Talk:Proto-Indo-European language/Archive 2

Numbers are way off

The numbers recently added bear no resemblance to the modern reconstructions seen in Lehmann and Sihler among others. They seem to have come out of an ancient pre-syllabic resonant, pre-laryngeal view of the language. That means that the source is from before Brugmann's Grundriss (1886), bizarre. I would correct this, but I won't have access to Sihler's New Comparative Grammar until Monday. Crculver 01:52, 3 September 2005 (UTC)

I'm updating the reconstruction of the numbers now. I'm also restoring dab's version of the reconstructed texts. --Angr/_{tɔk tə mi} 06:20, 3 September 2005 (UTC)

it will be rfc time for Nixer soon, his newbie bonus is almost used up, and he has done nothing but cause disruption, judging from his talkpage, not just here. I'm fine with removing even the 'sheep' and 'Verunos' texts, Angr, the sheep is discussed at Schleicher's fable, and the Verunos one could go to wikibooks: Over at wikibooks, we could add our own versions to it, as well as versions in other languages. dab (ᛏ) 06:32, 3 September 2005 (UTC)

What is the reason for the hypothesis that the (s) in the beginning maybe wasn't part of the original PIE word s(w)eḱs, I just wondered since it's found in nearly all(?) of the known daughter languages. Would it make more sense meaning-wise if the word lacked the s, or how goes the reasoning?

The evidence for an s-less version *weḱs comes from the Old Prussian ordinal uschts "sixth" and the Armenian vec‘, where loss of an original s is somewhat improbable. The Greek heks is not inconsistent with *weḱs, either, since PIE w occasionally shows up in Greek as h in word-initial position, especially where there's an s in the same syllable, e.g. *wespero- "evening" > hesperos, *wes- "to clothe" > *hesnūmi > hennūmi, and the Latin name Vesta vs. Greek Hestiā for the goddess of the hearth. It also ties in with the fact that in the languages with the s, the w after it is variable. The idea is that most languages altered *weḱs to resemble *septm by taking over the initial s, but in some of them the s was added to *weḱs, giving *sweḱs, while in others it replaced the w, giving *seḱs. The idea that a number can change its initial consonant to resemble that of its neighbor in numerical order has support elsewhere: the Balto-Slavic words for "nine" start with d, taken over from the word for "ten". In Germanic, Celtic, and Italic the words for "four" and "five" all start with the same sound, even though in PIE they started with different sounds. (Granted, *penkʷe > *kʷenkʷe in Italic and Celtic may be phonological since it happened with the root *pekʷ- "cook" as well.) --Angr/_{tɔk tə mi} 06:35, 29 September 2005 (UTC)

And what about Russian shest' and Hebrew shish? Seems it could be an ancient, pre-PIE root--Nixer 07:03, 29 September 2005 (UTC)

More likely a coincidence. The Russian word goes back to Proto-Slavic *šestь which in turn goes back to *s(w)eḱs < *weḱs (although the Balto-Slavic initial š is a little confusing). The Hebrew word goes back to a Proto-Semitic root *š-d-š, which in turn will go back to something in Proto-Afro-Asiatic. --Angr/_{tɔk tə mi} 07:45, 29 September 2005 (UTC)

OK, interesting to hear about. It also seems that the "et" mostly was changed inte a diphtong in Germanic, for some reason.

Text samples

The texts in "Example texts" are not renditions of the actual language, but merely linguistic guesswork made several thousands years after the language developed into the various separate branches of Indo-European. We tend to avoid longer text samples even in articles about living languages, and here it seems entirely inappropriate. Please incorporate it (briefly) in the article, but move the texts themselves to WikiQuote.

Peter ^Isotalo 09:47, 5 September 2005 (UTC)

I'm coming to this "party" late, and although I agree entirely with the sentiment of caution expressed above by Angr, dab and Isotalo with regard to the "reconstructed" texts, I nonetheless feel it important to state that such texts were very helpful to me at least when I was first learning about *PIE and attempting to grok the concepts involved, knowing then as I do now that the "reconstructed" Paternoster was in no small part guesswork and, of course, was never spoken in reality (having not yet been written, such would have been hard to do, to say the least--heh). For those of us of a haptic learning style, it is often difficult for us to wrap our minds around the important linguistic characteristics of new languages when we learn them unless we can actively speak and practice them. Thus, the speculative texts were useful in that I got a "feel" for how *PIE might have flowed, which I feel assisted me in learning various other aspects of modern *PIE-descendant languages, for example. Others might benefit similarly, if at least a conservative, short speculative recon text were included in this article. —Ryanaxp 19:06, September 6, 2005 (UTC)

of course, I agree entirely, that's why I have no problem with pointing to the texts from the external links section. It would be a violation of NOR, as well as other "what WP is nots", however, to copy-paste texts found on the Internet into the article. dab (ᛏ) 19:30, 6 September 2005 (UTC)

Wikipedia isn't a place where people can toy around with ancient languages as if they were conlangs, but rather a place to concisely present topics of scientific interest. Putting these text samples in means the entry is no longer concise. Since most Indo-Europeanists no longer believe that the proto-language can be reconstructed to the point of extended sentences, we would be presenting the quaint pseduoscience of yesteryear as current fact. I say no to text samples. Crculver 19:28, 6 September 2005 (UTC)

whoa, wait a minute. I don't know that reconstructed texts are considered any less useful than in Schleicher's time. I mean, Schleicher wasn't so naive as to believe to construct authentic Chalcolithic PIE. These reconstructions are as useful as ever. The point is, rather, that the construction shall not take place on Wikipedia. It takes place in academia. If there are texts notable in academia, there is no reason not to discuss it on WP. Like Schleicher's fable, the "Verunos" text could get its own article if it is not wanted here, but it certainly has a place on WP. The 'prayers' otoh, to the best of my knowledge have no academic credentials, and are internet jetsam. dab (ᛏ) 07:17, 7 September 2005 (UTC)

Why did you add the early, naive version by Schleicher, which is proven to consist many mistakes, along with the more modern version of the tale? Does the oldest version of the text add something valuable to the article? --Nixer 08:08, 7 September 2005 (UTC)

of all the texts here, it is the most notable. However, I already agreed to remove it, since it is already discussed at Schleicher's fable. We could leave "Verunos" here, or scrap the entire section, if that will buy us peace.dab (ᛏ) 08:12, 7 September 2005 (UTC)

I am oly about deleting the oldest version of the tale.--Nixer 08:16, 7 September 2005 (UTC)

and I am only about deleting texts that were found on the internet, without proper attribution who composed them and where they were published. Schleichers version is very instructive, placed against the 1979 version, because it shows how the understanding of PIE has evolved between the 1870s and the 1970s (although most of the progress dates to between the 1870s and the 1910s) dab (ᛏ) 08:34, 7 September 2005 (UTC)

since I agree with Angr that these texts need not be included here, and since I am getting tired of the debate, I have created The king and Varuna, and reduced the "Sample texts" section to links to subarticles. dab (ᛏ) 12:14, 12 September 2005 (UTC)

superscript

should we use ʷ, ʰ or ^w, ^h? In theory, they should be rendered identical, but they aren't. In my browser, the  version looks better. dab (ᛏ) 08:12, 7 September 2005 (UTC)

I suppose we really ought to be using ʷ, ʰ (the IPA template is necessary for the characters to appear correctly in all browsers), even though we have been using ^w, ^h up till now. --Angr/_{tɔk tə mi} 08:15, 7 September 2005 (UTC)

Without IPA template I can see transparent squares only.--Nixer 08:18, 7 September 2005 (UTC)

I am not so sure. this is related to the 'precomposed diacritics' issue. precomposed diacritics in Unicode is a compromise, strictly foreign to its basic philosophy. superscript h is still an h, with the additional property of floating in the air. Just like a-acute is still an a, with the additional property of having an acute. A workable solution is to accept precomposed diacritics of ISO 8859-1. It is cleaner to compose them, just like it is cleaner to use the sup-tag. Some browsers have problems rendering composed diacritics, especially in italics, but that is a technical problem that will go away in one or two browser generations, and we should not inform our decision by it. dab (ᛏ) 08:32, 7 September 2005 (UTC)

