The Great Proto-Indo-European Thread

[KathTheDragon]

are there any innovations that could be common to all IE languages except A/T? Other than assumed stop system changes like these.

According to Jasanoff, the s-aorist was originally only sigmatic in the indicative 3rd singular, and the whole subjunctive, a situation reflected in both Tocharian and Hittite, and was sigmatised throughout the whole indicative only in late PIE (leaving the optative still asigmatic, which iirc is the state in Sanskrit).

[WeepingElf]

Judging by how almost no two IE branches have similar treatment of the stops, it seems like we shouldn't be looking for an especially stable solution anyway.

A point which can be used, of course, to defend the traditional reconstructions (T D D^h, K^w K K') as well.

Right! The traditional reconstruction apparently was unstable - after all, no branch has preserved it unchanged! And many languages show some feature that is typologically rare - as long as it isn't impossible, it is possible

The other main reason I still find the standard system suspicious is that *Dʰ > *D does not seem like a "dog bites man" story. Do any precedents for an unconditional shift *Dʰ > *D exist?

Well, the answer to that question may have not much of a bearing on the issue, if it happens in a different system with different systemic pressures.

Fair. Each system has different systemic pressures, hence, different sound changes happen. (Hence, bogolangs are also unlikely, but that is another topic.) How similar is Cao Bang to PIE? Probably, not much.

Clearly enough the standard system would have been unstable. But as you can gather from the Cao Bang presentation, it seems that whenever voiced aspirates arise, they often continue on to voiceless aspirates. Aside from a couple of cases where this happened entirely on its own, this is also well-attested from languages that already have voiceless aspirates, such as in Mandarin, where the Middle Chinese *B series becomes sometimes plain voiceless, sometimes aspirated. But of the major IE groups, only Greek and probably Italic seem to have done this. It seems to me that we should expect this to have happened wider, if voiced aspirates once occurred everywhere in IE.

It is, of course, also conceivable that the voiced aspirates only were a feature of a "core PIE" after Anatolian and Tocharian split off - after all, the two resp. one series that they show may atually continue not the classical T D D^h system, but its precursor, whatever that was.

Sure. As I wrote yesterday, the *T D Dh system may even have been an innovation of the southeastern (Greco-Aryan) dialect area, while the northwest had something like *T D Ð (with voiced fricatives).

It does not seem very likely that the *T T' D system proposed by the glottalists shifted to *T D Dh in a single step. The *T' stops would have gone through an intermediate step, which may have been *T (plain voiceless; in this case, the *T stops would have to be "protected" from merger by some feature such as aspiration, like in Georgian and other Caucasian languages), or *D' (implosive). In a second step, this intermediate stage would have moved on to *D, pushing the old *D out of the way - to *Dh at least in Greco-Aryan, and perhaps to *Ð in NW IE.

It would be nice to know what the difference between the "voiceless" and the "voiced" stops in Hittite actually was, as Anatolian may have branched off during that intermediate stage. As I said earlier, there are some scholars who assume that it was actually a difference in aspiration rather than voice, which would point at *Th T D as the intermediate system.

The root structure constraints, however, seem to point at the *D grade having been a highly marked one in some stage.

(1) How so?
(2) By this, do you mean that *T and *Dh are more similar to eachother than either is to *D? The pairs *kap, *ghabh (take) and *tem, *dhem (dark, dim) come to mind, and make me wonder the same thing. But, the problem is, I haven't come across any other pairs like this, and 2 pairs doesn't seem strong enough to to be called a trend.

No two *D occur in a root (e.g., **deg- is not a possible root), while roots with two *T or with two *Dh are common, and indeed, *T and *Dh do not co-occur (e.g., *tegh- is not a possible root). This looks as if an assimilation rule was in force which removed *TeDh and *DheT roots from the language, but did not affect *D, which therefore probably did not occupy a middling position between the other classes, but was "marked" with a special feature that was typologically rare, and glottalization has been suggested as such a feature.

[KathTheDragon]

It would be nice to know what the difference between the "voiceless" and the "voiced" stops in Hittite actually was, as Anatolian may have branched off during that intermediate stage. As I said earlier, there are some scholars who assume that it was actually a difference in aspiration rather than voice, which would point at *Th T D as the intermediate system.

IMO, gemination is quite likely, as it would explain the consistent doubled writing, its application to almost all consonants, and the "lenition" laws (V­̄́CCV̀ > V­̄́CV̀ and V̀CCV̀ > V̀CV̀) very easily.

