Talk:Proto-Indo-European language

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search
Former good article nomineeProto-Indo-European language was a Language and literature good articles nominee, but did not meet the good article criteria at the time. There are suggestions below for improving the article. Once these issues have been addressed, the article can be renominated. Editors may also seek a reassessment of the decision if they believe there was a mistake.
January 9, 2007Good article nomineeNot listed
WikiProject Languages (Rated C-class, Top-importance)
WikiProject iconThis article is within the scope of WikiProject Languages, a collaborative effort to improve the coverage of languages on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 Top  This article has been rated as Top-importance on the project's importance scale.


Isn't a "desinence" always an ending, suffix, or terminator?

What Context is PIE surrounded by?[edit]

I would find it very helpful if there were clear links in the lead to the other major language "groups", with a quick overview of what the differences are. Hard to know what something is without knowing what it is not. Boundaries give form. Tx.. (talk) 19:18, 23 April 2019 (UTC) §

I also would like more context, especially dates of the developmental stages of Proto-Indo-European. Of course, dates would have to be estimates, but especially when stages are documented by written examples of languages, they would give perspective to the linguistic processes under discussion. Dates and locations of various linguistic groups, as well as information about migrations/relocations of peoples would be very helpful. I am surprised that all this linguistic evolution seems to have happened since about 4500 BC. Janice Vian, Ph.D. (talk) 07:19, 12 January 2020 (UTC)

How plausible is all this?[edit]

I have read one or two articles that claim that the reconstruction of Proto-Indo-European, as presently conceived, is pure fantasy, and most likely bears no accurate relationship to an actual common ancestor of the relevant languages. On the other hand, this article seems to present the reconstructions pretty much as a fact accepted by all. It would be useful for the article to mention the extent to which there is agreement or disagreement on the main body of the reconstruction, and to mention the dissenting voices, if only to dismiss them. Otherwise, it is hard for the non-specialist to choose between the different opinions. 2A00:23C5:4B91:AB00:ECD4:C5AB:54B3:691E (talk) 02:34, 19 November 2019 (UTC)

I don't know a modern linguist that rejects PIE as "pure fantasy". Some reconstructions are more secured than others: the "wheel" or "sky/god" are attested in almost every branch; while the word for "wine" is a debated issue. One misconception about linguistics is that it reconstructs words based on similar sounds. This is wrong. They are based on a common grammar (a series of grammatical cases in Indo-European), which makes it nearly impossible for the similarities to be due to chance or loans. Then we use sound laws to reconstruct the most probable form in the proto-language. This method has proven predictive: we had anticipated the existence of laryngeals before we discovered the Hittite language; and we found inscriptions in Celtic or Germanic that accurately fits in the proto-Celtic and proto-Germanic forms we had reconstructed. Now, linguists do not agree on everything, as it is the case in any field: but virtually all of them accept the Indo-European languages as a family, with a common ancestor for which we can reconstruct words. On the other hand, I know other scholars (like archeologists) who may debate the very existence of PIE. From what I read, they make basic mistakes in linguistics, like comparing sounds without looking at grammar/cases. We cannot therefore give voice to critics outside of the field of linguistics on this Wikipedia page. It would be like linguists debating the theory of relativity among citations of astrophysicists. Now for the accuracy of reconstructions, I would say that PIE words have not change a lot in 200 years of historical linguistics. The modifications are only on details that don't affect the general sound-form. It would be like linguists debating if the accurate reconstruction of the English word "dwelling" is "dyelling", "duelling" or "dwelling". Azerty82 (talk) 08:26, 19 November 2019 (UTC)
Which articles qualify PIE reconstructions as "pure fantasy"? Joshua Jonathan -Let's talk! 08:50, 19 November 2019 (UTC)
Jean-Paul Demoule's 2014 book has caused a controversy and debate in France (1[1], 2[2], etc.) But he's an archeologist and his book contains numerous linguistics mistakes. That said, his arguments that we cannot necessary equal a language with a people is legitimate, although the same could be said for archeology: they cannot equal a material culture with a people. Azerty82 (talk) 10:08, 19 November 2019 (UTC)
(OP) I believe that a statement in the introduction to the effect that the present reconstruction of PIE is broadly accepted by modern linguists would be beneficial, if this is indeed the case. This may seem obvious to people who already know about the subject, or may seem to be implied by the existence of the article at all, but it is in fact non-obvious to lay people who may have read sceptical opinions and may themselves instinctively feel that such a reconstruction, in any sort of detail, is far-fetched. 2A00:23C5:4B91:AB00:ACF3:885E:2E34:5B56 (talk) 00:17, 20 November 2019 (UTC)