I'm pretty sure ʷ, ʰ are treated as separate characters like ə and ʁ, which will also not be legible in all browsers without the IPA template. Even if the problem goes away in one or two browser generations, the problem exists now. Whatever we decide to do, the fact is that ʷ, ʰ (with the template) and ^w, ^h (with the sup tag) are legible in all browsers while unadorned ʷ, ʰ aren't and so shouldn't be used. --Angr/_{tɔk tə mi} 08:41, 7 September 2005 (UTC)

dab, My browser (IE 6) can not show these charasters, and the text becomes inreadable. Angr, your version is proper--Nixer 08:42, 7 September 2005 (UTC)

sure, sorry, this is not about the IPA template. The options are (a) IPA'd ʷ, ʰ, (b) sup-ed ^w, ^h. The cleanest solution, incidentially, would be a {{PIE|}} template, that would render PIE transcription centrally, and customizable via stylesheet. dab (ᛏ) 08:46, 7 September 2005 (UTC)

I have created Template:PIE for now. It is 'semantically' cleaner to use that, even if it's identical to IPA, since PIE notation is not IPA, strictly speaking. So I suggest we use that throughout. We can still choose between gʷʰ and g^wh, of course. dab (ᛏ) 08:51, 7 September 2005 (UTC)

Does the PIE template support ḱ and ǵ? I've been using the Unicode template for those because I wasn't sure the IPA template supported them, not being IPA characters. --Angr/_{tɔk tə mi} 10:53, 7 September 2005 (UTC)

Okay, I just opened this page in IE (the browser that makes the templates necessary) and I see that ḱ and ǵ work fine. But I don't like that the PIE template automatically puts everything in italics. Most fonts with advanced Unicode characters don't have pretty italics, they just sort of slant the roman face in a very ugly and difficult-to-read way. I'd much rather use regular roman face than italics. --Angr/_{tɔk tə mi} 11:04, 7 September 2005 (UTC)

well, it's just that italics are quite important to make the words stand out qua word, imho making the text much easier to read. I thik this trumps font prettiness, which is a browser issue anyway; in any case, if we want PIE in italics, that should properly be handled by the template, not case by case. But I see your point. ḱ is really also just a compromise, since the k̑ looks ugly in most browsers. This should all properly be customizable via stylesheets, but I can't be bothered to fiddle with that just now. dab (ᛏ) 11:10, 7 September 2005 (UTC)

Mistakes in examples

Your examples are inproper because does not use upper case and acute accent, which leads to impossibility to distiguish k^w from ḱ for example.

I replaced your version by mine.

If you want to keep your version of the examples in the article, please, fix the mistakes.

Nixer, this isn't a valid excuse to add a lot of stuff on which there is a consensus to leave it out. --Pjacobi 20:38, September 11, 2005 (UTC)

I don't even get the excuse. How are we unable to distinguish k^w from ḱ? We are doing that, with no problems, all over the article. dab (ᛏ) 21:43, 11 September 2005 (UTC)

But in the your version of the Schleicher's tale they both depicted one way.--Nixer 21:54, 11 September 2005 (UTC)

It's not our version of Schleicher's fable. The first version is as he wrote it, the second version is as Lehmann and Zgusta wrote it. Those are both published and verifiable. The version you keep wanting to restore is from an anonymous website. --Angr/_{tɔk tə mi} 22:31, 11 September 2005 (UTC)

Your version of the example contradicts the rules and spelling which you are protecting now and besides it is ambiquious. But the authors of course, published this using professional phonetic alphabet, not the way that is used in your example (which is obviously, adapted version, and adapted by non-professional). So, insert the original version from the sciencework.--Nixer 23:06, 11 September 2005 (UTC)

You're right. There are all sorts of problems with the reconstructed texts. That's why they shouldn't be there at all in the first place. But if they are there, they should be there in the spelling they were published with. That's just academic honesty. --Angr/_{tɔk tə mi} 23:20, 11 September 2005 (UTC)

Do you have the original paper? They can not be published with this spelling. Moreover - a important information was lost during the "adaptation" for the standard Latin charset.--Nixer 23:35, 11 September 2005 (UTC)

ffs, Nixer, when will you get it, It's not our version, it's Schleicher's.

It is not a version by Schleicher. It is not even version by Lehmann. Get original sciencework and look yourself.--Nixer 11:21, 12 September 2005 (UTC)

Maybe once you publish a major PIE Compendium yourself, we will consider citing your spelling preferences here. This is a wikipedia article, and not your personal notepad. Did it ever occur to you that you are a nuisance to other editors with your unshakeable convictions combined with your limited expression power? Angr, I support removal of Schleicher's fable. It does have its own article, after all. dab (ᛏ) 10:21, 12 September 2005 (UTC)

Nixer is in blatant 3RR violation now. I will smack him with a short block to show that we are serious. He has been warned before. I count six reverts by Nixer in the last 24 hours. Nixer has been shown far too much lenience in the face of his behaviour. Nixer, even if there is a spelling mistake in our Schleicher's version (which I doubt), why don't you point it out to us rather than reverting to your unsourced version, together with your unexplained numerals. dab (ᛏ) 11:30, 12 September 2005 (UTC)

dispute

we seem to be making slow progress. This is like dealing with a four year old, frankly. Nixer seems to be unhappy with our spelling conventions.

Your spelling conventions are not too bad, except the fact they hide ethymological connection between modern languages and PIE and too hard for non-professionals. But the issue is that the text examples that do you bring here do not match the conventions that you are protecting here.--Nixer 09:23, 13 September 2005 (UTC)

Why, for crying out loud, don't you make a coherent suggestion here, then? "There is another version" certainly isn't a comprehensible basis for a "dispute". "There is another version" of the numerals? Where? What? Why? Spelling? Phonology? What do you want? There are variants of reconstructions of practically everything in this article. If you were able to give a coherent overview of the disputed points, your edits would be most welcome.

Re-read the prevous topic. --Nixer 09:23, 13 September 2005 (UTC)

"There is another version" is not a coherent representation of an academic dispute. It is not even clear whether you are going about actual differences of reconstruction, or just spelling conventions. dab (ᛏ) 09:08, 13 September 2005 (UTC)

Nixer, the etymological connection between PIE and modern IE languages cannot be shown by any spelling conventions, so that is no argument. What is hard for nonprofessionals is to be confronted with several different spelling conventions at once. There is one spelling convention in use on this page, and that is the same one that users will find in the vast majority of works dealing with the reconstruction of PIE. If you don't want to confuse nonspecialists, stick to the accepted spelling conventions. There is no serious content dispute on this page, and I'm removing the "disputed" tag accordingly. --Angr/_{tɔk tə mi} 09:54, 13 September 2005 (UTC)

Nixer actually says the "spelling conventions are not too bad" (thank you! I am sure the past two generations of IE scholars will be very glad you approve!), but he objects to the texts given here. Hello? the texts were removed to sub articles. Your revert concerned your list of numerals. Nixer, what are you, a troll? You don't make any sense at all, even if one tries very hard to parse your English. Please stop wasting our time. dab (ᛏ) 10:42, 13 September 2005 (UTC)

Thorn stuff

Now I am tempted to put the "disputed" tag back up. I thought the whole idea of *kþ and *gð had been chucked out decades ago. --Angr/_{tɔk tə mi} 14:12, 13 September 2005 (UTC)

:) you mean the "metathesis" explanation is not proper. It was really intended as a bait to spur you (and others) into giving more detail... the equations Hittite tk, Sanskrit kS, Greek khth stand, of course, but I agree that it should be explained differently. But we'd then probably need to get rid of the "þ allphone" altogether, no? dab (ᛏ) 14:51, 13 September 2005 (UTC)

Oh, metathesis happened, it just didn't happen in PIE. Hittite and Tocharian show us that. Metathesis happened in Greek, and something happened in Sanskrit so that tk´ showed up as kS but kt stayed kt. It needn't have been metathesis in Sanskrit; I can easily imagine a sound change tk´ > tś (satemization) > kś (dissimilation) > kS (ruki rule). I don't know whether that's ever been published, though; it just occurred to me this second. By all means we want to get rid of the þ allophone for PIE altogether. --Angr/_{tɔk tə mi} 16:29, 13 September 2005 (UTC)

well, yes, I mean metathesis in PIE minus Tocharian and Hittite, of course, the examples I give, after all, show that no metathesis took place in Hittite. What's the problem with KT staying KT? It's only TK that is at all affected. It's not like TK becomes KT first and Kþ in a second step. TK goes to Kþ directly (I thought you were objecting to 'metathesis' because of that). dab (ᛏ) 18:41, 13 September 2005 (UTC)

