The Great Proto-Indo-European Thread

[Sleinad Flar]

Or Flar*.

The Celtic t-preterite continues many things, but is in no way parallel to the germanic weak preterite. Indeed it continues generalized 3 sg endings, to which other personal endings were added. E.g. Old Irish -biurt < PIE *b^hēr-s-t- + PCe 1sg thematic ending -ū. Mostly these continue s-aorists, but there might be some thematic aorists and imperfects in there as well (Lepontic KariTe < PIE *(e-)kr-ye-t-?). The same augmented s-aorists are also the origin of the Celtic s-preterites, as Gaulish legasit < PIE *leg^h-(e)h₂-s-t- + 3sg -it, shows.

So no, the Celtic t-preterites have got nothing to do with the Germanic dental preterites. Germanic doesn't even continue s-aorists AFAIK.

*Although I don't dream to claim to be as knowledgable in all matters Celtic as Dewrad, I think I know a thing or two about this matter.

[WeepingElf]

Or Flar*.

The Celtic t-preterite continues many things, but is in no way parallel to the germanic weak preterite. Indeed it continues generalized 3 sg endings, to which other personal endings were added. E.g. Old Irish -biurt < PIE *b^hēr-s-t- + PCe 1sg thematic ending -ū. Mostly these continue s-aorists, but there might be some thematic aorists and imperfects in there as well (Lepontic KariTe < PIE *(e-)kr-ye-t-?). The same augmented s-aorists are also the origin of the Celtic s-preterites, as Gaulish legasit < PIE *leg^h-(e)h₂-s-t- + 3sg -it, shows.

So no, the Celtic t-preterites have got nothing to do with the Germanic dental preterites. Germanic doesn't even continue s-aorists AFAIK.

*Although I don't dream to claim to be as knowledgable in all matters Celtic as Dewrad, I think I know a thing or two about this matter.

I see. Germanic indeed doesn't continue s-aorists.

[Terra]

EDIT: There still is the issue that Proto-Italic */f/ gives /b/ in medial position in Latin. So it may have been voiced in that position?

It was.

Note 1: The series of PIE stops traditionally termed 'voiced aspirate' yields voiceless
fricatives in word-initial position in Latin and Sabellic, voiced fricatives
word-internally in Sabellic, and voiced stops (merging with the old voiced stops)
word-internally in Latin. The Latin stops probably go back to voiced fricatives, as is
shown by the variant forms of Lat. ab- (see s.v.). This points to a complementary
distribution of Proto-Italic voiceless word-initial fricatives vs. voiced word-internal
ones. In other words, there was one fricative phoneme with two allophones (see
Stuart-Smith 2004: 196-198, and the critique of her phonemic interpretation by
Kortlandt 2007: 150). I will note voiceless fricatives in my Pit. reconstructions, but it
seems likely that they were voiced word-internally. After nasals and sibilants, the PIE
voiced aspirates probably did not change into fricatives, but remained stops.

I know that the Germanic weak preterite is understood as being a replacement of the PIE reduplicated forms, one of the few of which that survived was "did".

FYI, Gothic preserves reduplication.

I see. Germanic indeed doesn't continue s-aorists.

Is there a list of these s-aorists somewhere?

[Basilius]

There is PIE/PCe *tisres > Gaulish tidres (Old Irish teoir) "three" (fem.) I guess that's close enough.

Not really: as far as we can determine the <ð> of <tiðres> probably represented /ts/, not /θ/.

However, Skr. tisrás = teoir (and cátasras = cetheoir) doesn't (don't) seem to point to a *ts.

[Salmoneus]

I have read somewhere that the Celtic t-preterites came from a generalization of the 3sg. *-t throughout the paradigm. .

Such as, perhaps, when I said that earlier in this thread?

[Dewrad]

There is PIE/PCe *tisres > Gaulish tidres (Old Irish teoir) "three" (fem.) I guess that's close enough.

Not really: as far as we can determine the <ð> of <tiðres> probably represented /ts/, not /θ/.

However, Skr. tisrás = teoir (and cátasras = cetheoir) doesn't (don't) seem to point to a *ts.

Why should it? Gaulish is ancestral to neither Irish nor Sanskrit.

[Basilius]

There is PIE/PCe *tisres > Gaulish tidres (Old Irish teoir) "three" (fem.) I guess that's close enough.

Not really: as far as we can determine the <ð> of <tiðres> probably represented /ts/, not /θ/.

However, Skr. tisrás = teoir (and cátasras = cetheoir) doesn't (don't) seem to point to a *ts.

Why should it? Gaulish is ancestral to neither Irish nor Sanskrit.

So, PIE *sr -> Gaulish /tsr/?

[kanejam]

EDIT: There still is the issue that Proto-Italic */f/ gives /b/ in medial position in Latin. So it may have been voiced in that position?

It was.

