Wiktionary Request pages (edit) see also: discussions
Requests for verification English add new request | history CJK add new request | history Italic add new request | history Non-English (all other languages) add new request | history Requests for verification in the form of durably-archived attestations conveying the meaning of the term in question.	Requests for deletion English add new request | history CJK add new request | history Italic add new request | history Reconstruction add new request | history Non-English (all other languages) add new request | history Requests for deletion of pages in the main and Reconstruction namespace due to policy violations; also for undeletion requests.	Requests for deletion/Others add new request | history Requests for deletion and undeletion of pages in other namespaces, such as appendices, templates and modules.	Language treatment requests add new request | history Requests for changes to Wiktionary's language treatment practices, including renames, mergers and splits.
		Requests for moves, mergers and splits add new request | history | archives Discussion of proposed moves, mergers and splits of entries or other pages.	Category and label treatment requests add new request | history Requests for changes to Wiktionary's categories or labels, including additions, deletions, renames, mergers and splits.
		Requests for cleanup add new request | history | archives Cleanup requests, questions and discussions.
{{attention}} • {{rfap}} • {{rfdate}} • {{rfquote}} • {{rfdef}} • {{rfeq}} • {{rfe}} • {{rfex}} • {{rfi}} • {{rfp}}

All Wiktionary: namespace discussions 1 2 3 4 5 - All discussion pages 1 2 3 4 5

This is a combined verification/deletion page for any reconstructed entries, i.e. those in the Reconstruction: namespace. This includes reconstructed entries in languages for which some attestation exists, such as Latin and Old English.

See the following table for other entries:

Language	For verification	For deletion
English	Wiktionary:Requests for verification/English	Wiktionary:Requests for deletion/English
Chinese/Japanese/Korean	Wiktionary:Requests for verification/CJK	Wiktionary:Requests for deletion/CJK
Italic (Latin, Romance, etc.)	Wiktionary:Requests for verification/Italic	Wiktionary:Requests for deletion/Italic
other	Wiktionary:Requests for verification/Non-English	Wiktionary:Requests for deletion/Non-English

Adding a request: To add a request for deletion, place the template {{rfd}} or {{rfd-sense}} to the questioned entry, and then make a new nomination here. The section title should be exactly the wikified entry title such as ]. The deletion of just part of a page may also be proposed here. If an entire section is being proposed for deletion, the tag {{rfd}} should be placed at the top; if only a sense is, the tag {{rfd-sense}} should be used, or the more precise {{rfd-redundant}} if it applies. In any of these cases, any editor, including non-admins, may act on the discussion.

Closing a request: A request can be closed once a month has passed after the nomination was posted, except for snowball cases. If a decision to delete or keep has not been reached due to insufficient discussion, {{look}} can be added and knowledgeable editors pinged. If there is sufficient discussion, but a decision cannot be reached because there is no consensus, the request can be closed as “no consensus”, in which case the status quo is maintained. The threshold for consensus is hinted at the ratio of 2/3 of supports to supports and opposes, but is not set in stone and other considerations than pure tallying can play a role; see the vote.

• Deleting or removing the entry or sense (if it was deleted), or de-tagging it (if it was kept). In either case, the edit summary or deletion summary should indicate what is happening.
• Adding a comment to the discussion here with either RFD-deleted or RFD-kept, indicating what action was taken.
• Striking out the discussion header.

(Note: In some cases, like moves or redirections, the disposition is more complicated than simply “RFD-deleted” or “RFD-kept”.)

Archiving a request: At least a week after a request has been closed, if no one has objected to its disposition, the request should be archived to the entry's talk page. This is usually done using the aWa gadget, which can be enabled at WT:PREFS.

If there are no non-North Germanic cognates, this should be moved to an Old Norse entry. @Knyȝt --{{victar|talk}} 23:20, 28 March 2020 (UTC)Reply

Why? @victar — Knyȝt 09:10, 29 March 2020 (UTC)Reply

@Knyȝt: Because it can be formed by dalr +‎ -ð, making it's existence in PG questionable with no other cognates. --{{victar|talk}} 17:19, 29 March 2020 (UTC)Reply

@victar: That would render a **dald, which cannot be the ancestor of the descendants listed. The PG -i- is needed for the umlaut. — Knyȝt 19:42, 29 March 2020 (UTC)Reply

@Knyȝt: Fair point, so an unattested ON *del, from *daljō + -ð, which actually fits better semantically. --{{victar|talk}} 20:12, 29 March 2020 (UTC)Reply

@Skiulinamo Could you please create some kind of parent entry for these forms so I can delete this without losing info? Thadh (talk) 15:43, 29 December 2022 (UTC)Reply

Inaccurate reconstruction and meaning. -- Gnosandes (talk) 07:50, 6 April 2020 (UTC)Reply

What makes you say it is inaccurate? 70.175.192.217 01:12, 26 September 2021 (UTC)Reply

@Ентусиастъ Is there any reason to doubt this reconstruction and meaning? This, that and the other (talk) 06:20, 23 April 2022 (UTC)Reply

@This, that and the other Per me, it should be something more like *vьsъ. I doubt that there is any need of <ś> and there definitely has to be a final /*ъ/ in order for the word to correspond properly to the Lithuanian and the BPSl ending. It's very rare for a Proto-Balto-Slavic *-as to give Proto-Slavic *-ь. An example is Proto-Balto-Slavic *-āˀjas (whence Lithuanian *-ojas, Latvian -ājs) which gave Proto-Slavic *-ajь. There is no need of <ś> in the reconstructed Proto-Balto-Slavic *wiśas too, as this would have given a Lithuanian *višas, but it's Lithuanian visas instead. Both this and the Balto-Slavic reconstructions are wrong in this regard, esp. the former. Ентусиастъ (talk) 09:19, 23 April 2022 (UTC)Reply

@Ентусиастъ: West Slavic data clearly points towards *vьśь (Polish wszystko, wszelki, Czech všechno, Slovak všetok) and I don’t see how they could derive from **vьsъ, OCS вьсь, вьсꙗ / вьсѣ (vĭsĭ, vĭsja / vĭsě) (cf. gorazd), Russian весь (vesʹ) too points at least towards the final soft yer. It comes from older *vix- by progressive palatalization. This unpalatalized *x is actually attested in Old Novgorod forms like вхоу. Derksen in Etymological Dictionary of the Slavic Inherited Lexicon also reconstructs *vьśь and explains the -śь as originally locative plural ending (*-xъ in Slavic, generalized from PBSl *-šu < PIE *-su in ruKi contexts) and Lithuanian lack of š by levelling from forms to which ruKi did not apply:

The origin of this etymon may be a Lpl. *uiṣu. In Lithuanian, the š < *ṣ may have been replaced with s when the variant -su of the Lpl. was generalized (F. Kortlandt, p.c.). Slavic generalized the ending -xъ < *-ṣu in the Lpl., which is why the pronoun has *ś < *x as a result of the progressive palatalization. In North Russian, we still find forms with x (cf. Vermeer 2000: passim).

// Silmeth ^@talk 11:26, 23 April 2022 (UTC)Reply

@Silmethule I know about the West Slavic analogues. Also cope:-) Ентусиастъ (talk) 20:01, 23 April 2022 (UTC)Reply

To be fair, how does a hard stem explain them? Asking honestly. Vininn126 (talk) 20:11, 23 April 2022 (UTC)Reply

Lemma should be renamed to *vьxъ with note (in descendants) about West Slavic *vьšъ and East and South Slavic *vьsь (third palatalization). Sławobóg (talk) 12:32, 20 September 2022 (UTC)Reply

 Done Sławobóg (talk) 17:38, 20 January 2023 (UTC)Reply

@Sławobóg Are you going to move *pěnędzь to *pěnęgь, *kъnędzь to *kъnęgъ, *otьcь to *otьkъ, etc. to make Wiktionary consistent with this change? It makes no sense to keep them if we allow vьxъ without the progressive palatization… I’d rather revert this move. // Silmeth ^@talk 18:23, 20 January 2023 (UTC)Reply

No, for these we simply just add "from earlier X", like I did with *kъnędzь. Sławobóg (talk) 18:27, 20 January 2023 (UTC)Reply

Then why not add “from earlier *vьxъ” to *vьśь, in order to keep consistency? If your argument is “because it’s not the earliest reconstructible form because Old Novgorodian doesn’t have the palatalization in this word”, then we should also move *xlěbъ and other o-stems, because Old Novgorodian did not have the final yer, its хлѣбе (xlěbe) never had the *-as > -ъ change like the rest of Slavic. So… should we also move *xlěbъ to *xlaibas or something? Current Wiktionary Proto-Slavic reconstructions are not the earliest forms (and perhaps a bit anachronistic), but at least somewhat consistent in the features they show. // Silmeth ^@talk 18:37, 20 January 2023 (UTC)Reply

Because ś didn't exist, it was made up by some linguists, meanwhile most dictionaries reconstruct this word as *vьxъ (care to see references?). vьśь was not helpful in any way. Comparing it to хлѣбе (xlěbe) is false analogy. Sources don't even mention Novgorodian (besides Derksen), it is not final argument (just supporting one), main arguments are sound changes and Baltic cognates. Third palatalization is pretty late. If we don't like -x- for some reason, we need to make separate lemma for East/South and West Slavic, and that is nonsense. Just google "vьxъ" and "vьśь" and see the results. Sławobóg (talk) 19:02, 20 January 2023 (UTC)Reply

Care to share the recordings of native late Proto-Slavic speakers proving that “ś didn't exist”? Something like *Ix → *Iç → *Iɕ which then in different branches merges either with ʃ (west) or sʲ (south, east) seems like a reasonable model of the palatalization to me (in which *ś /ç ~ ɕ/ would be a real phoneme at some point). Also, we do use the *ś notation for 2nd palatalization too, eg. in inflection of *duxъ (loc. *duśě, nom.pl. *duśi) or *muxa (*muśě) which have different reflexes in west and rest of Slavic too.

References like Boryś, WSJP, Vasmer, or Melnychuk are right to list two forms, earlier *vьxъ and later *vьs/šь as they’re etymological dictionaries of specific modern languages, where the word went through those stages. But Wiktionary isn’t for specific Slavic language/branch, and generally has been treating *ś as a separate phoneme thus far, *vьxъ as the main form is inconsistent with this.

Of course we could change all -śi, -śě, etc. resulting from 2nd regr. palatalization to -xi, -xě, etc. too – but that’d be a bigger change and, I guess, a longer discussion. // Silmeth ^@talk 20:34, 20 January 2023 (UTC)Reply

Then change it back. But before that, can you explain how PBS -as gave PS -ь? Sławobóg (talk) 20:48, 20 January 2023 (UTC)Reply

You have the development in Derksen’s: *wiš- (PBS *š due to ruKi) > *vix- > *vьxъ > *vьśь (prog. palatalization changing the consonant which in turn influences the vowel). Same thing as with *-ingaz > *-ingas > *-in/ęgъ > *-ędzь in *kъnędzь, *pěnędzь.

