Talk:Proto-Indo-European language/Archive 1

From Wikipedia, the free encyclopedia

Archive This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

To the tune of Yankee Doodle:

Proto-Indo-European
Was the mother tongue,
But everyone who used to speak it
Is now dead and gone.

An original composition by Kbh3rd

Ha! --MerovingianImage:Atombomb.gifTalk 23:04, Sep 14, 2004 (UTC)
http://math.uchicago.edu/~wald/lit/anything_goes.txt
And though I'm just a steppe parparian,
In two weeks I'll pe Tocharian,
And it shows.
Anything Koes.

Like lightning (Dye:us forgive the simile),
The Teutons are switching, Grimmily,
All three rows.
Anything Goes.
dab 12:58, 29 Oct 2004 (UTC)

I'm planning to eventually give an overview over the reconstructed grammar. But it will take some time. dab 12:58, 29 Oct 2004 (UTC)

OK, this is useful, because it is difficult to find both on and outside the internet. But for neadability purposes, could you use paradigms rather than just endings? --Caesarion 17:20, 26 Feb 2005 (UTC)

Contents

What's that star thingy?

I came to the PIE page hoping to find some explanation of the ubiquitous asterisk symbol at the beginning of many PIE-reconstructed words. But, sadly, I didn't.

Wouldn't this be a good page to mention it on?

Steverapaport 18:45, 5 Feb 2005 (UTC)

The meaning of the asterisk is exactly "This is a reconstructed word [i.e., not a word that was ever actually recorded anywhere in this form]." AJD 19:08, 5 Feb 2005

well, you would have found iit, had you taken it upon yourself to read as far as line 3 where we say "The standard convention is to mark unattested forms with an asterisk" ... dab (ᛏ) 22:13, 5 Feb 2005 (UTC)

baltic or caucasian?

I have come across conflicting information on wikipedia on whether modern baltic (esp lithuanian) or modern caucasian (esp georgian) languages most resemble PIE. What if anything is the consensus?

Baltic languages are Indo-European. Caucasian languages, including Georgian, aren't. AJD 23:54, 16 Feb 2005 (UTC)

See the following excerpt from the main PIE article: "Other works have tried to show that the Caucasian languages, particularly the Northwest Caucasian family, spoken in Georgia and Turkey, may be the closest relatives to the Indo-European stock. While these are not widely-held theories, substantial evidence investigated by the linguist John Colarusso seems to support their theory. In particular, the one-vowel hypothesis which has been put forward for Indo-European would be borne out by the usage of substantial secondary articulation like that found in the Northwest Caucasian languages and, indeed, in the hypothesized PIE. Also, the Northwest Caucasian languages preserve a large number of guttural phonemes which may be the modern equivalents of PIE "laryngeals"."

The article makes no reference to the more common idea that the baltic languages, Lithuanian in particular, most closely resemble PIE. I never claimed Caucasian languages were Indo-European.

Isn't that idea nowadays considered obsolete?

By the way, I read somewhere about similarities between IE and Caucasian. For instance, In IE, n was used for negation, in Proto-caucasian the similar nasal m, etc. There were a few similarities, like that.

Velars

The existence of the plain velars as phonemes separate from the palatovelars and labiovelars is disputed. In most circumstances they appear to be allophones of one of the other two series, and none of the daughter languages (with the possible exception of Albanian ) has reflexes of them that differ from the other two series.

