Talk:Grammatischer Wechsel

merge?
why keep them separate? dab (&#5839;) 18:36, 23 January 2006 (UTC)


 * Well, it certainly is handy to have a separate focus on the synchronic element, but I wouldn't mind them being in one article if this was a separate section at the bottom. But the Verner's law article, despite a couple of recent improvements, strikes me as a little chaotic.  This merger would have to be done thoughtfully.  --Doric Loon 20:33, 28 January 2006 (UTC)

I'm removing the merge tag as four months on no-one has done anything about it. Perhaps later, when the Verner's Law article has been improved, this can be raised again. --Doric Loon 17:35, 5 May 2006 (UTC)

Dutch: was/waren
Should it be included that the s/r alteration is also visible in the Modern Dutch copula, i.e. not just in the Modern English one? Chrispcam 10:19, 1 December 2006 (UTC)

Table
I've been playing around with the table. It was difficult before to see the developments easily, because you had to keep skipping columns to follow the same phoneme through its history. I think this is better, but the information is so complex that it is difficult to get it just right. The problem now is to make it easy for the eye to see how the rows are grouped in two. I've tried using blank rows to create double lines, but this is not ideal. Does anyone with more experience with tables have a better idea on the aesthetics? Colour? --Doric Loon 23:25, 1 December 2006 (UTC)


 * I've done something to the aesthetic end by marking the PIE column as key information. That's not especially necessary, though, as the association between the PIE consonant and its two descendants is clear enough from the row spanning. ᛭ LokiClock (talk) 17:23, 15 October 2010 (UTC)

No examples with original *p?
I'd say the example word that's given is actually pretty good if you use Dutch instead:

heffen (inf), hief, hieven, geheven

It clearly shows the alternation with ff/v. The article is wrong to suggest that the voiced alternate would be b, because as far as I know intervocalic -b- becomes -v- everywhere in Dutch (i.e. zeven, drijven rather than German sieben, treiben). Geminated f is a result of an original following -j-, but otherwise this example demonstrates it quite well. So, use or not? CodeCat 00:30, 13 November 2007 (UTC)


 * I would say be careful. The article already mentions (last sentence) that Dutch can give the impression of GM when in fact it is only reflexing terminal devoicing.  In the case of heffen, that would not explain -ff- in the infinitive, but it does suggest that this can be complex.  See also High German consonant shift where medial v>b is discussed as one of the related shifts, and mention is made of the problem of levelling in strong verbs obscuring the sound shifts. I would feel more comfortable if you could find a historical grammar of Dutch which specifically calls this particular example GM.  Then, fine.  --Doric Loon 10:00, 13 November 2007 (UTC)


 * I see what you mean. Assuming that heffen comes from Proto-Germanic *hafjan hebban, *hafjan is an anomalous verb in its own right. It has the -jan ending characteristic of Class I weak verbs yet shows a strong (ablauting) preterite in both old and modern languages. The conjugations in attested languages are as follows:


 * OHG: heffen huob huobum gihaban
 * OE: hebban hōf hōfon hafen
 * Gothic: hafjan hōf hōfun hafans
 * This gives PG: *hafjan *hōf *hōfum *hafans


 * The -j- would have caused gemination of the consonant to -ff- through West Germanic gemination, and followed by umlaut this would give *heffjan, with later loss of the -j-. OHG shows a regular voicing of non-geminated /f/ > /b/ word-medially in the non-present forms, and Dutch shows the same though with /v/ as a result. However, Old English 'hebban' led to modern English 'heave', so what I'd like to know is why OE has -bb- < -ff- there in the first place and what caused it to be lost again. I'm also not sure whether hebban belongs in the article at all for this very reason. CodeCat (talk) 15:37, 29 November 2007 (UTC)

Well, if you are right about the PG reconstructed forms, then this is not GW, and the verb does not belong here - quite correct. You haven't given a source, but I take it you have those reconstructed forms from a standard text book? GW presupposes that the consonants were already distinct with Verner's Law, i.e. the alternation should be visible in PG. If you can definitely confirm that that is not the case here, please delete the example. Incidentally, the answer to your question about the -bb- in Old English may be found by comparing the weak verb to have in all of these languages. It also takes a -bb- in the Old English infinitive, and loses it again in modern English. --Doric Loon (talk) 22:12, 29 November 2007 (UTC)

Revision of table by Codecat
I see the table has just been greatly expanded. My first impression is that this is very impressivley detailed, and it has cost a lot of work, but on reflection, it contains far too much information. The purpose of the table is to show PAIRS of consonants which then alternate. We have other articles on Grimm's Law and Verner's Law, which give the whole story. This is a specific article on the phenomenon of consonant alteration in verbs, and its purpose is purely to explain that. Now we have a table which gives FOUR consonant variants for each PIE radical, two of which are irrelevant for the purpose, and the reader cannot easily see which two it is that alternates. I am hesitant simply to revert, but I will do so if we don't get a consensus here. --Doric Loon (talk) 14:17, 16 October 2010 (UTC)
 * I did it to be able to explain the alternation seen in Old English hebban. But now that I think about it, it's probably the only example, as GW only affects strong verbs, which as a rule don't have a -j- suffix in Germanic. So maybe I went a bit overboard on that one. I've removed the extra table rows again, but left the other changes as they were. CodeCat (talk) 21:26, 16 October 2010 (UTC)

Be careful not to confuse Grammatischer Wechsel (affecting strong verbs) with Germanic spirant law (affecting weak verbs) - they look similar. --Doric Loon (talk) 08:08, 18 October 2010 (UTC)

*f ~ *b in Modern Dutch?
In the current article I read the statement: "*f ~ *b (no examples in the modern languages)[citation needed] Old English: hebban – hōf hōfon hafen ("to lift", cf. heave)".

But what about Dutch: hebben (to have), ik heb (I have), jij hebt ((sing.) you have), hij/zij/het heeft (he/she/it has), u heeft ((honorific (sing. + plur.)) you have)?Redav (talk) 21:00, 6 July 2018 (UTC)


 * Hi @Redav, that is terminal devoicing and developed in relatively recent history, not a remnant of Grimm-Werner. We're talking historical linguistics here, not typological comparisons, so the fact that this might look similar in the modern language is not the point. Doric Loon (talk) 14:11, 15 June 2023 (UTC)

r–s in German
Like Dutch, modern German also has this alternation in (aus)erkiesen–(aus)erkoren. However, the infinitive is rather rare. Albrecht J. (talk) 13:11, 15 June 2023 (UTC)


 * Hi @Albrecht J., if you can find a German who actually uses kiesen - kor - gekoren (with any prefix you like) as inflected forms of a single verb today, I will take my hat off to you and we can put that person's dialect into the article. I am not aware of it in any form of German since the courtly novels. Doric Loon (talk) 14:15, 15 June 2023 (UTC)