Celts

[Sailgios]

A deep study of catalan names, province by province, would be helpful to discern any pre roman, pre gaulish and pre iberian IE layers (if there were any) in catalonia.

Curchin kind of did that. Although I don't have his work for Catalonia I do have his classification of toponyms for Roman era Comunidad Valenciana and I think the results shouldn't be too different. I'll quote below, although it's in Spanish.

Ibéricas (9 nombres = 14% del total): río Saetabis;
poblados ¿Edeta?, Ilerda, Ilici, otobesa, (Portus) Illi-
citanus, Saetabicula, Saetabis, Saguntum.

Púnicas (2 = 3%): poblados Cartalias, Carthago
nova.

Indoeuropeos (22 = 35%): ríos Alebus, Lesyros,
Pallantia, Sicanus, Sucro, Tader, Turia, Udura; llanura
Dêra; ¿promontorio Tenebrion?; poblados Bernaba,
Celeret, ¿Elo?, ¿Kelin?, Leiria, Menlaria, namnatius
(Portus), Sarna, Sebelaci, Sicana, Sucro, Tyris.

Griegos (12 = 19%): llanura Spartarion pedion;
isla Strongyle; promontorio Trete; poblados Akra Leu-
ke, Alonai, Aspis, Cherronesos, Crabasia, Hemerosko-
peion, ¿Herna?, onussa, ¿Thiar?

Latinos (11 = 18%): promontorios Ferraria, Scom-
braria; poblados Ad Aras, Ad Leones, Ad Statuas, Ad
Turres, Alterum, Dianium, Lucentum, oleastrum, Va-
lentia.

Inciertos (7 = 11%): río Sorobis; poblados Ad nou-
las, Arse, Beleia, Hyops, Lassira, Lauro.


Dado que la región de contestania y edetania ya había tenido un proceso de urbanización y establecimiento de ciudades (oppida) en la época prerromana (grau Mira, 2002; Almagro-gorbea, 2003), no es sorprendente una toponimia ibérica de proporción igual a la latina. Además, la colonización griega y el nombramiento de rasgos físicos por navegantes griegos explican el número amplio de topónimos griegos. Pero lo que destaca principalmente es la gran proporción de nombres indoeuropeos, que deben de representar una o varias lenguas indoeuropeas desconocidas, relacionadas con un asentamiento de hablantes del indoeuropeo, quizás anterior a la llegada de los iberos. La presencia extensa de topónimos indoeuropeos en regiones previamente consideradas como ibéricas, ya ha sido señalada por villar (2000); ahora se puede documentar su predominio en la región contestano-edetana.

Naturally Catalonia, being to the north of Comunidad Valenciana, might have had some Celtic toponymic influences considering it bordered Gaul, but it seems more plausible that it had a similar story as CV - at a time IE-speaking (or partially populated by such) that became part of the Iberian-speaking sphere probably long before Greeks and Romans came along. Celtic-speakers might have by-passed the region, maybe settling in less populated areas of the Meseta and then moving West in the following centuries. Around here they might not have arrived until the 5th century, with the earliest ones probably in the south, then moving north along the coastline maybe a couple of centuries afterwards. Celtic-speakers would still be arriving in the NW during the Roman period, hailing from the Meseta, at least as far as I know personal names and theonyms suggest so.

[Cejo]

1.  Around 500AD a completely new form of Irish Gaelic called ‘old irish’ became the accepted new across dozens of tribes including many who were enemies. Prior to this they had spoken archaic/primitive Irish which was much closer to iron age Celtic. This spread uniformly through Ireland despite no political power that could explain it, migration etc.

...I'm not sure about this specific example. ...

...Otherwise they’d be packing of the future princes to be fostered but find them returning home at 16 unable to speak their own biological tribe’s dialect. I actually suspect this weird practice of the elites temporarily swapping children was designed to keep dialect converging and not diverging. ...