This isn't still true, is it? I didn't think there was still any disagreement about this among scholars of Indo-European, and didn't Craig Melchert remove all doubt that these must have been different phonemes in Indo-European by showing that they have three distinct reflexes in Luwian?:

*ḱeyor 'lies down' > zīyar(i)
*kes- 'comb' > kisāi-
*kwid 'what' > kui

AJD 01:56, 11 May 2005 (UTC)

No. people will not accept a phoneme in the Proto-Language based on a single Luwian word. Also, in your example, there is nothing to stop us from saying it was *kuid rather than *kwid. From my own (unpublished) lexicostatistical studies, I tend to believe that *kw was *ku in early PIE, and I can cite a number of published voicings of that opinion, if you really press me :) dab (ᛏ) 05:52, 11 May 2005 (UTC)

I think "mainstream" Indo-Europeanists (whom I thoroughly subjectively define as Jay Jasanoff and the people he respects) are convinced by Craig Melchert's Luwian evidence. But that doesn't mean the matter is now settled beyond all dispute. --Angr/comhrá 10:38, 20 May 2005 (UTC)

I encountered a recent article on the internet about Balto-Slavic phonetics giving a detailed reconstruction of the development of Baltic and Slavic from PIE, where the author questions the "three-tectal" theory; so I don't think this is beyond dispute.

Furthermore, even if Luwian did have three different velars, this is no proof of PIE having it. There is no reason why such a development wasn't particular to a subset of PIE dialects; in fact, the fact that Lithuanian disagrees in a number of cases with other satem languages in this respect argues in favor of this. Benwing 3 July 2005 07:03 (UTC)

Is this correct??

Let me see if I got all 10 of these correct:

  1. oi-no
  2. dwo
  3. trei
  4. kwetwer
  5. penkwe
  6. sweks
  7. septm
  8. oktou
  9. neun
  10. dekm

And I believe 20 is wikmti and 100 is dkm-tom. How about 11 to 19, 30 to 90, and 1000?? Georgia guy 00:17, 20 May 2005 (UTC)

"one" is tricky; some languages seem to come from *oinos; but others come from *oikos or *oiwos and others from *sem-. The others are basically okay, though people might quibble about the endings and might want some more laryngeals and more palatal marks stuck in. I don't think 11 to 19 are reconstructable because there's too much divergence among the attested languages, and too much possibility for analogy. According to R.S.P. Beekes' Comparative Indo-European Linguistics, 30-90 were probably something like:
  • trih2-dḱomth2
  • kwetur-dḱomth2
  • penkwe-dḱomth2
  • sweks-dḱomth2
  • septm-dḱomth2
  • h3eḱth3-dḱomth2
  • h1neun-dḱomth2
Some people have reconstructed *ǵheslo- for 'thousand' but it's pretty controversial. It looks good as the source of Sanskrit (sa)hasram (where sa is from *sm- for 'one'), and maybe Tocharian A wälts/B yaltse. Greek χίλιοι is supposed to be from *ǵhesl-yo-, which is plausible enough. Some claim Latin mīlle is supposed to be from *smī-ǵhslī- (again with *sm- for 'one') but I don't think too many people believe that. The Germanic and Balto-Slavic words (Gothic þusundi, Lithuanian tūkstantis, OCS tysǫšti') seem to be from *tuHs-ḱmti- and mean something like 'swollen hundred'. --Angr/comhrá 10:34, 20 May 2005 (UTC)

is there any evidence that the /d/ in dḱomth2 really belongs? The main evidence from Beekes seems to be some highly suspect claims about the "glottal stop in the /d/ causing lengthening of the previous vowel" [since he believes in the controversial glottalic theory]. Benwing 3 July 2005 07:03 (UTC)

I don't know if there's direct evidence for it. I'm willing to believe it was there in the oldest stage of the language, because it does seem likely it's the same root as *dekm, but I'm also willing to believe the d was lost without a trace before the language broke up. --Angr/tɔk tə mi 3 July 2005 08:25 (UTC)

controversial claims

the stuff in the introduction about supposed PIE-Caucasian connections sounds extremely fringy to me, and really doesn't belong there. i would bet that Greenberg's Eurasian hypothesis (i.e. that PIE, Uralic, and Altaic, perhaps also Korean, Japanese, and Eskimo-Aleut, maybe a few others, were sister stocks) is *far* more accepted, but it is not even mentioned here.

also, AFAIK the reconstruction of PIE case endings is disputed; even Beekes, for example, who is fairly opinionated, is quite tentative about suggesting dative/ablative/instrumental plural endings along the lines of what's given here.

i also think that something that would go along way towards helping clear the FUD going on here is to emphasize in this page the fact that "PIE" refers to a period of 1000-2000 years, over which there were (obviously) many different stages and changes. [it should also be explained that the "Proto" period of a language is usually assumed to be divided into two large stages: an Early stage of total uniformity and a Late stage in which dialect variations exist, but changes for the most part still happen in parallel across all dialects.] IMO, much of the confusion in PIE studies stems from these fundamental facts getting ignored by far too many PIE researchers.

