaul, as well as the dialect they spoke, the con- clusion is inevitable that repeated advances of the invading Galli forced the Celtae from their home in Gaul. As the Galli kept coming into Gaul in increasing numbers, the earlier inhabitants were confined within ever narrower limits, which is the explanation of Livy's statement that they left their homes on account of their abundans multitudo. The account given by Livy seems very clear and straightforward, but unfortunately the question is complicated by three considerations. 394 Richard Wellington Husband First, Livy's statements have been thought to be inconsistent in some particulars. Second, certain other ancient writers do not agree with Livy as to the date of the migrations. Third, it is difficult to locate the migrating tribes in Italy after their arrival there. Livy states clearly (v. 34) that in the reign of Tarquinius Priscus the Gauls made their first appearance in Italy, i.e., shortly after the year 600. But he tells us that there was another story current (v. 33. 1-5), that a certain Arruns, a citizen of Clusium, induced the Gauls to migrate to Italy, and to make an attack upon the city of Clusium in 391, and thence the Gauls marched to Rome and burned it in 390 (traditions differ slightly as to the exact year). This story Livy mentions only to express his disbelief in it : equidem haud abnuerim Clusium Gallos ab Arrunte seu quo alio Clusino adductos; sed eos, qui oppugnaverint Clusium, nonfuisse qui primi Alpes transi- erint, satis constat. Ducentis quippe annis ante quam Clusium oppug- narent, urbemque Romam caperent, in Italiam Galli transcenderunt (v. 33. 4-5). In several other passages Livy seems to imply that the Gauls were new arrivals at the time of the siege of Veii in 396. Thus he says : maxime in ea parte Etruriae gentem invisitatam novos accolas Gallos esse, cum quibus nee pax satis fida nee bellum pro certo sit (v. 17. 8). The important thing here is not that he calls them novos accolas, but that he characterizes them, as if the people of Etruria had never heard of the Gauls prior to this time, or at least had never known anything definite about them. He makes similar statements also in two other passages (v. 35. 4; v. 37. 2). Mullenhoff assumes (Deutsche Altertumskunde II, 251) that, apart from the account of Livy, there is a somewhat uniform tra- dition among the ancients that the Gauls entered Italy shortly before 396, and almost immediately afterward made their attacks upon Clusium and Rome. But instead of this we find that the authorities are fairly evenly divided. Thus Pliny (NH. iii. 17. 125) says that the Etruscan city of Melpum was captured on the same day as that on which Camillus captured Veii. And Appian (Celtica ii) gives an exact date, 392, for their coming. Dionysius also (Ant. Rom. xiii. 10. 14-17) informs us that the Gauls crossed the Alps in the expectation of attacking Clusium. On the other hand, Polybius (ii. 17-18) and Plutarch (Camillus 16-17) are in agreement with Kelts and Ligukians 395 Livy in stating positively that the Gauls made their attacks upon Clusium and Rome long after they had first entered Italy. Livy and Appian are the only writers who attempt to assign an exact date for their arrival. Justinus (xx. 5. 8; xxiv. 4. 2) contradicts himself, and may therefore be neglected. Livy explains his several statements that the Gauls were a new and strange race to the Etrus- cans by the use of the phrase maxime in ea parte, by which he means new in the neighborhood of the fanum Voltumnae, situated not far from Volsinii. And he interprets this later by saying: nee cum his primum Etruscorum, sed multo ante cum eis, .qui inter Appenninum Alpesque incolebant, saepe Gallici pugnavere (v. 33. 6). Livy repre- sents that five successive migrations took place, and Plutarch gives the information that the Gauls subdued eighteen different cities belonging to the Etruscans in the valley of the Po. Certainly no small amount of time was necessary to drive back the powerful Etrus- cans from this large territory. Livy may have placed the date a little too early, he probably makes a slight error, but without doubt he is closer to the truth than Mullenhoff and others who assume the later date for the first inroad of the Gauls. (Cf. D'Arbois op. cit. I, 166; Lavisse Histoire de France I, p. 25; Hirschfeld Timagenes und die gallische Wandersage; Czoernig Die alten Volker Oberitaliens, pp. 158-66; Skutsch, in Pauly-Wissowa VI, 750). Miiller-Deecke (Die Etrusker I, pp. 144 ff.) and Desjardins (Geographie de la Gaule Romaine II, pp. 204 ff.) argue for the accuracy of Livy. The arguments for and against Livy's narrative may best be given by an examination of the last thoroughgoing and hostile criticism. Niese (Zeitschrift fur deutsches Altertum, NF. 30, pp. 129 ff.) holds that this account of the migrations is unworthy of credence on the following grounds: (1) The assumed connection of the wandering Kelts with the founding of Massilia is a mixture of two distinct stories. Thus Plutarch (Solon ii), Justinus (xliii. 