Talk:Ligurian language

From Wikipedia, the free encyclopedia

Placing a considerable quantity of useful discussion on Ligurian here, where it can be readilly found. It was squirrelled away on user talk pages. --Nantonos 20:13, 11 July 2005 (UTC)

I happen to know Lepontic probably as well as anyone else, and I doubt very much that it is even a Celtic language, let alone a close relative of Gaulish. I happen to believe it is closely related to Ligurian, a member of a separate branch of Indo-European. (There is a note to that effect in the article on Lepontic.) And given that we know even less about Noric than we know about Lepontic, who knows... You draw your conclusions.

Cheers,

Pasquale 8 July 2005 23:49 (UTC)

The notion has been around a long time that the dialect of the Lepontii (Lepontic), together with those of the Salassi and Taurini (the three have been jointly referred to as North Ligurian or Subalpine), were affiliated with Ligurian (an Indo-European language originally spoken not only in Liguria proper but also in the Western Alps, the South of France, both sides of the northern Appennine ridge, as well as remote valleys of North-Eastern Italy and the Eastern Alps). Traditionally the main source was Joshua Whatmough (in Robert Seymour Conway, Joshua Whatmough, and Sarah Elizabeth Jackson Johnson, The prae-Italic dialects of Italy, 1933). While this work may be outdated in many respects, it still remains an extremely useful source. Personally, I wrote a 1973 thesis at the University of Rome entitled "La lingua delle iscrizioni leponzie" (unpublished) in which I weighed all the evidence and took a position slightly in favor of the Ligurian hypothesis (I'll spare you the details). Meanwhile, in 1970, Michel Lejeune had published a long article in which he took a stand in favor of the Celtic hypothesis. His entire argument, however, hinges on a sleight-of-hand. It is based on one important inscription, whose attribution to Lepontic is doubtful. Many think it may be in Cisalpine Gaulish. (The same alphabet of Lugano used for Lepontic was later also used for Cisalpine Gaulish.) In this inscription there is a personal name Uwamogotsis, or "highest host". Lejeune says the inscription cannot be in Gaulish because *gotsis is nowhere else attested in Celtic. So, it has to be Lepontic. Ergo, Lepontic is closely akin to Gaulish, because this personal name clearly shows Celtic phonology (*p > 0, *st > ts). Never mind that with this sleight-of-hand, *gotsis does become attested in Celtic! (Furthermore, personal names are not all that significant because, having fallen under Celtic influence, the Lepontii may have used Gaulish personal names, something for which there is some evidence also among the early Germanic people.) It all pretty much boils down to whether Indo-European *p was preserved in Lepontic (as in Ligurian) or not. The most commonly attested word in Lepontic is PALA (possibly for *palla, since geminate consonants were apparently never written), which means 'stone, tombstone'. Does this show an Indo-European *p? Whatmough seemed to think so. I favor that view as well. But, having chosen the Celtic interpretation, Lejeune, of course, had to reject that etymology. One of the problems with the Lepontic inscriptions is that no distinction is made between voiced and voiceless stops, so that PALA can stand for *pala, *palla, *bala, *balla. BTW, in the Alps, there are lots of mountain peaks called Palla-this and Palla-that. By folk-etymology, these seem to have been interpreted as "ball" and even translated into German as "Kugel" (e.g. Pallabianca = Weisskugel). But these mountains don't look like balls at all, they look more like... well, rocks!

Pasquale 22:58, 9 July 2005 (UTC)

As for the Todi inscription, indeed, when I worked on it, it was universally considered to be Cisalpine Gaulish. The Lepontii were just a small indigenous mountain tribe of Canton Ticino and adjoining areas of northern Lombardy and Piedmont. Well before the Gaulish invasion of northern Italy, they developed the alphabet of Lugano, one of five alphabets of northern Italy traditionally known as "North Etruscan" alphabets because they were derived from the Etruscan alphabet (Lugano, Sondrio, Magrè, Bolzano, Este — with additional subtypes and variants). There is no presumption, however, that to each alphabet corresponds only one language. In particular, the alphabet of Lugano was clearly used for Cisalpine Gaulish.

