European languages before Indo-European

[linguoboy]

Assumptions of in-progress irregularity are a problem for chain shifts as well. Sound change has no memory. If we assume that e.g. the Great Vowel Shift originally rampaged thru English words one by one, should we not find abundant examples where a word underwent one change such as /aː/ > /ɛː/ during an early period, then got hit by a wave of /ɛː/ > /eː/, slightly later by /eː/ > /iː/, and finally was targetted by a late torrent of /iː/ > /əi/? Nothing such happens at all.

That's not true. I can't think of any examples for the GVS, but I can for the Second Germanic consonant shift. PGmc *þūsundī should've yielded Standard German **dausend, but it didn't. Even though the attested OHG form is expected dūsunt, its modern descendent is tausend. Similarly, PGmc *dwergaz yields OHG twerc but StG Zwerg (not *Twerg).

[Tropylium]

Assumptions of in-progress irregularity are a problem for chain shifts as well. Sound change has no memory. If we assume that e.g. the Great Vowel Shift originally rampaged thru English words one by one, should we not find abundant examples where a word underwent one change such as /aː/ > /ɛː/ during an early period, then got hit by a wave of /ɛː/ > /eː/, slightly later by /eː/ > /iː/, and finally was targetted by a late torrent of /iː/ > /əi/? Nothing such happens at all.

That's not true. I can't think of any examples for the GVS, but I can for the Second Germanic consonant shift. PGmc *þūsundī should've yielded Standard German **dausend, but it didn't. Even though the attested OHG form is expected dūsunt, its modern descendent is tausend. Similarly, PGmc *dwergaz yields OHG twerc but StG Zwerg (not *Twerg).

I admit that tausend is an intriguing case (I have seen various attempts at explanation) — but Zwerg is actually quite regular, due to a secondary shift /tw/, /dw/ > zw in Middle High German times. Cf. *θwinganan > OHD dwingen > zwingen, and the general lack of tw-, dw- in modern German. This might well have originated as a case of the High German shift partially "wrapping on itself" (the English "second" vowel shift of ea /ɛː/ > /eː/ further on to /iː/ also comes to mind), but in any case it did not operate randomly.

Labov is on my reading list, yes, but I haven't heard of his results in this area having been replicated very well. I actually recently happened on a similar study that explicitly supports the environment-by-environment model:
https://www.academia.edu/7981563/A_seem ... plications

The general bottom line I guess is that "irregularity" is not an explanatory phenomenon to itself, it is an admission of ignorance on the causes of a phenomenon.

[Salmoneus]

No. It was a dogma of the Neogrammarians, who prevailed in the late 19C over the earlier view that sound changes were merely tendencies. But the evidence on ongoing sound changes is that they diffuse word by word, and sometimes the process runs out of steam, leaving a few exceptions.

Larry Trask discusses this in his Historical Linguistics (ch. 10), and William Labov talks about it in his massive Principles of Linguistic Change.

It's still nearly regular on historical timeframes, and it's also the best rule of thumb: i.e., assume that a change is regular till proven otherwise, because if you don't look for subtle, conditioned regularities you probably won't find them.

I am skeptical, FWIW. Sound changes in progress certainly diffuse. But do they do it on a word-by-word basis?

Yes. Probably most people here are able to observe that in their own speech: the same phoneme in the same context in different words is often subject to sub-phonemic variation, and this in turn results in strange differences between words that 'should' be in the same class. This is often differentiation along a spectrum of usage frequency.

On example that springs to mind that's been discussed here recently is canadian raising in some American dialects. For many speakers, this is very irregular and hard to generalise about, often down to the level of particular words being anomolous (eg iirc 'spider' lacks raising for many speakers, and 'hydrogen' has it for some)

If they really did this, it is quite puzzling why sound changes still do run to completion, historically. And even more puzzling why you sometimes find what looks like a diffused sound change that in language A has been applied wider than in language B. We'd expect instead sound changes to leave leftovers all over the place, and to stop in all sorts of random places instead of nicely conditioned ones.

Sound changes often don't run to completion, historically. This is a big reason why there are irregularities in English spelling.