Benwing 3 July 2005 06:52 (UTC)

You're right; possible external connections of PIE shouldn't be mentioned in the introduction. Maybe you could write a section on the various theories (like Nostratic, AFAIK the most popular of all the crackpot theories) and stick it at the end of the article. And I'd be happy to see more tentativeness in the reconstructions of the endings given; I had been meaning to do it myself someday but have never gotten a round tuit. --Angr/tɔk tə mi 3 July 2005 08:31 (UTC)
I also agree it has no place in the intro. Imho, Nostratic is fringy, but not "crackpot", there can well be a section about these theories, properly qualified as speculative. dab (ᛏ) 3 July 2005 09:01 (UTC)
To me, Nostratic is crackpot, but I'll concede some pots (like Proto-World language) are more cracked than others. --Angr/tɔk tə mi 3 July 2005 10:44 (UTC)

In this thread [1] you can read controversial claim why PIE can be Adamic language, as suggested by Catherine Emmerich revelations.

Table of Correspondences & Rules List

I think you should also add a table of regular sound correspondences and a list of sound laws or rules. Perhaps, every language article should contain a list of changes that would describe its developement from its ancestor tongue in detail, so, for example, Modern German would include some Old High German information, and Old High German would also contain some Proto-(West-)Germanic information, etc. What do you think?

As for fonts, I'd recommend "Arial Unicode MS". I'm using it very often, since it contains all the necessary characters, including IPA, Greek, Cyrilic, Devanagari, Arabic, and Hebrew, lots of diacritics, whatever.

your browser (or style sheet) chooses your font for you :) all we do is encode stuff in unicode. Yes, your suggestion is obviously an ultimate aim here, but well, feel free to begin. It's a tall order. dab (ᛏ) 7 July 2005 13:57 (UTC)

That's right.

As for the table, I've actually started already :). I'm ready to send the "first draft" to those who are interested. It's by no means exhaustive, a lot of information remains to be added and some information might be incorrect, but if anybody wants... :)--Pet'usek 7 July 2005 14:11 (UTC)

I'd like to see that! You know, by the way, that there is an overview table of such things in Calvert Watkins' The American Heritage dictionary of Indo-European Roots, 2nd edition, Boston & New York 2000.
As for the changes, yes it would be good to have brief info in the various language articles, but I feel that sound shifts deserve to be dealt with fully in their own articles. To take that example of Germanic → Old High German, we have a new article on that at High German consonant shift, which I would modestly suggest might be a model. But yeh, why not summarise it in the German article too? --Doric Loon 7 July 2005 14:28 (UTC)
Pet'usek, take a look at Proto-Celtic; is that the sort of thing you mean? --Angr/tɔk tə mi 7 July 2005 15:34 (UTC)
Well, yes, that's it - almost. It's a good start anyway. Since there might be various conditions and environments in which the phonemes might change in different ways, I'd use superscript numbers as links to detailed explanations, but the phonemes themselves might function as links - either to the expl. or to examples. Imagine, for instance, a general rule like this in a hypothetical language "A": VsV > VrV. As this rule took place in various languages, even in various language families, I think creating a separate referential article - a list of rules or something like that - might be useful. A question arises here what format such a list should have, but reciprocal linking to languages where these rule could be found should certainly be included. Now, if we know that in some languages certain consonantal clusters get simplified (e.g. /*str-/ > /c/ or sim.), there should be a link to this rule at all consonants affected (i.e. *s, *t, *r), otherwise I don't know how exactly we could make the information easier to access, keeping the summary or table transparent and well-arranged at the same time... For those who'd like to see my FIRST DRAFT table, [2] (In the "Notes" part, an attempt at a more detailed summary of PIE > LIE laryngeal transformation is also present; it's made in the MS Excel format, but I'm going to prepare the HTML version)
As for the High German consonant shift, I like it :), but still, there could be - just for an easy and quick reference - a little more simpler table (e.g. at the very beginning of the article). I must still think about it - the combination of transparency and exhaustive information is crucial I think... Another question has just come across my mind: perhaps, we should have three or two tables in every language description - the correspondences between the ancestor and the language described (1), between the language and its sister languages (2), and between the language and its daughter languages ;) (of course, 1 & 2 might be in a single table)? I don't know... :)

