So, not to sound like a crank, but... I find a lot of details about reconstructed PIE a little hard to believe

[bradrn]

Equally: is it just reconstructed PIE, or do other reconstructed proto-languages seem typologically odd in several ways, like proto-Uralic, proto-Semitic, or proto-Vasco-Caucasian?

Wouldn’t know about the first two, but ‘Vasco–Caucasian’ is surely rubbish.

Fair. It is a model of a language which is not known first-hand. There are probably features from different dialects and time stages in it, which means that it probably was never and nowhere spoken precisely the way we reconstruct it. But it probably gets quite close to what kind of language the IE languages descend from. Yet, the fact that the reconstruction was changed several times since it was first done tells us that we should expect further changes to it as linguists find out more about the IE languages and their relationships - if they did not get it right, say, 100 years ago, why should they have got it right now?

The thing is the classical genealogical tree model is always a simplification, because it doesn't take into account lateral relationships (i.e. substrate/adstrate), and in the case of the IE family, a huge one. Thus the reconstructed PIE is a kind of "Frankestein" whose materials come from several sources.

We’ve gone over this already. ‘PIE’ as usually considered may well be overreconstructed — massively so, perhaps — but its basic phonological characteristics are uncontroversial and well-confirmed by numerous systematic correspondences between IE languages.

[Talskubilos]

Wouldn’t know about the first two, but ‘Vasco–Caucasian’ is surely rubbish.

Just as other posited macro-families such as "Nostratic" and so on. Conflating Wanderwörter and other loanwords with genuine inherited lexicon is a guaranteed path to failure.

We’ve gone over this already. ‘PIE’ as usually considered may well be overreconstructed — massively so, perhaps — but its basic phonological characteristics are uncontroversial and well-confirmed by numerous systematic correspondences between IE languages.

I'm afraid the problem doesn't lie on the "accuracy" of the reconstruction but on the adequacy of proto-languages and the classical genealogical trees as valid models for describing language relationships.

[bradrn]

We’ve gone over this already. ‘PIE’ as usually considered may well be overreconstructed — massively so, perhaps — but its basic phonological characteristics are uncontroversial and well-confirmed by numerous systematic correspondences between IE languages.

I'm afraid the problem doesn't lie on the "accuracy" of the reconstruction but on the adequacy of proto-languages and the classical genealogical trees as valid models for describing language relationships.

Well, the discussion in this thread concerned the phonology of PIE, so the phonological accuracy of PIE is all that is really relevant here. And the comparative method works even in cases where the tree model is inaccurate, which is why we can be so sure of the phonology of PIE even though there is some uncertainty about its lexicon and subgrouping.

[Talskubilos]

Well, the discussion in this thread concerned the phonology of PIE, so the phonological accuracy of PIE is all that is really relevant here. And the comparative method works even in cases where the tree model is inaccurate, which is why we can be so sure of the phonology of PIE even though there is some uncertainty about its lexicon and subgrouping.

The thing is PIE can't be anything like a real language but a theoretical construct.

[bradrn]

Well, the discussion in this thread concerned the phonology of PIE, so the phonological accuracy of PIE is all that is really relevant here. And the comparative method works even in cases where the tree model is inaccurate, which is why we can be so sure of the phonology of PIE even though there is some uncertainty about its lexicon and subgrouping.

The thing is PIE can't be anything like a real language but a theoretical construct.

I’m confused about what you mean by ‘theoretical construct’. Are you suggesting that a ‘Proto-Indo-European’ language never existed and the ‘Indo-European’ languages have no common ancestor? Or are you saying that such a language existed, but its actual phonology was very different to how we have reconstructed it? (Personally, I accept PIE and its phonology as we have reconstructed it, though certainly it would have had differences in both consonant and vowel series; I am more sceptical about its lexicon, but I am not discussing the PIE lexicon here.)

[Talskubilos]

I’m confused about what you mean by ‘theoretical construct’. Are you suggesting that a ‘Proto-Indo-European’ language never existed and the ‘Indo-European’ languages have no common ancestor?

Not exactly. Although it's obvious IE languages share some common inheritance, I don't think it could be adequately described in terms of the genealogical tree model without incurring in an oversimplification. Even sub-groups such as Indo-Iranian, Balto-Slavic or Italo-Celtic are (in increasing order) more or less problematic.

IMHO, the IE family is the result of a series of expansion and replacement processes over a time-spam of several millenia.

[bradrn]

You haven’t answered the question. In your opinion, do the IE languages have a common ancestor, yes or no? (And I feel confident in saying that this is one of the very few times in linguistics where there is a yes/no answer.)