As I understand it, Lithuanian having s suggests the original form might have been loc. pl. *wišu (not *wišas) and that Baltic replaced *-šu with *-su (variant of the ending outside of ruKi contexts, which was generalized in Baltic, compare Slavic *vьlcěxъ < *-šu with Lithuanian vilkuose – though I don’t know what the story exactly is here, I don’t know much about Baltic) // Silmeth ^@talk 21:15, 20 January 2023 (UTC)Reply

@Useigor, ZomBear, Rua, Ivan Štambuk: pinging y’all cause you have been more active in Proto-Slavic than I have, maybe you have better input (or maybe I’m arguing under a non-issue and *vьxъ for main Wiktionary lemma is OK). // Silmeth ^@talk 21:25, 20 January 2023 (UTC)Reply

This is said to be a neuter i-stem, but such nouns have a lemma in *-i, while *-iz is reserved for non-neuters. Either the gender or the inflection is wrong. —Rua (mew) 12:36, 17 April 2020 (UTC)Reply

And none of the alleged Germanic descendants is in Wiktionary! The Finnic loan is present, though. RichardW57 (talk) 13:31, 17 April 2020 (UTC)Reply

Kluge reconstructs a z-stem as the ancestor to the OHG and ON. --{{victar|talk}} 22:21, 17 April 2020 (UTC)Reply

I think this can be deleted in favour of more recent reconstructions. Just let me make sure we don't lose and descendants or break any links first. Leasnam (talk) 18:19, 18 April 2022 (UTC)Reply

As @Dbachmann wrote in the entry in 2017, this was not really a word in Proto-Semitic, but rather a wanderwort that had spread from Arabia by the dawn of the Common Era. No serious modern lexicon of PS includes this word. —Μετάknowledge^{discuss/deeds} 04:27, 8 August 2020 (UTC)Reply

Delete. How will we avoid the lengthy cognate lists? Fay Freak (talk) 12:49, 8 August 2020 (UTC)Reply

I'm not sure. I imagine that a Proto-Arabic is the ultimate source of the wanderwort. We could therefore conceivably host everything in a separate list at جَمَل (jamal), although this would require a good explanation to make it clear that we're not talking about attested Arabic. —Μετάknowledge^{discuss/deeds} 18:54, 8 August 2020 (UTC)Reply

But the Old South Arabian cannot be from Proto-Arabic, innit? And the Ethio-Semitic forms will also be earlier borrowings from the times when the Ethio-Semitic speakers settled in Southern Arabia. Similarly Modern South Arabian, a niece-language group of Old South Arabian. Host at Reconstruction:Undetermined 😆? Fay Freak (talk) 19:59, 8 August 2020 (UTC)Reply

@Fay Freak: You make a very good point. There's also Proto-Berber *a-lɣəm, which is thought to be a very old borrowing from a Semitic source that underwent metathesis, and is apparently the source of Hausa raƙumi and various other words. Now, this is a very unorthodox solution, but what if we created a page like Appendix:Semitic wanderwort gamal (or an alternate title; I'm sure there's a better phrasing) to discuss the problem, stick in a couple references, and host the descendant list? —Μετάknowledge^{discuss/deeds} 04:29, 10 August 2020 (UTC)Reply

Hmm, is there a reason the list couldn't just be in the etymology section of one of the words (e.g. Proto-Arabic) with an appropriate qualifier, like "the ultimate origin is a Semitic wanderwort which was also the source of " ? - -sche (discuss) 06:10, 13 September 2020 (UTC)Reply

Keep: Granted, it may not have existed in PSem., but I think that it better to have a central entry and explain its existence in the reconstruction notes or etymology. Should be moved to PWS though. --{{victar|talk}} 22:57, 11 November 2020 (UTC)Reply

I've moved it to Reconstruction:Proto-West Semitic/gamal-, which at least is better than having it at PSem. @Metaknowledge, Fay Freak --{{victar|talk}} 23:34, 11 November 2020 (UTC)Reply

I'm still not sure that it can be safely reconstructed to PWestSem, and I don't see any references for that statement (besides the lazy authors who simply consider it to be PS, which we know is untenable). We know it is a wanderwort; I suppose a defensible lie is better than an indefensible one, but I was hoping for a more honest solution. Note to closer: all the incoming links still have yet to be fixed. —Μετάknowledge^{discuss/deeds} 23:48, 11 November 2020 (UTC)Reply

I'm not saying it's a solution -- I still stand by my original reasoning to keep -- but since this is only found in WSem. it belongs as a PWS entry, regardless. --{{victar|talk}} 00:22, 12 November 2020 (UTC)Reply

Germanic *aiþaz is often cited at being an early Celtic borrowing. Regardless, given how it's disputed, a PIE entry isn't warranted. @AryamanA --{{victar|talk}} 16:39, 8 November 2020 (UTC)Reply

• Delete; I remember that one has argued not without reason that this is a Celtic borrowing, and the likelihood of a Germanic-Celtic isogloss in comparison to a borrowing heavily speaks against this reconstruction. Fay Freak (talk) 01:17, 24 December 2020 (UTC)Reply

Merge to *h₁ey-, which in any case ought to mention at least Celtic *oytos. A loan is a likely possibility, and these details would be better discussed somewhere else such as on the PG and PC pages (and the latter does not even exist yet!). Note though also Greek οἶτος (oîtos) as another suggested cognate. --Tropylium (talk) 18:41, 3 February 2021 (UTC)Reply

In view of Beekes' own preferred etymology from *h₂ey- (“to give”), merging to *h₁ey- (“to go”) seems no less speculative. I suggest that we add the Greek as well as Avestan 𐬀𐬉𐬙𐬀 (aēta, “punishment”), which is another perfect match, and so move the page to *Hóytos, where the two etymologies can be elaborated. In my opinion Beekes is right that the Greek evidence strongly points to *h₂ey- as the root, and the semantics fit better (“*that which is given” > “lot, fate”, “punishment”, “oath”; it is not even clear what a nominal in *-tó- from the semantically intransitive “to go” should mean). Perhaps other authors avoid connecting the Germanic and Celtic with it because this root has not otherwise survived in those branches or is somewhat obscure. — 69.121.86.13 16:11, 30 January 2023 (UTC)Reply

• Delete. Two words don't warrant a PIE reconstruction. Highly speculative. Ghirlandajo (talk) 21:17, 7 June 2022 (UTC)Reply
• Delete as probably nonexistent in PIE. Even Kroonen said that "it is unlikely that the formation goes back to Proto-Indo-European, only to surface in two neighboring branches at the far end of the IE-speaking area. It is more probable that the word somehow arose in a shared cultural zone with similar legal traditions." I have just merged whatever salvageable from this page to the root page, so now we can delete this. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 01:36, 19 September 2022 (UTC)Reply
• Delete. I have argued for high likelihood of fictivity below regarding a Proto-Semitic gloss “oath” at *yamīn-. Fay Freak (talk) 20:18, 22 May 2023 (UTC)Reply
  • @Fay Freak You already voted above. Chuck Entz (talk) 05:16, 23 May 2023 (UTC)Reply
• Keep: I do not understand why this request has not been closed after 3 years, 4 unanimous close votes and 1 suggestion to merge with the root. So, I vote contrarian to confirm the status quo. The reconstruction can be cited from reliable sources, who admit alternatives. The question of borrowing is undecidable according to {{R:goh:EWA|eid}} (1998). PIE *oitos, as they put it, is formally acceptable for both; *h₁óytos is laryngealist cosmetics (that the online edition does usually add laryngeal notation when it is secure is not decisive). Adams & Mallory 2006 clearly state *h₁óytos and add Tocharian B aittaṅka (“directed towards”) to the comparison. Sluis/Jørgensen/Kroonen 2023 discuss the Celto-Germanic isoglosses and do not improve on the status quo but confirm that *h₁óy-to- is a possibility, to conclude: "The Bell Beaker maritime network is not the only archaeological context for which the diffusion of linguistic features between these early IE groups can be hypothesized." (The Indo-European Puzzle Revisited 2023: 207). On the other hand, the evidence of Oscan eítuns is highly unreliable and that is just not good enough to reject Greek αἶνος, ἀναίνομαιor (Peters) and Hittite ḫai- (Puhvel) . Adams and Mallory posit *h₂⸝₃ehₓ-, compare Latin omen, Hittite ḫai- (“believe, take as truth”) and note that "some would also include the Celtic and Germanic words for ‘oath’" (A&M 2006: 323). In sum, the entry is warranted but sufficient hedging will be needed. If you disagree entirely, you should be able to offer a better solution. I am looking at different options but this is not exactly an improvement, and so I cannot disagree. DurdyWendy (talk) 20:53, 8 March 2024 (UTC)Reply

Schwebeablaut isn't a root "variant", it's an environmental change. --{{victar|talk}} 00:31, 11 December 2020 (UTC)Reply

Appears to actually be attested, though in New Latin, not Vulgar Latin. Hazarasp (parlement · werkis) 02:54, 21 February 2021 (UTC)Reply

Indeed: , , . Should we interpret the New Latin use as "inherited" from Vulgar Latin? Otherwise, there are basically two homographic lemmas: a non-attested Vulgar Latin one, reconstructed from its descendants, and a re-invented one in New Latin.  --Lambiam 10:13, 22 February 2021 (UTC)Reply

@Hazarasp, Lambiam The word isn't attested before the 14th century or so (DMLBS). We can't very well say that the Medieval Romance forms derive from New Latin, where the word is in fact reborrowed/calqued from Romance. Therefore these are separate, and the Proto-Romance (Vulgar Latin) entry has to stay. Brutal Russian (talk) 23:57, 3 April 2021 (UTC)Reply

Hmm, I'm inclined to disagree; I would expect just one entry, which could explain that the word is inferred from Romance terms to have existed in Vulgar Latin but is not attested until later recoined in New Latin. By comparison, AFAIK we don't and shouldn't have separate reconstruction entries for attested terms in other languages (say, English or Ojibwe) where they can be inferred to have existed at some period before their actual attestation; we don't even have separate etymology sections for English words which existed in early modern English (Latvian, etc) and were later independently recoined in modern times. (OTOH, we still have a redundant reconstruction entry on lausa.) What do we do in comparable situations in Chinese or Hebrew? AFAICT we don't have any reconstructed Hebrew or Chinese entries, which suggests we may be handling "late (re)attested" words in those languages in mainspace only, as I would expect. - -sche (discuss) 06:58, 21 June 2021 (UTC)Reply

Just noting that this has been moved to Reconstruction:Latin/cosutura (without n) and changed to derive from *cōsō, an unattested variant of cōnsuō. On one hand, this could all be handled in the mainspace by just mentioning the hypothetical n-less forms in etymology sections without spinning off whole entries, and I think such things are handled that way in situations were an n-form is attested earlier than an n-less variant; OTOH, it's now at least less stupid than having both *consutura and consutura... - -sche (discuss) 03:31, 22 March 2022 (UTC)Reply

Ditto. --{{victar|talk}} 22:22, 31 March 2021 (UTC)Reply

Should be saved, but moved to *ǵar- —caoimhinoc (talk) 19:36, 18 August 2021 (UTC)Reply

No consensus. No further action. —Caoimhin ceallach (talk) 15:29, 1 June 2024 (UTC)Reply

Improper closing of an RDF. -- Sokkjō 00:52, 3 June 2024 (UTC)Reply

Why is it an improper closing of an RFD? "A request can be closed once a month has passed after the nomination was posted." It's been years and there's no consensus. Same with Wiktionary:Requests_for_deletion/Reconstruction#Reconstruction:Proto-Indo-European/énu. —Caoimhin ceallach (talk) 10:31, 3 June 2024 (UTC)Reply

This entry invents a completely new phoneme, that I have not seen in any sources, in order to reconstruct an entry. I know that Wiktionary gives some leeway for its own research, but a whole phoneme goes too far IMO. This is definitely something that needs to be sourced. —Rua (mew) 10:25, 6 May 2021 (UTC)Reply

The Old Saxon, Old High German, and Old Dutch forms could maybe point *krūtsi, if I'm remembering the relevant historical developments correctly. The Frisian form is apparently a borrowing from Low German; if it was native, we'd expect *crēce. The OE form should maybe be excluded; unlike the other Gmc. forms; the consonantism may point to a loan from a dialect where Latin -c- before front vowels gave /tʃ/, not /ts/. If Old English crūċ is removed and Old Frisian crioce is relegated to borrowing status, a case could be made for the page to be kept as *krūtsi. This would be a sensible adaptation of pre-Old French /ˈkrut͡se/ into the late common WGmc. phonological and morphological framework. Hazarasp (parlement · werkis) 11:53, 6 May 2021 (UTC

Could it be that Proto-Germanic and Proto-West Germanic had phonemes only occurring in loanwords? Like the voiced postalveolar fricative occurs in German only in loanwords but in most familiar items like Orange and Garage. No doubt either the Proto-West Germans were able to pronounce . You have not seen it because of restricted use then, and by reason that Indo-Europeanists like to deal with inherited terminology rather than to sully themselves with language contact. Fay Freak (talk) 12:53, 6 May 2021 (UTC)Reply

