I'm watching this: https://www.youtube.com/watch?v=wcCx43I2Vio

What do you make of his suggestion near the beginning that the second generation speaking the newly adopted language, who grow up speaking it "will acquire this newly adopted language as their first language ... they end up as native speakers of a version of the second language which has a load of features leaked over from the first language"?

That is standard substrate theory, but it does depend on e.g. how intense the new language environment is. E.g. an Indian in England or Ireland is surrounded by English all the time and can grow up speaking it totally, but someone in England during the Anglo-Saxon invasion might not have been totally immersed in Old English. There was no TV then, and many areas will have had few Anglo-Saxons.

Well, the fact that kids will acquire as their native tongue the language that they hear, warts and all, is pretty uncontroversial -- we've seen it happen lots (c.f. Indian English, African English varieties etc). The question of how much of it demonstrates a "substrate" of the parents' actual language(s) -- that's where the debate lies.

French-based creoles are remarkably similar the world over, and the argument is over whether that's because the slaves in those areas came from similar linguistic backgrounds, or whether it somehow points in some way to what are the "natural" basics of language. Several English-based creoles have notable similarities too, but others show clear indications of a substrate. (eg Polynesian languages normally make a distinction between inclusive and exclusive "we" -- ie whether or not "we" includes "you" -- and this is carried into the creoles, along with a singular/dual/plural distinction (oh look, back on topic for this forum!!))

I didn't watch the video, so I don't know what examples he uses, but there has been a lot of debate about this stuff, and I think a lot of people are taking unnecessarily extreme positions -- ie "well X is definitely of Celtic origin, so Y that could be of Celtic origin certainly is, and Z that looks like it's actually of Germanic origin must really be Celtic" vs "well Z is definitely of Germanic origin, so Y that could be of Celtic origin certainly is not, and X that looks like it's of Celtic origin must really be Germanic."

I've spoken to enough Indian and Pakistani people to know that their English is dramatically marked by an Indic substrate, and Indic languages are remarkably similar in their underlying logic to the Celtic language family. The idea that languages only have a single ancestor is pretty naive.

_________________
A language belongs to its native speakers, and when you speak it, you are a guest in their homes.
If you are not a good guest, you have no right to complain about receiving poor hospitality.

How is that? I don't know much about the Indic language family, but I'm fluent in Irish and Welsh, so I'd be interested in hearing more.

_________________
The dialect I use is Cork Irish.
Ar sgáth a chéile a mhairid na daoine, lag agus láidir, uasal is íseal

Celtic languages are VSO, Indic are SOV, which might appear to be huge difference, but it really is pretty superificial. Another superficial difference is the tendecy to use "postpositions" rather than prepositions, but there's a school of thought that says qualifiers naturally tend to all be the same direction to the words they qualify, and that anything which has some qualifiers on one side and others on the other is unstable, and shows the language is in a transitional phase.

eg. English has prepositions (to him, by her) and pre-qualifying adjectives (good example, similar example); but Celtic languages have prepositions (i ni, aig Sìne) but attributive adjective commonly after the noun (dyn mawr, té bheag)

...but I'm off on a tangent.

The things that are most similar are the tense and aspect system -- there's a tendency to use verbal nouns/gerunds and there's effectively only a present progressive and not a present simple, so a nabitual tends not to be "present" at all.

There's also a destinction between alienable and inalienable possession -- "mo" is "mera" in Hindi, but "agam" is "mera pas"... = (in) my proximity.

There are lots of very similar things where English verbs are equivalent to phrasal construction. Irish and ScG "is love at me on you"; Hindi "love (in) my proximity towards you is"

_________________
A language belongs to its native speakers, and when you speak it, you are a guest in their homes.
If you are not a good guest, you have no right to complain about receiving poor hospitality.

In fact, I'm sure i saw one puported family tree that suggested the Indic and Celtic families were the earliest to split off and migrate from the IE homeland, because they were seen to share features that were thought to represent original features that were later lost in all the other branches (or to put it another way, they appeared to have fewer neologisms). However, I can't say whether that gained any widespread acceptance. Certainly, the Italo-Celtic hypothesis is still kicking around (that Italic and Celtic were the result of a single migration that split into two families. My personal feeling is that this is likely a bit of an old-school ethnocentric view and that the link between Celtic and Indic was a newer theory originating from this century based on the notable shared linguistic features.

The fact that I've not found a family tree similar to the one I saw about 15 years ago suggests that either the new idea was largely found to be false, or the old false assumptions have just been published as truth so often that people accept them uncritically.

_________________
A language belongs to its native speakers, and when you speak it, you are a guest in their homes.
If you are not a good guest, you have no right to complain about receiving poor hospitality.

Celtic languages are not similar to Indic languages.

An early split in PIE was between the centum and satem languages (https://en.wikipedia.org/wiki/Centum_an ... _languages). Irish is a centum language; Hindi is a satem language.