I’m confused about what you mean by ‘theoretical construct’. Are you suggesting that a ‘Proto-Indo-European’ language never existed and the ‘Indo-European’ languages have no common ancestor?

Not exactly. Although it's obvious IE languages share some common inheritance, I don't think it could be adequately described in terms of the genealogical tree model without incurring in an oversimplification.

OK… so in that case, how would you describe IE, if not in terms of the tree model? (In enough detail as possible, please. It seems we have a habit of misunderstanding you when you try to summarise your ideas in a couple of sentences.)

IMHO, the IE family is the result of a series of expansion and replacement processes over a time-spam of several millenia.

So are all other language families. This doesn’t invalidate the tree model, the wave model, the comparative method or anything else we’ve been discussing. (The tree model, I should note, is invalid, but not because of stratum influence, and in cases like IE it still seems a reasonable approximation.)

[Seirios]

re: /i/, /u/, perhaps a strange question. but why not treat /j/ /w/ as the consonantal allophones of /i/ /u/, i.e. the other way around? This analysis is also common in languages where /i/ /u/, /j/ /w/ don't contrast and perhaps alternate in some ways.

[WeepingElf]

re: /i/, /u/, perhaps a strange question. but why not treat /j/ /w/ as the consonantal allophones of /i/ /u/, i.e. the other way around? This analysis is also common in languages where /i/ /u/, /j/ /w/ don't contrast and perhaps alternate in some ways.

I think the main reason to treat the vowels as syllabic allophones of the consonants is uniformity with the "other" syllabic resonants, i.e. nasals and liquids. There are good reasons to treat them as one class; their behaviour is very similar.

[Travis B.]

Not this again...

/facepalm

[Travis B.]

(To reject PIE is to reject modern historical linguistics, and yes, while substrata, adstrata, and Wanderwörter are present, they do not call into question the fact that a PIE can be reasonably reconstructed based upon and is a natural conclusion of present-day linguistic knowledge.)

[WeepingElf]

(To reject PIE is to reject modern historical linguistics, and yes, while substrata, adstrata, and Wanderwörter are present, they do not call into question the fact that a PIE can be reasonably reconstructed based upon and is a natural conclusion of present-day linguistic knowledge.)

And we have a pretty good idea now where, when and by whom it was spoken.

[Talskubilos]

And we have a pretty good idea now where, when and by whom it was spoken.

Well, answering your previous question, I don't think PIE was a real language spoken by actual people, but rather a kind of collage from several ancient languages. At best, the IE languages would descend from a dialectal continuum who evolved and interacted with other languages over a time-spam of several millenia.

IMHO, PIE should be seen as a convenient fiction for comparative purposes.

[Talskubilos]

(To reject PIE is to reject modern historical linguistics, and yes, while substrata, adstrata, and Wanderwörter are present, they do not call into question the fact that a PIE can be reasonably reconstructed based upon and is a natural conclusion of present-day linguistic knowledge.)

It would be necessary to study and identify these "substrata, adstrata, and Wanderwörter". To quote some examples already mentioned, words like "horse" and "wheel" are likely candidates for Wanderwörter.

[alice]

Equally: is it just reconstructed PIE, or do other reconstructed proto-languages seem typologically odd in several ways, like proto-Uralic, proto-Semitic, or proto-Vasco-Caucasian?

Wouldn’t know about the first two, but ‘Vasco–Caucasian’ is surely rubbish.

Ah, but does that make it "typologically odd"?

[Ketsuban]

Well, answering your previous question, I don't think PIE was a real language spoken by actual people, but rather a kind of collage from several ancient languages. At best, the IE languages would descend from a dialectal continuum who evolved and interacted with other languages over a time-spam of several millenia.

All you're doing here is making the comparative method sound disreputable by describing it uncharitably. Yes, Proto-Indo-European is a reconstruction spoken by no actual human - it has to be, we have no written records from the region and period during which we believe the language(s) ancestral to the Indo-European family were spoken - but that doesn't mean the language(s) ancestral to the Indo-European family (which for the sake of brevity we call Proto-Indo-European knowing that they are almost certainly not identical to the reconstruction) didn't exist.

However, you're not just uncharitably describing the comparative method - you're building a strawman to justify calling PIE "a kind of conlang" and say things like "the problem [lies] on the adequacy of [...] the classical genealogical trees as valid models for describing language relationships". We're not nineteenth-century Darwinists trying to draw up the family tree of Clostridium botulinum here - we're aware of substrate and adstrate influence and Wanderwörter, we just require extraordinary evidence for your extraordinary claim that genealogical relationships between languages (whether tree-like or wave-like) are definitionally invalid.