[Salmoneus]

It would be nice to know what the difference between the "voiceless" and the "voiced" stops in Hittite actually was, as Anatolian may have branched off during that intermediate stage. As I said earlier, there are some scholars who assume that it was actually a difference in aspiration rather than voice, which would point at *Th T D as the intermediate system.

IMO, gemination is quite likely, as it would explain the consistent doubled writing, its application to almost all consonants, and the "lenition" laws (V­̄́CCV̀ > V­̄́CV̀ and V̀CCV̀ > V̀CV̀) very easily.

But those could also equally apply to aspiration or voicing.


Regarding the root structure issue, an alternative could be that the symmetrical roots are a result of a conditioned split.

We could, for instance, imagine two stop series: Th vs D. Then imagine that there are two types of vowel: oral and nasal.
Add the rule that stops are prenasalised next to a nasal vowel, and then have prenasalisation slide into voicing. These being, after all, very closely-connected distinctions. And drop contrastive nasal vowels.

roots that once had an oral vowel could be ThVD, ThVTh, DVD, or DVTh.
roots that once had a nasal vowel could be DhVD, DhVDh, DVD, or DVDh.
in other words, you could never have Th and Dh stops in the same root, but both would be compatible with D stops. Which is what we see.

Not that I'm saying this is more likely than your 'three stop series with bidirectional voicing assimilation and an unassimilable series' option. Although I would point out that voicing assimilation is more often in one direction onely, rather than in both.

[Nortaneous]

Breathy-voiced stops are often reconstructed as intermediates in (S)EA -- what happens with them?

Do we see anything like voicing assimilation anywhere?

[Richard W]

Well, I meant voiceless/voiced/implosive specifically without a fourth or fifth series in addition. The Cao Bang slideshow shows that such systems do seem to exist, but they're very rare, and probably unstable. Most languages in Africa and SE Asia either have more than three series, or have three but one of them has only a few members.

The 'original' voiceless aspirates accompanying voiceless/voiced/'implosive' tend to be fairly marginal. We've seen the (highly plausible) theory that SW & Central Tai voiceless aspirates are recent developments from clusters with /r/ (a cycle that has repeated itself in some of the dialects). In Khmer, apparently the voiceless aspirates should be analysed as consonant clusters.

Also, this is just a general comment, not directed specifically at any argument here or elsewhere, but I really wish PIE linguists could just get over their addiction to unconditional sound shifts, such as "all voiced stops become aspirated unconditionally in all positions". Shifts like that just dont happen very often in the real world. Apparently they happen at least sometimes, such as Cao Bang, but it doesnt have to be the only explanation or even the most popular explanation.

In most (all?) languages of the area, including Tai languages such as Cao Bang, phonation contrasts on consonants are restricted to initial position. I also strongly suspect that Cao Bang is overwhelmingly monosyllabic.

[Richard W]

Breathy-voiced stops are often reconstructed as intermediates in (S)EA -- what happens with them?

The breathiness winds up as a vowel feature. It can affect vowel quality, as in Khmer, where breathiness itself is restricted to a few dialects. In the Tai languages, it has affected the tone, doubling the basic tone count from 3 to 6. It can, independently, be interpreted as aspiration, as seen in Thai and Lao and some languages under their influence, including some of the Mon-Khmer languages (Nyah Kur, some Khmu).

Do we see anything like voicing assimilation anywhere?

Well, there's rightward register spreading, whereby the breathiness/non-breathiness of associated with an initial occlusive is copied to the second, stressed syllable if it begins with a non-occlusive. This is known in at least Thai (Tai-Kradai), Khmer (Austorasiatic) and Cham (Malayo-Polynesian). In Thai it manifests itself as exceptions to the simple rules for deducing the tone from the spelling of words.

[Pabappa]

Another very minor comment: I just meant to say earlier i suspect that among languages that have three stop series and one of them is implovsives, most of them will have no plain voiced stops. That is, it is more like Swahili and Vietnamese where the three series are implosive/voiceless/voiceless aspirated. I'd accept though that the plain voiceless stops could easily be heard as plain voiced stops by the speakers of a language where all the voiceless stops are aspirated.

[marconatrix]

IMO, gemination is quite likely, as it would explain the consistent doubled writing, its application to almost all consonants, and the "lenition" laws (V­̄́CCV̀ > V­̄́CV̀ and V̀CCV̀ > V̀CV̀) very easily.

Could you explain those vowel diacritics. Was Hittite a tone language??