However, I admit I have been away from the most recent scholarship. It appears that there is a trend now to consider Lepontic and Cisalpine Gaulish not as discrete languages, but as closely related dialects of "Cisalpine Celtic" (e.g. Joseph F. Eska, The linguistic position of Lepontic, 1998). This position seems to be gaining ground, although it plays fast and loose with both history and data. On line, you can find this article by Eska, which will give you an idea of what he's up to: www.bol.ucla.edu/~maom/ research/Eska_Mercado_Vergiate.pdf. Notably, he has to reject the identification of Lepontic KARITE (which can be read as *karrite) with Gaulish karnitu, showing different 3sg preterite morphology and the possibility of a progressive assimilation of -rn- to -rr-, something quite alien to Celtic (or simply of an otherwise well-attested pre-Celtic root *kar- or *karr- meaning 'stone'). Notice that the Lepontic corpus is, by and large, older than the Celtic invasion of norther Italy. I don't know how these Celticists square that away.

After 400 BC the powerful Gaulish tribe of the Insubres held sway in what is now Western Lombardy, centered around Mediolanum (Milan). They may have commingled in part with the indigenous Lepontii, their northern neighbors, but to say that they were one and the same or that their languages were dialects of one and the same language, or even the same language, is, IMO, absurd and ignorant. To then go as far as say that the Insubres spoke Lepontic because they used a variant of the alphabet of Lugano (as Paul Russell seems to do) is plainly ridiculous.

By the way, the assertion that Celtiberian is closely related to Gaulish is also way off the mark. Celtiberian is probably the most divergent of the Celtic languages. Among other things, it is the only Celtic language to preserve final -m -- other than Lepontic, of course (but not Cisalpine Gaulish, which shows -n).

Pasquale 02:16, 11 July 2005 (UTC)

Conway, Whatmough & Jackson Johnson 1933 still remains a fundamental reference not just for Lepontic, but for all the prae-Italic languages of Italy, a kind of editio princeps, if you will. If you get a chance to look at it, you'll see what I mean. Whatmough 1933 carefully weighs all the evidence and leans slightly towards the Ligurian affiliation for Lepontic. But he certainly would have been shocked by the current claims identifying Lepontic and Gaulish, which seem totally to ignore the historical reality of the Gaulish invasion of northern Italy of approx. 400 BC, the fact that this invasion most probably came from Noricum (!), that there is no other evidence of Celtic peoples in northern Italy prior to that invasion, that large swaths of the South of France were Celticized even later than northern Italy (Provence was still Ligurian-speaking possibly as late as 200-150 BC and the dede bratou dekantem inscriptions are most probably Ligurian, not Gaulish). The Lepontic inscriptions, on the other hand, date as far back as 600 BC, if not even earlier.

Personally, having studied all the Ligurian evidence in depth (substratum terms, toponymy, and the few extant Apuanian inscriptions, written in the Etruscan alphabet), I have no doubt that Ligurian was Indo-European. And, BTW, that's what the Wikipedia stub for the Ligurian language used to say, until an anonymous contributor (172.153.145.157) changed the wording from "Very little is known about this language which is believed to have been Indo-European" to "Very little is known about this language, which is believed to have been non-Indo-European, as evidenced by certain place-names in the region where it was spoken." What place-names? Genua, nearly identical to Genava (also alternatively known as Genua in ancient times)? Briga? Langa? Alba? Album (e.g. Album Intemelium, Album Ingaunum)? I have never seen more Indo-European-looking place-names. No, in fact, Ligurian represents a little-known extinct branch of Indo-European, akin but different from Celtic (and not simply for the preservation of *p), and sharing some features with Osco-Umbrian (e.g. the full labialization of ALL the labio-velars, but NOT of the velar plus glide clusters (e.g. *ekwo-, etc.), a very different situation than in Celtic. There is ample evidence of the previous presence of this Ligurian branch not only in northern and central Italy, the Alps, and at least the south of Gaul, but also in the Iberian peninsula, and Corsica and Sardinia. There is even, in all likelyhood, a considerable Ligurian adstratum in Basque!