3. 6), and Aristotle (Fragm. 508) tell of the friendship existing be- tween the Phocaeans of Massilia and the Kelts. Livy without justification assumed that these Kelts were the ones who finally settled in Italy. That Mullenhoff (II, 252) is right in believing that the Kelts moved after the time of Herodotus, and that Miiller-Deecke are wrong in their statement that the Etruscans were driven back 396 Richard Wellington Husband from northern Italy prior to 524, based upon a misinterpretation of Dionys. 7. 3. (see Mullenhoff I, 179). (2) The names of the tribes of Gaul were unknown before the time of Caesar, and Livy derived his names from Caesar's Commentaries (cf. D'Arbois II, 301-4; Hirschfeld 331 ff.). Further, that when the Gauls reached Italy these names disappeared. That Livy, in saying that Ambigatus was king of Celticum, knows only of Caesar's section of central Gaul, whereas there was a kind of political unity among the Kelts extend- ing from western Spain to the Scythians, and Ambigatus could properly be called king of this whole territory (D'Arbois II, 303, n. 2, and 301). But what right have we to assume that tribal names were unknown before Caesar's campaigns ? Any one of a large number of ancient historians whose works have been lost may have mentioned them. And what right have we to assume that there ever was a political unity that would justify the name Celticum, as applied to the whole Keltic continental territory? The sole evidence, upon which D'Arbois lays so much stress, that of similarity of language, is trifling. And since he wrote, this is materially weakened by the discovery of strong dialectic variations, which, as indicated above, lend valuable support to Livy's narrative. His meaning is perfectly clear, that the migrating tribes belonged to the section of Gaul where, in his own time, these tribes lived. It should be stated that if the relation I have assumed between the Celtae and the Galli is correct Livy should not have included the Aedui and the Ambarri. (3) The names Ambigatus, the very wise, Segovesus, having knowledge of victory, and Bellovesus are poetical names, which Livy in accordance with his tendency took from some Gallic poem. Others have raised the same objection, but it is well to say that no trace of any such Gallic poem exists, nor are the names more fictitious, on the ground of their appropriateness, than are Pericles, Sophocles, Alexander, Hilarius, and Prudentius. (4) Taking a great multitude across the Alps is a story in imitation of the journey of Hannibal. But they certainly got across the Alps, as did many others. These objections to the story of Livy are of no great importance, except the first, which has to do with chronology, and does not invalidate the remainder of the story to the slightest degree. And even on this point Livy is the only writer who does not palpably con- Kelts and Ligukians 397 tradict himself. Pliny confuses three swarms of Gauls in the attack on Melpum. Nor do we know where Melpum was. Modern guesses have placed it near Milan, yet it might as well be anywhere else. The four successive migrations occupied much time, and the last, that of the Senones, may not have been much earlier than 400, but the first long antedated that time. If we were obliged to give up the story told by Livy, where should we turn for information ? Niese replies by pointing to a passage in Polybius (ii. 17), where it is said that the Gauls looked with envious eyes upon the fertile land of the Etruscans, and finally with but slight provocation moved against them and drove them from the valley of the Po. This is repeated by Dionysius (xiv. 113). These must replace Livy and others as authorities. The only place where the relation mentioned by Polybius could exist is to the north in the valley of the Etsch. It could not have been in the northwest, where Ligurians and not Kelts would have been their neighbors. It was the Galatae (Diod. v. 32. 5) of the north and east who captured Rome and robbed Delphi. Later Gauls followed the tracks of the earlier by this route, says Polybius (ii. 18.4; 19. 