Rasmus Bjørn has created the webpage Prehistoric loan relations: Foreign elements in the Proto-Indo-European vocabulary; the introductory text is self-explanatory:
This page allows historical linguists to compare and scrutinize proposed prehistoric lexical borrowings from the perspective of Proto-Indo-European. The first entries are all (135 in total) extracted from my master’s thesis “Foreign elements in the Proto-Indo-European vocabulary” (Bjørn 2017). Comments are encouraged at the bottom of each entry. New entries will be added, also on request.
Take this not as the conclusion, but an invitation to join the conversation.
The thesis is available for download here (PDF).
Via this LH comment by David Marjanović; thanks, DM!
The English could do with a bit of cleaning up.
(His English is, however, some orders of magnitude better than my Danish.)
The actual substance of the thesis looks fascinating. I will be interested in what the many LH experts on the matter have to say.
It’s long, seems thorough, and the subject really needs this kind of work. I’ve read the first couple of chapters and skimmed the rest, so I probably should hold back my comment for a day or two, but I want to keep the thread alive.
What I glean from the concluding chapters is that there have been borrowing in both directions, with Uralic as well as with other families, but there are no clearly identifiable adstrates (i.e. whole semantic classes with a shared identifiable origin). I’m slightly disappointed with that, But that’s not Rasmus Bjørn’s fault, it’s the Indo-Europeans who failed to provide him with the data to support it, I suppose “Old European Farmer’s Language” could turn into that, but it would take a lot more work with the substrate.
On a more critical note, I don’t think he’s using the latest accepted reconstructions for all families. I don’t know much about the current consensus on the various Caucasian families, but Uralic-Yukaghir is thoroughly dead. Or at least not relevant on the timescale of this work. Borrowings from Uralic into Yukaghir would be another matter.
Thanks for the preliminary report!
Badly edited before posting, though. I was supposed to do housework.
I’ve finally taken a look at the thesis itself. My impression from the website is unfortunately confirmed: the Uralic data are from the Uralisches etymologisches Wörterbuch, which dates from 1988 and was already outdated (by a few years) when it came out and contains many unreliable reconstructions; it’s apparently worse than Pokorny’s for IE – search this page for UEW. Nothing more recent is listed among the sources at all.
My impression from our recent discussion here is that it remains one of several possibilities – but that’s quite irrelevant, because the only cited source is from 1940. The several papers from the 21st century have evidently escaped the author’s attention, even though he cites the etymological dictionary of Yukaghir, which dates from 2006.
Oh dear. Thanks for the sad update!
Uralic-Yukaghir
I think I’m building my impression chiefly on Ante Aikio’s The Uralic-Yukaghir lexical correspondences: genetic inheritance, language contact or chance resemblance?. And by “thoroughly dead” I don’t mean that Uralic-Yukaghir is impossible. Nothing is impossible, and geography alone calls for an occasional look in that direction. But it does seem to be far beyond retrieval. And the common source would certainly be much deeper than Uralic, so it wouldn’t matter for loanwords in PIE.
And of course this post on Protouralic, where I got the link to Aikio’s paper.
Citing the UEW instead of hunting down etymological corrections made in specialist literature is not necessarily an indication of something being wrong, though. Many of the Uralic “core” vocabulary reconstructions that come up here like *kala ‘fish’, *mete ‘honey’, *mura ‘cloudberry’ have been well enough established already since the mid-1900s. The references also contain plenty enough newer works on IE-Uralic contacts, which to me is a good sign that the author working with established material, not just data-stripmining Uralic dictionaries (for which the UEW would indeed not be the proto-lexicon you want to rely on).
re Yukaghir: I’ve started indexing Nikolaeva’s book sometime back, and it’s interesting how the Uralic-Yukaghir hypothesis actually looks much better from the Y side. The documented word material is scarcer than it first looks like — yes, there are 2659 entries, but most of them not even common Yukaghir entries, but rather words attested only from a single dialect, and/or recent Russian, Yakut or Evenki loanwords. (I get the impression that further fieldwork on the Yukaghir varieties would be valuable, as long as they still are there.) Looking only at word groups that are well-attested across the various dialects, there is an at least decent proportion of Uralic parallels. But I also share the impression by Häkkinen and to a degree Aikio that they would have to be in several layers, the newest being loans…
I had an impression that Yukaghir was isolated from contact with Uralic speakers for over 3500 years.
So that would make these “newest” layer rather old.
Sorry if I gave that impression – it merely increases the probability of something being wrong.
Interesting indeed.
Hi everybody,
I’m the author behind the thesis, and I thoroughly appreciate your comments. As a young aspiring scholar, it is a certain privilege to see oneself discussed in the third person 🙂
I agree with much of the criticism presented here, especially on the relationship between Uralic and Yukaghir, and trust that you do not discredit the work as a whole from this. I am well aware of the problems with Redéi and I use him primarily like Pokorny; a common reference point for the items used. I sincerely urge that you use the comment section on the blog to help me update faulty data. The site is as much all of yours as it is mine (oh, how I would have loved to know all there is to know about all these languages – hopefully, one day!), and the blog is merely an ordered way to collect and present all the data floating around.
So comments are more than welcome (if word specific preferably on http://loanwords.prehistoricmap.com/).
Best regards, and thanks again!
Rasmus G. Bjørn
Thanks very much for joining the conversation, and I hope you get some useful comments!
So this is a good place to post this attempt to explain PIE accent & ablaut not only together – which has often been tried, always with very unconvincing details until now – but in an Indo-Uralic perspective. As far as I’m qualified to tell, it all makes sense now.
Interesting.
Yup! And I should add that the compensatory lengthening before “second-syllable *i” may make more phonetic sense than it may seem. According to most reconstructions, Proto-Uralic only distinguished two vowel (archi)phonemes in non-first syllables, an open one (*a/*ä depending on vowel harmony) and a more closed one (variously notated as *i, *e or *ə; all reconstructions agree there was no *ə in first syllables).
This has happened again in German verbs. Today, there are two infinitive endings: -en, and the -n in -eln, -ern and tun. Go back a thousand years or up two thousand meters, and it becomes evident that the -e- used to belong to the stem, representing a merger of several distinct vowels, and the infinitive ending used to be -n alone.
@David Marjanović: I have always wondered though, why the vast majority of verb stems in German apparently ended in unstressed vowels. Moreover, does that mean that the imperative form that until within living memory preserved the –e ending (thus looking like the first person singular) was actually the original infinitive (i.e. stem)?
…Well, given stress on the root syllable, that’s not surprising…?
The imperative was, usually, the original stem – not the original infinitive, which had an extra -n and still does. Except of course in Alemannic in the wide sense and generally in Dutch, where -en is [ə].
The latest on Indo-Uralic: two regular metatheses reveal a number of new potential cognates, or regularize previously proposed vague lookalikes.