Note 1: The series of PIE stops traditionally termed 'voiced aspirate' yields voiceless
fricatives in word-initial position in Latin and Sabellic, voiced fricatives
word-internally in Sabellic, and voiced stops (merging with the old voiced stops)
word-internally in Latin. The Latin stops probably go back to voiced fricatives, as is
shown by the variant forms of Lat. ab- (see s.v.). This points to a complementary
distribution of Proto-Italic voiceless word-initial fricatives vs. voiced word-internal
ones. In other words, there was one fricative phoneme with two allophones (see
Stuart-Smith 2004: 196-198, and the critique of her phonemic interpretation by
Kortlandt 2007: 150). I will note voiceless fricatives in my Pit. reconstructions, but it
seems likely that they were voiced word-internally. After nasals and sibilants, the PIE
voiced aspirates probably did not change into fricatives, but remained stops.

Um I don't think there's any evidence for the intervocalic fricatives in the Sabellic languages being voiced, other than /s/ voicing to [z] in Oscan (and on to /r/ in Umbrian). I suppose though that, since this was always written <s> in the native Oscan alphabet, and none of the other alphabets used distinguished /v/ or /G/ so you could argue that they might have been voiced. I'm sure there's some evidence against it, like appearing in voiceless consonant clusters after epenthesis, but maybe that could be a result of voicing assimilation.

Anyway the other option is that the fricatives were voiceless in all places in Proto-Italic and voiced intervocalically in Latin, subsequently merging with the voiced stops.

That is if Proto-Italic was a valid grouping; is there any strong evidence for or against this? I've seen older texts argue that the resemblence is through contact.

[Drydic]

Last I checked there was a weak majority in favor of unity. The evidence against is more lack of many widespread strong unifying factors than decisive differences. But you're right Latino-Faliscan and Osco-Umbrian are starkly different branches.

[Salmoneus]

De Vaan says "the concept of Proto-Italic has been challenged at various occasions, but agreement seems to be increasing on the view that there was indeed such a common prestage". And by Proto-Italic he means the ancestor of both Latino-Sabellic and Venetic. He notes that the inclusion of Venetic is debateable, particularly in light of the lack of evidence, but he doesn't even mention the possibility that Sabellic and Latino-Faliscan don't share a common ancestor. He also gives common PIt changes - none of them look like they couldn't be borrowed or independently arrived at... but then, what exactly is the difference between two neighbouring dialects that undergo the 'same' change because they're dialects of the 'same' proto-language, and two neighbouring dialects that undergo 'different' but formally identical changes due to areal influence? I'm not sure this distinction is all that meaningful.

[Drydic]

Exactly. It doesn't have the obvious unity that say Germanic does, but without significant new evidence detailing why it should not be a grouping, it's got validity.

[WeepingElf]

Um I don't think there's any evidence for the intervocalic fricatives in the Sabellic languages being voiced, other than /s/ voicing to [z] in Oscan (and on to /r/ in Umbrian). I suppose though that, since this was always written <s> in the native Oscan alphabet, and none of the other alphabets used distinguished /v/ or /G/ so you could argue that they might have been voiced. I'm sure there's some evidence against it, like appearing in voiceless consonant clusters after epenthesis, but maybe that could be a result of voicing assimilation.

Anyway the other option is that the fricatives were voiceless in all places in Proto-Italic and voiced intervocalically in Latin, subsequently merging with the voiced stops.

That the voicing is not indicated in the Sabellic inscriptions doesn't mean that it wasn't there: most writing systems do not distinguish allophones. For instance, you'll almost never meet an ä, an ö or an ü in an Old High German text; but that doesn't mean that the language did not yet have any umlauted vowels - it just means that they were still allophones of the back vowels, conditioned by an /i/ in the following syllable, and therefore needed not be indicated in writing.

[Richard W]

NE: although this comes with the caveat that this etymology is not exactly widely reported and to my knowledge initial vowel loss is not usually a thing in Greek, so this is either one of a few isolated cases or actually a loan with uncertain provenance. In any case, the wide majority of Greek words with initial voiced stops are loans.

The loss of initial, unaccented vowels is a regular development from Classical to Modern Greek. Examples include psari 'fish' from opsarion and the variable survival of the augment. Other examples can be found in John Wells' blog on initial clusters.

[hwhatting]

Having started to browse the dead-tree copy of NIL that was on my gift heap this year, I've written up some thoughts on PIE *bhag- "share" on my blog. Thought it might interest some of you.

[Morrígan]

Having started to browse the dead-tree copy of NIL that was on my gift heap this year, I've written up some thoughts on PIE *bhag- "share" on my blog. Thought it might interest some of you.

Cool read. I say that I've read others positing that bog- in Slavic is a loan from Indo-Iranian. I'm not sure I can comment beyond this, since I don't recall seeing *bhag- presented as a root with *a

[Chagen]

I honestly have to ask where I can find books on PIE. They don't seem to be very common. Obviously being a reconstructed proto-lang there'll never be a reference grammar, but I wanna know just what, say, Dewrad used to make Wenetic.