1). Niese thinks there is one piece of evidence to show that Polybius is correct. The Gauls in Italy needed help against the Romans, and at the battle of Telamon they were allied with the Taurisci, a tribe from the north (Polyb. ii. 28. 4; 30. 6; cf. 15. 8). But their especial allies were the Gaesatae (Polyb. ii. 22. 1; cf. 28. 3; 34. 2), who came from the valley of the Rhone. These facts show that all connection of the Italian Gauls with their kinsmen was toward the north, and not toward the west. The Taurisci certainly came from the north, but we have no knowledge elsewhere of the Gaesatae. Niese fur- ther mentions that when Hannibal entered Italy he was opposed by the Gauls on the western side of the Alps, although he was in alliance with the Insubres and Boii in Italy, and therefore there was no con- nection between the Kelts of Italy and those of Gaul. But Niese forgets that this alliance was made after Hannibal reached Italy, and only after consideration as to whether it would be advisable to help the Taurini or the Boii in their quarrel (Livy xxi. 39. 6). It is interesting also to notice that Niese's main argument depends upon a passage in Polybius that has materially damaged his reputa- 398 Richard Wellington Husband tion as a geographer. His ideas of the position of the Alps and the Rhone display such ignorance as to discredit utterly his authority in regard to this event. Again, his statement of the envy in the minds of the Gauls because of the possession by the Etruscans of the fer- tile valley of the Po sounds like a rationalizing of the old story of the wine and the oil (Livy v. 33. 1-5; Pliny xii. 2. 5; Dionys. xiii. 10. 14). Against such paucity of genuine information, the reason- able account given by Livy must still be credited. And yet Hirt is carried away by the arguments of Niese, and feels them to be so cogent that he would follow Niese even if he did not have the sup- port of Polybius {Die Indogermanen 1, 171). It is noticeable that E. Meyer does not follow the lead of Niese, but accepts the story of Livy, except for the date, which he places extremely late (Gesch. des Altertumsl, 2, p. 794). Coming next to the question of the routes followed in the four successive migrations to Italy, it is to be noted, as even Niese points out, that the earliest writers state that the Gauls in most ancient times had their home in the extreme west, or northwest, of Europe. These are Heraclides Ponticus (Plut. Cam. 22), Clitarchus (Strabo vii. 293), and Callimachus (Hymnus in Delum 174). Their stories coincide with Livy and Pausanias (i. 4. 1), who represent the Gauls as coming from the ocean and the ends of the world. This is indeed very indefinite, but it assures us that the most ancient writers were unanimous in thinking that the Keltic migrations originated in Gaul. The first horde was composed of Bituriges, Arverni, Senones, Aedui, Ambarri, Carnutes, and Aulerci. Of these, the Aedui and Ambarri were Galli, the others were Celtae. (It was pointed out above that Livy is probably in error in including any Galli in the first migration.) Livy says that they made their way through the Taurini, and, according to the carefully considered last edition of Weissenborn, by way of the Duria. The farthest point reached eastward was where Mediolanium was founded, and without doubt all followed closely behind. Nicholson (Keltic Researches, p. 166) gives an interesting etymology for Mediolanium, which makes it clearly Keltic. The names Rigomagus and Bodincomagus in the heart of the district of the Taurini prove the presence of Kelts among the Taurini. The route by way of the Cottian Alps has also left Kelts and Ligukians 399 its trace in the names Quadiates (CIL. V, 7231, 3) and Quariates (CIL. XII, 80; Pliny iii. 35), preserved today in the name Queyras in the department of the Hautes-Alpes. The gentile name Quia- melius occurring among the Deciates (CIL. XII, 226) has the same origin. The two modern villages Quarlasco and Quassasco, the first near Turin and the second near Ivrea, indicate the presence of the same tribes. In these names, therefore, both those with the very common Keltic suffix -magus, and those with the qu of the Celtae, we have the best of proof that the Celtae of the first migration settled among the Ligurians from the pass of Mont Genevre eastward. If there is need of further evidence, we have the name of Semigalli applied to the Taurini by Livy (xxi. 38), although the Taurini in the estimation of the ancients represented the true old Ligurian stock, antiqua stirps (Pliny iii. 123, etc.; see TAP A. XL, p. 74). It is impossible at present to do more than guess why the names of the Keltic tribes were not preserved. The second migration of Gauls was that of the Cenomani, a division of the Aulerci, followed by the Libui and the mixed race of the Saluvii. These, according to Livy, took the same route as the earlier movement. This harmonizes with the statement of Cato, to the effect that the Cenomani had once lived in the vicinity of Massilia: auctor est Cato, Cenomanos iuxta Massiliam habitasse in Voids (Pliny iii. 19. 