Well there may or may not have been a PIE minus Hittite, but I certainly don't believe in a PIE minus Tocharian. There's no problem with KT staying KT at all. All I'm saying is that whatever the various languages decided to do with TK, they decided to do it after PIE was over. The only reason þ was ever postulated was because the Neo-Grammarians, not knowing Hittite and Tocharian, wanted something to cover the equation Greek kt = Sanskrit kS. It couldn't be PIE *kt, because that was already Greek kt = Sanskrit kt (or ST in the case of k´t, e.g. aSTau 'eight'). So they picked something that was dental like t but a fricative like S and landed on þ. Once Hittite and Tocharian were discovered, it turned out that kþ words had tk in PIE, so you had to get from *tk to Greek kt, Sanskirt kS, and whatever the other languages do. Greek has simple metathesis; Sanskrit has something funny that might or might not have involved metathesis; but no language requires the postulation of an allophone þ. The only allophonic variation reconstructable for PIE as far as I know is that /t/ has an affricate allophone [t^s] before another /t/ (which probably only happened at a morpheme boundary, a root ending in t followed by a suffix beginning with t); this [t^st] sequence became tt in Indo-Aryan, st in Greek, Germanic, and Balto-Slavic, ss in Italic and Celtic, and I don't remember what elsewhere. --Angr/_{tɔk tə mi} 19:14, 13 September 2005 (UTC)

fair enough, by all means insert this into the article. I am happy to accept the þ stuff as an early 3rd millennium development. I do believe in a, well, post PIE dialect continuum, minus Tocharian. I.e. the Tocharians lost contact before 3000, and that mattered. I don't know when the Anatolians said goodbye, maybe around 3500. By 3000, PIE may have been over, but people, minus Anatolian and Tocharian, stayed in touch, so to speak. dab (ᛏ) 20:38, 13 September 2005 (UTC)

Whatever the factual merits, it seems to me that this is rather too detailed for the place where it is now--in the presentation of PIE phonology. If it is to be retained, a separate section would be in order--or perhaps better yet a separate article. --teb728 21:37, 13 September 2005 (UTC)

that's how we work :) start hacking things together, and export them when they begin to take shape. In any case, I found a subsection with the content t, d, dh is a bit pathetic. If we do want subsections for each row of stops, they need to be systematically discussed, not just enumerated. dab (ᛏ) 05:56, 14 September 2005 (UTC)

subscripts

I'm fine with the superscript ʷʰ -- but the subscript ₃₂₃ render ugly in my browser, plus they don't align with ₄, _x if these occur once in a while (see the king and the god). I suggest we use h₁, h₂, h₃. dab (ᛏ) 06:28, 14 September 2005 (UTC)

For me, the problem is the italics. I had to open the edit window just to see what you meant by ʷʰ, although it works fine as ʷʰ. That's why I argued against having the {{PIE}} template automatically put everything in italics in the first place. The same is true for the number subscripts. ₁₂₃ are almost illegible but ₁₂₃ are fine. --Angr/_{tɔk tə mi} 07:07, 14 September 2005 (UTC)

hm, ʷʰ and ʷʰ look exactly the same for me, since my broswer refuses to italicize these. But you have a point, maybe we should remove the italicization from {{PIE}}. We can always add it back in half a year or so if/as standard fonts get better. dab (ᛏ) 07:11, 14 September 2005 (UTC)

The standard way of typesetting PIE reconstruction is in italics. We should do the same here. Wikipedia is not a place for original research, which includes how forms are presented. Our job is to mirror the handbooks as closely as possible. CRCulver 07:12, 14 September 2005 (UTC)

that's not an argument. We need to decide on one single standard, rather than copying various conventions indiscriminately from handbooks. Decisions on layout and naming conventions, and on how to organize an encyclopedia in general, is "original research" performed on Wikipedia. Most of IE literature is not set in html. If we find that current browsers don't render PIE notation in italics satisfactorily, we will, of course, opt for some solution that does render satisfactorily. —Preceding unsigned comment added by Dbachmann (talk • contribs) 07:26, 14 September 2005

I agree the italics are quite illegible. I would like to try boldface or an increased font size.--JWB 23:27, 5 October 2005 (UTC)

Articles on PIE vocabulary

A number of articles on PIE vocabulary (eg Keu-3) were recently proposed for deletion in a debate which is not concluded but is tending towards the articles being merged into a single article, Proto-Indo-European roots. Esperanto vocabulary is suggested as a possible parallel article to take some inspiration from. Rd232 18:01, 19 October 2005 (UTC)

we shouldn't have articles on individual roots. There can conceivably be articles about individual words, such as Dyeus. But there is no need to delete these, either. Just make them redirects. dab (ᛏ) 20:47, 11 November 2005 (UTC)

How would you all feel about proposing a new Wiktionary for PIE? It could take the Indo-European root as the head-word and list the derivatives. In interesting cases it could also have discussions of interesting points, even if Wiktionaries don't normally have much discursive text. --Doric Loon 21:40, 11 November 2005 (UTC)

yes, but that would still leave Copyright issues. We couldn't just make up our own forms on wiktionary, we'd have to go by Pokorny. Ultimately, we'll just reproduce Pokorny, which will be a copyright violation (not that it isn't online, but that isn't wikimedia's problem). dab (ᛏ) 21:54, 11 November 2005 (UTC)

It sounds like a good idea to me. And I shouldn't think it would be all that close to Pokorny; for one thing it would have laryngeals. I would expect it to be as similar to Watkins as to Pokorny. --teb728 04:55, 14 November 2005 (UTC)

By the way, Dyeus - incorrect. Correct form is deivos.--Nixer 21:45, 11 November 2005 (UTC)

no it isn't. It's a different word. Dyeus is the chief god. deivos is an adjective meaning "celestial". Same root, different words. dab (ᛏ) 21:54, 11 November 2005 (UTC)

Heh How about rēḱsdeiwos-kʷe? :-) --Nixer 22:11, 11 November 2005 (UTC)

what about it? It isn't about Dyeus, but about Verunos (or Lughus). Dyeus is a *specific* god; deivos just means "divine one", i.e. "god". Seeing your blasphemous attitude, I would advise you to stay indoors during thunderstorms over the next few weeks :p dab (ᛏ) 22:20, 11 November 2005 (UTC)

No, this is not an adjective, this is noun.--Nixer 23:37, 11 November 2005 (UTC)

The translation of the tale name is "The king and the god", not "The king and the light one". By the way, "light one" (adjective) is leucos which is also exist in the text. How would you translate "vequet leucos deivos Verunos" ?--Nixer 23:40, 11 November 2005 (UTC)

Nixer, you know what? Wikipedia isn't "the Free University", and I'm not paid to tutor you. I suggest you go read some books. What do you want? You have been trolling these articles for months, and you only managed to make clear that your knowledge of the subject is superficial at best. dab (ᛏ) 11:22, 14 November 2005 (UTC)

Stop inserting the bullshit here!