However, on second thought, the logic I employed further up is somewhat shaky; Dutch/Low German -s- isn't necessarily reconcilable with Low/High German -ts-. Additionally, the pre-OF form would be /ˈkrot͡se/; the note at kryds about the vocalism in /uː/ (> later /yː/) pointing to a late date of borrowing is spot-on. If the borrowing was late, it's not necessary to posit a common WGmc. source; separate borrowings in each WGmc. language could've easily resulted in similar phonological forms. In short, the WGmc. "cross" words seem to be separate, but interlinked borrowings from after the common WGmc. period (though this isn't watertight). However, the idea that some of the Germanic forms are borrowings from others is worth considering (Old High German → Old Saxon?). Hazarasp (parlement · werkis) 12:59, 6 May 2021 (UTC)Reply

WG would have most certainly come into contact with Latin speakers that began to palatalize velars. The question in my mind isn't *if*, it's *how* we deal with these words. Using *c as a stand-in for an otherwise foreign sound is probably the easiest way to go about it. Compare also *unciju, which which has some interesting Old English variants. To quote what I wrote on the talk page, "1) even if the term entered one branch and quickly spread throughout WG, it's impossible to pinpoint the source, and 2) we're calling Frankish PWG so even if the word was adopted into Frankish and spread from there, that's still PWG yielding the word in every branch." --{{victar|talk}} 16:17, 6 May 2021 (UTC)Reply

That doesn't really address the issues that I raised (I never said anything about the palatalisation of velars being a problem!), which mainly concern the vocalism (which could be seen as indicative of a later loan) and the discrepancy in the consonantism. Positing a PWG *c to cover for the discrepancy left me a bit skeptical with only one example, but now that you've found another, I'm a bit more open to the idea. You'll need to find one or two more to really convince me, though. Hazarasp (parlement · werkis)

By the way, shouldn't *unciju be ōn-stem *uncijā? Hazarasp (parlement · werkis) 18:12, 6 May 2021 (UTC)Reply

@Hazarasp: Here's a fun one: *palancijā. As for the ō-stem, I was going by Köbler. --{{victar|talk}} 01:45, 7 May 2021 (UTC)Reply

OHG and Old Frisian both have ō-stems, but Middle Dutch and OE both have a ōn-stem. This kind of stuff makes me suspicious that they're seperate borrowings. Hazarasp (parlement · werkis) 02:39, 7 May 2021 (UTC)Reply

Meh, that's WG for you. --{{victar|talk}} 05:13, 7 May 2021 (UTC)Reply

Little details like these are important if the basis for assuming a common source is tenuous in the first place. Hazarasp (parlement · werkis) 09:14, 7 May 2021 (UTC)Reply

@Hazarasp: ō- and ōn-stems were effectively merged by late Old Dutch, so you can't base any conclusions on that. This merger likely affected other dialects in the area, as we can see that modern German has it too. —Rua (mew) 08:30, 7 May 2021 (UTC)Reply

Good to know, though I probably should've checked that; I'm not too well-versed in what happened to the Gmc. continental breakfast. That still leaves the OE form difficult to explain, though. Of course, such a problem can be sidestepped if we see them as seperate borrowings (as we probably should; at the very least, the OE form is not easily connected) Hazarasp (parlement · werkis) 09:14, 7 May 2021 (UTC)Reply

*strūcijō (“ostrich”). --{{victar|talk}} 05:52, 25 May 2021 (UTC)Reply

• In case people missed it, there is a ongoing discussion on the take page: Reconstruction talk:Proto-West Germanic/krūci. --{{victar|talk}} 16:17, 6 May 2021 (UTC)Reply

There is also a related entry *krūcigōn, derived from this noun. Leasnam (talk) 16:22, 6 May 2021 (UTC)Reply

Given that, I note that the assumption that the word is “clearly Christian vocabulary” is not without doubt. Primarily this relates to a kind of punishment, applied to slaves, and one damn well imagines that Romans made Germans acquainted with it even in the 1st century CE, so just for hyperbole and indifference argument I point out that it could be Proto-Germanic no less than postdating Proto-West Germanic. The real etymon of this verb is crucifīgere, from crucī (af)fīgere, in Medieval Latin written crucifīcāre (which is also in the TLL from some gloss, together with cruciāre). Formally, the common haplology points to proto-date, as also the parallel with German predigen, Old English predician, Old Saxon predikōn from which the North-Germanic forms like Danish prædike, Swedish predika, Icelandic prédika, Norwegian Nynorsk preika are derived, as obviously the Christians vexed man with preaching from the very beginning, “very far before the religion itself started to become a thing”. Fay Freak (talk) 12:56, 7 May 2021 (UTC)Reply

• Whatever it's worth, the EWN calls it an "old Germanic borrowing" (so likely before Old Dutch) and the NEW vaguely calls it "a word of the conversion", which could mean a quite early date in relation to the Franks. ←₰-→ Lingo ^Bingo _Dingo (talk) 18:13, 6 May 2021 (UTC)Reply

I found this paper https://scholarlypublications.universiteitleiden.nl/access/item%3A2971774/view which deals with describing and dating various sound changes of the Latin dialect that West Germanic speakers would have been in contact with. It uses Germanic borrowings as part of its data as well. This seems very useful for figuring out how old they are, based on what Latin sound changes they have carried over. In 3.23 and 3.24, three changes are discussed that would have produced an affricate /ts/:

• From Vulgar Latin /tj/: 2nd century. Reflected also in Gothic, e.g. 𐌺𐌰𐍅𐍄𐍃𐌾𐍉 (kawtsjō).
• From Vulgar Latin /kj/: not found in early Germanic loanwords, late enough for the velar to feed into the West Germanic gemination, e.g. Old Saxon wikkia (< vicia). Velars from Germanic feed this in Old French, e.g. Proto-West Germanic *makkjō > maçon.
• From Vulgar Latin /k/ before front vowel: not found in early Germanic loanwords either, e.g. Proto-West Germanic *kistu, *kaisar. Again, Germanic loans into Old French feed this, e.g. cion. Attested in inscriptions from the 5th century onwards.

Since PWG is considered to have ended around 400, this places it before the palatalisation and therefore forms like the one being discussed here are an anachronism. It is of course possible for Vulgar Latin /tj/ to end up in PWG as an affricate, but I find it unlikely that speakers would treat this as its own phoneme, since to their ears it would have sounded like a sequence /ts/ (compare how western European speakers nowadays hear Slavic c). So if we do want to denote this sound, I think a sequence *ts should be used and not *c. —Rua (mew) 08:21, 7 May 2021 (UTC)Reply

Ending PWG at the 5th century excludes Frankish, and since we merged Frankish into PWG, that date needs to be pushed forward. --{{victar|talk}} 00:43, 8 May 2021 (UTC)Reply

I'm not in favour of Wiktionary following a different standard from what is linguistically agreed upon. Wiktionary should be a linguistic source. —Rua (mew) 13:15, 8 May 2021 (UTC)Reply

Academics don't even agree that there was a single Proto-West Germanic, let alone on a date when all its descendants diverged. I feel like we keep having the same conversation about finite PWG vs. a WG continuum. If a dialect absorbs a word and it spreads through the dialects, that's still the language absorbing the word. --{{victar|talk}} 19:02, 8 May 2021 (UTC)Reply

@Victar: Seriously? It's exactly an academic linguist's freaking job to be precise and distinguish between different layers of loanwords, and the existence or non-existence of PWG doesn't make one whit of difference to this fact. --Florian Blaschke (talk) 10:48, 4 June 2025 (UTC)Reply

With an unknown etymology and only Old Norse einir as a descendant, this entry should be deleted. @Rua --{{victar|talk}} 16:05, 9 May 2021 (UTC)Reply

Delete. The Latin term is a borrowing, though from another Italic language, so you can’t well reconstruct from it but other things than inheritance become more likely for the other languages as well. The Old Norse may be back-formed from a compound with ber, Icelandic einisber, borrowed from the Latin jūniperus in the form *iēniperus we have as the Vulgar Latin form, which is already suggested by Schiller-Lübben 1875. Fay Freak (talk) 08:41, 2 February 2022 (UTC)Reply

Also note that in westrobothnian, all dialects show an initial j- (jên) which is not the expected outcome of *einir (it ought to be êin). For some reason whoever is keeping the westrobothnian wiktionary wrote it with a g but it doesn't change anything. If this word was truly native and occcured natively in old norse or proto germanic, it must have become êin. But no dialect shows this to be true. I am of the belief that it is a loanword aswell, and not even one old enough to have been nativized the same across scandinavia clearly. — This unsigned comment was added by Fay Freak (talk • contribs).

I don't know if everyone just didn't check, or you consider the Low German and modern (High) German terms to be borrowings from Old Norse. I'm guessing the former since these cognates went unmentioned, but now that I brought it up you might consider the latter. This borrowing + reanalysis + back-formation scenario you propose is an intriguing possibility, but you must admit it seems convoluted or at at the very least unnecessary. Even giving the two options equal evidential weight, there is no a priori reason to favor the Latin origin, on either phonetic or historic-cultural grounds: Junipers are abundant throughout Europe, and if anyone was selling juniper (for beermaking??) to the others it would likely have been the Scandinavians, who had more of them available. Having very niche, sparsely documented uses, both the wood and berries have had little economic value. Since that eliminates all but the formal evidence in Germanic (which, though patchy, is unproblematic on its own), and both sources cited on the entry believe in the Proto-Germanic word, we would be foolish to make such an extreme judgement on speculative grounds. — Ganjabarah (talk) 08:18, 15 February 2025 (UTC)Reply

Sorry, I see that Fay Freak did consider the West Germanic items and has elaborated. — Ganjabarah (talk) 08:25, 15 February 2025 (UTC)Reply

It seems kind of perverted to refer the idea of borrowing to the realm of speculation while not reconstruction, inherently speculative, including an Anatolian term whose meaning is not known and a Latin term whose demonstrated borrowed nature is ignored. Citing Orel for a formally and semantically shaky Middle Irish parallel (bulrush for juniper) does not make it better.

Neither borrowing nor reconstruction is a default assumption, though it is easy to observe the latter as such; in the first or second semester of linguistics they pick some word lists from New Guinea and reconstruct aught to learn methodology, which has no respect for the likelihoods and laws of reality, unfortunately some already make a living from these protolinguistics unworthy of publication: a star is a hole. Fay Freak (talk) 09:23, 15 February 2025 (UTC)Reply

(1) You don't seem to have paid attention to my reasoning. The Indo-European cognates are at best tangential to whether the Proto-Germanic term existed and utterly irrelevant to whether it deserves an entry. But let's say that we are going your route and only considering horizontal relations: I could almost as easily posit that the Latin word is a very early borrowing from Germanic, requiring equally many assumptions and odd phonetic shifts. The fact is, when you have multiple possibilities but no clearly best scenario, it is subjective to choose one over the others. What you misunderstand is that having a reconstruction entry with sufficient explanation as to its uncertainties is the default when such a reconstruction is plausible, whereas deleting or prohibiting the reconstruction because of one alternative is a clear indication of arbitrary preference. Wiktionary's job is to lay out the evidence in an unbiased way, not push for whatever theories its busiest few editors happen to enjoy at the moment. That is why we keep such reconstructions as Proto-Indo-European *kápros and *plúsis on even shakier grounds.

(2) Thanks for the entertaining side remarks.

— Ganjabarah (talk) 10:09, 15 February 2025 (UTC)Reply

@Ganjabarah: It is true that for clarity and neatness a reconstruction page with caveats can be preferrable, as e.g. I wrote warnings at كَأْس (kaʔs) and put related terms at the Proto-Semitic, which looks fair enough, if we consider the alternatives. Preferences, as a part of making knowledge-based judgements, rather than making arbitrarily life decisions—because unlike science, human life is pointless—, have reasons, such that intersubjective conviction of them is principally possible. In this case I looked into the data and came to the conclusion—though the exact origin of it be up to debate, as usually borrowings are not explicitly documented especially in past centuries—, that dialectological data, see the narrow coastal attestation area, supports borrowed origin in Low German. The High German term you just added doesn’t exist. It is a High-Germanized item found only in few etymological word list, as a represention of a Low German term, as people were unsure how to write the dialect but could presume it available knowledge, case in point Vollmann (1908) Wortkunde in der Schule auf Grundlage des Sachunterrichtes III p. 178 “Niedd. Bez sind Einbeerbaum”, apparently intending to emburden no elementary school child with academic spellings, in contrast to today’s spelling anarchy in the first education years. This your shortsight shows again the lack of training and critical thinking I argue as pervasive in academia. It is perfectly viable to become a professor of of most fields, including this one, by having a head too big for the body engaging in the lower levels of Bloom's taxonomy. Balancing, weighing, picking probabilities, let alone criticizing, as opposed to rearranging the material, is always in danger not to be well received.