130). We must assume that the Cenomani fol- lowed the route of the earlier tribes south almost to Massilia, then turned east over the Alps by Mont Genevre, along the Duria, and eventually settled north of the Po, where they founded the cities of Brixia and Verona. About forty-five inscriptions come from the terri- tory occupied by them in Italy (Pauli Die Inschriften nordetruskischen Alphabets, nos. 38-85). The Libui seem to be mentioned only twice in addition to the present passage, and both times by Livy. Here he says they settled behind, that is west of the Cenomani, implying also that they were near the Ligurian Laevi, who lived at the junc- tion of the Ticinus with the Po. Pliny (iii. 124) tells us that the Laevi and Marici founded Ticinum. The situation compels us to place the Libui and the Saluvii north and west of these, and that is the exact situation of the tribe known to the Romans as the Libici. Now Pliny in the same passage says that Vercellae was in the terri- 400 Richard Wellington Husband tory of the Libici, and was founded by the Saluvii. This makes the Libui but a pagus of the Saluvii, who as we have already seen were a Ligurian tribe largely mixed with Kelts. So Livy calls the Libui Gauls (xxi. 38) very properly, and contrasts the Libui and the Laevi with the true Ligurians (xxxiii. 37. 6). In this connection it is necessary to notice also a further fact told by Pliny, that Novaria was founded by the Vertamacori, a pagus of the Vocontii, and not, as Cato thought, by Ligurians (iii. 124). This is important, in view of the fact that one of the longest Keltic inscriptions of Italy was found near Novaria, and belongs rather to the Celtae than to the Galli (in Pauli op. tit., no. 25; Rhys Celtic Inscriptions, no. XXXIV). The third migration was that of the Boii and the Lingones, who entered Italy by the Pennine Alps. This necessarily brings them through the valley of the Aosta, along the river Duria. Here the Salassi lived, and the inference is fair that when these migrating tribes found the valley of the Po already thickly settled by Gauls they moved slowly onward to their final home south and southeast of the Po. There cannot be much doubt that many of them remained behind among the Salassi, and from these the Salassi derived the many Keltic elements in their language. This is the situation that caused the Roman writers to hesitate as to whether they should class the Salassi among the Kelts or among the Ligurians (TAP A. XL, pp. 74, 75, 78, 81). We have no evidence as to the route taken by the Senones, the last of the migrating tribes. Their ultimate home was in the east of Umbria, where they were known by the Romans for many years. An important Keltic inscription has been found at Todi, the ancient Tuder, which probably originated with this tribe (Conway The Italic Dialects II, p. 528; Pauli op. tit., no. 26). There are some very archaic inscriptions from the vicinity of Pisaurum, which are commonly said to be early Latin, but under some other influence. Conway is inclined to think the peculiarities are Umbrian, owing chiefly to the fact that the Senones were subdued by the Romans, and the colony of Sena Gallica was planted among them as early as 283 b.c. But the planting of a colony does not stop the speaking of the native tongue, and this fact in addition to the finding of a Gallic inscription not far away makes it almost certain that the influ- Kelts and Ligurians 401 ence upon the inscriptions of Pisaurum was that of the Senones. Some remarkable forms of verbs and nouns link them closely with inscriptions found in southern France. Kiepert (Lehrb. der alien Geogr. 341) holds that Bononia is the only city south of the Po having a Keltic name, but it is not at all improbable that names current among the Senones, e.g., Pisaurum, Sena, Aesis, Sarsina, Suasa, are also Keltic. Certainly they are not Latin. Schulze (Lat. Eigennamen 569) thinks Sarsina is of Etruscan origin. The history of the others is not available, but they are too old to be called Umbrian. Dartmouth College