If you prepare your preliminary table in wiki formatting, you can make a subpage for it on your user page here, which will make it easier for others to view and make suggestions on. Just create a red link like User:Petusek/IE chart or something like that, then click on it to edit. So long as it's still in your user space it doesn't matter if isn't not ready yet. --Angr/tɔk tə mi 7 July 2005 19:13 (UTC)

I see. OK. I'll try. :) I'm a newcomer here, so thanks for this piece of advice. :)
The idea seems complex, but quite interesting, It would be nice to see how a root has changed through the millennia in different languages.

Example texts

it's nice to have some example texts, but the source (author) must be attributed. Also, what is the "Aquan nepot" supposed to be? Please state a source for these texts. dab (ᛏ) 15:30, 22 August 2005 (UTC)

This was a 19th-century thing, wasn't it, writing texts in PIE? I'm told Jacob Grimm wrote fairy tales in his reconstructed language. Scholars of the 20th century became highly sceptical about our ability to reconstruct the language that well. So I am in two minds about including these. They are interesting as a testimony to 19th-century academic smugness, and possibly they really do have value in giving an impression of what the language MIGHT have been like, but this is not the way the discipline operates today, and these texts are not in the modern transcription system. There needs to be a very clear caveat given here. --Doric Loon 23:28, 22 August 2005 (UTC)
no, such texts are still written, for fun, and to illustrate the author's idea of PIE. Since every scholar has his own idea of the language, it is imperative that the authors of these texts are noted. So far, we have only Schleicher, so I'll cut the other texts to talk for now, until references are brought forward. dab (ᛏ) 10:04, 23 August 2005 (UTC)
Hey, Aquan Nepot is a prayer to Neptune, The Apostles' Creed, Heil Mary and Pater Noster are the classical Christian prayers. About authors of the tale Deivos Verunos you know from the other posting. Why do you commit vandalism? --Nixer
WP:CITE! You are not writing this article for me. The Verunos text is apparently reconstructed Proto-Indo-Aryan. For the others, we still have no reference. If they are your own, do consider wikibooks. dab (ᛏ) 13:24, 23 August 2005 (UTC)
This is NOT Proto-Indo-Aryan. This is Proto-Indo-European, written with another phonetic system, which is much more readable. Look at this:
http://www.grsampson.net/Q_PIE.html - This is the same text --Nixer

Nixer, your link doesn't work. I think what you meant to have as the link is http://www.grsampson.net/Q_PIE.html, or simply http://www.grsampson.net/Q_PIE.html. (see source code) -- D. F. Schmidt (talk) 13:57, 23 August 2005 (UTC)