[Richard W]

It would be nice to know what the difference between the "voiceless" and the "voiced" stops in Hittite actually was, as Anatolian may have branched off during that intermediate stage. As I said earlier, there are some scholars who assume that it was actually a difference in aspiration rather than voice, which would point at *Th T D as the intermediate system.

IMO, gemination is quite likely, as it would explain the consistent doubled writing, its application to almost all consonants, and the "lenition" laws (V­̄́CCV̀ > V­̄́CV̀ and V̀CCV̀ > V̀CV̀) very easily.

I can't tell from this whether you are aware of the following complicating facts:

• The Akkadian writing system from which the Hittite writing systems developed did not express a phonation contrast - an orthographic correspondence between Sumerian and Akkadian phonation had not yet been developed. (Later Akkadian writing systems did develop such a link, and so became capable of expressing the Akkadian phonation contrasts.)
• Hurrian, with which at at least one point Hittite had strong cultural links, also appears to use orthographic gemination to express a phonation contrast.

[KathTheDragon]

IMO, gemination is quite likely, as it would explain the consistent doubled writing, its application to almost all consonants, and the "lenition" laws (V­̄́CCV̀ > V­̄́CV̀ and V̀CCV̀ > V̀CV̀) very easily.

Could you explain those vowel diacritics. Was Hittite a tone language??

Sure. The acute represents an originally stressed vowel, and the grave indicates an unstressed vowel, in keeping with usual Indo-European notation. Hittite itself may not have had contrastive stress, but Proto-Anatolian, which is when these lenition laws took place, certainly did.

[KathTheDragon]

It would be nice to know what the difference between the "voiceless" and the "voiced" stops in Hittite actually was, as Anatolian may have branched off during that intermediate stage. As I said earlier, there are some scholars who assume that it was actually a difference in aspiration rather than voice, which would point at *Th T D as the intermediate system.

IMO, gemination is quite likely, as it would explain the consistent doubled writing, its application to almost all consonants, and the "lenition" laws (V­̄́CCV̀ > V­̄́CV̀ and V̀CCV̀ > V̀CV̀) very easily.

I can't tell from this whether you are aware of the following complicating facts:

• The Akkadian writing system from which the Hittite writing systems developed did not express a phonation contrast - an orthographic correspondence between Sumerian and Akkadian phonation had not yet been developed. (Later Akkadian writing systems did develop such a link, and so became capable of expressing the Akkadian phonation contrasts.)
• Hurrian, with which at at least one point Hittite had strong cultural links, also appears to use orthographic gemination to express a phonation contrast.

I was under the impression that it was not known exactly where the Hittites acquired cuneiform from, but that Hurrian was unlikely, and that Akkadian did express a voice contrast with the Sumerian voiced vs voiceless stops. Do you have a source?

[WeepingElf]

Regarding the root structure issue, an alternative could be that the symmetrical roots are a result of a conditioned split.

We could, for instance, imagine two stop series: Th vs D. Then imagine that there are two types of vowel: oral and nasal.
Add the rule that stops are prenasalised next to a nasal vowel, and then have prenasalisation slide into voicing. These being, after all, very closely-connected distinctions. And drop contrastive nasal vowels.

roots that once had an oral vowel could be ThVD, ThVTh, DVD, or DVTh.
roots that once had a nasal vowel could be DhVD, DhVDh, DVD, or DVDh.
in other words, you could never have Th and Dh stops in the same root, but both would be compatible with D stops. Which is what we see.

One moment - things are a bit unclear to me. I guess your Th later becomes T where it is not changed by your rule, and your D and your Dh are as traditional. So I can at least guess what you are trying to say, but I may be misguessing. What I guess is that Th turns into Dh in a root with a nasal vowel, and into T in a root with an oral vowel. Right? But you list DVD as a possible outcome - but it is forbidden. You still need a rule against two Ds in a root to start with.

[Tropylium]

It would be nice to know what the difference between the "voiceless" and the "voiced" stops in Hittite actually was, as Anatolian may have branched off during that intermediate stage. As I said earlier, there are some scholars who assume that it was actually a difference in aspiration rather than voice, which would point at *Th T D as the intermediate system.

IMO, gemination is quite likely, as it would explain the consistent doubled writing, its application to almost all consonants, and the "lenition" laws (V­̄́CCV̀ > V­̄́CV̀ and V̀CCV̀ > V̀CV̀) very easily.

At least Kloekhorst seems to support this as well:
The Proto-Anatolian consonant system: An argument in favor of the Indo-Hittite hypothesis?