Pasquale 19:25, 11 July 2005 (UTC)

If the published evidence indicates that Ligurian was more likely Indo-European rather than non-IE, then obviously the majority view gets more representation. On the other hand, since some scholars believe otherwise, then they must also be cited. Decius 23:18, 11 July 2005 (UTC)

In the meanwhile, I found an individual named Claudio R. Salvucci making this statement on this webpage:[1] "The use of the term Ligurians is confusing, since it is used both for a Pre-Indo-European language whose chief evidence is place-names, as well as an Indo-European language (not Celtic or Italic) from names in an inscripton near Genoa. The language of the historical Ligurians was probably the second of those." Decius 00:12, 12 July 2005 (UTC)

1 Gauls in Italy
2 Comment March 2006
3 Delamarre
4 Ligurian
5 Translation question
6 Interlinks

[edit] Gauls in Italy

Pasquale, regarding the 400BC invasion, why do you say that "no other evidence of Celtic peoples in northern Italy"? Looking at Livy, History of Rome Book V: The Veii and the Destruction of Rome by the Gauls

http://www.perseus.tufts.edu/cgi-bin/ptext?lookup=Liv.+5.34

it says

and when they learnt that the country in which they had settled belonged to the Insubres, a name also borne by a canton of the Haedui, they accepted the omen of the place and built a city which they called Mediolanum.

which seems to argue that the Insubres were already there when the Gauls invaded. Why do you say they came from Noricum, by the way? "the Bituriges, the Averni, the Senones, the Aedui, the Ambarri, the Carnutes, and the Aulerci" are all Gaulish, surely?

Also, could you explain why you consider the dede bratou dekantem inscriptions are most probably Ligurian? Meid for example states that they are Gaulish. I agree they have a different distribution to the Gaulish material written in Latin script, but its not different to the distribution of other Gaulish material written in Greek script, is it? --Nantonos 02:43, 14 July 2005 (UTC)

Well, the date of the Gaulish invasion of Italy may be slightly earlier, in any case in the fifth century BC, as I recall. (As you may know, Brennus attacked Rome in 390 BC, on the heels of that invasion.) The Wikipedia has virtually nothing about the history of Ancient Italy, and I don't have other Web references to give you right now, but it's all pretty well known. Certainly around 600-500 BC the Etruscans ruled the entire Po Valley, including the Milan area, Bologna (then known as Felsina), Spina (near Ferrara), etc. There were no Gauls in Cisalpine Italy then. Livy's account seems to be largely fantastic. Of the tribes he mentions, I believe only the Senones are attested in Italy, centered around Sena Gallica (modern Senigallia). Any good ancient history book will give you a detailed account of the names and geographical distribution of the Cisalpine Gaulish tribes: Insubres (around Milan), Libici, Boii (around Bologna), Cenomani, Lingones, Senones (I am quoting from memory). Nobody knows exactly what route they came from, but since the south of France (e.g. Provence) and the Alps (e.g. Savoy) seem to have come under Gaulish control considerably later, the notion that the Gauls came from Noricum is certainly appealing and has been supported by some historians (sorry, I can't give any sources now). The small region of Cadore, a natural entry route into northern Italy, was inhabited by and takes its name form the Catubriges, a Celtic tribe. The Cisalpine Gaulish tribal names are of little help. The Boii, for example, are attested in Transalpine Gaul, but do we know when they got there? However, they also gave their name to Bohemia, which may have been their original homeland. The tribal names mentioned by Livy were indeed all from Transalpine Gaul, but, again, AFAIK, only the Senones are actually attested in Italy. Pasquale 20:30, 14 July 2005 (UTC)

I'm not asking because I am ignorant of history or because I can't be bothered to read an introductory work. I'm asking what your sources are because your version of the history differs from the one I have read; rather than just say you are wrong I pointed to some contemporary evidence that could be interpreted counter to your claims and asked what your other evidence was. Yes, around 400 some Gaulish Celts crossed the Alps southwards; my point is that when they did, Livy indicats that they met other Celts (I didn't call the Gauls, I called them Celts). Merely saying 'its well known' or 'some historians' is not a compelling argument. Please cite sources, as I did. --Nantonos 17:28, 15 July 2005 (UTC)

The possibility that the dede bratou dekantem inscriptions may be in Ligurian rather than Celtic is an old idea of mine, which I have never published, but hopefully some day may. They are all pretty old and from the South-West of France, an area supposed to be Ligurian until relatively late (maybe 200-150 BC, but that's just a guess). For example, the Salyes, near Massilia, were most probably Ligurian. Now, dede has a 3sg preterite -e ending, attested in Lepontic, whereas Gaulish shows a 3sg preterite -u ending; furthermore, an inscription in the Lugano alphabet that has been considered either Lepontic or Cisalpine Gaulish (the Prestino inscription) has dedu. For me, the Prestino inscription is clearly in Gaulish, for several reasons, not just dedu. Also, consider that dede in dede bratou dekantem seems to mean 'gave', while dedu in the Prestino inscription seems to mean 'placed'.