On my part, the insight into كَأْس (kaʔs) also only came by getting better at a higher level related to Semitistics (i.e. ex post), while I had ignored the existing claim when reading Fraenkel, only adding word stories to Wiktionary I understand myself, whereas another author played easy by citing him in order to outsmart contemporary colleagues: after his 2022 book suddenly we see that van Putten ascended to a professor, congratulations to him, surely it is less sinful than his previous one as a videogame developer confusing our youth. Fay Freak (talk) 18:04, 15 February 2025 (UTC)Reply

No reason to assume this term survived into OE outside of the compound ynneleac which is inherited from WG *unnjalauk. --{{victar|talk}} 07:27, 23 May 2021 (UTC)Reply

There is one attestation as yna, the genitive plural of *yne (“onion”). I've added it to *unnijā. It's listed as one of the Alternative forms of *ynne. Perhaps we simply need to move it to a non-reconstructed entry (?) Leasnam (talk) 11:02, 23 May 2021 (UTC)Reply

Why is it *unnijā and not *unnjā as it is in *unnjālauk ? I would expect *ūnijā (ancestor of yne, ene) and *unnjā (ancestor of ynne). *unnijā appears un-etymological, as the latin does not have a double n, but double n in GMW-PRO results later from gemination caused by j Leasnam (talk) 11:07, 23 May 2021 (UTC)Reply

Works for me. Please move and delete. And thanks for catching my typo. --{{victar|talk}} 17:11, 23 May 2021 (UTC)Reply

@Leasnam Can you please take care of this? Thanks! -- Sokkjō 03:33, 27 February 2024 (UTC)Reply

Keep, this term, though related to ȳne, has a distinct history. Leasnam (talk) 04:45, 15 May 2025 (UTC)Reply

@DerRudymeister This is just a variant of *maganōn that might not even go back to PWG. The descendants should just both live on there. --{{victar|talk}} 21:33, 24 May 2021 (UTC)Reply

@Victar I'm not sure why you would want to create one page when both verbs and also the nouns that they derive from, have distinct descendants (either having or lacking the umlaut). I'm also not sure which one of them is the 'original' form, what makes you think that only *maganōn would be a valid reconstruction? If you want to keep *maganōną/*maganōn this creates an inconsistency in the fact that we have a PGM page for the noun *maginą, so we should change this to *maganą and list both variants there also. --DerRudymeister (talk) 17:26, 25 May 2021 (UTC)Reply

I agree with the above. Descendants of *maginōn are distinct from those of *maganōn, and I prefer to keep such variations separate as well. Leasnam (talk) 07:14, 10 March 2023 (UTC)Reply

Iffy semantics, janky root shape. --{{victar|talk}} 05:36, 31 May 2021 (UTC)Reply

• Keep. I don't see anything wrong the semantics. The root shape is rare, but compare *bʰuH-. De Vaan says "The restriction to Gm. and It., and the pervading zero grade, may cast doubts on a PIE origin; yet there is no decisive argument against it"; we can certainly add a note along those lines too, but I see no reason whatever to delete. —Mahāgaja · talk 07:25, 31 May 2021 (UTC)Reply

  All I see here is a Latin verb fruor (“to enjoy”) with several deverbal nouns, and a total left field guess of Germanic *brūkaną (“to make use of”) being related to it. The connection is worthy of mention in an etymology, at best, not a full PIE root. --{{victar|talk}} 01:16, 1 June 2021 (UTC)Reply

  @Victar: Our entry for fruor is lacking, but its meaning is actually quite close to that of uti, utor (“to use, to make use of”). L&S even glosses it this way: "to have the use and enjoyment of a thing, to have the usufruct". 2A02:2788:A4:205:4172:C92D:3A2E:3CF 13:33, 28 June 2021 (UTC)Reply

  If Ancient Greek φάρυγξ (phárunx) and Old Armenian երբուծ (erbuc) are from this root, then it is attested in two more families. --Vahag (talk) 09:43, 22 August 2021 (UTC)Reply

  I added these roots, but I could have done it wrongly. Please check. I'm minded to close this as keep unless @Victar comes back and says something. This, that and the other (talk) 07:14, 25 June 2022 (UTC)Reply

• The similiarity of Germanic and Latin words does not constitute a valid PIE lemma. Their connection with Greek and Armenian words is highly fanciful. Delete. --Ghirlandajo (talk) 21:09, 23 July 2022 (UTC)Reply

  Kölligin 2019 adduces *brūkaną < *pro- + *h₂ewg- (IF 2019: 115-136). He has also argued "From ‘throat’ to ‘self’?" (MSS 2022: 41-59), which is a bit much, but it nicely parallels Ancient Greek αὐτός (autós) < *h₂ew-, Ancient Greek φάρυγξ (phárunx, “throat”) and persona (IMHO). Point being, he does not insist on the prefix per se, discussing fruor, frux separately, cf. "While semantically convincing, the phonological side of the equation is problematic" (IF 2019: 126; emphasis mine). The required sound changes are speculative. How is that "no decisive argument against it" (de Vaan, etc.)? Anyway, I'm saying that Ancient Greek φάρυγξ (phárunx) is probably not too far away. IncrediblyBendy (talk) 14:35, 23 March 2024 (UTC)Reply

• Keep. De Vaan and Rix both note it, albeit with some qualifications which can be explained in the entry. In general, I find it profoundly weird that Wiktionary aspires to be the vanguard of etymological science and that the opinions of a couple of anonymous users with no proven credentials are supposed to override those of top specialists in the field. Sorry, folks, but it is more important for readers to know what De Vaan and Rix think than what Victar and Ghirlandajo think.--Anonymous44 (talk) 17:35, 24 January 2025 (UTC)Reply
• Keep. The fact that by some chance of history only zero-grade derivations of the root have survived does not mean the reconstruction is not valid, and neither does the fact that only Italic and Germanic descendants are attested. There are much more sparsely attested PIE roots and no one complains about them. The connection to Ancient Greek φάρυγξ (phárunx) and Old Armenian երբուծ (erbuc) is, in my opinion, highly doubtful; these should be assigned a question mark. The reconstruction however, even if based on Italic and Germanic alone, is both phonologically and semantically conclusive and plausible. --ElNuevoEinstein (talk) 07:58, 15 March 2025 (UTC)Reply

The variety of Old English forms shows that it was borrowed after the language no longer had a diphthong -au-, and also after umlaut. Otherwise it would have had the regular -īe-. In the other languages, the occurrence of stem-final -i must also point to a post-PWG borrowing, because a PWG -i would have been lost after a long syllable. This means that it cannot have been borrowed into the common ancestor. —Rua (mew) 08:37, 8 June 2021 (UTC)Reply

I absolutely agree that this is a funky burrowing, but I think it's the result of two things: 1. borrowings from Latin sometimes resulting in diphthong braking, compare *lēō ~ *lewō, so *kāul- ~ *kawul- seems a highly plausible vacillation 2. as with many plant words in WG, *-jā probably played a role in its descendants. --{{victar|talk}} 06:22, 9 June 2021 (UTC)Reply

The High German forms also have -t-, whereas from this PWG form *buzzera would be expected, giving modern *Busser. Compare *watar which does have the expected development. This means that the word must have been borrowed after the change t > z was productive in OHG, and therefore cannot be of PWG date. —Rua (mew) 14:47, 20 June 2021 (UTC)Reply

I agree. Besides, Old High German had at least 5 other native words meaning "butter", hinting at the likelihood that butira was fairly new and had not yet ousted out the other terms. Leasnam (talk) 20:26, 30 October 2021 (UTC)Reply

The following "Reconstructed Latin" entries are not actually reconstructed. Both are attested in Latin.

1) scepticus 2) absedium — This unsigned comment was added by The Nicodene (talk • contribs) at 06:16, 20 June 2021 (UTC).Reply

@The Nicodene: If they're attested, please cite them on their entries. --{{victar|talk}} 08:15, 21 June 2021 (UTC)Reply

google books:"scepticorum", google books:"sceptici" "et", google books:"scepticus" "et" turn up both capitalized and uncapitalized citations from the 16-, 17-, 18-. 19- and 2000s, although many suggest a definition more like "sceptic" (n. or adj.) than the entry's current definition "the sect of skeptics". For absedium, I find only a very few modern examples, some of which call it out as an error, like this and

• 1916, Archivio per la storia ecclesiastica dell' Umbria, volume 3, page 424:

  Parmesius Michus ad absedium (3) Saxarię mansit, / Et magnum bellum a latere dextero dedit, / Et per vim terram Saxariam indixit (4).

PS, we're missing declension information and an inflection table at obsidium. - -sche (discuss) 18:16, 21 June 2021 (UTC)Reply

This concerns an issue that comes up repeatedly: #1, #2. absedium is Medieval Latin, which is a language that postdates the emergence of Medieval Romance languages. If it also seems to underlie some Romance forms, then the reconstructed entry must stay unless the Romance formes are clearly borrowed from Medieval Latin. The problem is that no Romance descendants are given at *absedium, only two borrowings. The entry might be entirely in error and in that case should be moved and converted into mainspace Medieval Latin. Compare asedio, which leads to asediar which appears to be borrowed from Medieval Latin. In that case absedium would be a Medieval Latin borrowing from Medieval Romance such as that Spanish form. Du Cange has absedium as well as obsedium.

L&S gives Scepticī as an entry, but this is in fact erroneous since the word seems to be attested in the Greek script in the PHI corpus. It wouldn't be in TLL in any case because it's a proper name and these currently go up to the letter D. It doesn't seem to be attested in Medieval Latin either, so it should be moved to mainspace as a New Latin word. Brutal Russian (talk) 22:35, 25 June 2021 (UTC)Reply

"If it also seems to underlie some Romance forms, then the reconstructed entry must stay unless the Romance formes are clearly borrowed from Medieval Latin": I disagree, for reasons I express in the RFD of Reconstruction:Latin/consutura, above. (We don't have separate entries or etymology sections for every time a nonce is re-coined, either, or for early modern English use of foobar as a survival from Middle English vs modern revival / borrowing of it from Middle English, etc.) The one entry can explain when it was attested. For etymology sections to be able to link to the term with an asterisk, one could either use piping (*foobar) or make scepticus a redirect, but I see no reason to duplicate the content. - -sche (discuss) 17:26, 26 June 2021 (UTC)Reply

Anyway, I put citations at Citations:scepticus (and there are plenty more to be found using the searches linked above), so if someone would like to fix the definition (and add the right temporal label), that entry can be moved to mainspace. - -sche (discuss) 17:53, 26 June 2021 (UTC)Reply

I moved Reconstruction:Latin/scepticus to mainspace given that it is amply attested. - -sche (discuss) 22:11, 12 July 2021 (UTC)Reply

Duplicate of *hagustaldaz. --{{victar|talk}} 19:36, 24 August 2021 (UTC)Reply

(Notifying Rua, Wikitiki89, Benwing2, Mnemosientje, The Editor's Apprentice, Hazarasp): Thoughts on this? — Fytcha〈 T | L | C 〉 03:54, 26 January 2022 (UTC)Reply

@Fytcha I personally think hagastaldaz is probably garbage and that any terms that appear to require such a reconstruction are later re-formations of hagustaldaz, but I'm not a Germanic languages expert. Benwing2 (talk) 05:01, 26 January 2022 (UTC)Reply

Keep, at least for now, I don't see any reason for deletion. --Astova (talk) 21:32, 10 March 2022 (UTC)Reply