Just to give a particularly egregious example, English a-broadening. In Ireland this was, at least originally, a new phoneme confined to just a single word (father): in the US, it was before coda /r/, plus in the word 'father'. Later, many dialects irregularly analogised other words to 'father'. Still later, a whole heap of words got moved into this category, while others didn't, with little concern for environment. So, for instance, it's /pAs/ but /b{s/, yet /grAs/ but /kr{s/. /mAsk/ yet /m{skQt/. /pAst3/ yet /p{st@/, /p{sti/ and /p{st@l/.All words with /a/ followed by /S/ now have /{/, with the sole exception of 'moustache', which has /A/. Even between two words that were once phonemically the same: /k{nt/ vs /kAnt/, for instance. There are also words where speakers of the same dialect may differ. I, for isntance, have a flat vowel in 'lather', even though other SSBE speakers have a broad vowel - despite the fact that this sound change happened centuries ago, and predates the trap-bath split that I do have. Or take trans-, where we're seeing secondary analogisation: originally this prefix had broad A before a voiceless consonant, and flat before a vowel or voiced consonant, but this wasn't regular, and in any case the flat vowel (and sometimes even the voiced fricative) is gradually spreading back to the broad contexts: words like 'translate', 'transport', 'transfer' and so on can be heard with either broad or flat A.

Similarly old /au/: hence now /slAnt/, /l{nt@n/ and /flOnt/, all of which once had the same vowel in them. Similarly raising in America distinguishing 'can' from 'can', and the bad-lad split that applies to no words with /{d/ except specifically for 'mad', 'bad' and 'glad' (but may be more widespread now for some speakers).

Etc.

And the very concept of "irregular change" is epistemologically troubling. Clearly something must decide if a word in a given speaker's idiolect changes its shape or not. If it's not a change in the underlying phonology, then what? Are we to attach half-lifes to individual instances of phonemes? — And of course, does this not imply that "regular changes" consist of iterated irregular changes, and therefore the apparent regularity is merely an accidental epiphenomen of some sort?

No, because phonemes exist. The existence of probabilistic processes is not epistemically troubling at all.

Assumptions of in-progress irregularity are a problem for chain shifts as well. Sound change has no memory. If we assume that e.g. the Great Vowel Shift originally rampaged thru English words one by one, should we not find abundant examples where a word underwent one change such as /aː/ > /ɛː/ during an early period, then got hit by a wave of /ɛː/ > /eː/, slightly later by /eː/ > /iː/, and finally was targetted by a late torrent of /iː/ > /əi/? Nothing such happens at all.

That assumes that both the shift and the phonemic merger happen before the second part of the shift. In reality, it's not that /a:/ suddenly 'became' /E:/, it's that /a:/ moved toward the position where /E:/ was - that needn't mean that any lagging /E:/ words would necessarily merge with vanguard /a:/ words! Although I strongly suspect that you could find a few examples of this sort of thing if you looked.

Regularity can of course be greatly mixed up afterwards by dialectal or sociolectal mixture. But such complications do not seem like sufficient evidence to conclude that all the historically known examples of regular conditioned sound changes happened as unsystematic "double-slit experiment" cascades of a word here, a word there, with no guiding overall principle observable during any one change.

Strawman!


EDIT: it's not that irregularities have no cause, it's that those causes are not generalisable.

Most of the above mess around /a/ can be explained pretty simply:
- more commonly used words were more likely to broaden
- words that were perceived as more elevated, as words of foreign ancestry, or as more technical words were less likely to broaden
- words were more likely to broaden if they were semantically or phonologically associated with words that broadened; likewise, they were less likely to broaden (or more likely to un-broaden) if they were associated with words with flat vowels.

Unfortunately, both differentiation by context and change through analogy are inherently particular processes - we can make generalisations, but not absolute rules.

[Richard W]

The easiest to understand word-by-word changes are mergers, where the change can often be simply conceived of as a reanalysis. For some mergers, there may be a sudden rush at the end as new learners simply fail to learn the phoneme difference. Alternatively, the change may simply stop, as happened with Latin /ae/ > /eː/.

[WeepingElf]

I have taken another look at Kroonen's paper. The phenomenon one could call the 'alpha mobile' - the existence of doublets with and without an initial *a - is IMHO best explained by assuming that the donor language had a definite article *a. Words from languages with articles are sometimes borrowed with the article, sometimes without. Just look at all those Arabic loanwords beginning with al-; there is a nice doublet in English: alchemy < al ximija vs. chemistry < ximija (using an IPA-based ad hoc transcription for Arabic here). In German, we have the expression aus der Lameng 'easily' with Lameng from French la main 'the hand'. As a side note, I decided years ago, at a time when I wasn't aware of the alpha mobile, that Old Albic, my "re-creation" of a lost language related to the substratum language discussed here, has the definite article a!

