6 posts tagged “syncope”

She-Wolves and Godesses in Sanskrit are and odd bunch. You have two types of ī-stems in Sanskrit (and also in Indo-european) the hysterodynamic and proterodynamic ones.

vṛkī- 'she-wolf' is one of the Hysterodynamic ones (which is quite rare).

nom. vṛkīs  ( < *-iH-s)
acc. vṛkyam ( < *-iH-ém)
gen. vṛkyas ( < *iH-ós )

devī- 'godess' is Proterodynamic

nom. devī (< *-iH)
gen. devī-m (< *-iH-m)
acc. devyās (< *-iéH-s)

The most striking of this is, that 2 perfectly feminine words, perfectly animate and all, have two different flections and on top of that, one takes the nominative marker *-s while the other doesn't.

I'm imagining that at some earlier indo-european stage some cluster *Hs must have assimilated or something along those lines. But I have not quite figured out how these paradigms would work pre-syncope. And rather than leaving you all in the dark, I thought I'd post this up, and see if any readers have bright ideas where the nom. *-sg comes from, or why it is absent.

Beekes doesn't reconstruct it for PIE as far as I can tell. But then we would have to assyume quite a bizarre analogy. But any thoughts are welcome!

I've finally inserted all verb roots mentioned in LIV and s the results are much better than expected! I found 19 very sound verbal root pairs that differ at least in lenition of consonant, and sometimes also in vowel position as if it the process of schwebeablaut. Then there's also still a whole bunch of pairs which are semantically a bit harder to link.

The result thus is that in all probability I will be writing about the consonant lenition in verbal roots in Proto-Indo-European and trying to give a Pre-Indo-European solution for it.

I hope to keep you guys updated on my findings. My findings so far though, seem to match perfectly with Kortlandt's theories of Indo-Uralic. This might sound like a good thing, but if Kortlandt's theory turns out to work, and I can find no solution the penultimate-stress syncope way, it will mean I'll have to discard that theory.

There's still plenty fo room to attack Kortlandt's theory as well though. There's some roots that should  be pairs but are impossible to link semantically. If Kortlandt is right and all voiced aspirates come from voiceless consonants this should be impossible.

Plenty of stuff to think about and I still have a lot to research, but that's good. Whatever the result it will surely be enough top get my Bachelor degree, and that feels good. :D

A while ago I wrote a post on Kortlandt's article on consonant gradation, and I'm sure I'm not the only one who didn't full understand it. As a result I sent the good man an e-mail that has cleared things up somewhat. Though I still have some questions, which I'll ask him soon if none of you can clear it up.

Following is a translation of my e-mail and his reply, leaving out such formalities usually written in a letter since they seem a bit pointless on a blog:

Me:

To me it seems that the consonant gradation you propose is fairly obvious, but I do have some questions about the specifics, since I can't get the forms that you give exactly. For the sake of example lets use *keim-. It stands to reason that the pre-syncope form then was *kéiem  for the nominative.

Rule I would lenite the *m which  doesn't happen in Indo-European
Rule II would lenite the *i which doesn't happen either
Rule III would shift the accent to the second syllable and lenite the first resulting in *g'eiém, then syncope would yield *g'iém which is not the nominative.

So I considered taking a protoform *kéim, this regularily yields *kéim, but then if we would try to make the genitive/ablative:
*kéimes
Rule I to III wouldn't apply which would yield *kéims, not the form we are looking for either.

Could you please explain where I went wrong with my reasoning.

Kortlandt:
You are right that the rules do not work, my formulation of rule III is incomplete. Before rule III applies at least apocope takes place.

Nom: *keiem > *keim
Acc: *keiem(e) me > *g'iem m(e) For particle *me see: Hittite ammuk 'me' Comparable are Greek 1pl amme and 2pl umme.
Gen: *keyemeti > *g'imes For genitive ending *-ti see: The Indo-Uralic Verb

Besides that, I suggest you read Sammallahti's book that I mention in my article, which might help you get a better view of the situation.