Delete, seems clear-cut that *hagustaldaz was the original form. — Mnemosientje (t · c) 14:55, 24 January 2023 (UTC)Reply

A homegrown reconstruction by Fay Freak, which attempts to tell a story that can unite all the Semitic words for "dove". The unfortunate fact is that they are probably not actually related; an irregular Akkadian reflex, then an intentional misuse of the (not widely accepted) sound law proposed in Al-Jallad (2015), then a strange shift in Arabic from /h/ to /ħ/ because the latter is more "lovely", then an even stranger shift in Northwest Semitic to /j/ because that phoneme is "popular"... Even the maximalists in the Semitic reconstruction game don't let their imagination run away with them as wildly as this. As I have mentioned before, protolanguage reconstructions should be a serious attempt at documenting the common ancestor of attested languages, not a playground for our hypotheses in historical linguistics that no Semitist has endorsed, despite extensive attention to this group of words. —Μετάknowledge^{discuss/deeds} 06:40, 26 August 2021 (UTC)Reply

And he qualifies of unlikely a reconstruction, *yawn-, which would perfectly work for Northwest Semitic and was posited by Kogan and Militarev in a chapter specifically on animal names. Is there a taboo about doves in Semitic cultures which would account for irregularities from a common etymon? Malku H₂n̥rés (talk) 15:42, 26 August 2021 (UTC)Reply

This is a good question. I, for my part, found the forms more likely related than not, influenced by the acquaintance with taboos about animal names. Otherwise various terms for the dove are atomized without etymology, which is itself doubtful. Fay Freak (talk) 16:01, 26 August 2021 (UTC)Reply

• The Akkadian correspondence is not irregular. The most obvious examples for the correspondence are given, but it does not look like you even cared to read.
• You should elaborate why the sound law of /ʃ/ to /ʔ/ is misused or not widely accepted. It is absurd to claim that the sound change is not widely accepted—maybe the exact formulation isn’t, but this subtracts nothing of its applicability here. /h/ to /ħ/ occurs in Arabic sometimes, as with examples referenced, this is not “strange”. Compare also *purḡūṯ-, where Arabic بُرْغُوث (burḡūṯ) is “strange”, yet true. I stress that one often has to search for sources of contamination for Semitic etymologies. For the sound change of leading /ʔ/ to /j/ there are not few examples, compare Classical Syriac ܝܗܠܐ (yahlā) from *ʔahl-, or Old Armenian յիմար (yimar) borrowed from a form which began with glottal stop. As /w/ became /j/, the very distinction of the whole group of weak consonants in initial position became less relevant in Northwest Semitic at various stages.
• It was good that imagination was running wild a bit. I do my reps wild like an animal too to get a beast body. One has think like a Biblical patriarch a bit, and sometimes one has to sit down and tell a story 🔦. Not everything that can be seen is obvious, and not all that cannot be immediately seen cannot be seen at all. But all I told has analogues, confirmation of regular occurrence! I say, you get flustered by a great array of information too fast. So I do not see any actual argument, only insults for a great work that but combines insights endorsed by Semitists. Where is the “attention” anyway? Do you call Militarev and Kogan’s proto-stage variation Proto-Semitic *yawn-at ~ *wānay- (“dove”) attentive? You see I took great care. An unusual extent of, which was needed. Fay Freak (talk) 16:01, 26 August 2021 (UTC)Reply

  When I say "irregular", I mean it as a term of art. Not that it is unpredecented — although these are rare! — but that it does not follow the regular sound correspondences established by the comparative method. As for Al-Jallad's sound law, you need only read the paper to see that it doesn't apply to this word! (I said it's not widely accepted because the thought that it's morphologically conditioned still seems to be the mainstream idea; in that case, your application would still be a misuse, of course.) When you invoke multiple irregular sound changes to unite disparate forms, you are working against Occam's razor and it increasingly appears that you are justifying these sound changes by the fact that the semantics match, rather than because the sound changes are especially plausible. You say above that it's "doubtful" that the terms could be "atomized without etymology" — as I have told you before, our null hypothesis should be that there is no relation between any two given terms, and we should force ourselves to demonstrate beyond reasonable doubt that there is indeed one. You can work out however you like, and you can post your "great work" on a personal page, but we are trying to present solid reconstructions of Proto-Semitic. —Μετάknowledge^{discuss/deeds} 16:20, 26 August 2021 (UTC)Reply

  Pretty solid reconstruction here. You are setting up a false dichotomy of regular and irregular sound changes here, an outdated 19th century view of “sound laws”. “Irregular” and “regular” aren’t terms of the art, but subjective categories. The English and German strong verbs are called irregular, but to those who reconstruct Germanic they are regular categories.

  Occam’s razor is also no law in humans, they like to do defy rules and make things complicated. I thought I disentangled it?

  Obviously I read Al-Jallad’s and did not “see” (whatever that metaphor means) that it might not be applicable here. Nor do I know what “morphologically conditioned” is—sounds pretty esoteric. It’s just a sound shift that is known to happen bar certain constraints, boom.

  I didn’t force anyone. It is still left to the readers to doubt the form. But it is also to their convenience. This is the most orderly attempt of connection of the terms, without which manaman is left confused, cognates here and there with sound changes explained here and there. I would need anyway to mention examples of sound changes and it would be really messy to do that at the individual languages, so at some point the Proto-Semitic unification developed naturally. I chose the most likely form. More likely than a null-relation too. Not biased in the beginning towards a null-relation—methodological anarchism. Fay Freak (talk) 17:49, 26 August 2021 (UTC)Reply

  Again, you are intentionally missing the crux of the issue here. You say it is "left to the readers to doubt the form", but that's not how the dictionary works! We ask them to trust that all our mainspace entries are correct, rather than ask them to doubt (and though we have mistakes, we aim to eliminate them). A reconstruction is not truth, and we have a warning template to that effect, but we still ask readers to trust that we have presented them with something that is, beyond reasonable doubt, what we say it is. This may be "the most orderly attempt of connection of the terms", but it is still incredibly unparsimonious, and that is probably why, for example, Militarev & Kogan didn't try to unite them. You know full well they weren't ignorant of these terms, but they did not consider them to be related. I find this intriguing enough to merit discussion in an etymology section (probably under an autocollapsed box to save space), but we simply can't go around claiming these tenuous connexions as indisputable Proto-Semitic. —Μετάknowledge^{discuss/deeds} 18:06, 26 August 2021 (UTC)Reply

  Militarev & Kogan didn't try to combine them because they did not have the Amorite form, they were ignorant of this form—which comes right between the other Northwest Semitic and the Arabic form, plus this was already sixteen years ago before various treatises on certain sound changes like this š → h → ʔ one. This was what made me think. Then I found Rescher connecting the Akkadian with the Arabic, knowing even less forms. Militarev and Kogan in all seriousness postulate the form as one of some “complex protoforms accounting for both type of reflexes” (SED vol. 2 page L), they did unite them! For all them even less substance was sufficient to see a family. Where was the reasonable doubt? The way people copy starred forms they saw somewhere it makes little a difference whether we just mention a starred form in a collapsible side-note in the individual language or in new page, reason looks bleak anyway. But aesthetics look served better the way it is. Of course I think whether a Proto-Semitic page exists here is merely a detail. (After people voted for Proto-Albanian pages, in which you did not weigh in by your vote … you recognize that you idealized reason a bit?) Fay Freak (talk) 18:32, 26 August 2021 (UTC)Reply

  They were indeed ignorant of the Amorite form, but it is a bit dangerous — we can only assume its meaning and its status as a genuine Amorite word, given the cultural mobility of names. It adds another line of support, but not enough to reduce the tenuous nature of the reconstruction as a whole. Militarev in particular is given to motivated reasoning, so my point was that this connexion was too much of a stretch to even make it into the SED — that is a measure of how unparsimonious it is. I didn't vote regarding Proto-Albanian because I don't know anything about it; I do know about Proto-Semitic, and that's why I am careful to ensure a consistent level of security in our reconstructions. —Μετάknowledge^{discuss/deeds} 19:02, 26 August 2021 (UTC)Reply

Delete. — Mnemosientje (t · c) 12:52, 13 January 2025 (UTC)Reply

• Szemerényi, Oswald (1967) “Славянская этимология на индоевропейском фоне”, in В. А. Меркулова, transl., Вопросы языкознания (in Russian), number 4, page 23 after other references concludes that this has been borrowed separately into the subgroups of Balto-Slavic we reconstruct as *rugís (before Slavic palatalizations obviously) and thence into Proto-Finnic *rugis and Proto-Germanic *rugiz from the Thracian word *wrugya transcribed in Greek as βρίζα (bríza), if not a related word of the Trümmersprachen nearby.
• Wikipedia about the cultivation of rye: “Domesticated rye is absent from the archaeological record until the Bronze Age of central Europe, c. 1800–1500 BCE.” Fay Freak (talk) 20:57, 8 May 2022 (UTC)Reply

Proto-Germanic. It is probable that the Germanic word for "pipe" doesn't go back to Proto-Germanic; unlike the page says, Old Norse pípa is in fact attested, but it is late and scarce. This would seem to point to a borrowing from Medieval Latin and/or Middle Low German. (It also may be worth noting that when I created *pīpaną, Proto-West Germanic didn't exist, so the only option I had was to create the gem-pro entry.) Hazarasp (parlement · werkis) 05:11, 18 May 2022 (UTC)Reply

The edit history should merged into the PWG entry. --Sokkjo (talk) 15:46, 13 February 2023 (UTC)Reply

Old English. Like *ofermorgan, this is most likely fictitious; yestermorrow is first attested as late Middle English yestermorow (in Caxton's 1481 The history of Reynard the Fox), and there's no cogent reason to think that it existed any earlier. The adduced German gestern Morgen doesn't prove anything, as it could've easily been independently formed; c.f. modern English yesterday morning). Hazarasp (parlement · werkis) 09:32, 16 June 2022 (UTC)Reply

Delete. ·~ dictátor·mundꟾ 14:21, 13 July 2022 (UTC)Reply

It is hard to find any good sources claiming these existed in Proto-Slavic, it is usually believed these were early borrowings in South Slavic and borrowed to other languages durning christianization. Etymologies explained in Żyd, Rzym, krzyż. Sławobóg (talk) 20:01, 17 September 2022 (UTC)Reply

Incidentally, there should not be a stage like "Romance *Rọ̄ma", because that is not an reconstructed form at all, and considering the time that the borrowing would have occurred, most likely we are dealing with Late Latin, 'Vulgar Latin', or however one prefers to describe it. Nicodene (talk) 22:22, 17 September 2022 (UTC)Reply

Source must have been some Romance language and it had to sound *Rọ̄ma to give Slavic *Rymъ. Sławobóg (talk) 09:48, 18 September 2022 (UTC)Reply

1) That is exactly how the word sounded in Classical Latin. 2) Approximately when do you suppose the borrowing occured? Nicodene (talk) 12:33, 18 September 2022 (UTC)Reply

Boryś suggests 6th-7th century. Most sources I saw mention some Romance etymon reconstructed as *something and we should stick to that. Sławobóg (talk) 13:58, 18 September 2022 (UTC)Reply

"*Rọ̄ma" is simply wrong, as it implies that the form is unattested. Needless to say, Roma is one of the most thoroughly attested Latin words in history.