The second word bratou can be either, with its labialized labiovelar and a root amply attested both in Italic and in Celtic. But the final -m of dekantem ('tithe') can hardly be Gaulish, and shouldn't Gaulish show dekameton here anyway? Well, it's anybody's guess, but I vote for Ligurian. Pasquale 20:30, 14 July 2005 (UTC)

Thanks, that is helpful. --Nantonos 17:28, 15 July 2005 (UTC)

Kim McCone, Towards a relative chronology of ancient and medieval Celtic sound change (Maynooth 1996, p. 69) argues that Lepontic is part of Gaulish (and against Lejeune's and Schmidt's proposal that Lepontic is a separate branch of Celtic) thus:

The significance of Lep. -m vs. Cis. Gaul. -n (Lej. 432-5) has been completely undermined by the evidence of the Larzac inscription for fluctuation between -m and -n still at quite a late date in Transalpine Gaulish, and the suspiciously close proximity of Vercelli to the 'Lepontic' area is anyway sufficient to call the dichotomy into question. Since, moreover, the inscriptions around the Lakes are almost certainly earlier for the most part than the other three, the by any standards trivial -n for -m could be simply a matter of chronology as, for that matter, could the survival of an old -ei dative in the n-stems [referring to Lep. PIUONEI and ASILONEI where Trans. Gaul. usually has either -E or -I].

That leaves Lep. TETU /dedū/ 'set up' < *dedō < *d^hed^hoh₁e vs. Trans. Gaul. δεδε with -e taken over from the majority of perfects without final laryngeal (Lej. 446-52). Given original 3sg. -ū in some versus -e in most old perfects, the new 'weak' -it- preterite was free to adopt either or both. Consequently it would be rash to read too much into ('Lepontic') Vergiate KARITE, KALITE but (Cis. Gaul.) Todi KARNITU agreeing with Transalpine Gaulish καρνιτου. In addition to p < k^w, 'Lepontic' shares with Cisalpine Gaulish an assimilation of nd > nn not even found in Transalpine Gaulish and has an -it- preterite as well as an -ikno- patronymic otherwise definitely attested only in Gaulish among known Celtic languages.

--Angr/_{tɔk tə mi} 21:00, 14 July 2005 (UTC)

Well, that's one opinion. I don't think that simply because an opinion is published it becomes the gospel truth. And how does McCone square away the historical facts, or are those irrelevant?

I totally fail to get the point about Vercellae, well-known to be a Gaulish town (of the Libici or Anamares). It does not fall within the Lepontic territory. However, the Insubres (the most powerful Cisalpine Gaulish tribe) certainly did settle partly in Lepontic territory. Has McCone never heard of two people sharing (in part) the same territory?

The progressive assimilation of nd > nn is indeed found in Cisalpine Gaulish and it is a (presumably very recent) areal sound change that Cisalpine Gaulish (or at least Insubrian) shares with Lepontic (but also Osco-Umbrian!). And the patronymics in -ikno- are also attested in Italic (Paeligni, etc.).

Again, the Lepontic corpus is much too small to come to such cavalier conclusions. The first Indian to approach the Pilgrim Fathers in Plymouth a year after their landing (a Micmac from Maine who spoke some English and was on his way to Rhode Island) addressed them with the words "Hello Englishmen!". A corpus of two words. Ergo, the American Indians spoke English?

I am open to all possible conclusions regarding Lepontic but I am baffled by such gratuitous self-assurance. Pasquale 01:49, 15 July 2005 (UTC)

Oh, yes, and a map of Rhaetia [2] might help people understand just where the Lepontii lived and how small their territory was (basically Ossola and Ticino). Look for the Lepontii in the south-western corner of Rhaetia. The area to the South, including what was to become the Insubrian capital Mediolanum (modern Milan), was Etruscan around 600-500 BC, when the Lepontii began writing (mostly tombstones) in their alphabet (one of several Etruscan-derived alphabets in the Rhaetian territory). Interestingly, the small area occupied by the Lepontii wasn't even in Gallia Cisalpina, but in Rhaetia!