So there it is. I must admit, I haven't yet gotten round to reading Sammallahti's book. I'm also somewhat puzzled by the *(e) element Kortlandt conjures up in the Accusative, and the *e he show in the Genitive. My guess is these are epenthetic vowels, where the Accusative one might not be absolutely necessary.

But it's not 100% clear to me. Any suggestions? Otherwise I'll bother him with my uninformed ignorance again ;-)

I've been reading a lot of Kortlandt's articles lately, which are conveniently almost all available on the web. His idea's are interesting, his writing is tormenting. I spent hours trying to figure out his article Hittite hi-verbs and the Indo-European perfect which I think I understand now. To give the man some credit, he does write about really difficult subjects, which will never give birth to easy to understand articles, but a 57 word sentence should not be allowed, no matter how difficult the subject matter.

Anyway, I wasn't going to talk about that article, though it is an interesting read. I am going to talk about his article on Indo-Uralic Consonant Gradation. Why? Because it has a lot of interesting stuff, and also some stuff I have trouble accepting. But I believe he is on to something.

He first talks about the Proto-Uralic stress pattern, which accourding to him is:
(C)É(C)-CE(C)-CÈ(C)-CE(C)

Where acute stand for primary stress and grave for secondary stress. Any odd syllable is considered strong, while every even syllable is considered strong. A certain Proto-Uralic system of consonant strengthening and weakening has been found. Kortlandt, completely in style, gives absolutely no examples, so I'm left guessing here what it might have looked like.

Nevertheless he mentions two rules:
1. Consonants which follow the vowel of a weak syllable are weakened.
2. Consants which precede the vowel of a closed weak syllable are weakened.

Read the article for specifics.

Kortlandt aims to transpose such a weakening due to 'strong' and 'weak' syllable to Indo-European, his implication being, that if Indo-European has it, the process must be Proto-Indo-Uralic.

For this he choses to bring up The System of Nominal Accentuation in Sanskrit and Indo-European by my current Vedic Sanskrit teacher, Sasha Lubotsky.

Lubotsky takes a look at the barytona and oxytona in Sanskrit. Barytone is the accent on the penultimate syllable, oxytone is the accent on the ultimate syllable.

Looking at the data Lubotsky concludes that the oxytona almost exclusively have voiceless consonants as opposed to voiced aspirates, whily the barytona almost exclusively have the voiced aspirates and next to no voiceless consonants.

That's quite the discovery. The conclusion that the type of consonant is influenced by the type of accent is so blatantly obvious that I feel silly for mentioning it. I wasn't aware of this distribution.

Kortlandt tries to connect this weakening to the Uralic weakening. For this he needs a third rule.

This rule is the following: An open strong syllable becomes weak and loses its stress to the following syllable, which will become strong if it is closed.

And here the article loses much of its comprihensability. Partially because he starts *o juggling through analogies, while I tend to agree with Glen Gordon that it's more probable and thus more preferable that such *o's appeared through phonotactic developments (mostly).
 Also a lack of clear paradigms of his reconstructed proto-languages makes it hard to understand what is meant. This lack of giving examples is typical of Kortlandt's writing style, who will rather use a periphrastic explanation of what happens, rather than showing what happens. I guess I am more of a visual person than he is. Insights on the interpration of the rest of the article are appreciated.

I challenge anyone to figure out what is meant here.

Even if it somehow works out though, I think it's still more viable to with a penultimate stress based syncope, but one should keep in mind that a very early form of Indo-European must have had consonant gradation as a result of the position of the stress in combination with syllable shape. I need to overthink how this fits in with the penultimate stress syncope.

But before rethinking the model, I would like to understand what Kortlandt is trying to achieve, even though I might not agree. So any help is appreciated.