Going back to Ireland in 500AD, this article paints a different picture of how Irish patronage worked at that time. Essentially, it was cliental:

The delegation of children to subordinates would be
accompanied by the grant of a fief (rath), normally accounted in cattle,
awarded by lords in exchange for allegiance from their clients, as well as
stipulated renders of annual yields of livestock, together with military ser-
vices or hospitality dues to a lord among noble grades of “free clients”
(so´er-che´ile), or food-renders and labor services from “base clients”
(do´er-che´ile).

Fosterage would thereby play an intrinsic role in
underwriting corporate clientship in early medieval Ireland, underpinning
the whole hierarchy of rank-grades that organized the internal differentiation
of cene´la conical clans comprising its tu´atha petty kingdoms (Charles-
Edwards 1986).

The strategic fostering of offspring was an obvious way of consolidating clientship in com-
petition with rival patrons, where the size of one’s clientele determined a
man’s rank and honor-price. Obtaining fosterage of a lord’s child, on the
other hand, could be a prudent investment for a freeman’s social advancement:
particularly if he could secure the supremacy of his royal or noble fosterling
against the latter’s dynastic rivals, earmarked for slaughter

This leads me to believe that fosterage operated within the context of localized petty kingdoms, where it served a fundamental role in reinforcing the clientship structure and hierarchy that defined those kingdoms. It did not involve the exchange of children across large regions in a manner that would have accelerated the spread of a new dialect. It was more like sending the young prince down the street to live with the neighbors.

Basically, the fosterage system was vertical, but the dialect spread horizontally, so something else must have been at play.

[Strabo]

Some Iberian and potentially Belgic connections show up as well with the Suessetani tribe. Notice similar if not identical words show up in the lists above of Old Latium. Corbio, Setia, and also Suesse-Pometia (old Latin city not mentioned)

 
Speaking of the Suessetani from the Iberian penninsula...

"Replications of Gaulish Toponyms in Biscay: On the Etymologies of Gorbeia, Orobio and Orozko" (Journal of Celtic Linguistics 24, 1-34)

academia.edu/96997659/2023_Replications_of_Gaulish_Toponyms_in_Biscay_On_the_Etymologies_of_Gorbeia_Orobio_and_Orozko_Journal_of_Celtic_Linguistics_24_1_34_

Do not know why the author insists on a La Tene era date for the supposed arrival of Belgic Celtic speaking groups into northern iberia when he hiimself says IIRC that there is no or not much La Tene materials in the local archaeology of the proposed settlement zones. Perhaps an even earlier pre La Tene dating would be an option? The Parisii of Britiain are supposed to have arrive IIRC around the 8 th century bc, no?

[Strabo]

RSFO has an intesting dustrubtion, occupying the centre of France and therefore the uppper reaches of most French rivers but at the same time never displacing the northern or southern coast tribes who retained the coast and fhe lower party of the rivers. Which on one hand seems to leave the sea routes in the hands of others but it also means they would control the cross-land trade router because they occupied the area of central France where the upper parts of the rivers of France sort of converge. But they do not seem to try to displace the coastal tribes in the north. west or south. Then in Hallstatt C a weird thing happens. The flow reversed and much of the French hallstatt C material is of the British centred insular version and it penetrates deep into central France, even to Switzerland. That material must have streamed in via the northern French rivers upstream to the RSFO area using an established contact route. Why? Presumably because the collapse of urnfield must have cut off supplies from the east.

It seems from all this that the RSFO had friendly relations and no intention of conquering the sea facing tribes of northern France and that continued right up to the collapse of urnfield and rise of Hallstatt C.

The articles of Alain Henton that I shared have demonstrated TWICE that there were population movements from eastern france into north east France, Belgium and perhaps even Kent in the Urnfield era , followed by more movements into north east France, Belgium in the EIA-MIA period. The Urnfield period movements are not only coming from RSFO region but also the Main Souabe group in western Franconia and (western?) Baden Wurttemberg

[Cunobelinus_T]

That at first seems a very weird practice but it makes sense if it was specifically designed back in the bronze age as a means of keeping of keeping constant convergence and mutual intelligibility among elites spread all over. It’s also very useful for political alliances. I actually think fosterage is the magic ingredient that only very recently has become identified as very important in IE studies. It was discussed in the most recent book on the subject.