In the initial position this setup would be difficult to trace back to PIE though. The *T : *Dʰ contrast clearly existed in the word-initial position, but an interpretation as [tː] : [t] seems unlikely.

[KathTheDragon]

Kloekhorst seems to replace the "traditional" Sanskrit centricism with an Anatolian centricism. While I agree that the Anatolian system probably was tt::t rather than t::d, I think this can be taken from a post-PIE t::d system, via an intermediate tʰ::t, instead of necessarily reflecting the original PIE.

[WeepingElf]

Kloekhorst seems to replace the "traditional" Sanskrit centricism with an Anatolian centricism.

Fair. He is a specialist in Hittite, and you know that old saying about people with a hammer ... While it is tempting to regard things that Hittite does different from the other IE languages as archaisms, it must always be considered that Anatolian may have been the innovative branch.

While I agree that the Anatolian system probably was tt::t rather than t::d, I think this can be taken from a post-PIE t::d system, via an intermediate tʰ::t, instead of necessarily reflecting the original PIE.

Yes. Perhaps the "intermediate" tʰ::t system was the PIE one at the time Anatolian separated from Core IE. At any rate, the Hittite intervocalic geminates (if that's what they were) IMHO cannot be projected back to PIE.

As for Salmoneus's idea: it may be that an early stage had two types of stops: pulmonic *T and ejective *T'. At that stage, there was a constraint against two ejectives in one root. Permissible root types with two stops were three: TVT, TVT', T'VT, while T'VT' was forbidden. Then some (unknown) vowel feature caused some Ts to attain voicing, leading to the possible root types: TVT, TVT', T'VT, DVD, DVT', T'VD - the antecedents of the "Classical PIE" types TVT, TVD, DVT, DhVDh, DhVD, DVDh.

That would reduce the Pre-PIE stop grades from three to two, and thereby move the whole thing closer to Uralic, which has just one stop grade. Perhaps study of Indo-Uralic sound correspondences will reveal which factors conditioned the splits that happened in pre-PIE - or there were no splits in pre-PIE, just mergers in pre-Uralic.

[KathTheDragon]

Then some (unknown) vowel feature caused some Ts to attain voicing

Perhaps some phonation difference?

[WeepingElf]

Then some (unknown) vowel feature caused some Ts to attain voicing

Perhaps some phonation difference?

Maybe. As long as we don't know which difference was the cause, this idea doesn't explain anything. But it may hint at the direction where the solution is perhaps to be found.

[KathTheDragon]

Then some (unknown) vowel feature caused some Ts to attain voicing

Perhaps some phonation difference?

Maybe. As long as we don't know which difference was the cause, this idea doesn't explain anything. But it may hint at the direction where the solution is perhaps to be found.

True. But this would require some external relationship to be proven before further research can be done.

[Tropylium]

Then some (unknown) vowel feature caused some Ts to attain voicing

Perhaps some phonation difference?

Seems more likely than Salm's idea of a nasalization contrast, at least.

Kortlandt has also suggested in his paper Indo-Uralic consonant gradation more or less the same idea as here, but with stress as the trigger (and with widespread later levelling of the resulting *T ~ *Dʰ alternations).

As for Uralic correspondences, at least the PU gemination contrast (*p *t *č *ć *k : *pp *tt *čč *ćć *kk) could be cognate to one of the IE phonation contrasts (and most variations on Indo-Uralic / Eurasiatic / Nostratic propose that it is). But interestingly enough there's hardly any good evidence for a *tt in Proto-Uralic, in which light I suspect that the geminate series could instead result from some sort of earlier cluster simplification (e.g. *jp *jt *jk > *pp *ćć *kk, or *kp *pk > *pp *kk while *pt *kt remain).

[Salmoneus]

I suggested nasalisation as an example, not as something definitive. I chose nasalisation because it's known to have that effect (voicing adjacent stops), and because it doesn't involve a phonation contrast that could complicate matters. But any other contrast could have a similar effect, particularly a phonation contrast. [I'm a bit more skeptical about stress, but you never know]. Of course, the simplest would just be to have voiced and voiceless vowels...

The point was just that you didn't need to assume three series, with multiple bidirection assimilations at a distance. You could just assume two series and a contrast in the vowels. And WE rightly points out that you need something to block DVDs, but as he says this only requires, for example, ejectives with a unidirection dissimilation rule, which is very common. Actually, even unidirectional voicing dissimilation isn't that unusual.