The paper is also supportive of my guess that the language of Neolithic Central Europe, which most of the words in question are either from or mediated by, had a three-vowel system /a i u/, in which /a/ was much more frequent than the others.

[WeepingElf]

Would you mind taking your discussion of English pronunciations to somewhere else? It does nothing to shed light on the pre-Indo-European linguistic landscape of Europe.

[Terra]

Assumptions of in-progress irregularity are a problem for chain shifts as well. Sound change has no memory. If we assume that e.g. the Great Vowel Shift originally rampaged thru English words one by one, should we not find abundant examples where a word underwent one change such as /aː/ > /ɛː/ during an early period, then got hit by a wave of /ɛː/ > /eː/, slightly later by /eː/ > /iː/, and finally was targetted by a late torrent of /iː/ > /əi/? Nothing such happens at all.

That's not true. I can't think of any examples for the GVS, but I can for the Second Germanic consonant shift. PGmc *þūsundī should've yielded Standard German **dausend, but it didn't. Even though the attested OHG form is expected dūsunt, its modern descendent is tausend. Similarly, PGmc *dwergaz yields OHG twerc but StG Zwerg (not *Twerg).

Maybe 'Zwerg' is because German doesn't like the combinations "dw" and "tw"?

Anyways, here's two more German anomalies:
(1) 'Dunst' (cf 'dust') instead of 'Tunst'
(2) 'tauen' (cf 'thaw') instead of 'dauen'

[linguoboy]

Would you mind taking your discussion of English pronunciations to somewhere else? It does nothing to shed light on the pre-Indo-European linguistic landscape of Europe.

I have moved those posts specifically relating to English pronunciation to the "How Do You Pronounce 'X'" thread starting here. Please continue the discussion in that thread.

FYI, Tropylium has started his on spur discussion concerning the regularity (or lack thereof) of sound change here.

[WeepingElf]

Thank you, linguoboy! This step removed a lot of clutter from this thread.

[2+3 clusivity]

Anyone notice a similarity between the Etruscan -sa or -isa, "son of" patronymic suggested here: http://en.wikipedia.org/wiki/Etruscan_l ... n_literacy and the Eteo-Cypriot <a-ri-si-to-no-se a-ra-to-wa-na-ka-so-ko-o-se> "to Ariston (son of) Aristonax" suggested here http://en.wikipedia.org/wiki/Eteocypriot.

[TaylorS]

Anyone notice a similarity between the Etruscan -sa or -isa, "son of" patronymic suggested here: http://en.wikipedia.org/wiki/Etruscan_l ... n_literacy and the Eteo-Cypriot <a-ri-si-to-no-se a-ra-to-wa-na-ka-so-ko-o-se> "to Ariston (son of) Aristonax" suggested here http://en.wikipedia.org/wiki/Eteocypriot.

Glenn Gordon over at the Paleoglot blog has an amateur interest in a possible connection between Etruscan and Minoan, he would be a good person to ask.

[WeepingElf]

Anyone notice a similarity between the Etruscan -sa or -isa, "son of" patronymic suggested here: http://en.wikipedia.org/wiki/Etruscan_l ... n_literacy and the Eteo-Cypriot <a-ri-si-to-no-se a-ra-to-wa-na-ka-so-ko-o-se> "to Ariston (son of) Aristonax" suggested here http://en.wikipedia.org/wiki/Eteocypriot.

A relationship between Etruscan and Eteocypriot is of course not out of the question. Yet, Etruscan is not very well known (I have seen at least four different case paradigms posited for it, which means that the actual paradigm is not known yet, for instance), and Eteocypriot is almost completely unknown, and a single suffix with hardly known meaning simply is not enough evidence to posit a relationship.

@TaylorS: Glen Gordon ought to be taken with a few grains of salt. His ideas on Etruscan often do not agree with the scholarly mainstream, and he is so convinced of the relationship between "Aegean" and IE that he sometimes slips into circular reasoning. Also, he has a history of flaming people who do not agree with him and being banned from discussion groups on the ground of that. His ideas are not nearly as crazy as Octaviano's, but his social behaviour is not much better.

[TaylorS]

@TaylorS: Glen Gordon ought to be taken with a few grains of salt. His ideas on Etruscan often do not agree with the scholarly mainstream, and he is so convinced of the relationship between "Aegean" and IE that he sometimes slips into circular reasoning. Also, he has a history of flaming people who do not agree with him and being banned from discussion groups on the ground of that. His ideas are not nearly as crazy as Octaviano's, but his social behaviour is not much better.