Any idea what the book is that you're referring to here, Alan? Fosterage seems to me, too, to be a hugely plausible 'magic ingredient' in the BA/IE so it'd be great to follow this book up.

[alanarchae]

one thing i’m v skeptical about is the treating of the insular Celtic proto node as dating the split of insular Celtic breaking into branches. The British and Irish Celts had phases of strong interaction through the bronze age, late iron age, roman era and immediate post roman era. They imo were not at all seperate enough to use a divergence by time model. They had a lot of opportunity for phases of convergence.

As for the idea that insular Celtic broke off from all continental celtic before any of the latter broke up from each other, that is a tricky one. It would place its break earlier than Lepontic. The last wave of significant influence in the Lepontic area before the first recordings of it is Hallstatt D. However many link it to even further back Gollasecca (starting 900BC) and even Canagrate. Even Gollaseccs would push the date of the insular Celtic branch off to no later than the 10th century BC. Obviously if Lepontic pushed back to RSFO related urnfield Canegrate, placing Lepontic’s origins in the 1200sBC has s knock-on effect on the dating of the insular celtic node if you see the latter as branching off before the continental group split up. That’d push insular Celtic into the middle bronze/v early late bronze age and obviously that pushed insular Celtic deeper into the middle bronze age.


NB- i’m not entirely convinced by the model of insular Celtic as the first break off from the rest but I thought I would highlight the implications of that model when you also follow the usual explanation for Lepontic as linked to Canegrate.

FWIW I suspect the insular Celtic branch existed on the northern shores of France but was overlaid in iron age by standard Gaulish. I have a couple of reasons for thinking this. Firstly the interaction between Britain and northern France seems far too intense in the mid-late bronze age to imagine they didn’t speak the same dialect or at least greatly converge. Secondly (and this is a matter of great controversy as to whether this existed at the start or was a late insular development) the strange VSO word order that some imply was caused by a neolithic substrate is LEAST likely to happen in a place like Britain where the beaker folk basically wiped out/totally outbred the pre beaker population - much more do than in much of France/central Europe. So IF (a big IF) the VSO syntax is an original feature of insular Celtic then you’d imagine it was formed in a place with a much higher ENF substrate/mtDNA. One area where the beaker impact looks rather slight is NE France. Indeed we tend to forget how patchy beaker was with superconcentrations in a few nodal areas and vast gaps in between.

[Cejo]

...the strange VSO word order that some imply was caused by a neolithic substrate is LEAST likely to happen in a place like Britain where the beaker folk basically wiped out/totally outbred the pre beaker population - much more do than in much of France/central Europe. So IF (a big IF) the VSO syntax is an original feature of insular Celtic then you’d imagine it was formed in a place with a much higher ENF substrate/mtDNA. One area where the beaker impact looks rather slight is NE France. Indeed we tend to forget how patchy beaker was with superconcentrations in a few nodal areas and vast gaps in between.

What do you think of Booth et al. 2020 (Tales from the Supplementary Information: Ancestry Change in Chalcolithic-Early Bronze Age Britain Was Gradual with Varied Kinship Organization)?

This paper suggests that the transition in Britain may have taken as many as 16 generations, and was characterized more by "continuity and syncretism rather than disruption and replacement", but I've only just started reading through it.

[alanarchae]

...the strange VSO word order that some imply was caused by a neolithic substrate is LEAST likely to happen in a place like Britain where the beaker folk basically wiped out/totally outbred the pre beaker population - much more do than in much of France/central Europe. So IF (a big IF) the VSO syntax is an original feature of insular Celtic then you’d imagine it was formed in a place with a much higher ENF substrate/mtDNA. One area where the beaker impact looks rather slight is NE France. Indeed we tend to forget how patchy beaker was with superconcentrations in a few nodal areas and vast gaps in between.

What do you think of Booth et al. 2020 (Tales from the Supplementary Information: Ancestry Change in Chalcolithic-Early Bronze Age Britain Was Gradual with Varied Kinship Organization)?

This paper suggests that the transition in Britain may have taken as many as 16 generations, and was characterized more by "continuity and syncretism rather than disruption and replacement", but I've only just started reading through it.