For the sixth or seventh century CE, it is far too early to speak of distinct Romance languages. I would suggest mentioning 'borrowed in late antiquity' if that is necessary. Nicodene (talk) 22:06, 18 September 2022 (UTC)Reply

I think the best solution would be saying it came from Latin Rōma "via" Late Latin or Vulgar Latin. I suppose one could add "or an early Eastern Romance language", but that does seem early. My impression is that the transition from Latin to the Romance languages was more like a number of changes running more or less in parallel at different rates and beginning and ending at different times than a single process. That would make it very hard to say exactly when one became the other. It's probably better to be vague and equivocal. Chuck Entz (talk) 22:52, 18 September 2022 (UTC)Reply

Several sources say what they say. And this topic is about Proto-Slavic lemmas, not exact etymology of the modern words. My point is these lemmas didn't exist. Possibly all christian should be removed (*krьstъ, *krьstiti, possibly *cьrky but this requires additional discussion). Sławobóg (talk) 10:37, 19 September 2022 (UTC)Reply

Nearly all of these are entirely predictable from the lemma form and, in fact, automatically listed there. The only exceptions are *habutus, *boem, and *dire- which can be kept. I would delete the other 180 entries in Category:Latin reconstructed noun forms and Category:Latin reconstructed verb forms. Nicodene (talk) 07:37, 19 September 2022 (UTC)Reply

They're there for etymology reasons. Delete, and substitute (eg. 1st conjugation) every From ], infitive of ] with From ] to avoid the redlinks. On the Latin pages, links to declined forms (including also supine and past perfect) should not link to anything (maybe by making an argument like (eg.) |nolink=1 supported by {{la-verb}}). Catonif (talk) 10:25, 19 September 2022 (UTC)Reply

Update: Given that this RFD faced no opposition in 5 months, I've disabled links from the headwords of reconstructed Latin terms. All of the discussed entries are therefore orphans (excluding links from Romance etymologies, which are on their way to be removed). The only thing missing now is just actually deleting all the forms. Catonif (talk) 19:26, 8 February 2023 (UTC)Reply

Strewn together entry of semantically impossible words. @GabeMoore --Sokkjō (talk) 04:45, 30 September 2022 (UTC)Reply

@Skiulinamo How are these words semantically impossible? Seems like a pretty short semantic leap from a word meaning "tube" or "hollow" to go on, over thousands of years, to mean things like beehive, stomach, passage, flute, etc, not to mention of course that there's academic sources for each of them.

And I wouldn't say they're "strewn together", every word I listed as a descendant either has its own page in the academic works cited for it or is listed as a cognate in others. GabeMoore (talk) 04:54, 30 September 2022 (UTC)Reply

@GabeMoore: You think boat, beehive, hollow tube are solid cognates (**h₂ewlo- is also not a root BTW)? I'm seeing a lot of those really questionable entries you're adding, like Reconstruction:Proto-Indo-European/pulu- which has descendants which can't possibly be there and no PIE formations to show your work on any. --Sokkjō (talk) 05:04, 30 September 2022 (UTC)Reply

@Skiulinamo: I do actually, and so do the people who research this type of thing, whose work I'm citing. Very little of what I (or most people, I believe) do on Wiktionary is original research or speculation. Beehives are found in hollow spaces, i.e. tree hollows. Tubes are hollow. Boats are notably hollow to have buoyancy, although I even listed that olyi is only a possible derivative, as its etymology is not entirely clear. Still, Adams (2013) attributes it to *h₂ewlo-. If all these somewhat phonetically-similar words have the common theme of having a "hollow" something, whether they're tubes, beehives, passageways, stomachs, or flutes, it makes sense to reconstruct a common term for them, which is what numerous academics have done.

Also worth noting that alvus means visceral internal organs, the hull/hold of a ship, a hollow or cavity, and beehive, so I'd say that the semantic derivation there is already shown in Latin. Also compare English hole > hold (of a ship). GabeMoore (talk) 05:17, 30 September 2022 (UTC)Reply

@GabeMoore: The basis of the entry is Pokorny who should always be taken with a *HUGE* grain of salt, and trying to connect words together on an abstract concept like hollow is extremely tenuous, especially without any corresponding verbs. Latin alvus (“belly; womb; beehive; hull (of a ship)”), has also been connected to vulva (“womb”), which itself might be from *gʷelbʰ- (“womb”). Many of these words could be from different roots, if not substrate borrowings. --Sokkjō (talk) 18:51, 30 September 2022 (UTC)Reply

@Skiulinamo: Okay, but this is still your personal speculation. While Pokorny is pretty outdated nowadays, the fact remains that leading scholars in their respective fields such as Adams and Beekes have connected these words as a direct part of their work, which have become the go-to publication in these fields - whether or not Pokorny connected them first is irrelevant. You're saying that the entry should be deleted because their conclusions don't make sense to you, with no other support. The fact that several authors independently reconstructed this root is more than enough to earn it an entry. GabeMoore (talk) 05:57, 1 October 2022 (UTC)Reply

@GabeMoore: Firstly, one really has to pay attention to the publisher. Although Leiden is very highly looked upon, they're also a circle-jerk and you have to keep that in mind. Secondly, we're not beholden to any books or publications. In fact, as a project, we have our own reconstruction styles in PIE and other languages, and can decide if a reconstruction is worth an entry, or simply a mention in an etymology. --Sokkjō (talk) 06:30, 1 October 2022 (UTC)Reply

See Wiktionary:Votes/2013-10/Reconstructions need references. --Dan Polansky (talk) 09:19, 1 October 2022 (UTC)Reply

Right, doesn't this support what I did? My editing on Wiktionary goes off of the academic work for a subject, Wiktionary "isn't a place for novel ideas", as someone in that discussion said - Wikimedia sites are not primary sources. If academic authors are unanimous in the derivation for a word, one dissenting Wiktionary editor is not grounds to delete a page, that's contrary to the whole idea behind the citation-culture that defines Wiktionary and Wikimedia in general. GabeMoore (talk) 10:34, 1 October 2022 (UTC)Reply

@Skiulinamo: We aren't "beholden" to any particular publications, but we have to use academic sources to cite our work, which is what I've been doing. Your issue with this is your own speculation, and you haven't provided any actual sources for your stance. (Although I'm fairly sure that the guidelines are that pages should be created even if there's an academic dispute, and this dispute should be marked; irrelevant either way because there isn't any academic dispute here.) The terms I listed are pretty much unanimously attributed - and not just in Leiden - to *h₂ewlo-. GabeMoore (talk) 12:23, 1 October 2022 (UTC)Reply

"I'm fairly sure that the guidelines are that pages should be created even if there's an academic dispute": absolutely not the case. σπλήν (splḗn) comes to mind. --Sokkjō (talk) 02:05, 2 October 2022 (UTC)Reply

@GabeMoore: creating IE-proto pages is a very difficult work. You should not do it until you are much more experienced. That said, the relationship of these words as inheritance from a PIE word meaning "hollow tube-like object" is universally accepted and is not just Pokorny and Leiden. Because as Skiulinamo says HeRR is an invalid root shape, people assume *h₂ew-lo- and (for Hittite) *h₂ow-li-. --Vahag (talk) 11:54, 1 October 2022 (UTC)Reply

@Vahagn Petrosyan: I'm aware that it's quite difficult, and I usually spend nearly an hour on each PIE page, reading sources and finding cognates. I'm not just "stringing together" words I think sound similar - every single term I enter in any of my PIE pages has been attributed to its root by at least one author. This is the first time (at least since when I first started editing Wiktionary in high school) when anyone's raised an issue with my PIE pages; generally I think they've been a good addition to Wiktionary. GabeMoore (talk) 12:15, 1 October 2022 (UTC)Reply

When you're inexperienced, time does not equal quality. I'll sit on a PIE entry for weeks before publishing it sometimes and I consider myself pretty knowledgeable. Your entries make it very clear you don't understand the instructions at WT:AINE. --Sokkjō (talk) 19:01, 1 October 2022 (UTC)Reply

I've done quite a few PIE entries in the past few months and no one else has called me out for not understanding the basic instructions on how to make them. I may have created *h₂ewlo- under the wrong lemma here, and I make errors here and there, but I really don't see anything on that page that I routinely ignore. GabeMoore (talk) 21:09, 1 October 2022 (UTC)Reply

We don't have a team of people to check PIE entries, which is why we can't have inexperienced users creating entries. Reviewing your PIE entries, and I don't intend to be mean, but you clearly don't know what you're doing when you can't distinguish between a lemmatized noun and a root. You should try your hand at reconstructing Proto-Tocharian before embarking on PIE entries. --Sokkjō (talk) 00:25, 2 October 2022 (UTC)Reply

Assuming we keep this entry, *h₂éw-l̥ ~ *h₂w-én-s might be a better reconstruction. I'm still very skeptical of the inclusion of Armenian ուղի (ułi, “road; journey; passage”) and especially Hittite 𒀀𒌑𒇷𒅖 (a-ú-li-iš /⁠aulis⁠/, “a body part”), which Kloekhorst only guesses the meaning in an attempt to connect it here, while Puhvel speculates “milt, spleen”. --Sokkjō (talk) 19:01, 1 October 2022 (UTC)Reply

I found another proposed descendant of *h₂ewlós: Old Armenian աւղ-ո- (awł-o-, “ring”). I think we can securely reconstruct *h₂ewlós (“hollow object”). The formation of this word and its relationship to the etymon of Old Armenian ուղի (ułi), Proto-Slavic *ulica and the Hittite should be investigated further. Vahag (talk) 17:51, 2 October 2022 (UTC)Reply

Viredaz has connected Old Armenian աւղ (awł, “ring”) with ագ-ուցանեմ (ag-ucʻanem, “to insert into a ring”, transitive), ագ-անիմ (ag-anim, “to put on (clothes, shoes, rings)”, intransitive), and thus with the root *h₂ew- (“to put on (shoes, clothes)”) which others reconstruct as *h₃ew-. I think *h₂ew-los can be explained from that root as "hollow object into which something can be fitted". Vahag (talk) 18:20, 2 October 2022 (UTC)Reply

@Vahagn Petrosyan: "to put on (clothes)" => "hollow object into which something can be fitted" seems rather improbable. Maybe it's related to PIE *weh₂- (“to be empty, hollow”), whence Latin vānus (“hollow, devoid”). Perhaps even a chiming root with *ḱewH- (“to hollow out”), whence Latin cavus (“hollow, excavated”). 🤷 Anyway, I'll try and clean up the entry. --Sokkjō (talk) 07:46, 4 October 2022 (UTC)Reply

@Vahagn Petrosyan: do you think Proto-Slavic *ùlica (“passage, street”) could be borrowed from Old Armenian ուղի (ułi, “road, passage”) or vice-versa? Alternatively, both could be from *h₂éwl-ih₂ ~ *h₂ul-yéh₂-s, if I'm not mistaken. It seems more likely that they share a common root than their meanings being innovated twice. --Sokkjō (talk) 22:49, 7 October 2022 (UTC)Reply

@Skiulinamo: a borrowing between those two geographically far-apart groups is excluded. The relationship should be explained within the framework of inheritance from PIE. PS. *h₂ul-yéh₂-s would give Old Armenian **ուղջ (**ułǰ) Vahag (talk) 10:14, 8 October 2022 (UTC)Reply

@Vahagn Petrosyan: and what would a leveled *h₂ul-i(e)h₂ render in Old Armenian? --Sokkjō (talk) 21:33, 8 October 2022 (UTC)Reply

@Skiulinamo: it would give *ուլ (*ul). Vahag (talk) 07:05, 9 October 2022 (UTC)Reply

@Vahagn Petrosyan: մայրի (mayri) seems to support at least some cases of PIE -yeh₂- > Armenian -i. How would you derive ուղի (ułi)? --Sokkjō (talk) 07:26, 9 October 2022 (UTC)Reply

PIE *y becomes Armenian ǰ when preceded by *l, so մայրի (mayri) is not comparable. The formation of ուղի (ułi) is unexplained. The -ի (-i) is not a problem, it can be an inner-Armenian addition. The ł instead of l is a problem: people assume a formation containing PIE *n, possibly paralleling Ancient Greek αὐλών (aulṓn), but this is uncertain. Vahag (talk) 07:53, 9 October 2022 (UTC)Reply

@Vahagn Petrosyan: Forgive me, as I'm no expert in Armenian, but what I was referring to with my example was a case of, as Martirosyan refers to it, "metathesis or y-epenthesis", i.e. *h₂el-yo- > *aly > ayl ~ ayɫ. --Sokkjō (talk) 04:51, 10 October 2022 (UTC)Reply

• OK, I've rescued the entry and moved it to Reconstruction:Proto-Indo-European/h₂ewlós. --Sokkjō (talk) 23:05, 12 October 2022 (UTC)Reply