And this is a map of Liguria proper and areas to the north: [3]. Here you can see the location of Gallic Vercellae, between the territories of the Libui (also known as Libici) and Laevi, not at all in Lepontic territory. Pasquale 02:05, 15 July 2005 (UTC)

I'm not defending McCone's conclusions, and by no means am I calling them gospel truth. I'm just letting you (and anyone else who cares) know what one well-respected Celticist has had to say on the matter. And I don't think he sounds particularly self-assured (for him, anyway) or is coming to cavalier conclusions, he's just pointing out that the arguments that have been brought forward to show that Lepontic and Cis.Gaul./Trans.Gaul. are very different from each other can be explained differently. If anything, he's arguing against coming to cavalier conclusions. --Angr/_{tɔk tə mi} 05:41, 15 July 2005 (UTC)

Fine, Angr, but consider the following. McCone's statement (as quoted by you) "That leaves Lep. TETU /dedū/ 'set up' < *dedō < *d^hed^hoh₁e etc." takes both the long quantity of the final -u and the derivation from *d^hed^hoh₁e for granted. In fact, both are mere conjecture. Eska (whom I have criticized for gratuitously conflating Lepontic and Cisalpine Gaulish, but who knows his Celtic and Indo-European stuff) pretty much demolishes both hypotheses (see Eska & Mercado, www.bol.ucla.edu/~maom/ research/Eska_Mercado_Vergiate.pdf, pp. 17-18, and fns. 42-45). In other words, both the assumption of a final -ū in TETU and the derivation from *de-dō-e — whether from *de-deh₃-e (Markey & Mees) or from *d^he-d^heh₁-e (Bammesberger, evidently followed by McCone) — are highly improbable guesswork.

And that is precisely the sort of attitude I was referring to by "self-assured" and "cavalier conclusions". When Tom and Dick formulate a hypothesis (and a shaky one to boot) and then Harry refers to it as accepted fact, is that scholarship? In fact, there isn't even agreement on whether the *dhē- or the *dō- root is reflected in TETU and δεδε, EXCEPT that both must be derived from one and the same root, in the name of a prejudged Gaulish identity of these corpora (never mind that TETU seems to mean 'set up' and δεδε 'gave'). Again, is that scholarship? (Incidentally, what is the evidence for an -e ending "in most old perfects" in Gaulish, other than δεδε itself? To me, what seems "rash" is to conclude that there was some kind of free variation in Gaulish between 3sg. preterite endings -ū [again, totally unproven vowel length] and -e, something I don't recall seeing in any language.)

In the pre-Roman Iberian peninsula, there were certainly several dozen languages, belonging to a variety of language groups, some Indo-European, some non-Indo-European. Ditto for pre-Roman Italy. Curiously, when it comes to Continental Celtic studies, geographic labels such as "Trans. Celt." and "Cis. Celt." are assumed to identify languages.

And what about McCone's reference to the poorly understood Larzac inscription and its "fluctuation between -m and -n"? This inscription is dated to 90-110 AD, when Latin was certainly widely spoken in Gaul. Maybe the instances of final -m are a reflection of Gaulish/Latin bilingualism. How else can you explain this odd resurgence of final -m centuries after it had changed to -n? And then use it as evidence against the "significance of Lep. -m vs. Cis. Gaul. -n"! I am sorry, but I don't believe this is scholarship at all. Pasquale 17:41, 15 July 2005 (UTC)

[edit] Comment March 2006

This all seems very strange to include in Wikipedia, even for a discussion page. Pasquale seem to disagree with just about everything written by established Celticists on Lepontic and makes some very strong statements about their scholarship. Archaeological evidence makes it quite clear that there were Celts in Northern Italy before the famous invasion after 400BC and no one (apart from Pasquale) who has written on the topic seems to doubt Lejeune's conclusions. All subsequent scholars (US, Irish, Welsh, German, French, Italian) have supported his argument that Uvamogozis derives from *upamo-ghostis and means '(having) supreme guests', i.e. it is a superlative bahuvrihi very similar to the Irish name Fergus (probably '(having) super guests'). The curators at the museum in Como where the Prestino inscription is housed have no doubt that Lepontic is Celtic either -- in fact it is 'Ligurian' itself which is the dubious and problematic language, not Lepontic.