The closest thing I can think of is Dnghu's grammar and....well, it's not really a good grammar in the slightest.

[hwhatting]

I'm not sure I can comment beyond this, since I don't recall seeing *bhag- presented as a root with *a

Well, both LIV and NIL do; they obviously seem to accept /a/ as a PIE phoneme. If one accepts the reconstruction *b^hag (which, as you can see, I've come to doubt), it is one of those cases of /a/ near a plain velar. Of course, those who don't accept /a/ need to reconstruct something like *b^hH₂eg- or explain the attested forms as reduced grades of *b^heH₂g- , as Beekes seems to do (see Sal's comment on my blog).

[Particles the Greek]

I honestly have to ask where I can find books on PIE. They don't seem to be very common. Obviously being a reconstructed proto-lang there'll never be a reference grammar, but I wanna know just what, say, Dewrad used to make Wenetic.

The closest thing I can think of is Dnghu's grammar and....well, it's not really a good grammar in the slightest.

That really depends on what you want from such a book; I can think of at least three which might meet some of your requirements.

[Drydic]

And you'll need probably 3-4 books to get a good idea.

[WeepingElf]

The best one to start with is IMHO Indo-European Language and Culture by Benjamin W. Fortson IV. Then read Indo-European Linguistics: An Introduction by James Clackson, which treats some matters in more depth. Indo-European Language and Culture by Thomas Gamkrelidze and Vjačeslav Ivanov is not an introduction; it presupposes familiarity with the standard model and presents some bold changes to it which in some parts make sense, but only for a prestage, and in some parts do not. It nevertheless is an interesting read, and it does have a nifty PIE glossary in the second volume, but rewritten in their glottalist reconstruction.

What regards dictionaries, Pokorny's Indogermanisches Etymologisches Wörterbuch is not only in German, but also more than fifty years out of date. Unfortunately, a more up-to-date etymological dictionary has not appeared yet, so Morrígan's spreadsheet is still the best we have.

[Tropylium]

Focusing more on the "where": for one, several of Winfred Lehmann's books are available online via University of Texas, including PIE Phonology and PIE Syntax.

Also I am not sure how unofficial this is exactly (I'm guessing "quite") but there's this one site which appears to have Fortson's and Clackson's PIE monographs available as pdf. Also Michael Meier-Brügger's Indo-European Linguistics.

(And checking my files, I moreover seem to have wound up with a pdf copy of Mallory & Adams' The Oxford Introduction to PIE too somehow…)

[Dewrad]

but I wanna know just what, say, Dewrad used to make Wenetic.

Probably unlikely to be of much help, in honesty: so far I've referred to about eight or so books, none of which is actually a handbook of PIE. Partially this is because I have a fairly clear idea of what I want to emulate, and it's better for me to consult sources on these topics rather than a catch-all reference, and partly because I already had a reasonable grounding in PIE (in so far as any keen amateur and non-specialist does, I guess).

However, I would happily second WeepingElf's suggestions heartily. I would also suggest you avoid Beekes Comparative Indo-European Linguistics: An Introduction. That was my first introduction to PIE (because that's pretty much all there was readily available in English in the late 90s), and let's just say it remains one of the least accessible handbooks I've encountered. I wish I'd read either Clackson or Fortson beforehand.

Arguably, for conlanging purposes, I would still say that the rather traditionalist "Brugmannian" reconstruction given in Szemerényi's Introduction to Indo-European Linguistics is more useful. It stops budding PIE-conlangers getting carried away by all the laryngeals, if nothing else.

EDIT: Yeah, shit, I should get back up on that horse at some point.

[Morrígan]

I really need to make sure that I have some sort of Definitive Collection, since I keep discovering that my files are not all in one place.
Some Books

Like Dewrad, I first encountered Beekes, and while it was definitely overwhelming, I still found it more informative than Szemerényi. It's good, though people might complain about Beekes being overly willing to explain things using laryngeals, I'm not convinced that he being too liberal with them¹. I actually own Meier-Brügger, and I've found it an interesting supplement to Beekes for the more advanced reader, and provides a convenient (English) summary of classes described in LIV.

(1): There are clear cases where they are needed, i.e. where many possible **a are best explained as *h2e, and **a is otherwise so rare, the model is simplified by proposing that all cases of **a be reconstructed as *h2e until there is a compelling reason not to

[jmcd]

I was reading a description online of Proto-Germanic by Lehmann
"holders of the glottalic theory face the problem of accounting for a glottalized labio-velar."
What do yous think of this? Is this an appropriate argument against the new kids on the block?

[Nortaneous]

I was reading a description online of Proto-Germanic by Lehmann
"holders of the glottalic theory face the problem of accounting for a glottalized labio-velar."
What do yous think of this? Is this an appropriate argument against the new kids on the block?

No.