  1. ^ Garnier, Romain; Demoule, Jean-Paul. "« L'indo-européen a-t-il existé ? »". Le grand débat, published in La Recherche n°497 (mars 2015), 83-85.
  2. ^ Pellard, Thomas; Sagart, Laurent; Jacques, Guillaume (2018). "L'indo-européen n'est pas un mythe". Bulletin de la Société de Linguistique de Paris. 113 (1): 79–102. doi:10.2143/BSL.113.1.3285465.

Lineage Chart[edit]

Vishnu Sahib added a chart that shows the Prakrit and modern Indo-Aryan languages of India descended from Sanskrit. My recollection is that this is not accurate according to current research. Can you provide sources for this? There are two sources in the intro to the article that suggest it is accurate, but one is not definitive (saying 'most' modern Aryan languages are descended from Sanskrit) and the other is from a publication on a topic only indirectly related to linguistics. I've also started a discussion on the Commons page for the chart itself here. --Spasemunki (talk) 23:00, 14 December 2019 (UTC)


KIENGIR and ExperiencedArticleFixer disagree about the wording of the lead.

  • Old verson: Proto-Indo-European (PIE) is the linguistic reconstruction of the ancient common ancestor of the Indo-European languages
  • ExperiencedArticleFixer's version: Proto-Indo-European (PIE) is the reconstructed ancient common ancestor of the Indo-European languages.

I suggest as concise alternative:

  • Proto-Indo-European (PIE) is the reconstructed common ancestor of the Indo-European languages.

Drop "ancient" (which sounds weasel-ish), drop "linguistic" (which is redundant, per page title), PIE is reconstructed. Of course, PIE is also a reconstruction, since this word can describe both the process and its result. But "reconstructed common ancestor" sounds better and is the common wording in textbooks about historical linguistics. –Austronesier (talk) 09:09, 16 July 2020 (UTC)

I think the intro should make it clearer that PIE is theoretical, and clarify the relationship of the reconstructed elements. I propose:

Proto-Indo-European is the theorized common ancestor of the the Indo-European language family. No direct record of Proto-Indo-European exist. Its proposed features have been derived by linguistic reconstruction from documented Indo-European languages.

--Spasemunki (talk) 10:04, 16 July 2020 (UTC)