Sceptics to the consonant gradation due to stress, could maybe say the stress is dependant on the type of consonants, besides it being typologically unlikely this would actually make the series of stops difficult to explain, and it wouldn't have to find some really nice explanation for verbal pairs like *grebh- 'to grab' and *kerp- 'to pluck', which are a bit to similar for comfort in my opinion. And there's more pairs like it.

[Edit: I misread a bit of the article, I thought it said the third rule took place between the first and second, but it simply takes place after the two, I am still not getting the results I want though.]

The prehistory of indo-european *s-stems is by far the most problematic class of nouns. Today during Vedic Sanskrit class I once again ran into something terribly confusing.

The word in question here is *h₂eusṓs 'Dawn', not only an entity but probably also a goddess, even in Indo-European times. In Latin we find the word as aurora in Greek as ἠώς (Homeric) and ἕώς (Attic). In Sanskrit though we find uṣas उषस्, with the root in the zero grade. I'll first discuss how I would reconstruct the Early Indo-European form, and then discuss the difficulties that arise.

In my reconstruction of indo-european the paradigm of 'Dawn' would look like this.

Nom. *χǝwsásǝ=sǝ [1]
Gen. *χǝwsǝsá-s(y)ǝ

From this paradigm the *e grade in *h₂eusṓs is incredibly hard to explain. There's never a moment in the paradigm that one would expect an accent to fall on the first syllable. If there's never any form of grading in that part of the root, there's no reason to restore it, so it can be analogical, it must be paradigmatic.

The phonetic environment itself doesn't leave room for a epenthetic *e either, since Sanskrit proves that a it is absolutely possible for that syllable to be in zero-grade.

Now one idea is, that maybe the Genitive which, in later indo-european would end up as *h₂eus-s-ós, gave a heavy enough cluster to put in the e grade, but the you would have to assume that two *s next to each other would not merge into one, which is disproves by words like *h₁es-si 'You are', which is actually found as *h₁e-si in all indo-european languages.

Another problem with the putting in the e-grade in the genetive, is that it is generally thought that Sanskrit got its zero grade from the genitive, giving a paradigm like this:
nom. *h₂eusṓs
gen. *h₂us-(s)-ós

The only other way to solve this word, is to chop of a schwa at the ende of the *s-stem suffix *-ǝsǝ, and then to get *-ǝs, but that would give a completely new *s-stem and I would be contradicting myself completely from my last blog post about *s-stems, but just for the sake of it the paradigm would look like this:

Nom. *χéwsǝs=sǝ
Gen. *χǝwsás-s(y)ǝ

Yielding:
Nom **h₂éus-os
Gen **h₂us-ós-(s)

Then there would follow major accent juggling and analogy to get the correct Nominative out.

This is a difficult subject, and I've been stalling to post this post for some time now, since I still do not have a satisfactory answer. I would like to continue to think that the *s-stem suffix was *-ǝsǝ, and not assume a second form to explain nominals like this. But inherent to this type of suffix, a full grade in the root is almost impossible to achieve in a phonologically regular way.

There might be some tie-in here with Glen Gordon's recent thoughts that in fact the thematised  suffix variant besides the neuter non-thematic ones.

So *r next to *er for example. Glen Gordon argues that the latter is a later (post-syncope period) development. The only problem I have with this is that it would lead to an assumption that 'Dawn' was once inflected as a neuter noun, and was later 'thematisised'. I'm not sure how convincing I find the idea that a rather abstract thing was later thematisised; Even if it was a Goddess, which arguably is quite animate, you'd expect to still find reflexes of the inanimate phenomenon dawn itself in the Indo-European languages. Nevertheless this theory would explain the *e grade in the root with a paradigm somewhat like this

Nom. *χéws(ǝ)s
Gen. *χǝws-s-ás(y)ǝ / *χǝws-ás-s(y)ǝ

The *ǝ is put between brackets for it is epenthetic, but it's unclear whether it was vissible to the accent. This also accounts for the two variants of the genitive. If the epenthetic *ǝ was visible to the accent, you get the second variant, and if it wasn't, you'd get the first variant with an analogoes inserted á in the suffix of the genitive.