  Thanks for the great work. By the way, Pokorny's *awlós probably should not be presented as an alternative reconstruction. It is a pre-laryngeal notation of the same thing as *h₂ewlós. Vahag (talk) 07:25, 13 October 2022 (UTC)Reply

  @Vahagn Petrosyan: Hah, thanks. I beat you to *awlós. The school of "*a in PIE doesn't exist" is probably bunk and onomatopoeic roots with *a is probably a thing. --Sokkjō (talk) 07:45, 14 October 2022 (UTC)Reply

The Tocharian just requires any phoneme between the nasal and dental and the Indo-Iranian could be secondary. Both entries at Reconstruction:Proto-Indo-European/h₂en-. --Sokkjō (talk) 23:40, 3 October 2022 (UTC)Reply

• Sidenote: @GabeMoore, it's not recommended to even use {{R:ine:LIPP}} as a source; his work is that problematic. --Sokkjō (talk) 00:28, 4 October 2022 (UTC)Reply

Keep. I disagree that LIPP is problematic. I agree though that Tocharian is a poor match, better off at with on, ἀνά (aná), etc. —Caoimhin ceallach (talk) 17:37, 13 December 2023 (UTC)Reply

No surprise there. Regardless, presuming an entry at RC:Proto-Indo-European/h₂en- is enough supposition. -- Sokkjō 18:50, 13 December 2023 (UTC)Reply

No consensus. No further action. —Caoimhin ceallach (talk) 15:47, 1 June 2024 (UTC)Reply

Improper closing of an RDF. -- Sokkjō 00:53, 3 June 2024 (UTC)Reply

Proto-Germanic. This shows only one descendant fedelgold (“gold-leaf”); however the OHG word has an alternative form which is pfedelgold which would mean the PGmc term would have an intial p- not f-. Leasnam (talk) 06:35, 13 October 2022 (UTC)Reply

Extra note: this should preferably be history-merged to guerra per the discussion below. —Al-Muqanna المقنع (talk) 02:44, 29 October 2022 (UTC)Reply

This has some similarities to Reconstruction:Latin/circa above, which I don't have strong feelings about, but it strikes me that the "Keep" argument is much less tenable in this case. First, guerra/werra is one of the most frequently used novelties in Medieval Latin, attested copiously across a wide area down to the early modern era (the citations in DMLBS go up to 1545) and as a term of art in medieval law, werra ending up as the standard spelling in English Latin, guerra on the Continent. So unfortunately this series of edits by Nicodene labelling it non-standard, sporadic, and early medieval was incorrect. I've largely reverted them (and added some quotations).

But, maybe more importantly, we have the following observation by Laury Sarti, Perceiving War and the Military in Early Christian Gaul (ca. 400–700 AD) : " vocabulary, as far as it can be reconstructed, however does not contain a synonym for the Roman term bellum to refer to warfare. The word that comes closest is werra. This term only gained this specific meaning in the course of the Merovingian Age, at the earliest." (Note that Frankish *werru does not mean "war".)

This means that the term's acquisition of its dominant Latin meaning is virtually coincident with its appearance in the sources (9th century at the latest per Niermeyer). I'm generally more conversant/interested in historical work with Medieval Latin documents than the diachronic linguistics, but it seems to me, at least, to make little sense then to distinguish a reconstructed Vulgar Latin entry from the well-attested Medieval Latin one. —Al-Muqanna المقنع (talk) 12:42, 27 October 2022 (UTC)Reply

Hi Al-Muqanna. I am in favour of removing this 'reconstructed' entry and moving its contents to the attested entry. 'Ninth century at the latest' is certainly early enough for a first attestation.

I'm not convinced that guerra or werra can really be considered standard, since the literary Latin term, the one regarded as most 'correct', remained bellum throughout the Medieval Ages, to the best of my knowledge. That is, it did not stop being used, only to be revived later. But if a citation can be found that does directly support the notion of guerra/werra becoming the standard word for 'war' in literary Latin, at least temporarily, I'd have no further objection. Nicodene (talk) 12:56, 27 October 2022 (UTC)Reply

Thanks Nicodene. Bellum was in use as well, of course, but of the two sources I've added, one is a monastic chronicle and the other is a formal treaty document. We also have other learned sources like Baldus de Ubaldis, one of the most prominent late medieval jurists, ad Digest 14.2.2 " qui tempore guerrae propter defensionem vadunt " (cited here, but also just search guerra in that book). That seems fairly standard to me. —Al-Muqanna المقنع (talk) 13:07, 27 October 2022 (UTC)Reply

@Al-Muqanna I see- that is a higher degree of acceptance than I'd expected. Perhaps we can simply add a note such as 'coexisted during the Middle Ages with the Classical equivalent bellum before being eliminated during the Renaissance'. Nicodene (talk) 15:22, 27 October 2022 (UTC)Reply

I don't mind that usage note, though it might be redundant? Given the humanist reformation of Latin-writing during the Renaissance I don't think there should be any expectation that novel terms from Medieval Latin will be carried over into New Latin; at least I would (and do) treat it as "opt-in" and mark it as "Medieval Latin, New Latin" if it is. —Al-Muqanna المقنع (talk) 15:55, 27 October 2022 (UTC)Reply

Yeah, true. Nicodene (talk) 23:29, 27 October 2022 (UTC)Reply

• If this page is going to be moved to the mainspace, it needs to be moved, not deleted. --Sokkjō (talk) 01:54, 29 October 2022 (UTC)Reply

Since a page already exists at the target it cannot be moved (werra, main entry: guerra). Content-wise, thanks to Nicodene's work, everything has already been moved except the PItWRom pronunciation (which is potentially problematic given the timeframe of semantic evolution and the loan to Latin). If the problem is preserving attributions the relevant procedure would be a history merge, though there doesn't seem to be a process for those on WT and a talk page note should also be sufficient. —Al-Muqanna المقنع (talk) 02:20, 29 October 2022 (UTC)Reply

There's no process for history merges, but we certainly do them here. It's been a while since I've done one, but I know how to do them. Chuck Entz (talk) 02:33, 29 October 2022 (UTC)Reply

Nice, will add a note to the top about it in that case. —Al-Muqanna المقنع (talk) 02:44, 29 October 2022 (UTC)Reply

@Nicodene just going ahead and deleting an RFD page without giving people time to comment first is really bad form. Do you also plan on edit-waring me about it? --Sokkjō (talk) 07:27, 30 October 2022 (UTC)Reply

@Skiulinamo No, I didn't 'delete the page' (?), I moved its now-redundant content and left a link. And no, you can't 'move' the reconstruction page when the mainspace already has the corresponding entry, which you should have seen, considering that I both linked it and referred to it in the RFD comment box.

The 'reconstructed' page is inevitably going to be deleted, considering that the term is extensively attested (and early on), even in that exact spelling. If you want to stubbornly edit-war away my edit to the page, and stalk through my profile, have at it. Nicodene (talk) 12:45, 30 October 2022 (UTC)Reply

@Nicodene You neglected to read the above. We have the ability to merge revision histories. --Sokkjō (talk) 23:44, 30 October 2022 (UTC)Reply

@Sokkjō You neglected to re-read your own comment where you demanded that the page be moved. And falsely claimed I deleted it. Nicodene (talk) 23:51, 30 October 2022 (UTC)Reply

Conclude this. The Latin word is first attested 858 CE (borrowed from Romance obviously) but the descendants are Italo-Western. So, keep the reconstruction page and add a Medieval Latin descendant. Kwékwlos (talk) 11:13, 7 March 2023 (UTC)Reply

@Kwékwlos: It's not obvious at all, and all the sources I can find treat its use as continuous, not a re-borrowing: the FEW treats it as a Merovingian borrowing, Strecker's Introduction to Medieval Latin lists werra as a typical example of Medieval Latin originating in Merovingian usage (p. 31), the more recent Medieval Latin: An Introduction and Bibliographical Guide likewise lists werra as a Latin term borrowed in the Merovingian era. Seems far more likely to me that the Carolingian example from 858 is continuous with earlier Merovingian usage and not a novel re-borrowing. —Al-Muqanna المقنع (talk) 11:51, 7 March 2023 (UTC)Reply

The ninth century is quite early in any case and a written <werra> can be taken simply as the orthographic representation of the popular term, rather than a borrowing from Romance into Latin (which had only just barely begun to be distinguished). Nicodene (talk) 12:56, 7 March 2023 (UTC)Reply

Proto-Semitic. — Fenakhay ^{(حيطي · مساهماتي)} 19:01, 2 November 2022 (UTC)Reply

@Fenakhay: This reconstruction is based on non-standard meanings of أَبَى (ʔabā), which are copiously shown in Landsberg and Nöldeke to which I have updated the Arabic entry, however the latter author prefers بَغَى (baḡā) as their origin, and on my first glance the Akkadian is also related rather to this, as also our Assyriologist Profes.I. thinks, and with this alone the Indo-European relations fall away. For obscure أَبَّ (ʔabba, “to prepare etc.”) I point to theories at أُهْبَة (ʔuhba), of which root it can be assimilated. The present Proto-Semitic page should be moved to Proto-West Semitic *ʔabay-. Fay Freak (talk) 21:13, 2 November 2022 (UTC)Reply

What do people think of the proposed PIE etymology? It does have a source, but the meanings in PIE and Proto-Semitic are pretty different, even if you can posit "love" as an intermediate meaning or something. Also, it's one of only two entries in Category:Proto-Semitic terms derived from Proto-Indo-European. 98.170.164.88 22:39, 2 November 2022 (UTC)Reply

Chronologically and formally, it would make most sense (I rate the chances higher) to assume Semitic having borrowed later, perhaps from an an unattested Old Iranian descendant of the Proto-Indo European into the oldest West Semitic languages, there being other basic borrowings like 𐡆𐡌𐡍𐡀 (zmnʾ, “time”), maybe dying it out in the Iranian branch of Indo-European unlike the Indian one owing to its vulgarity. In any case it remains that it is better to move the entry to Proto-West Semitic, as I seem to have not explicitly enough suggested, @ZomBear, Fenakhay, as its presence in there is yet somewhat better evidenced for it than for Proto-Semitic, even if we decide that it is not enough either; the entry as Proto-West Semitic seems to be quite bearable for me though now, giving an etymology to various Semitic words including Ethiosemitic ones and a borrowing in the Middle Kingdom of Egypt. Fay Freak (talk) 22:51, 4 March 2023 (UTC)Reply

Proto-West Germanic *biggō (“piglet”). Old Dutch has figga, fig, and Middle Dutch vigge, vigghe, but forms with b- do not appear till almost Modern times, and may be due to conflation of the aforementioned and Middle Dutch bagge. If this is the case, perhaps the reconstruction can be moved to the Alternative form listed on the entry; but that doesn't explain the Low German and Frisian terms, they may be simply borrowings from the Dutch. 06:36, 29 October 2022 (UTC) Leasnam (talk) 03:51, 11 November 2022 (UTC)Reply

This should be either deleted or moved to *tḱóymos (see *tḱey-). Likely the only term that belongs to this root is Greek κοιμάω (koimáō). @Sausage Link of High Rule, Florian Blaschke. --Sokkjō (talk) 00:08, 3 December 2022 (UTC)Reply

Can you offer some evidence for this, or even the existence of a root *tḱóymos?  --Lambiam 18:11, 4 December 2022 (UTC)Reply

For what it's worth, Kroonen and Derksen support the derivation from *ḱey-, and likewise Matasović, while not as explicit, reconstructs an original anlaut *ḱ- rather than *tḱ-, so he doesn't seem to think of *tḱey-, either. --Florian Blaschke (talk) 00:29, 5 December 2022 (UTC)Reply

Proto-Germanic + the 2 presumed descendants (Old English and Old Norse). SoP. Also, very likely calqued between the two languages, or formed on similar model. Leasnam (talk) 18:13, 18 December 2022 (UTC)Reply

Also, the PGmc translates as "fall into the corpse/battlefield" (which doesn't make much sense). The Old Norse is "fall into the dead/slain", the Old English as "fall into the slain", which do make sense Leasnam (talk) 18:25, 18 December 2022 (UTC)Reply

• Delete all three. --Sokkjō (talk) 04:08, 19 December 2022 (UTC)Reply
• @Leasnam. Delete. —Caoimhin ceallach (talk) 14:11, 20 October 2023 (UTC)Reply

Keep all three. I've changed my view. I don't see why this is necessarily calqued, compare *tō banini werþaną and the like. —Caoimhin ceallach (talk) 16:01, 1 June 2024 (UTC)Reply

• Delete the Proto-Germanic, but keep the Old English and Old Norse phrases, as they are idiomatic. —Mahāgaja · talk 14:23, 20 October 2023 (UTC)Reply

I think you're right. I was under the impression that valr simply meant battle and val could be dative. Do a-stems always have -i in the dative in Old Norse (unlike Icelandic)? —Caoimhin ceallach (talk) 17:54, 20 October 2023 (UTC)Reply

Proto-Brythonic.