The numbes without d letter - is an evident bullshit and certaily became in use far after the PIE times in certain germanic languages.--Nixer 12:53, 14 November 2005 (UTC)

The forms quoted are from reputable English-language handbooks. Please stop arguing against the views of legitimate scholars. Wikipedia is not a place for original research or crank views. CRCulver 13:12, 14 November 2005 (UTC)

Nixer: WP:CITE, WP:NOR. If you have read these, you haven't understood them. Make your own webpage about your own theories and stop pestering Wikipedia, which is about reporting academic publications, it is not an academic publication itself. Get your views published in some reputable journal, and we would at least be able to mention them. dab (ᛏ) 14:02, 14 November 2005 (UTC)

Which books do you consider reputable and which - not? If Starostin's publications for you are not reputable, than what is?--Nixer 14:32, 14 November 2005 (UTC)

Sarostin doesn't seem to be universally appreciated, and I find that his name rarely comes up in bibliographies nowadays. I'd recommend sticking to the authors of the standard English-language handbooks: Lehmann, Beekes, Sihler, and (if one must) Szemerenyi. CRCulver 02:04, 16 November 2005 (UTC)

This form without d letter is from proto-centum language.--Nixer 14:33, 14 November 2005 (UTC)

Nixer, here's something else for you to read and understand before you start reverting again: Wikipedia:Blocking policy. --Angr/_{tɔk tə mi} 14:13, 14 November 2005 (UTC)

it would help if you actually quoted Starostin. Starostin was a serious linguist, even if he had a penchant for the speculative. You have misquoted him before, so you can be sure that we won't take your word for "Starostin said this or that". To begin with, I would even like a quote that he believed in a "Proto-Centum" langauge (I don't think it is impossible that he postulated this, even if phylogenetic Centum-Satem separation is certainly not a mainstream opinion, and I have yet to find a proponent; people are shruggingly agnostic about it at best, nobody to my knowledge is positive about a "Proto-Centum" language) dab (ᛏ) 15:17, 14 November 2005 (UTC)

rather than follow my invitation and actually quoting Starostin (rather than just namedropping him), Nixer is out on a "3RR-aware" revert war (4 reverts in 27 hours) -- this is clearly bad faith now, and if I could be bothered, I would open an rfc, or ask other admins to chastise him or whatever. But since there isn't really anything to comment about, I suppose clicking 'rollback' every few hours is cheaper for now. Nixer, I do work on WP anyway, it is really the same to me to roll you back every now and then, so why do you even bother? dab (ᛏ) 16:15, 15 November 2005 (UTC)

I have blocked him for 24 hours for gaming the system. Violating the spirit of 3RR is just as punishable as violating its letter. --User:Angr/talk 17:10, 15 November 2005 (UTC)

I am not too happy with this Angr -- since you are clearly involved. I suggest you post your action on WP:AN/I for others to review, so people can unblock him if they disagree. dab (ᛏ) 17:36, 15 November 2005 (UTC)

Done. --User:Angr/talk 18:33, 15 November 2005 (UTC)

In reply to CRCulver, I do not think that we necessarily should restrict ourselves to a handful of prestigious scholars. Starostin imho definitely deserves to be referenced, if necessary with the caveat that he is not mainstream. However, we haven't even reached the stage where the question is whether or not to quote Starostin, because Nixer hasn't quoted Starostin on anything. He just kept waving the name around. dab (ᛏ) 10:50, 16 November 2005 (UTC)

May be there was some form without d sound, but it's like English twony or Russian шеесят - the shortened casual forn in fast speech. The d sound, though existed in PIE because it is preserved in Russian until now.--Nixer 11:40, 17 November 2005 (UTC)

The [d] sound of двадцать, тридцать, пятьдесят, шестьдесят, семьдесят as well as the Germanic [t] of twenty, thirty, forty, fifty, sixty, seventy, ninety is NOT inherited from Proto-Indo-European. These words are secondary formations within Slavic and Germanic; they are transparently re-built from the simple digits plus a form of the word for "ten", and as such they CANNOT be used in the reconstruction of Proto-Indo-European. The word-internal *d can only be presumed as likely in oldest PIE, but it cannot be reconstructed on the basis of words in existing IE languages. Nixer, please leave the list of numerals alone now. --User:Angr/talk 14:40, 17 November 2005 (UTC)

Nixer may have a point. I can tell you that in Albanian, those numbers are: dhjetë (ten), njëzet (twenty), tridhjetë (thirty) dyzet (forty), pesëdhjetë (fifty), jashtëdhjetë (sixty), shtatëdhjetë (seventy), tetëdhjetë (eighty), nëntëdhjetë (ninety). Do you see all the dental consonants. Is it a mere coincidence that they are present here as well? You do I presume know the place of Albanian within the IE language family. Rex^(talk) 18:09, 17 November 2005 (UTC)

Yes, Albanian is one of the most recently attested and least conservative branches of Indo-European. The dentals in Albanian are certainly secondary, just as they are in Slavic and Germanic. More conservative branches (Greek, Sanskrit, Latin, Celtic, Armenian, Tocharian) do not show any trace of a *d in the parts of the words that correspond to English -ty. Re-forming one's words for "twenty, thirty, forty", etc. along the lines of "two-tens, three-tens, four-tens" is trivial and can happen at any time. The fact that the Greek, Sanskrit, etc. words do not look like this proves they must be original. --User:Angr/talk 18:35, 17 November 2005 (UTC)

I beg to differ. In Greek they are: deka, endeka, dodeka, dekatria, dekatessera, dekapente, dekaexi, dekaephta, dekaochto, dekaenia. So now we have a pattern:

     CENTUM < GREEK & ENGLISH
PIE <
     SATEM < ALBANIAN & RUSSIAN

All these language have a dental in those numbers. Where did they get if from? From their common root (PIE) of course. Rex^(talk) 19:00, 17 November 2005 (UTC)

May I take it that your silence indicates approval and that the article will be edited accordingly? Rex^(talk) 19:45, 17 November 2005 (UTC)

what are you even talking about? this isn't about teens, it's about tens (20, 30, 40 etc.). could you at least check out what the subject is? nobody disputes that that "ten" was dek'm-, that's not the bleeding point. Can we just go back to citing references rather than setting up strawmen please? dab (ᛏ) 21:38, 17 November 2005 (UTC)

Forty-five minutes of waiting is not enough to assume I'm being silent. dab is right, we're talking about the decades: 20, 30, 40, 50, 60, 70, 80, 90. That's what you glossed the Albanian examples as. The Greek words in question are eikosi, triakonta, tessarakonta, etc., with a decidedly non-dental [k]. --User:Angr/talk 22:00, 17 November 2005 (UTC)

You are speaking about modern Greek, not ancient. Do you know the word dodecahedron, which is from ancient Greek? By the way, Armenian IS NOT conservative.--Nixer 22:12, 17 November 2005 (UTC)

No, I'm not. I'm speaking about Ancient Greek. Dodecahedron comes from the Greek word for 12, not the word for 20, and is therefore irrelevant. --User:Angr/talk 22:53, 17 November 2005 (UTC)

My apologies, I meant eikosi, trianda, saranda, peninda, exinda, ebdominda, ogdonda, eneninda. Anyway, do you have any references that Albanian is conservative. That may be just your POV. And what about Latin: viginti, triginta, quadraginta, quinquaginta, sexaginta... There are dental consonants all over the place! Rex^(talk) 22:24, 17 November 2005 (UTC)

Also in Sanskrit: vimsatih, trimsat, catvarimsat, pancasat etc. It's obvious that all these dentals are from PIE. Rex^(talk) 22:28, 17 November 2005 (UTC)

In Latin it's g. But here is Sanskrit: vinshati, trinshath, chatvaarinshath... n is dental consonant--Nixer 22:34, 17 November 2005 (UTC)

PS can't the ancient Greek ones be interpreted as: eikosi, triakonta, tessarakonta, etc There's dentals all over the place. I trust that the article will be edited accordingly... Rex^(talk) 22:32, 17 November 2005 (UTC)

No, of course.--Nixer 22:34, 17 November 2005 (UTC)

Wrong, darling - proof: in Armenian the same numbers are kessan, yeressoun, karassoun, hissoun, vatssoun, yottanassoun, outssoun, innissoun. Who can do the maths? I see dentals! Rex^(talk) 22:42, 17 November 2005 (UTC)

PS, /s/ is a dental. Rex^(talk) 22:44, 17 November 2005 (UTC)

Compare Russian, Sanskrit and PIE:

dvadsat - vinshai - dvidcmt,

tridsat - trinshath - tridcmt

(old Russian) chetverdsat - chatvaarinshath - quetverdcmt--Nixer 22:45, 17 November 2005 (UTC)

Similarities - this is further proof that those fairy tales about *wīḱm̥t-, *trīḱomt-, kʷetwr̥̄ḱomt- etc should be binned. Rex^(talk) 22:50, 17 November 2005 (UTC)

You're looking at the wrong place in the word. Consider the entry for "thirty": "*trīḱomt-; originally perhaps *tridḱomt-". The question is whether the *d before the *ḱ can be safely reconstructed for the proto-language. None of the conservative languages indicates that it can; it just seems likely to have been there at some early point on the basis of the words' meanings. The dentals in trimśat, triakonta, etc. are from the *t after the *m, which is uncontroversial. The nasal consonant of the Sanskrit forms cannot come from *d and so is also irrelevant for the current discussion. Armenian /s/ is dental, but it comes from *ḱ, not *d, so for purposes of reconstruction it counts as an originally nondental consonant. --User:Angr/talk 22:53, 17 November 2005 (UTC)