This idea would also have a couple of other side-effects:

- if you assume the T series is also aspirated (not rare for voiceless stops), and then some are voiced, then that instantly explains why you end up with rare 'voiced aspirates'. Rather than having to suppose an unconditional shift from commonplace plain voiced stops to rare voiced aspirates

- if you do assume that the second series are ejective, this would also explain why they end up as voiced. If they DID voice in the voicing situation, you'd end up with plain ejectives vs perhaps implosives, perhaps creaky voice, perhaps 'voiced ejectives' (voiced onset, voiceless ejective release), etc. So temporarily there'd be four series. But these two series would be quite 'weird' and not that commonly distinguished (if we assume the ejective series starts out rarer than the plain), so would then merge again. The 'voiced ejective' variant could become plain voiced (if the other voiced series are actually breathy), leading the ordinary ejectives with them. This seems more motivated than a direct switch of ejectives to voiced stops.

[Richard W]

I was under the impression that it was not known exactly where the Hittites acquired cuneiform from, but that Hurrian was unlikely, and that Akkadian did express a voice contrast with the Sumerian voiced vs voiceless stops. Do you have a source?

Gelb, in 'Old Akkadian Writing and Grammar', states that the writing system for Old Akkadian did not distinguish voiceless, voiced and emphatic. By 'Old Akkadian' he means up to the end of the Ur III dynasty.

During the Ur III dynasty, Sumerian was being used for writing by many Akkadian speakers (Gelb, ibidem).

Turning to my opinions, which are largely second-hand, it is therefore likely that Sumerian was widely spoken with an Akkadian accent, and this is highly likely to have affected the pronunciation; thus the native Sumerian phonation contrasts, if different to Akkadian, will have been replaced by Akkadian contrasts. Once this had happened, Sumerian signs had an associated voicing state, and were available to record the contrasts when writing Akkadian.

Of course, it may just have been that when writing Old Akkadian, dropping voicing contrasts made the writing simpler. Gelb suggests that sign differences may have been employed rather to record the laryngeals that are apparently absent from the writing system, though they do seem to have incidental effects. Real writing systems are notorious for omitting phonemic contrasts, and introducing them into a writing system can be an up-hill task.

[KathTheDragon]

But then does that not just beg the question of why the Hittites did not simply follow the Akkadian practice of not indicating their phonation contrast, rather than introducing a practice (iirc) not present in Akkadian?

[Tropylium]

I chose nasalisation because it's known to have that effect (voicing adjacent stops),

It is? I know of nasalization in nasal+stop clusters (which you could easily get from breaking of preeding nasal vowels), but not of voicing of stops before nasal vowels.

if you assume the T series is also aspirated (not rare for voiceless stops), and then some are voiced, then that instantly explains why you end up with rare 'voiced aspirates'.

Sounds good in principle, but again, are there any actual precedents?

What kind of voicing shifts should be assumed to be probable is an empirical question, dependent on what is attested elsewhere in the world.

- if you do assume that the second series are ejective, this would also explain why they end up as voiced. If they DID voice in the voicing situation, you'd end up with plain ejectives vs perhaps implosives, perhaps creaky voice, perhaps 'voiced ejectives' (voiced onset, voiceless ejective release), etc. So temporarily there'd be four series. But these two series would be quite 'weird' and not that commonly distinguished (if we assume the ejective series starts out rarer than the plain), so would then merge again. The 'voiced ejective' variant could become plain voiced (if the other voiced series are actually breathy), leading the ordinary ejectives with them. This seems more motivated than a direct switch of ejectives to voiced stops.

What sounds like the weakest link in this chain is the idea that plain ejectives and "voiced ejectives" would merge as the latter and not as the former.

[KathTheDragon]

I think a lot more research needs to go into the actual phonetic realisation of the three series. After all, voiced aspirates are only attested in a single branch (Indic), and aspiration is attested in what could well be an actually genetic subgroup (Graeco-Aryan). In the north-west, both Germanic and Italic point to a realisation of this series as voiced fricatives in their histories, which should also be taken into account. Moreover, Germanic and Italic are unlikely to be more closely related than a common European subgroup, probably also including Celtic and Balto-Slavic (and perhaps also Armenian? Iirc the affinities of Armenian are uncertain). Speaking of Armenian, it's an interesting puzzle. PIE *t is uniformly reflected as Armenian tʰ, while *d is reflected as d in 4 dialects and t in the other 3, and *dʰ is reflected as dʰ in 2, d in 2, t in 2, and tʰ in the last (according to Wikipedia), which makes one wonder exactly how this came to be, and exactly what the pre-Armenian realisation was.