Well, Bmees, for one thing, I never said UVAMOKOZIS was not Celtic. In fact, I never doubted it was Celtic. You obviously misread what I wrote. For me, UVAMOKOZIS is undoubtedly Celtic. What is not sure is that UVAMOKOZIS is "Lepontic". It could be Cisalpine Gaulish, i.e. of the Gaulish tribe of the Insubres, who took over the area of Milan and surrounding plains, chasing the local tribe of the Lepontii up the mountain valleys of Ticino and Ossola. This inscription is written in a later variant of the Alphabet of Lugano that is quite different from the original one used in the older inscriptions from 700-400 BC, which are surely Lepontic. My contention is that Lepontic and Cisalpine Gaulish are different languages. By the way, Lejeune says the same thing, but the difficulty is in determining exactly which inscriptions belong to one and which to the other language, given how short some of them are. It's interesting that you cite Fergus as an example of *ghosti- in Celtic. In 1970, Lejeune wrote that the reason this inscription has to be assigned to Lepontic (rather than Gaulish) is that *ghosti- is nowhere else attested in Celtic, ergo it cannot be Celtic, ergo it must be Lepontic. And then he adds, ergo Lepontic is also Celtic, because of *p > 0 and because of *-st- to -ts- (or tau gallicum or whatever). It is precisely that syllogism that I have referred to as sleight-of-hand. As for me, I would simply say that this particular inscription is in Cisalpine Gaulish (Celtic), not in Lepontic, and that the affiliation of Lepontic remains in doubt. And, by the way, I never said I have conclusive evidence that Lepontic belongs with Ligurian (which was probably a non-Celtic Indo-European language). All I am trying to say is that the evidence is inconclusive one way or the other. Only by forcibly conflating Lepontic and Cisalpine Gaulish into one language (simply because they are written in closely related variants of the same alphabet, a principle nowhere else held as evidence in linguistics), can you then assign Lepontic to Celtic. I call that another sleight-of-hand. Pasquale 23:08, 28 March 2006 (UTC)

I should add, by the way, that the name Fergus is usually not derived from *ghosti- 'guest', but from Proto-Celtic *gustu- 'force', in other words, *Uφer-gustu- (cf. other personal names such as *Kuno-gustu-, *Oino-gustu-; see Proto-Celtic - English Glossary). So, the etymology offered by Bmees for Fergus is most probably incorrect, and Lejeune may still be correct that *ghosti- is nowhere else attested in Celtic. Nonetheless, his 1970 argument remains circular. Once again, it was precisely Lejeune's 1970 argument that (1) attributed the Prestino inscription to Lepontic rather than Cisalpine Gaulish simply because *ghosti- (in the proper name UVAMOKOZIS) is attested nowhere else in Celtic; then (2) assigned Lepontic to Celtic on the basis of *p > 0 and *-st- > -ts- (in the very same UVAMOKOZIS). However, Lejeune 1970 considered Lepontic a completely different Celtic language than Cisalpine Gaulish (as distant from Gaulish as, say, Celtiberian). My understanding is that, on that point, Lejeune's thinking has not changed. In other words, he would completely disagree with latter-day Celticists, such as Eska, who quite arbitrarily conflate Lepontic and Cisalpine Gaulish into one and the same language. Pasquale 18:15, 29 March 2006 (UTC)

[edit] Delamarre

The article right now is still very unbalanced giving us mostly Delamarre's view that Ligurian was a Celtic language. I haven't studied Ligurian much, nor do I have references on the Liguran corpus, but someone who does should add the counter-Celtic arguments into the article. Decius 14:27, 12 July 2005 (UTC)

Yes, and if I come across contrary views I will add them; but yesterday the artice had little info and zero sources. I was about to add an unreferenced template, but then came across at least one referencable opinion, so added it. Which was better than what was before. --Nantonos 18:39, 12 July 2005 (UTC)

I have some problems with Delamarre's statement. For one thing, both Geneva and Genoa assert that their name comes from the IE root *genu- that means "knee", not the one that means "cheek". Secondly, even if they do come from the "cheek" word (the "chin" meaning seems to be restricted to Germanic), what's the evidence this word was used in Celtic for the mouth of a river? In Brythonic, anyway, the word seems to be *abber-, as in Aberystwyth and Aberdeen. In Old Irish gin "mouth" is used for a mountain pass, not a river mouth. Also, according to Germanic tribes and Marius, the Ambrones were a Germanic tribe, not a Celtic one. --Angr/_{tɔk tə mi} 17:44, 12 July 2005 (UTC)