Yes, thank you --Nixer

then why did you not provide the link from the beginning, saying "Geoffrey Sampson, S.K. Sen, E.P. Hamp"? Click on WP:CITE. Read it. No, read it, this is Wikipedia policy, and we are bound by it here. You cannot just come here and copy-paste text into the artice without comment, you need to say where it is from; and not only on the talk page, after being implored and begged to do it. I'll revert you again, and this time please do it properly, like I have done in Schleicher's example. I mean the Verunos text; you still have to provide references for the prayers. dab (ᛏ) 14:13, 23 August 2005 (UTC)


cut texts

  • Deivos Verunos
To regs eghest. So nepotlus eghest. So regs sunum evelt. So toso ceuterum precsquet: "Sunus moi guenhota!" So ceuter tom reguem evequet: "Ihgesuo deivom Verunom". So regs deivom Verunom uposesore nu deivom ihgeto. "Cluthi moi, pater Verune!" Deivos Verunos cata divos egveght. "Quid velsi?" "Velmi sunum." "Tod estu", vequet leucos deivos Verunos. Reguos poteni sunum geguenhe.
this appears to be a PIE-ified Proto-Indo-Aryan text, composed by one S.K. Sen, see Talk:Proto-Greek language.
  • Pater naseros (the Paternoster)
    • (Version 1)
      Pater naseros cemeni, nomen tovos estu cventos, reguom tevem guemaght ad nas, veltos tevem cvergeto cemeni ertique, edom naserom dagheres do nasmebhos daghei tosmei letodque agosnes nasera, so lemos scelobhos naserobhos. Neque peretod nas, tou tratod nas apo peuces. Teve senti reguom, maghti deiromque bhegh entom. Estod.
    • (Version 2)
      Pater naseros cemeni, nomen tovos estu iseros, reguom tevem guemaght ad nasmens, veltos tevem cvergeto cemeni ed eri, edom naserom dagheres do nasmebhos tosmei daghei ed le agosnes nasera, so lemos scelobhos naserobhos. Neque gvedhe nasmens bhi perendom, tou bhegve nasmens melguod. Teve senti reguom, maghti ed deirom eneu entom. Estod.


  • Aquan Nepot
Puros esiem. Deivons aisiem. Aquan Nepot dhverbhos me rues. Meg moris me gherdmi. Deivos, tebherm gheumi. Vicpoteis tebherm gheumi. Ansus tebherm guemi. Nasmei gvertins dedemi. Adbherome ci sime guerenti! Dotores vesvom, nas nasmei cerddhemes. Aquan Nepot, dhveronus sceledhi! Dheghom Mater toi gheumes! Dheghemia Mater, tebhiom gheumes! Meg moris nas gherdmi. Eguheies, nasmei sercemes.

These example texts should not be included. Even correctly cited they are unencyclopedic in the extreme. We should mention that some people from Schleicher up to the present day have attempted to write texts in reconstructed PIE, and provide links to some of them, and explain why this practice is no longer considered scientifically feasible today. --Angr/tɔk tə mi 03:58, 2 September 2005 (UTC)

Nixer

why didn't you just link to [3]? I couldn't find the texts via google since they are given as jpgs. Now, you are in violation of WP:CITE, WP:NOR, AND GFDL (the last only unless you are Raphael Sawitzky yourself). You practically violate every principle of Wikipedia by your stubborn re-insertion of these texts. You still haven't given a reference for the Neptune prayer, and the Christian ones were just copied off some Christian website. This is not what we do here. I will revert your additions as vandalism from now on, unless you lighten up and start respecting policy. dab (ᛏ) 11:05, 25 August 2005 (UTC)

TEB728

TEB728, Thank you for the link. dab, I am not Raphael Sawitzky, but you asked a link - a I give. But his translation different. And, of course, he is not the author of the translations in http://www.christusrex.org/www1/pater/JPN-proto-indo-europ.html. it is obvious because different technics and level.

TEB728 have given us the right link to the source. --Nixer 15:02, 26 August 2005 (UTC)

ok Nixer, I realize that you did not discuss your edits because you are not fluent in English; it is, however, very important that you understand Wikipedia policy if you want to contribute. You also replaced my annotated version of Schleicher's tale with your unannotated one (why?). Your texts are linked to via external links. That should be good enough, why do you insist so much that they are inserted verbatim and without comment in the article? dab (ᛏ) 15:32, 26 August 2005 (UTC)