Fair enough.

[TaylorS]

I'm thinking of maybe creating a fictional Western European isolate that would be a relic of the languages spoken on the Atlantic fringe before the Neolithic. Essentially the substratum beneath WeepingElf's Hesperic languages


My own gut feeling is that things like VSO and SVO word order as well as definite and indefinite articles are very old areal features of Western Europe and this language would have these features.

[Dewrad]

For what it's worth, I honestly do not think that the VSO word order attested in Goidelic and Brythonic can be ascribed to substratal effects: it's entirely explicable within a solely IE framework, without recourse to substrata.

[WeepingElf]

For what it's worth, I honestly do not think that the VSO word order attested in Goidelic and Brythonic can be ascribed to substratal effects: it's entirely explicable within a solely IE framework, without recourse to substrata.

Fair. The Semitic substratum theory has few adherents today, as the Insular Celtic languages, apart from VSO word order and a few typological correlates of it, are not much like Semitic languages at all (and there is otherwise no good reason to assume an Afrasian language to ever have been spoken in the British Isles); and invoking an unknown substratum doesn't really explain anything, so an IE-internal explanation is better.

This of course did not prevent me from creating, with Old Albic, a language that could have been a substratum language "responsible" for the aberrant typology of Insular Celtic. But hey, it is just a conlang and nothing else.

EDIT: One reason why some scholars posit an unknown substratum is that the two insular subbranches of Celtic appear to have developed nearly the same aberrant typology seemingly independently from each other (there doesn't seem to have existed a "Proto-Insular-Celtic" distinct from Proto-Celtic), but maybe they just form a Sprachbund within the Celtic branch.

[Dewrad]

EDIT: One reason why some scholars posit an unknown substratum is that the two insular subbranches of Celtic appear to have developed nearly the same aberrant typology seemingly independently from each other (there doesn't seem to have existed a "Proto-Insular-Celtic" distinct from Proto-Celtic), but maybe they just form a Sprachbund within the Celtic branch.

Given that VOS word order is only the tip of the iceberg when it comes to common features between Goidelic and Brythonic which were most likely due to language contact, I'm more in favour of this as an explanation.

[WeepingElf]

I am bumping this old thread because I have changed my mind on some things.

I used to maintain the idea that the Beaker culture was a movement of refugees from Central Europe, fleeing the (Indo-European) Corded Ware invasion and spreading a Central European para-IE language across Western Europe, and that this language has left its traces in the Old European Hydronymy.

This is wrong.

By now, it has been pretty firmly established that the Beaker people originated in the Iberian Peninsula, where the oldest C14 datings have been found. Also, their presence in Central Europe is younger than that of Corded Ware. So much, then, on the "refugee" scenario. This makes it unlikely that they spoke a language closely related to PIE, unless an earlier east-west movement had brought that language into the Iberian Peninsula. So,

either the language of the Old European Hydronymy is not related to PIE
or the Old European Hydronymy is unconnected to the Beaker culture
or both.

Actually, I now think that the OEH has nothing to do with the Beaker culture, the close resemblance of the ranges of both entities notwithstanding. The Beaker culture never seems to have been the majority population in Western Europe, but either a conquering elite or - perhaps more likely - a diaspora, like the Jews and Gypsies of later times. Conquerors might have imposed their language on their domain, but not necessarily so, and if yes, probably not to the degree that all the rivers were named in the conquerors' languages. A diaspora probably wouldn't have made any dint on the river names.

On the other hand, the Old European Hydronymy still looks as peri-IE as it used to do.

So I now think that the Beaker culture and the Old European Hydronymy have nothing to do with each other, but the OEH was connected with the spread of Neolithic agriculture across Western Europe before the time of the Beaker culture.

[jal]

By now, it has been pretty firmly established that the Beaker people originated in the Iberian Peninsula (...) So I now think that the Beaker culture and the Old European Hydronymy have nothing to do with each other, but the OEH was connected with the spread of Neolithic agriculture across Western Europe before the time of the Beaker culture.

So the Beaker culture people where proto-Basks? :)


JAL

[WeepingElf]

By now, it has been pretty firmly established that the Beaker people originated in the Iberian Peninsula (...) So I now think that the Beaker culture and the Old European Hydronymy have nothing to do with each other, but the OEH was connected with the spread of Neolithic agriculture across Western Europe before the time of the Beaker culture.

So the Beaker culture people where proto-Basks?