In the first century CE and earlier, 'German' was frequently used as a geographical label (people in the area bounded by the Rhine, the Elbe, the north sea and the Alps) without necessarily indicating a linguistic definition. But I will look for more info on the Ambrones. --Nantonos 18:39, 12 July 2005 (UTC)

According to Harpers Dictionary of Classical Antiquities, 1898 (Ambrones entry available online at Perseus), the Ambrones were a Celtic tribe. Yet I know from handling Dardani and Scordisci articles that ancient writers often gave conflicting ethnic identifications of many of these tribes. Decius 18:44, 12 July 2005 (UTC)

Well, in the absence of any attestations of the Ambronic language, what criteria can be used to determine what they were? And if we don't really know for sure whether they were Celtic or Germanic or even something else, then the fact that they could talk with the Ligurians doesn't help us decide what kind of language the Ligurians spoke. --Angr/_{tɔk tə mi} 20:17, 12 July 2005 (UTC)

My understanding of Plutarch's account is that, at the Battle of Aquae Sextiae, the Ligurians and the Cimbri/Teutones, fighting on opposite sides, discovered that they used the same self-designation "Ambrōnes" (at least, I believe it's a long ō), roughly meaning 'men', as a battle-cry. Like: What is it that you're shouting over there? "Men"? But that's what we call ourselves! I don't want to sound like a broken record here, but I have a working hypothesis: ambrones is the Ligurian word for 'men'. And I further suspect (OK, just a guess) that Umbri has the same root (the term seems to have originally referred to the pre-Umbrian population of Umbria -- the historical Umbrians called themselves Safins, i.e. Sabines). Pasquale 20:30, 12 July 2005 (UTC)

I looked up the quote from Plutarch in my own copy of Plutarch (Dryden translation), and there is no indication in Plutarch that the Ligurians and Ambrones began discoursing with each other at all, let alone in mutually intelligible tongues: rather, the Ambrones were shouting "Ambrones!" as a battle-cry; and the Ligurians, on hearing this term, found it to be identical to "an ancient name in their country", so the Ligurians, as they advanced to engage the Ambrones, also began to shout "Ambrones!". Both sides kept shouting this, enraging one another, till the Ligures finally fell upon them, joined by the Romans---and all of the Ambrones were slaughtered. There was no "chit-chat" going on. Decius 20:51, 12 July 2005 (UTC)

Postscriptum, I found the Dryden translation of Plutarch's Caius Marius online:[4]. If Dryden's translation is accurate, then Delamarre's interpretation is childish, unrealistic, and is not substantiated. Decius 21:59, 12 July 2005 (UTC)

And on top of all this, I don't think it is even known with any certainty whether the Ambrones were even Celts---but even if they were, the translation doesn't indicate that any actual discourse went on at all between the Ligures and Ambrones. Decius 22:13, 12 July 2005 (UTC)

I am sorry, Decius, but I was just trying to inject a little humor into the story. That is why I prefaced my imaginary quote by "Like:" (a vulgarism meaning "as if"). I thought it was obvious. In any case, the Plutarch text you quote is pretty much what I remembered. I didn't say they spoke the same language. What I said is that they "discovered that they used the same self-designation "Ambrōnes", roughly meaning 'men', as a battle-cry." Well, more or less, in any case. Indeed, it is not known with certainty if the Ambrones were Celtic-speaking, Germanic-speaking, or neither. It is also not known if they were just another name for the Cimbri and Teutones, or a third tribe that had joined the Cimbri and Teutones in their peregrinations. Plutarch's text allows for the interpretation I put forth, namely that the term was a self-designation of the Cimbri/Teutones, but it does not require it. Pasquale 22:38, 12 July 2005 (UTC)

Yes, I agree that Ambrones may well have meant "Men" or something similar. It baffles me how Delamarre was able to extrapolate such a scenario from Plutarch. In the meanwhile, I'm trying to find more references for the article. Decius 22:45, 12 July 2005 (UTC)

[edit] Ligurian

Interesting; why not write Battle of Aquae Sextiae while you're at it? So Pasquale, are you saying Ligurian is the missing link between Celtic and Italic? Are you an Italo-Celtist? I must say it seems likely there was a lot of prehistoric contact between the groups (see Talk:Vates). Maybe the tree model just breaks down here, and Ligurian was "half Italic, half Celtic"? I have been taught that Lepontic is a Gaulish dialect without reasonable doubt. But I'll have to go and review the evidence now. dab (ᛏ) 07:26, 13 July 2005 (UTC)