Concur with Spasemunki.(KIENGIR (talk) 15:59, 16 July 2020 (UTC))
"No direct record of Proto-Indo-European exist. Its proposed features have been derived by linguistic reconstruction from documented Indo-European languages" is nice and terse, and pretty much the same as stated in the last phrases of the second paragraph, just less flowery. I would swap the order of the two statements, putting "no direct record of Proto-Indo-European exists" at the end. And of course trim the second para to avoid repetition. –Austronesier (talk) 17:35, 16 July 2020 (UTC)
Go on, and we'll see how it looks.(KIENGIR (talk) 04:07, 17 July 2020 (UTC))
Good. My whole point was logical: the lede was conflating a language with the reconstruction of a language, which are two different things. I think it looks better now. --ExperiencedArticleFixer (talk) 08:37, 17 July 2020 (UTC)
But PIE is the language, not the reconstruction. The reconstruction is of PIE. PIE would have existed with or without our attempts to reconstruct it. Rua (mew) 09:43, 17 July 2020 (UTC)
My point, precisely. --ExperiencedArticleFixer (talk) 09:48, 17 July 2020 (UTC)
The existence of PIE is a theory proposed to explain the relationship of the IE languages. We can't describe the existence of PIE independent of the reconstructions as a fact. The reconstructions argue for the plausibility of it existing, but aren't sufficient to establish that it did. --Spasemunki (talk) 10:12, 17 July 2020 (UTC)
My point is not about whether we can talk of PIE has having existed or not. That is irrelevant. Even if it did not exist, when the article says "PIE is estimated to have been spoken as a single language from 4500 BC to 2500 BC", it is talking about the language (even if it never existed), not about the reconstruction (which we know exists). My point is categorial and logical. The fact that the language, even if fictional, is conceptually different from its reconstruction should be obvious by the fact that the language is believed to have existed thousands of years ago, while everybody knows that the reconstruction is only modern. --ExperiencedArticleFixer (talk) 11:15, 17 July 2020 (UTC)
Indeed, Spasemuki has right, despite what you say, no contradiction, since an unverified hyphothesis obviously may be interpolated by a modern reconstruction (though both are hyphothesis')(KIENGIR (talk) 21:24, 17 July 2020 (UTC))
Indeed, what Spasemunki said is mostly correct, but irrelevant to my point. Please explain the meaning of "an unverified hyphothesis [sic] obviously may be interpolated by a modern reconstruction". It does not make any sense to me, not to mention the misspellings. --ExperiencedArticleFixer (talk) 23:20, 17 July 2020 (UTC)
I was replying to Rua referring to the existence of the language. Tried to use the indent level to make that clear, but Talk isn't great for keeping threads straight. I think everyone is in agreement on the wording of the intro sentences. --Spasemunki (talk) 23:30, 17 July 2020 (UTC)
Yes, I'm happy with it. --ExperiencedArticleFixer (talk) 23:41, 17 July 2020 (UTC)
Sorry for the misspelling. I wanted to point out there is not anachronism, and unverified hypothesis is naturally may be simulated by a late reconstruction, but both of them are imaginary.(KIENGIR (talk) 23:27, 17 July 2020 (UTC))
I think you fail to see my point. As many logical points, it is both extremely simple and sometimes difficult to see. Think: both the PIE hypothesis and the reconstruction are modern, so none of them can be dated to 4000 BC. Ergo, the PIE referred to in the PIE hypothesis is not a reconstruction, but the actual language that is hypothesised to have existed in 4000 BC. --ExperiencedArticleFixer (talk) 23:37, 17 July 2020 (UTC)
I understand you, I think just missing words may confuse us. You have right in your exact sentence now, but we should not forget hypothesised to have existed, so we may not refer anywhere that it have existed or needed to be existed, juts because a reconstruction is assumed (or assumed to be equal with the target of the hypothesis).(KIENGIR (talk) 00:19, 18 July 2020 (UTC))
What missing words are you talking about? Anyway, I agree with this now, except that I think that the mainstream scientific position is that it indeed existed. --ExperiencedArticleFixer (talk) 00:27, 18 July 2020 (UTC)
Like in the beginning with estimated. Well, indeed we should be enough careful not to confuse mainstream scientific positions with facts.(KIENGIR (talk) 00:34, 18 July 2020 (UTC))
Oh, but if you understood my point, you'll see that it is irrelevant whether it existed or is only estimated to have existed. Either way, the PIE hypothesis does not refer to the modern reconstruction (although, as you said, the reconstruction could be believed to be equivalent to the hypothesis' target). --ExperiencedArticleFixer (talk) 00:39, 18 July 2020 (UTC)
If the mainstream of linguistic sources says that PIE existed, then that is the majority position Wikipedia should take as well. It would be WP:OR or WP:UNDUE to do otherwise. Rua (mew) 09:04, 18 July 2020 (UTC)

The existence of PIE is not a standalone hypothesis, it's a trivial corollary of accepting that IE is a language family. It is axiomatic in historical linguistics (and not only mainstream linguists), that a language family presupposes a common ancestor. If there is no common ancestor, it is not a family, that simple. It is the essence of historical linguists that a language family is defined by shared features that derive from a common ancestral source (I am not aware of alternative defintions of "language family" outside of linguistics, and wouldn't be interested if there were). The fact that the comprative method only is able to create an approximation of how this ancestral source looked like (due to the facts of linguistic entropy), does not mean that the mere existence of such a source is in doubt. And of course, the consensus view among historical linguistics is that IE is a language family (including the corollary that PIE was a real entity at some stage in history).

Another corollary: we cannot somehow attribute a higher factuality to the existence of a language family than to the existence of the ancestral source. So if you attach the label "hypothetical" to the proto-language itself (its reconstructed shape will always be hypothetical), you must also call the decendants a "hypothetical language family". Does anybody really want this kind of wording for the IE languages? :) –Austronesier (talk) 11:23, 18 July 2020 (UTC)

Regarding this last point, now we are saying "theorized", which I hope you find to be better than "hypothetical", since a theory in the broad sense is just how we see something that is currently unobservable. --ExperiencedArticleFixer (talk) 13:51, 18 July 2020 (UTC)