[Talskubilos]

All you're doing here is making the comparative method sound disreputable by describing it uncharitably. Yes, Proto-Indo-European is a reconstruction spoken by no actual human - it has to be, we have no written records from the region and period during which we believe the language(s) ancestral to the Indo-European family were spoken - but that doesn't mean the language(s) ancestral to the Indo-European family (which for the sake of brevity we call Proto-Indo-European knowing that they are almost certainly not identical to the reconstruction) didn't exist.

However, you're not just uncharitably describing the comparative method - you're building a strawman to justify calling PIE "a kind of conlang" and say things like "the problem [lies] on the adequacy of [...] the classical genealogical trees as valid models for describing language relationships". We're not nineteenth-century Darwinists trying to draw up the family tree of Clostridium botulinum here - we're aware of substrate and adstrate influence and Wanderwörter, we just require extraordinary evidence for your extraordinary claim that genealogical relationships between languages (whether tree-like or wave-like) are definitionally invalid.

By their own definition, proto-languages are a special type of conlangs. IMHO, the problem lies on the notion of a "common ancestor" language for large and complex families like the Indo-European one. On the other hand, I don't think these "substrate and adstrate influence and Wanderwörter" have been adequately studied yet.

[Imralu]

At best, the IE languages would descend from a dialectal continuum who evolved and interacted with other languages over a time-spam of several millenia.

Whence come dialect continua? Latin was a language which spread and went through a phase of dialect continua before those dialects drifted far enough apart from each other to be separate languages. No one is disputing that PIE split into daughter languages and, of course, before a distinct split, there would have been a dialect continuum or several. Do not dialects of a language have a common ancestor?

Your arguments are basically like "Tyrannosaurus rex certainly didn't look exactly like these reconstructions. We've probably got the colours wrong. Maybe the posture and shape of the fleshy tissues. Therefore, Tyrannosaurus didn't really exist and was, at best, a cluster of related species."

[Talskubilos]

At best, the IE languages would descend from a dialectal continuum who evolved and interacted with other languages over a time-spam of several millenia.

Whence come dialect continua? Latin was a language which spread and went through a phase of dialect continua before those dialects drifted far enough apart from each other to be separate languages. No one is disputing that PIE split into daughter languages and, of course, before a distinct split, there would have been a dialect continuum or several. Do not dialects of a language have a common ancestor?[/quot

Your arguments are basically like "Tyrannosaurus rex certainly didn't look exactly like these reconstructions. We've probably got the colours wrong. Maybe the posture and shape of the fleshy tissues. Therefore, Tyrannosaurus didn't really exist and was, at best, a cluster of related species."

The thing is PIE is more like a "Frankenstein" with parts from various sources.

[keenir]

All you're doing here is making the comparative method sound disreputable by describing it uncharitably. Yes, Proto-Indo-European is a reconstruction spoken by no actual human - it has to be, we have no written records from the region and period during which we believe the language(s) ancestral to the Indo-European family were spoken - but that doesn't mean the language(s) ancestral to the Indo-European family (which for the sake of brevity we call Proto-Indo-European knowing that they are almost certainly not identical to the reconstruction) didn't exist.

However, you're not just uncharitably describing the comparative method - you're building a strawman to justify calling PIE "a kind of conlang" and say things like "the problem [lies] on the adequacy of [...] the classical genealogical trees as valid models for describing language relationships". We're not nineteenth-century Darwinists trying to draw up the family tree of Clostridium botulinum here - we're aware of substrate and adstrate influence and Wanderwörter, we just require extraordinary evidence for your extraordinary claim that genealogical relationships between languages (whether tree-like or wave-like) are definitionally invalid.

By their own definition, proto-languages are a special type of conlangs. IMHO, the problem lies on the notion of a "common ancestor" language for large and complex families like the Indo-European one. On the other hand, I don't think these "substrate and adstrate influence and Wanderwörter" have been adequately studied yet.

and how do we study the waderworts and sub/ab-strates that are even more unknown and speculative than PIE itself?

Your arguments are basically like "Tyrannosaurus rex certainly didn't look exactly like these reconstructions. We've probably got the colours wrong. Maybe the posture and shape of the fleshy tissues. Therefore, Tyrannosaurus didn't really exist and was, at best, a cluster of related species."

The thing is PIE is more like a "Frankenstein" with parts from various sources. [/quote]

so, you're saying this PIE Tyrannosaurus has parts of Giganatosaurus and Megalosaurus and Triceratops?