No, I am not saying either that Ligurian is the missing link between Celtic and Italic or that it was "half Italic, half Celtic", nor do I believe that the classification of Ligurian impinges on whether there ever existed an Italo-Celtic branch within Indo-European. As a matter of fact, I have serious doubts about the solidity of an Italic branch within Indo-European, let alone Italo-Celtic (I suspect the apparent Italic unity is due at least in part to later convergence, and would separate out from Italic not only Venetic, but also the Sabellic group, leaving within Italic only Latino-Faliscan with other, poorly attested languages of Southern and South-Central Italy: Ausono-Sicel ? Hernico-Apulian ?).

I consider Ligurian a separate branch of Indo-European, close but different from Celtic, and with affinities also to Sabellic and possibly Illyrian. This separate branch of Indo-European was originally flung far and wide across Western Europe, from Portugal to Slovenia and perhaps from Jutland to Abruzzi and Sardinia. It was overrun in part by the spread of Celtic: Iberian Celtic first, Gaulish later.

The problem with recognizing the Indo-European nature of Ligurian is not only its very limited attestation, but also the use of the same term to refer to a bundle of substratum lexical items identified with Liguria and neighboring regions of Italy, France, and Spain. These pre-Indo-European lexical items may very well have been actual words in Ligurian, but that does not a priori make Ligurian non-Indo-European any more than the very substantial pre-Indo-European lexicon in Latin, Greek, Germanic, etc. makes those languages non-Indo-European.

As far as Lepontic is concerned, the consensus through decades of (still valid) scholarship, up to about 1970, was a cautious leaning towards a "para-Celtic" rather than Celtic affiliation. The term "para-Celtic" was used (instead of "Ligurian") in part in order not to prejudge the separate issue of whether Ligurian was indeed Indo-European, in part because Indo-Europeanists are leery of setting up a separate Indo-European branch around a poorly attested language (the same hesitation applies to Daco-Mysian, Thraco-Phrygian, etc.) and will obviously not wish to use a LESS attested language (Ligurian) as a point of reference for a (slightly) better attested language (Lepontic).

The above situation changed with the publication of Lejeune's Lepontica (1970). While Lejeune's review of the extant inscriptions is very thorough, his rationale for ultimately identifying Lepontic as a Celtic language is rather flimsy. In any case, Lejeune never confuses Lepontic with the later Cisalpine Gaulish, on the contrary the distinction between the two is actually essential to his argumentation. But since the two languages use the same alphabet, it is often hard to tell in which of these two languages the extremely short texts of most of these inscriptions are written. (The preservation of word-final *-m in Lepontic, but not in Gaulish is one element that, as far as I know, no one has cast into doubt.)

More recently, others (Eska in particular) have gone as far as conflating the two languages, but this seems totally preposterous to me, since it completely ignores the history of who the Lepontii were (a small Alpine tribe whose presence in the area precedes the Gaulish invasion of northern Italy certainly by several centuries), just when Cisalpine Gaul was settled by the Gauls, in other words, the history of the area and people involved.

The Lepontic corpus is simply too small to come to a decision. (The Cisalpine Gaulish corpus is equally small, if not smaller, but no one doubts that it was Gaulish.) For Lepontic, only a clear case of Indo-European *p preserved (hence, not Celtic) or lost (hence, Celtic) can definitively decide the issue. The only supposedly solid example that Lejeune could muster is the proper name Uvamokozis in the Prestino inscription, an inscription written in an unusual variant of the alphabet of Lugano which allows for a voice distinction between /t/ and /d/ and has a Gaulish-looking verbal form dedu (with 3sg. preterite ending -u, instead of Lepontic -e). And since the same character for /p/ also represents /b/, one can always read PALA ('tombstone, stone, rock', the most common word in the Lepontic corpus) as /bala/ and avoid seeing a /p/ in that word. (Never mind that there is substantial evidence of a substratum lexical item *palla with similar meanings, both in the Alps and the Appennines.) So, ultimately, as you can see, it becomes sleight-of-hand. Very sad indeed.

Pasquale 18:44, 13 July 2005 (UTC)