Possibly. They may have spoken a language related/ancestral to Basque, and may have spread some words widely across Western Europe. This may be behind the "Vasconic" lexical resemblances people like Vennemann seem to see everywhere in western European languages.

[KathTheDragon]

If they're more recent than Indo-European (which seems to be what you imply) might they also be the source of the Germanic substrate?

[WeepingElf]

If they're more recent than Indo-European (which seems to be what you imply) might they also be the source of the Germanic substrate?

A substratum is, if I am not mistaken, by definition an older layer. Now if Bell Beaker is later than Corded Ware, it may be an adstratum, though. But I don't expect very much influence from that language on what was to become Germanic. A few words, yes, but neither hydronymy nor the overall structure of the language. The substratum would be mainly Aquan, which I no longer connect to the Bell-Beaker complex, as I wrote yesterday.

Of course, it is hard to tell whether a given "orphan" word in Germanic is from the Aquan substratum, the Bell-Beaker (and possibly Vasconic) adstratum, yet another source, or actually inherited from PIE but simply not found out yet from which PIE word. Etymology sometimes takes strange turns.

[KathTheDragon]

If they're more recent than Indo-European (which seems to be what you imply) might they also be the source of the Germanic substrate?

A substratum is, if I am not mistaken, by definition an older layer. Now if Bell Beaker is later than Corded Ware, it may be an adstratum, though.

Ah, I wasn't aware of the distinction.

[WeepingElf]

At any rate, I don't expect much Bell Beaker influence in western European IE languages beyond a few loanwords, perhaps in the field of metalworking, which the Bell Beaker people were involved with spreading across western Europe. If the Bell Beaker people, as they seem to have been, were a diaspora spread across a Western European population of different languages, their influence on the latter languages wouldn't be stronger than that, for instance, of Yiddish on Polish. While Polish has a number of Yiddish loanwords (especially pertaining to Jewish culture, such as sztetl), its phonological and grammatical structure don't seem to be affected by Yiddish, and Polish rivers don't bear Yiddish names.

So far, I see no reason to abandon or radically reform my Aquan hypothesis, only that it probably has no longer anything to do with the Bell Beaker people, and a few words I ascribe to Aquan may actually be Bell Beaker words.

Indeed, the Bell Beaker languages would have functioned as an adstratum not only in the western part of Corded Ware IE (approximately between the Rhine and the Vistula), but also to the Aquan languages west of the Rhine, which was the western limit of the Corded Ware.

[TaylorS]

I am bumping this old thread because I have changed my mind on some things.

I used to maintain the idea that the Beaker culture was a movement of refugees from Central Europe, fleeing the (Indo-European) Corded Ware invasion and spreading a Central European para-IE language across Western Europe, and that this language has left its traces in the Old European Hydronymy.

This is wrong.

By now, it has been pretty firmly established that the Beaker people originated in the Iberian Peninsula, where the oldest C14 datings have been found. Also, their presence in Central Europe is younger than that of Corded Ware. So much, then, on the "refugee" scenario. This makes it unlikely that they spoke a language closely related to PIE, unless an earlier east-west movement had brought that language into the Iberian Peninsula. So,

either the language of the Old European Hydronymy is not related to PIE
or the Old European Hydronymy is unconnected to the Beaker culture
or both.

Actually, I now think that the OEH has nothing to do with the Beaker culture, the close resemblance of the ranges of both entities notwithstanding. The Beaker culture never seems to have been the majority population in Western Europe, but either a conquering elite or - perhaps more likely - a diaspora, like the Jews and Gypsies of later times. Conquerors might have imposed their language on their domain, but not necessarily so, and if yes, probably not to the degree that all the rivers were named in the conquerors' languages. A diaspora probably wouldn't have made any dint on the river names.

On the other hand, the Old European Hydronymy still looks as peri-IE as it used to do.

So I now think that the Beaker culture and the Old European Hydronymy have nothing to do with each other, but the OEH was connected with the spread of Neolithic agriculture across Western Europe before the time of the Beaker culture.

I can't remember if I have already mentioned it, but archaeologist Jean Manco in her book Ancestral Journeys argues that Italo-Celtic speakers had already dispersed as far as Iberia by the time that the Bell Beaker style developed (a dispersion marked by the spread of Yamnaya-type anthropomorphic stelae across southern Europe) and that the distinctive beakers were ultimately derived from Yamnaya pottery styles but with some "native" Iberian influence. She says that Lusitanian may be a relic of this initial dispersion, most of which was submerged by later Celtic migration.