Russian numerals can be derived from PIE ones with trivial rules: in Satem languages m became a, c became s or ts (Russian ц), so all the Russian numerals simply deriving with regular phonetic rules.--Nixer 22:59, 17 November 2005 (UTC)

Albanian numbers are derived from PIE, I thought Franz Bopp was clear enough on that point. These strawman arguments have got to stop. The article as it currently is is misleading. Rex^(talk) 23:05, 17 November 2005 (UTC)

Then how do you explain the Lithuanian: dešimt, dvidešimt, trisdešimt, keturiasdešimt, penkiasdešimt, šešiasdešimt, septyniasdešimt, aštuoniasdešimt, devyniasdešimt? Coincidence? Rex^(talk) 22:56, 17 November 2005 (UTC)

I already said above: "Re-forming one's words for 'twenty, thirty, forty', etc. along the lines of 'two-tens, three-tens, four-tens' is trivial and can happen at any time." Shall I say it again? Words meaning "twenty, thirty, forty" etc. are especially prone to being tossed out and re-formed when they stop being transparent. It happened in Balto-Slavic, Germanic, and Albanian. These languages show a reflex of PIE *d before the reflex (if any) of *ḱ not because a cluster *-dḱ- is actually reconstructable for PIE, but merely because they re-formed the words on the basis of their words for "ten". The Lithuanian, Russian, Germanic, and Albanian numbers for 20, 30, 40 etc. are not the direct descendants of the PIE words for 20, 30, 40, etc. --User:Angr/talk 23:13, 17 November 2005 (UTC)

So, your point is: in Russian it should be двацать, not двадцать, трицать, not тридцать (dvatsat, not dvadsat, tritsat, not tridsat) etc. Then try to pronounce it differently - you hardly will sucseed.--Nixer 23:21, 17 November 2005 (UTC)

I think that the evidence that the t in twenty etc is derived from the PIE word for the same number is overwhelming. I can't understand what is so objectionable about that. Rex^(talk) 23:32, 17 November 2005 (UTC)

What you or I or Nixer or dab thinks is irrelevant. This is an encyclopedia, we report on what the published sources say. The published sources say the reconstructed forms are *wīḱm̥t-, *trīḱomt-, *kʷetwr̥̄ḱomt-, etc., which presumably (but not provably) come from even older forms *dwidḱm̥t-, *tridḱomt-, *kʷetwr̥dḱomt-, etc. I have cited two sources who agree on this fact. No published source I can find thinks that the t of the Germanic forms or the d of the Slavic forms goes back to Proto-Indo-European. We must stick to what the published sources say. Removing valid, sourced information from pages is vandalism, so don't do it again, either of you. --User:Angr/talk 06:53, 18 November 2005 (UTC)

yes guys. It's nice you have an interest in PIE reconstruction, but this is a serious article, and not like theism or something where anybody can insert 'some think', 'others believe', 'yet others fancy' etc. This is a discipline with more than a century of serious research. You cannot think you can brainstorm on a wikipedia talkpage for 20 minutes and think you have contributed anything of note to Indo-European linguistics. Go to your library. Find an introduction to IE linguistics. Check the reconstructions given there. If they are different from what we have here, come back, cite author, title and page, and the form given. Then we'll have something to talk about. Angr was very kind to give you some basic advice. If you have any more questions, come to WP:RD/L. Thanks. dab (ᛏ) 07:37, 18 November 2005 (UTC)

No controversial information

We all agree that in the PIE there was form with d sound. I even can agree that form without d sound could existed in fast casual speech. I even can agree that German (and only German) forms were rebuilt. Though existance of these forms in later PIE still controversial. So we should remove the contraversal information from the article until the full consensus--Nixer 10:08, 18 November 2005 (UTC)

Nixer, stop it. We have asked you to cite sources dozens of times now. Wikipedia:Verifiability is policy. If your opinion is challenged, you have to back it up with citations, otherwise your edits are in violation of policy. As long as you don't cite a source, we are not even in dispute over anything, you are simply trolling this article. Once you cite a single source, we can begin a reasonable dispute. If you continue to fail to recognize this basic principle of Wikipedia, I will issue short blocks from now on. dab (ᛏ) 10:21, 18 November 2005 (UTC)

I've added the disputed notice until this dispute has been resolved. Rex^(talk) 13

05, 18 November 2005 (UTC)

I will agree that there is a dispute once you have cited your first reference. All belligerence and no sources does not make for a dispute. dab (ᛏ) 13:49, 18 November 2005 (UTC)

It is difficult to AGF at this point Rex, but maybe you are just a little slow. So once again: You may dispute information in an article, but if you do so, you need to cite credible references. You may not have bothered to note this, but Angr has provided references for the mainstream position:

Beekes 1995, 212–16; Sihler 1995, 402–24

Nobody is trying to censor minority opinions here. We are actually begging you to come forward with such views, so we can quote them and be done. You cannot slap "disputed" tags on articles just on a whim. Cite your sources. Once you do so, and still are turned down, then you will be in a position to add warning tags. Until you have not cited a single reference for whatever is your view, I will regard further reverts as plain vandalism. dab (ᛏ) 15:20, 18 November 2005 (UTC)

I've been following this discussion with fascinated bewilderment, but now it has reached the point of sheer silliness. Angr and Dab are right: it's time you two dropped this. You are coming over as amateurish, and if your only basis is what seems subjectively most plausible to you, then you would be better to call it a day. --Doric Loon 15:27, 18 November 2005 (UTC)

Perhaps you haven't noticed, but there is an edit war going on. In such circumstances, the factual accuracy and/or neutrality of the article is disputed. As far as I can see, the placement of the tag is necessary in such circumstances. However, I removed it, because I have been in an edit war over a tag in the past and I know that users don't like it when the article is in their version. That's why I removed it. Rex^(talk) 15:28, 18 November 2005 (UTC)

It's called reflection. Check the edit history, after having had second thoughts over placing the tag as it would appear silly given that the article is copied from those books and the content of the books is not disputed. That's why I removed the tag at 15:19, but you failed to notice and placed your bad tempered message here at 15:20 (check the edit history). Rex^(talk) 15:32, 18 November 2005 (UTC)

wow. so you reflected. After pestering us for two days. Thank you so much for reflecting, please consider reflecting before editing next time. You call this an 'edit war'? I call it blatant trolling. Nixer has been doing this for ages, only god knows why, and you just came along for the ride after we discussed Nixer's block on AN/I. 'fascinated bewilderment' is the right term for what I'm experiencing here. I cannot even guess what sort of contorted argument is going on in Nixer's brain, but Rex really seems to be here just for the spectacle. This isn't even a controversial point in PIE linguistics. I mean there are literally dozens of topics worth disagreeing over, but even there you'd need provide references. Just pointing out that certain words contain dentals is no substitute for citing actual academic discussions. dab (ᛏ) 15:40, 18 November 2005 (UTC)

Petty sarcasm benefits no one. I think I have already told you, that Wikipedia:Trolling involves bad faith and is a deliberate attempt to disrupt. That is not what has been going on here. Disputes should be resolved by compromise - clinging to your position (ring a bell, dab?) may constitute trolling. As far as I can see, Nixer is the only one who has even attempted to reach a compromise by altering his position. You have stubbornly clung to yours. WHat does that tell you? Rex^(talk) 15:49, 18 November 2005 (UTC)

I stubbornly cling to WP policy, which says that as long as you don't cite sources, you don't have a case, and are only wasting people's time. Why would a sane person do that in good faith? Nixer need only have opened Beekes, and asked us to mention that he reconstructs -dk- for early PIE, voila, no objections. Alternatively, you can pester people, without citing a single author, until somebody gets frustrated enough to do your job for you. Works every time, but it's not what I call 'good faith'. The -dk- forms were in the article all the time. Nixer wants the information gone that the -d- is a conjecture that cannot directly be reconstructed. Why he wants that information gone, nobody knows. Why you decided to support him in that I can only guess at, maybe because you decided you wanted to get my goat, over at AN/I? This may not be the case, but I honestly see no other reason why you would get involved in such a silly revert war. dab (ᛏ) 21:23, 18 November 2005 (UTC)