With this both the zero-grade and full-grade in the root have been established, and a later thematisation is accounted for because Dawn was a goddess and thus animate.

Nevertheless I feel like I'm cheating the system a bit by using this explanation. It's almost unfalsifiable, but sadly, also almost impossible to prove with our current knowledge.

[1] I use *χ for *h₂ because I think there's a good chance it was /χ/ and it is typographically more pleasing. Also *h₁ will be *x, and *h₃ will be *χʷ. These were the original values of the Laryngeals I suspect, but many of these shifted, *x was almost certainly [h] or [ʔ] very early on.

In this stage of indo-european there were no vocalic allophones of *w and *y, so I will no longer transcribe them as *i and *u, as I usually do, in this stage.

A while ago I posited the idea that the paradigm for 'foot' in Indo-European might have a phonotactic schwa insertion. Glen Gordon spook out against this, saying it was paradigmatic levelling rather than a phonotactic constraint. I didn't want to believe him, simply because phonotactic constraits are so much nicer than paradigmatic levelling, but I found some examples which shows that he's probably right.

So let us look at the paradigm for 'foot'. We have the root *ped- With a paradigm like this:

Nom. *pṓd-s
Gen. *ped-ós

The ō in the nominative is due to mono-syllabicity of the word. The vocalism of this word is then perfectly acceptable.

The genitive though is slightly more problematic. From the idea of syncope, we'd rather see a word like **pd-ós, which we clearly don't. My idea was then that a *e was inserted (which was previously a schwa), due to phonotactic constraints, after all /pd/ is a rather difficult cluster to pronounce initially, even Greek, known for it's sometimes rather strange initial clusters doesn't allow such a cluster. Glen Gordon argued that this is not so much a phonotactic constraint, but rather the tendency of Indo-European to avoid asyllabic roots alternating with syllabic roots in one paradigm.

At first I did not see why one would assume this, and then I suddenly remembered the root *peth₁- 'to fly'. This root is well attested in many languages, including English (feather). But the Greek word πτερόν 'wing' is the one I'd like to talk about. As we can see, Greek has a perfect zero grade in the root for this word, a Indo-European reconstruction of this word would be:
*pth₁ró-m
By the phonotactic constraint I first proposed such a cluster would be forbidden. I would expect a form **peth₁ró-m, so I have to re-evaluate my theory. I'm not quite feeling into giving into Glen's theory yet, and would like to change my theory a bit;

When there's a initial cluster of two consonants of which one of the two elements is voiced, a schwa is insterted between the two.

I believe Greek agrees with me on this one, since there's no clusters like bd, gd, pd, bt etc. to be found in the Greek language initially. Nevertheless I'm going to have to search for words to disprove me again; If you're feeling in a particularly productive mood, I encourage my readers to do the work for me!

Just a little side not to finish this post of, it might be interesting to know that Sanskrit in fact does have the phonotactic constraint that I posited the first time, the word for 'wing' in Sanskrit is patram पत्रम्. Which is a problematic word as well since it brings up some of Sanskrit's sporadic laryngeal vocalisation, but we'll leave that for some other time.

Sanskrit does allow such clusters medially though. The third person plural perfect in Vedic Sanskrit of patati पतति 'to fall/fly' is paptuḥ पप्तुः While in Classical Sanskrit by a bizarre analogy from sad- 'to sit' it became petuḥ पेतुः. To sit in 3ppl perf. is seduḥ < *sazduḥ < *sasduḥ. This e-vocalism rather than a medial CC cluster was generalised to all reduplicating perfects that have identical initial reduplication (Labials, dentals and fricatives). Seems like Sanskrit wasn't particularly fond of the medial zero-grade roots after a while either.