I’ll have a read of that shortly. Looks interesting. I do agree that in some places like Britain and Ireland it’s very much easier to see it as a gradual convergence on network based thing. There is no archaeological horizon in the bronze and iron age that looks like it came in an took over the place. Instead there is constant elite interaction with the near continent. And although there appears to be signifant autosomal change in the c.1200-800BC era this is not reflected in yDNA to any significant degree. Even high status graves still look very much L21dominated. Unless you arrive at that info with a bias towards a particular model, it’s obvious that the best gut interpretation is networking with a constant modest gene flow along the network. Some migrations likely did happen but I think that was not the main process. I think the model is sometimes called cumulative Celticity. 

There is an interesting lasting frontier between the RSFO and other urnfield groups and the north French and coastal Belgic groups. Basically for centuries the Urnfield groups on the upper reaches of the french rivers and the Atlantic/English channel french groups in the lower reaches of the rivers kind of butted each other in the middle reaches. But they seem to have had trade relations from the start of RSFO in north/central all the way to 700BC. In fact Hallstatt C saw a very strong insular phase of influence in central France. So there seems to have be a long relationship. The RSFO spread across the upper to mid stretched of the northern rivers leaving the lower reaches. It seems to me that quite a lot of the RSFO spread within France was in central areas that were peripheral to the main centres of earlier settlement. Kind of infilling. 

I don’t take the ‘game of thrones’ type view of cultures always hell bent on destroying each other. I think it’s possible a largely landlocked culture like RSFO could have had a symbiotic friendly relationship with maritime cultures who were masters of the sea. That might have been the view RSFO groups had of the Atlantic/channel coastal tribes of France and possibly also the relationship they would have had with proto Ligurian cultures on the nearby Med coasts. The human level of it could be intermarriage and fosterage to seal political relationships etc. Its hard to say the effect on elite dialects but it’s possible there was one. Of course there could also be hostile relationships and invasions but I don’t think that would be the normal baseline. I think it’s rather silly to think of the movers and shakers of the bronze age as psychopaths constantly risking death in fights with similarly armed enemies.

[alanarchae]

...the strange VSO word order that some imply was caused by a neolithic substrate is LEAST likely to happen in a place like Britain where the beaker folk basically wiped out/totally outbred the pre beaker population - much more do than in much of France/central Europe. So IF (a big IF) the VSO syntax is an original feature of insular Celtic then you’d imagine it was formed in a place with a much higher ENF substrate/mtDNA. One area where the beaker impact looks rather slight is NE France. Indeed we tend to forget how patchy beaker was with superconcentrations in a few nodal areas and vast gaps in between.

What do you think of Booth et al. 2020 (Tales from the Supplementary Information: Ancestry Change in Chalcolithic-Early Bronze Age Britain Was Gradual with Varied Kinship Organization)?

This paper suggests that the transition in Britain may have taken as many as 16 generations, and was characterized more by "continuity and syncretism rather than disruption and replacement", but I've only just started reading through it.

A very interesting read and i’ve certainly always believe the speed of genetic turnover is exaggerated by the fact that only a small minoruty received recoverable burial and the native pre beaker burial tradition was low key cremation anyway. I’ve no doubt that even in Britain it took many centuries to outbreed the locals and there must have been living largely in parallel with each other for some time without intermarrying much. Though I still believe that the lack of mixing is significant when thinking about languages and would suggest a fairly unadulterated dialect of Rhenish derived beaker would have existed. That’s rather different to areas which absorbed a ton of ENF by taking mostly ENF wives.,Or somewhere like nordic europe where there was a mix of single grave, battle axe and bell beaker. 

That all said, in the post-beaker BA there was then a several phases of elite interaction and we don’t know what effect that had on dialect c. 2200-800BC. It might be the elites has a lingua franca over vast areas and the prestige of it spread it down the social scale to entire populations. That factor would be more powerful from say 1300BC on when hillfort central places controlling crafts, materials and exchange may have created a powerful mechanism.

[pelop]

There was an abstract announced recently that might be of relevance to the discussion here