I'm surprised you actually see it that way. If you check which articles I have edited you will see that a substantial number of them are language related. I have an interest in this area and at first sight, with what I know (however limited it is), Nixer seems to make more sense. Of course I have a long history of imagining links and cognates, such as saying that the English word "milk" is somehow linked to the Greek word "Αμέλγω" and the English word "door" to the Greek "θύρα". However, I find it odd that you think that I wanted to get your goat (sic). Don't you remember on Talk:Arvanitic language (another language related article), where you were one of the selective few who exhibited some signs of sanity and recommended that Greek propaganda be dropped in favour of neutral wording in line with every single source we had. Rex^(talk) 21:56, 18 November 2005 (UTC)

PS I have recently cleaned up the Albanian language article, and I would like your opinion on certain issues. People have been saying that they don't like the Indo-European cognate table, the Tosk-Geg table, and here are the odd ones: I have an Albanian grammar book here and it says that there are six cases in Albanian: Nominative, Genitive, Dative, Accusative, Ablative and in certain words a Vocative. People have been saying that as Genitive is the Dative with a preposition, then they are the same case (see Balkan linguistic union) and that is why the Genitive was dropped from the table - what do you think. Also, People have been suggesting that Albanian doesn't have palatal plosives, but they are something else (postalveolar affricates) (q and gj). What is your opinion? Rex^(talk) 21:56, 18 November 2005 (UTC)

Nixer, for your information, Beekes 1995 (p.214) lists the Old Church Slavic decads (partly Romanized) as “dъva desęti, trьje desęte, četyre desęte, pętь desętъ, sestь desętъ, etc.” So apparently the Slavic decads were also rebuilt. --teb728 17:08, 18 November 2005 (UTC)

About Albanian q and gj I have read that – exactly like the neighboring Macedonian ќ and ѓ – they are pronounced as palatalized velar plosives ([kʲ gʲ]) in the south (I've read Modern Greek has the same or something similar) and as alveolopalatal ( = palatalized postalveolar) affricates ([t͡ɕ d͡ʑ]) in the north (the latter being identical to the northern neighbor Serbocroatian's ć/ћ and đ/ђ. Balkan Sprachbund at its best. :-) Unfortunately I don't remember the source, and I've never been to Albania or Macedonia.

Maybe palatal plosives ([c ɟ]) are used somewhere in the Albanian-speaking area, too, I have no idea, but generally, palatal plosives are rare, at least in Europe.

David Marjanović | david.marjanovic_at_gmx.at | 17:06 CET | 2006/11/18

Proper form

Look at this:

[1]

In this page Azzurro describes the proper forms of numerals. Also here [2] he places the proper form of the Schleicher's tale.--Nixer 22:41, 19 November 2005 (UTC)

Sihler did not say what you have attributed to him. You are sort of getting the right idea--citing a source. But it has to be a published source by a recognized expert; a submission on an online forum doesn't count. If you find a recognized expert who disagrees in a published source with what Sihler or Beekes writes, you might be able to add it. But don't remove what Sihler or Beekes write (unless, of course, you can show they are incorrectly quoted). You may not agree with Sihler and Beekes, but that does not negate what they write or the fact that they are experts. --teb728 04:10, 20 November 2005 (UTC)

The sources are Semereni and Beekes.--Nixer 08:39, 20 November 2005 (UTC)

Nixer, I've reverted your edit. Why? Because you didn't simply add content of your own, you obliterated the valuable content that was there before. The page lists a generally-accepted reconstruction with laryngeals, so why are you replacing this with something that has a very out-of-date view of PIE phonology? I think it is about time that we took you to Requests for Arbitration, since you are not only putting forth very odd theories not widely accepted by the scholarly community, but erasing the hard work that we've already done. CRCulver 09:53, 20 November 2005 (UTC)

There are no generally accepted larynegals in the your version. My position is: place here reconstructions with larynegals or place without them. But you are placing here the reconstruction without all the larynegals, partly displaiing them. This is invalid reconstruction. Next, there IS difference in the ending between the numeral for twenty and thirty, for example. You place here numerals without this difference.--Nixer 10:09, 20 November 2005 (UTC)

The laryngeals here are generally accepted. "Nine" has a prothetic vowel in Greek, therefore there was a laryngeal (alternative theories about prothetic vowels passed away in the 1960s), "one" and "eight" would have to have laryngeals to fit in the standard shape of PIE roots as set forth by Benveniste. These are all matters that were decided years ago. Do you not have access to the latest handbooks? It certainly seems that way if you are citing amateur websites and Szemerenyi. Please avail yourself of up-to-date resources and be prepared to quote specific pages before joining the discussion. CRCulver 11:43, 20 November 2005 (UTC)

Nixer, in a recent edit summary you asked, “Why you dont accept ANY changes of mine?” But when you have good ideas, they are accepted. Let me remind you: On 2 September, you introduced the Numbers section, and with changes to the reconstructions to conform to modern scholarship it is still there. On 13 September (when you stopped insisting on a different spelling system) you added another reconstruction of thousand, and it’s still there just as you entered it. But in the current (or hopefully recent) revert-war you have only erased the widely accepted reconstructions even though the reconstructions you favored were always there as alternates (except the differences in the endings for twenty and thirty). By engaging in revert-wars you gain a bad reputation and obscure good ideas you may have.

With regard to the points in your last reply, I don’t have access to Sihler, but I see that in Beekes laryngeals are used consistently, and twenty (duidḱmti) and thirty (trih₂dḱomth₂) have different endings. --teb728 11:58, 20 November 2005 (UTC)

Nixer has cited a source? Ah, I see, it's just one "Azzurro", a promising "Newbie" posting on the "Разное" section of a Russian internet forum. Still, that's quite am improvement over citing no sources at all, go Nixer, you must be dizzy from such academic heights! Maybe one of these days he'll find out there are libraries? And cite actual books? You never know. In the meantime, I fully support Nixer in this! Why bother with losers like Beekes when we could quote Azzurro in our article. dab (ᛏ) 11:41, 20 November 2005 (UTC)

dab, quit the sarcasm: it doesn't beome you --teb728 11:58, 20 November 2005 (UTC)

sorry, I don't see any way to discuss this with a straight face anymore. That said, Azzurro seems to know his stuff, and if Nixer discusses his edits with Azzurro beforehand, we'll have less trouble. The thing is that he seems more prepared to listen to Azzurro than to us, I don't know why, maybe because the discussion is in Russian. The thread is enlightening in other respects, too. It appears Nixer has free reign on ru:Праиндоевропейский язык, with half the article consisting of his "Примеры текстов", and somehow cannot stomach our standards here. dab (ᛏ) 12:16, 20 November 2005 (UTC)

If you have anything against, say. Why the examples posted by Azzurro are incorrect?--Nixer 13:06, 20 November 2005 (UTC)

No one is saying those are wrong. The point is, Sihler is a published source, so we include his reconstructions on this page. Beekes is also a published source, so we include his reconstructions on this page. Azzurro is just some guy on an internet forum, so we don't include his reconstructions on this page. --User:Angr/talk 14:56, 20 November 2005 (UTC)

This is not HIS reconstructions. He gave them from the sources.--Nixer 16:38, 20 November 2005 (UTC)

Which sources? --User:Angr/talk 17:20, 20 November 2005 (UTC)

*Absolute Silence...* Rex^(talk) 18:09, 20 November 2005 (UTC)

"References: Azzurro (quoting teh sources)" dab (ᛏ) 19:22, 20 November 2005 (UTC)

In his last post to the first thread of this section (just above dab's unhelpful rant), Nixer made a couple of interesting points: Laryngeals seem to be shown inconsistently, and the reconstructions shown lack the different endings of twenty and thirty. Accordingly I have changed the article to show all of Beekes' reconstructions, which use laryngeals consistently and include the endings of twenty and thirty. Someone with access to Sihler should verify the reconstructions attributed to him, as I doubt that he reconstructed two with both 'u' and 'w' (aren't they alternate spellings of the same thing?) and I doubt that he reconstructed eighty with two o-grades in one word. --teb728 23:07, 20 November 2005 (UTC)

I think Beeks is enough for the purpose. We should delete the first column.--Nixer 02:33, 21 November 2005 (UTC)