Irish is definitely part of the Western PIE grouping. What is more debatable is that Celtic and Italic originated together in an Italo-Celtic family (https://en.wikipedia.org/wiki/Italo-Celtic). This theory fell out of favour, and has come slightly back into favour (https://www.jstor.org/stable/30007451), but is still a debatable theory.

But what Celtic languages are not is close to Indian languages.

I'll play the devil's advocate here; I suppose that a distinction should be made between a claim like "Celtic languages are close to Indic languages" and "Celtic languages are similar to Indic languages". The former is demonstrably untrue, the latter may be demonstrated to be more or less true relative to other language groups.

In other words, two languages or language groups may have similar features, but this does not necessitate that they are more closely related diachronically or philologically to each other than they are to other languages which do not share the same features. Irish and Welsh, for example, are VSO languages, which is relatively unusual. They share this feature with Biblical Hebrew, Classical Arabic, Filipino and Māori, but obviously, they are not as closely related to any of these languages as they are to, say, Spanish or French, which do not have this feature.

In a sense, how early two languages diverged is not necessarily relevant in a comparison of features between those languages. A sort of "convergent evolution" of linguistic features is always possible without the resulting features necessarily resulting some shared historical feature.

With that being said, I don't know what this stuff about the Indic and Celtic families being "the earliest to split off and migrate from the IE homeland" is about:



News to me if it did. I've never heard anyone make a plausible argument for it.



It's not just kicking about. It's a very defensible position still, and has been argued for as recently as 2007: "Kortlandt, Frederik H.H., Italo-Celtic Origins and Prehistoric Development of the Irish Language, Leiden Studies in Indo-European Vol. 14, Rodopi 2007", so your personal feeling that it is "old-school" is at least a bit off. As for "ethnocentric", the arguments have little to do with ethnicity, or at least originally didn't. Even if some genetic research has come into play recently, I don't think we can consider this supportive of an "old-school ethnocentric view". The argument is primarily linguistic, based on common linguistic features, located geographically not very far apart. This is indicative of a language continuum, as you'd expect to see where a single language group has existed for some time across a relatively large area. Looking at Gaulish, Lepontic, and even and Archaic Irish, it's very clear that the earliest attested Celtic Languages were very closely related to Latin. This is the reason that scholars of philology have posited that they may have been part of a single language grouping, Italo-celtic, which only separated from each other quite recently compared to other branches like Germanic, Slavic, etc.



I think the family tree you saw about 15 years ago may need its branches pruned. One argument is significantly more plausible than the other.

Well then my Hindi must be even worse than I thought!


That's a hypothetical concept for which there is a some fairly strong evidence, but the evidence doesn't change the fact that it's hypothetical -- until and unless we invent time-travel and plant recording devices with atomic batteries in places of social importance to our ancestors, nothing will ever be provable.


The problem is that a heck of a lot of historical linguistics follows a pretty outmoded way of thinking -- the whole notion of a divergent model of language, where every branch on the tree is of languages getting more different from a single shared parent. The divergent model denies the existence of substrates, whereas we have seen bucketloads of evidence of substrates -- eg Bislama: yestedei blong yumifala (yesterday belongs to us=you,me,and another 3rd person or persons); see also Scottish English which merges features of Modern English with features of Scots.

Hell, even the Insular Celtic languages seem to be a subfamily that split far too recently for them to be genuine members of the continental P and Q Gaelic families. Insular Celtic theory suggests that Brythonic only took up the word-initial P instead of Q (Welsh Pwy vs Irish Cé & ScG Cò) for socio-political reasons, positing that people wanted to sound like the druids, who were definitely from a very different group of continental Celts, and from hints to the geography, very probably true P-Celts.

The resistance to the Insular Celtic theory seems to rely quite heavily on the received wisdom of the divergent model saying that substrates don't exist... despite the increasing evidence from the empires showing that it's a very real thing. And if a member of the Q-Celtic language family could turn into a phonetically P-Celtic language through contact with a member of the P-Celtic family, well, people with dorsal consonants might stop using them when in contact with people without them. There are lots of examples of mass language change in written history. The Vikings invaded northern France and picked up the local language, but couldn't master the initial G (hence "guerre" becoming "war") and became "Normans". By the time the Normans invaded England they had almost definitely lost their initial H under French influence, leading to posh people talking about "an hotel", rather than "a hotel". Indian people started talking English during the Raj, but until recent years, very few had alveolar T, D and N, and instead tended to use retroflex ones (because that sounded closer to their ears than the apico-dental equivalents).

The whole thing is really messy and murky, and it's unfortunate that linguists tend to get very invested in looking for a knowable "truth", when there really is no such thing.

Mera pas Hindi accha nahi he, na?

_________________
A language belongs to its native speakers, and when you speak it, you are a guest in their homes.
If you are not a good guest, you have no right to complain about receiving poor hospitality.