First I want acknowledge the fact that your edit was not destructive this time; that is a change for the better. However, I disagree totally with your comment above. We should not delete the first column. For one thing, it is more in line with the thinking of mainstream linguists than Beekes. Secondly part of the “purpose” is to show that there is disagreement over some reconstructions. --teb728 03:13, 21 November 2005 (UTC)

you are right, my rant may not have been helpful. If Wikipedia means putting up with the clueless and deranged, talking to them in soothing friendly tones and praise them if once every couple of months they make a non-destructive edit, Wikipedia may not be for me after all. I prefer to work with educated peers who agree to put up their references or shut up, and I don't expect to be treated any differently myself. dab (ᛏ) 09:12, 21 November 2005 (UTC)

Proto-World language

I do not consider rolling back Nixer's stuff revert warring so much as tedious cleaning, but others seem eager enough to jump in and block me for it, so here's a pointer to Proto-World language (edit | talk | history | protect | delete | links | watch | logs | views) where Nixer is doing his inscrutable work at the moment. dab (ᛏ) 22:02, 22 November 2005 (UTC)

Whatever. There's nothing encyclopedic that can be said about that topic anyway, so let the cranks have their crank playground. --User:Angr/talk 22:13, 22 November 2005 (UTC)

I suppose you have a point. I still liked having a responsible article about an irresponsible topic :) feel free to ignore the topic, of course dab (ᛏ) 22:21, 22 November 2005 (UTC)

"root" articles

what shall we do with Ker-5 and friends? They could all be redirected to List of common Indo-European roots, which is not too bad (but why "common"? -- it could be moved to Proto-Indo-European roots, the article there badly needs merging too. But we need to establish how to go about this dictionary material. Is it ok for Wikipedia to be a PIE dictionary, too? What about copyright issues, can we just dump Pokorny and be done? dab (ᛏ) 12:40, 2 January 2006 (UTC)

Ack. As far as I'm concerned, Wikipedia is not a dictionary, including a dictionary of hypothetical roots. Transwiki them to Wiktionary if they'll have them. Obviously we can't just dump Pokorny there, not only for copyright reasons, but because Pokorny was already forty years out of date when it was published, and is even worse now. At most, all we should have here in Wikipedia is Proto-Indo-European roots, which should not be a list but rather a discussion of what IE roots look like (full grade, zero grade, o-grade, extended grade; the fact that *tek-, *teg-, *dek-, *degh-, and *dheg- are all possible root shapes but **tegh-, **dhek-, and **deg- aren't; the fact that the two consonants can't have the same place of articulation; the fact that CeRT verb roots can take a nasal infix but CeT roots can't, etc., etc.). --Angr (t·c) 13:07, 2 January 2006 (UTC)

Actually, I just re-discovered Wikipedia:Articles for deletion/Indo-European root word articles, where it was decided they should be merged to Proto-Indo-European roots, but no one's done it yet. I myself was one of the ones who voted for that solution. The people from Wiktionary said they didn't want them, since they're hypothetical roots, not attested words. I also like your suggestion at Talk:List of common Indo-European roots#Focus? that they could be put at Wikibooks. --Angr (t·c) 13:26, 2 January 2006 (UTC)

well, I agree. But can we just delete ker-1 through to ker-5 (and all the other k-roots in Category:Indo-European linguistics) then? Imho they should have been speedied, it is stupid to have an article entitled "ker-5", it appears these were blindly copied out of Pokorny; obviously, this root is not known as "ker-5", the five is just a random ordinal it got in Pokorny or some other dictionary. Shall we delete away, or embark on a multiple AfD for them all? dab (ᛏ) 13:38, 2 January 2006 (UTC)

There's already been a multiple AfD for them all: Wikipedia:Articles for deletion/Indo-European root word articles, and the result was that they should all be merged into Proto-Indo-European roots. It's just no one's done it yet. --Angr (t·c) 14:30, 2 January 2006 (UTC)

that's fine, I'll do it some time. But it seems exceedingly pointless to have a redirect called "ker-5" left hanging around. dab (ᛏ) 14:52, 2 January 2006 (UTC)

Well, take it to WP:RFD then. --Angr (t·c) 15:01, 2 January 2006 (UTC)

Indo-European menu item in edit tools box

There is now an "Indo-European" menu item in the edit tools box, making it possible to quickly and easily add the following characters: ḱ ǵ ʰ ʷ h₁ h₂ h₃ m̥ n̥ l̥ r̥ ē ō þ ð ƕ. (I know the last one is used for Gothic, not PIE, but it seemed silly to have a separate menu item for Gothic consisting solely of þ ƕ, so I stuck it here.) --the preceding comment is by Angr 01:17, 14 January 2006. (Added {{unicode}} --teb728 03:32, 21 May 2006 (UTC))

Where? I don't see any "Indo-European" menu item. --teb728 04:04, 21 May 2006 (UTC)

No, the whole idea of having menu items in the edit tools box got undone a few days after I posted the above message. It was a great idea, one most other Wikipedias as well as Wiktionary have, but it was confusing and frightening people here. Angr (t • c) 09:35, 21 May 2006 (UTC)

---

Confusing people? What people? Who are these wingnuts? How can it be confusing? There are plenty of far more confusing things on Wikipedia than this. Ugh. Somebody's taken the Think out of GroupThink again. --Glengordon01 19:19, 12 July 2006 (UTC)

Root articles revived

The idea of having root articles does not need to be dead. Have a look at the proposals at User:Doric Loon/PIE Roots project page and see what you think. --Doric Loon 16:30, 31 March 2006 (UTC)

Attribution

In that table listing nominal suffixes, the right side refers to "Ramat 1998", but there is no such reference in the References section. Can someone who has that book add a reference? Wtrmute 01:59, 2 May 2006 (UTC)

Geminates in PIE

Are there any geminated consonants in the Proto-Indo-European language? --84.61.54.143 17:11, 12 June 2006 (UTC)

Geminates aren't usually reconstructed for the protolanguage, no. There does seem to have been a degemination process active: the second person singular of *h₁es- "to be" ought morphologically to have been *h₁essi, but reconstruction from attested languages gives *h₁esi instead. Angr (talk) 18:50, 12 June 2006 (UTC)

• • ...but that's of course one extremely frequent word. we do reconstruct geminates where a root-final consonant meets the same suffix-initial consonant, and it is a matter of taste whether you describe these as regularly preserved, as 'morphologically protected' against de-gemination, or as restituted. But for some reason, there are indeed no KeK roots that would become geminates in zero grade. dab (ᛏ) 22:22, 12 June 2006 (UTC)
  • Gemination in IE is non-distinctive and even a signal of syllable boundary. So in *h₁essi, we see two esses and we know they are in different syllables. As dab says, there are no "*KeK roots" that could be zerograded into a geminate within a given syllable either. In fact, in words like *h₁édti "he/she/it eats", we even see sibilantization to further seperate the two stops (/'ʔed^sti/). Let's face it: IE just doesn't like gemination. --Glengordon01 19:37, 12 July 2006 (UTC)

New history section

Thanks to dab for the new history section. One thing that got deleted in the change, however, is the mention of the pivotal role of Sanskrit in the discovery of PIE. Indeed I think that the mention of Sanskrit should be expanded in this article and/or in the detail article to note that early philologists believed that the proto-language was Sanskrit and that this belief accounts for peculiarities of reconstruction like voiced aspirates. --teb728 08:02, 7 August 2006 (UTC)

the section is summarized from Indo-European studies to replace the unfocussed paragraph we had before. I do not think that the history section here should be any longer. Shorter, if anything. I do not know of any 'early philologists' who believed such a thing about Sanskrit, certainly not Jones in 1782. Such ideas, if ever held, would belong to the pre-history of the field. I suppose it is nevertheless true, although not very notable, that Sanskrit was considered rather more archaic than it acutally is in the early days, maybe this is what you mean. Also, do you mean voiceless aspirates? I don't see how the reconstruction of voiced aspirates could have a connection with the idea of Sanskrit being PIE (especially since in that case there would be no need for any reconstruction) dab (ᛏ) 08:08, 7 August 2006 (UTC)

Yes, voiceless aspirates. I believe Lehmann's Theoretical Bases has a mention of how early scholars reconstructed voiceless aspirates for the proto-language only because of Sanskrit words like phala. I am unfortunately travelling at the moment and don't have access to the work, but surely someone here has it and could add some references. I should mention that Lehmann, at least, believes that understanding the history of IE reconstruction as a whole is vitally important for IEists, and so things cannot be just cosigned to "the prehistory of the field". Even if superseded, they still provide lessons. CRCulver 08:20, 7 August 2006 (UTC)

no no, it is perfectly valid to discuss reconstructions of voiceless aspirates based on Sanskrit evidence. The prehistory comment referred to the idea "Sanskrit is PIE". Building reconstructions on Sanskrit evidence does not amount to such an assumption. There is some evidence for voiceless aspirates outside Sanskrit too, and the topic may by all means be discussed. dab (ᛏ) 08:49, 7 August 2006 (UTC)

I believe that Franz Bopp stated that Sanskrit was either PIE or very, very close to it. In any event, it's probably mentioned in Theoretical Bases. Really hurting from lack of a library here. CRCulver 08:51, 7 August 2006 (UTC)

I don’t think the history section is too long, but it may be too detailed—it certainly begins too abruptly with Bopp. (Remember that our audience includes people who know nothing about the subject.) I agree that the original paragraph was unfocused, but it had the merit of noting that the idea of PIE began with the appreciation of the similarities of Sanskrit with Greek and Latin.

It’s possible I was mistaken in thinking I had read in Theoretical Bases that the pioneers (perhaps linguists as opposed to philologists) believed the proto-language was Sanskrit as opposed to something very like it. I’m pretty sure it said at least the latter. (I have it in front of me, but I can’t find either right now.)

I was referring to voiced not voiceless aspirates. (The voiceless aspirates were dropped from the traditional reconstruction after the discovery of Hittite and so are not very notable). But (despite glottalic theory) the traditional reconstruction (or at least its transcription) still has voiced aspirates. And that is one reason I think the early overestimate of the primitiveness of Sanskrit is notable. --teb728 01:15, 8 August 2006 (UTC)

---

teb728: Indeed I think that the mention of Sanskrit should be expanded in this article and/or in the detail article to note that early philologists believed that the proto-language was Sanskrit [...]

Perhaps you misread the good Sir William Jones when he says about Sanskrit, Greek and Latin that "[...] no philologer could examine them all three, without believing them to have sprung from some common source, which, ***perhaps, no longer exists***." Here, "some common source" refers to Indo-European (or in his words, Japhetic), not Sanskrit.

Actually I was not referring to Jones but to later pioneers, some or all of whom (as I admitted in my 8 August post) might better be called linguists. --teb728 07:56, 11 August 2006 (UTC)

Suggestion: Smaller details of "IE History" may be redirected towards Indo-European studies. However on Proto-Indo-European language, we could simply try to summarize its history as accurately as possible. A kind of teaser for those that are interested in more.

teb728: "But (despite glottalic theory) the traditional reconstruction (or at least its transcription) still has voiced aspirates."

"Voiced aspirates" are simply a name now, divorced of phonetics. One might be tempted to call them plain ol' voiced stops with early voicing onset (as with French /b/, /d/ and /g/) without necessarily adopting the Glottalic Theory for the last stage of IE. (That's another story.)

Yes, Sanskrit had an impact on the modern transcription of IE but again this detail should be directed to Indo-European studies, no? Here, we're simply discussing the Indo-European language, reconstructed as of today, 3:15 PM GMT ;)

--Glengordon01 08:27, 10 August 2006 (UTC)

Think from the point of view of our audience: readers who read the PIE article to learn about the subject (rather than to watch for disimprovements). Now you and I know that in a PIE context dʰ, for example, refers to a consonant of disputed value. And our audience reader will learn a little about this uncertainty from the text of the “Consonants” subsection. I think such a reader would find it interesting also to learn where this curious reconstruction came from, but I don’t think it would ever occur to him to look in Indo-European studies for an explanation. I think that if this historical background were mentioned in the “History” subsection, it would be a better teaser to visit the detail article than the dry details that are there now. --teb728 07:56, 11 August 2006 (UTC)

the uncertainty of phonetic values is imho a red herring. the glottalic argument should be discussed on glottalic theory. I am not sure what point you are trying to make regarding voiced aspirates. The debate on voiceless aspirates is much more interesting (for being phonological, not merely phonetical), and no, the voiceless aspirates are not completely out of the picture, their PIE status remains a minority opinion. dab (ᛏ) 09:37, 11 August 2006 (UTC)

My mention of “uncertainty” referred to the beginning of the “Consonants” subsection, where a table of consonants, including bʰ dʰ ǵʰ gʰ gʷʰ, is followed the explanatory text, “The table gives the most common notation in modern publications. … Raised ʰ stands for aspiration. According to the glottalic theory … the "voiced aspirated stops" may not have been voiced.” (When you say that “the uncertainty of phonetic values is … a red herring” and should be relegated to glottalic theory, that seems to imply that this last sentence should be deleted; I trust that’s not what you mean.) The reason I mentioned this uncertainty was to point out that the article does not say anything like “"Voiced aspirates" are simply a name now, divorced of phonetics,” as Glengordon01 put it.

The point which I am trying to make about voiced aspirates is this: I think that when a reader reads the beginning of the “Consonants” subsection, the voiced aspirates will seem peculiar to him. (They certainly seemed peculiar to me when I first learned about PIE. Do the digraphs bh and dh occur in [transliteration of] IE languages outside of Indic?) I think the reader would be interested to learn that the voiced aspirate notation reflects a (possibly old-fashioned) traditional reconstruction. Perhaps the solution is something as simple as changing the first sentence to “The table gives the traditional reconstruction as reflected in the most common notation in modern publications.”

The debate over voiceless aspirates may be interesting but is not relevant here because, unlike voiced aspirates, they are not mentioned in the article. Is there a discussion of that debate on Wikipedia? (I hope the answer is not that it is here in the PIE article, and I overlooked it.) --teb728 01:25, 13 August 2006 (UTC)

---

Jones did say *perhaps*, and I do think the idea "PIE=Sanskrit" was around at the time (Jones was not the first to suggest Greek-Latin-Sanskrit relationship, the idea goes back to the early 18th c.). If Crculver finds his Bopp reference, we can mention it here, but any further detail should, I agree, be discussed in the sub-article. dab (ᛏ) 10:58, 10 August 2006 (UTC)

I’m not asking for further detail: I think there may be too much detail already. I think the first paragraph would be improved by being made a little broader and substantially less deep: perhaps like this:

The idea of an Indo-European proto-language began in the 18th century with the appreciation of the similarities of Sanskrit to Greek and Latin. Systematic study began in the early 19th century, with Franz Bopp. And by the early 1900s a traditional reconstruction of the proto-language had been developed, which in its main aspects is still accepted today. The work done in the 20th century has been cleaning up and systematization, as well as the incorporation of new language material, notably the Anatolian and Tocharian branches, unknown in the 19th century, into the Indo-European framework.

--teb728 07:56, 11 August 2006 (UTC)

---

We can't say it's "traditional" when this notation is standard and currently used. Maybe we could say something more like "The table gives the most common notation in modern publications. Due to differences of opinion on exact phonetics, it is best to regard a few of these symbols as divorced of their phonetic reality."

Right now, the Phonology section awkwardly segments things according to sound type. However sections like "Resonants" and "Semivowels" are redundant because there isn't any noteworthy controversy with those and it simply goes about repeating what we already see in the table above. Ick!

So maybe we can revamp the Phonology section so that we focus on the standard IE notation. Then we just go through all phonemes whose phonetics are currently under debate. Those that aren't an issue can be left out. Certainly, laryngeals are a matter of debate still, as well as the subsystem of stops due to the Glottalic theory, then maybe palatals because of markedness issues. Just brief summaries with links to their respective articles if possible.

Wouldn't that be more ordered and relevant to this article's focus? --Glengordon01 02:55, 13 August 2006 (UTC)

yes, the phonology section could do with a revamp, complete agreement. I see no contradiction between 'traditional' and 'current standard' though. If it wasn't current, it couldn't be 'traditional' but at best 'historic'. dab (ᛏ) 11:29, 13 August 2006 (UTC)

Hmm. Seeing the definition of "traditional" on Answers.com, I stand corrected. Funny, I perceive "traditional" only in a past perfective kinda way. Perhaps it's because my rabid liberalism makes me perceive of traditions as outdated bumpkinisms preserved only by the walking undead. --Glengordon01 12:00, 14 August 2006 (UTC)