A UtherPendrogn creation. @M.Aurelius.Viator has questioned this on the grounds that it is a modern Welsh spelling and that Proto-Brythonic would have been too early for this Roman borrowing to have entered the language at that stage. I don't deal with Celtic languages myself, so I'm bringing it here. Also pinging @Mahagaja. Chuck Entz (talk) 18:44, 13 January 2023 (UTC)Reply

Schrijver, given as a citation, has the Brythonic as *lewū . —Al-Muqanna المقنع (talk) 18:47, 13 January 2023 (UTC)Reply

It looks fine to me. Considering the enormous number of Latin loanwords in Proto-Brythonic, I have no idea why anyone would think it was too early. As for the spelling, Schrijver follows different conventions than we do. He shows the final syllables as they were in Proto-Celtic, while we show the latest Proto-Brythonic form after final syllables were lost. We also show the fortition of l and r in initial position, rendering them as ll and rr (see WT:ACEL-BRY). This spelling is entirely in keeping with our conventions for Proto-Brythonic. —Mahāgaja · talk 19:07, 13 January 2023 (UTC)Reply

You (and Wiktionary as a whole) are using the wrong terminology. Proto-Brittonic would be the ancestor of Brittonic - the ancient Celtic language of Britain attested from the 4th century BC through the mid-6th century AD, after which it transitioned into Neo-Brittonic. The time period of Proto-Brittonic would be somewhere between the break-up of Proto-Celtic (in the early Iron Age) and the 4th century BC. This is way before Rome conquered Britain and British speakers borrowed this word for lion, so it makes no sense to speak of "Proto-Brittonic" *lewu:. As the word (along with most other Latin loans) was borrowed during the Roman era in Britain (1st century through early 5th century AD) t would properly be Brittonic, not Proto-Brittonic. M.Aurelius.Viator (talk) 04:34, 14 January 2023 (UTC)Reply

> We also show the fortition of l and r in initial position, rendering them as ll and rr.

This is completely wrong for Brittonic (no less Proto-Brittonic)! This sound change only occurred in medieval Welsh. M.Aurelius.Viator (talk) 04:36, 14 January 2023 (UTC)Reply

It was discussed here: Wiktionary_talk:About_Proto-Brythonic#Fortis_l_and_r. Note that the spelling "ll" in the context of Wiktionary's Proto-Brythonic transcriptions isn't supposed to be the sound of modern Welsh <ll> : rather, it's supposed to represent a "fortis" as opposed to extra-short ; likewise, and are represented as <rr> and <r>. The argument given, which seems sound, is that reconstructing the loss of final vowels to Proto-Brythonic requires the liquids to have already developed some kind of contrast, since mutation is based on the historical presence of final vowels. Unless you argue that Proto-Brythonic did still have final vowels, or that the mutation of these consonants in Welsh is an innovation introduced only by analogy to other consonant alternations, the chronology seems to require this.--Urszag (talk) 04:48, 14 January 2023 (UTC)Reply

Sorry, it's just wrong and not a single professional Celticist uses such spellings! And, once again, you guys are using completely incorrect terminology! When you say "Proto-Brythonic" you actually mean to say "Neo-Brittonic" - i.e., the new stage of the British Celtic languages that starts around the middle of the 6th century AD, when old Brittonic final syllables were now completely gone via apocope and the internal compound vowel was lost or greatly weakened via syncope. Proto-Brittonic - which is not a term that is typically used by Celticists - would be the immediate ancestor of Brittonic (1st millennium BC through the mid-6th century AD) and daughter language of Proto-Celtic (2nd millennium BC). How could so many of you here get this basic terminology so wrong?? It's so infuriating! M.Aurelius.Viator (talk) 17:53, 14 January 2023 (UTC)Reply

"The mutation of these consonants in Welsh is an innovation introduced only by analogy to other consonant alternations" may indeed be the case; David Stifter posits this exact scenario for Old Irish's "un-"lenited l, r, and n. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 17:21, 15 January 2023 (UTC)Reply

(Proto-Brittonic/Brythonic, like other proto-languages, is the common ancestor of the Brythonic languages, i.e. the phase up to the ~6th century AD; Wiktionary's terminology is perfectly accurate. It does not mean "the ancestor of the Brythonic language", singular, any more than Proto-Indo-European is the ancestor of some other language called Indo-European. —Al-Muqanna المقنع (talk) 04:56, 14 January 2023 (UTC))Reply

You don't know what you are talking about - this is not the terminology used by any professional Celticist today! The language tree for the British Celtic languages is as follows:

Proto-Indo-European > Western Indo-European > Italo-Celtic > Proto-Celtic > Common Insular Celtic > Proto-Brittonic (which not a commonly used term!) > Brittonic > Neo-Brittonic (also called Common Archaic Neo-Brittonic) > Old Welsh/Cornish/Breton/Cumbric > Middle Welsh/Cornish/Breton > Modern Welsh/Cornish/Breton. M.Aurelius.Viator (talk) 18:01, 14 January 2023 (UTC)Reply

It's pretty easy to find in recent literature actually. —Al-Muqanna المقنع (talk) 00:59, 15 January 2023 (UTC)Reply

I have been a student of Celtic historical linguists going on 40 years now - numerous well known scholars in the field have consulted with me on various linguistic problems and cited/thanked me in their papers. M.Aurelius.Viator (talk) 21:52, 15 January 2023 (UTC)Reply

@Mahagaja, Urszag are absolutely correct on all points. This entry looks fine to me. --Sokkjō (talk) 20:32, 14 January 2023 (UTC)Reply

No mainstream Indo-Europeanists or Celticists believe in "Western Indo-European" or "Italo-Celtic" as actual nodes in a family tree; they're just polyphyletic regional groupings. There is no point whatsoever in reconstructing "Proto-Brittonic", "Brittonic", and "Neo-Brittonic" as separate protolanguages since they all have the same attested descendants. The latest stage of the language ancestral to the Brythonic languages is just barely attested, which is why we do have a few Proto-Brythonic lemmas that are not reconstructions. Other scholars might prefer to use a different name for an attested language, but we have chosen not to. (Likewise for Proto-Norse, which also has a few attested terms.) —Mahāgaja · talk 20:47, 14 January 2023 (UTC)Reply

I don’t know one example of anyone using the term “Proto-Brythonic”, common terminology should be used as is stated above, Proto-Celtic > Proto-insular-Celtic > Brittonic > Neo-Brittonic. Adopting this approach would be less confusing to people researching the subject and bring Wiktionary closer with other works on the subject. The arguement of not using separate terminology because they have the same attested descendents is complete nonsense, by that logic Old, Middle and Modern Welsh should not have separate terminologies. The sheer number of phonological, morphological and grammatical changes during the time between the 1st and 6th centuries would have rendered the language unrecognisable to speakers at either end and therefore I think its completely justified using separate terminologies for the respective eras.

My opinion on transcribing Neo-Brittonic is to use spelling conventions from earliest attested Medieval Brittonic languages look at the most common shared spellings and the rest decide as a collective which is best to use, we could obviously have an IPA transcription next to the chosen orthography. Silurhys (talk) 18:49, 15 January 2023 (UTC)Reply

What the heck are you talking about? Google Scholar Google Books Back to the subreddit from whence you came! -- Sokkjō (talk) 21:23, 15 January 2023 (UTC)Reply

The term isn't commonly used - but if you even bothered to read the links you shared (or even understood what they are saying!) you would realize that they are using the term to refer to the ancestor of Brittonic (i.e., a language that pre-dates the earliest attestation of Brittonic in the 4th century BC), not Wiktionary's idiosyncratic (and idiotic) use of the term to refer to the early medieval, immediate ancestor of Welsh/Cornish/Breton - a term that should properly be Neo-Brittonic (or, if you insist on being Welsh-centric), Neo-Brythonic. This is the standard among all Celticists today. M.Aurelius.Viator (talk) 21:58, 15 January 2023 (UTC)Reply

The only two people here who have a clue what they are talking about is @M.Aurelius.Viator and @Mellohi!, you should certainly take guidance from them. They are clearly knowledgeable in up to date terminology and obviously have a much better understanding of Celtic Historcal Linguistics in general than the rest of you here. At the end of the day, Wiktionary isn’t about your ego, it’s about presenting up to date, academic information in the best way possible. It seems clear to me that there is a major lack of care, ignorance and refusal to learn here. Take the Uindorix entry for example, it has been put under Proto-Celtic when it clearly does not belong there and the answer to trying to fix the issue is essentially “stop being pedantic”. It’s not pedantic, there is nothing wrong with a high level of detail, it increases the quality of the entries and will be much better for other people who come to Wiktionary for information in the subject. It seems to me, Celtic is a complete mess on Wiktionary at the moment. Silurhys (talk) 22:32, 16 January 2023 (UTC)Reply

Mahagaja's point was strictly about reconstructing them as proto-languages. A proto-language is a reconstruction based on the comparative method. The stages of Welsh are all attested- we don't need to reconstruct them.

If you only have one set of data, how can you reconstruct multiple stages? It doesn't matter how different they had to have been, you have no way to even guess at how many there were, or their relative chronology. Chuck Entz (talk) 21:27, 15 January 2023 (UTC)Reply

We do have attestation from several periods though? We have an abundance of names attested (place, personal, tribal etc.) from different periods, we have inscriptions on gravestones, boundary markers, Christian monuments and the bath defixiones (which granted is possible it’s Gaulish but there is no evidence either way). Of course we have an idea of the scale of sound change and a relative chronology , Jackson (Language and History in Early Britain) particularly has done loads of work on chronology (and sound change) and McCone (A a relative Chronology of Ancient and Medieval Celtic Sound Change), Schrijver (Studies in British Celtic Historical Phonology) has done massive amounts on sound change, as have countless others. Silurhys (talk) 22:23, 15 January 2023 (UTC)Reply

Once again, the issue is that Wiktionary is using completely wrong terminology. Proto-Brittonic (or Proto-Brythonic, if you insist on using a Cambro-centric term) would be a daughter branch of Proto-Celtic and the immediate ancestor of Brittonic - the ancient Celtic language of Britain which is attested from the 3rd century BC and lasted until the mid-6th century AD, after which Celticists refer to its new form (post-apocope, post-syncope, et al.) as Neo-Brittonic. Wiktionary is using Proto-Brythonic where they should actually be using the term Neo-Brittonic, because all the "Proto-Brythonic" reconstructed forms they are using show apocope - something that didn't occur in Brittonic languages until at least the late Roman era and was not complete until the mid-6th century (using Kenneth Jackson's dating, which is still the standard). As Silurhys states, we have plenty of onomastic evidence and even some limited inscriptional evidence for Brittonic and for Archaic Neo-Brittonic (mid 6th through 8th century AD). The next stage in the British Celtic languages is the break up into separate daughter branches, Old Welsh, Old Cornish, Old Breton, and Cumbric, by the early 9th century AD. M.Aurelius.Viator (talk) 20:17, 16 January 2023 (UTC)Reply

By the way, I see a lot of vague argument from authority without any details- basically name dropping without the names. The entry in question has references- do you? Chuck Entz (talk) 21:39, 15 January 2023 (UTC)Reply

@Mahāgaja Do you have any background in Celtic historical linguistics?? Because you are saying absolutely ridiculous things! Italo-Celtic is absolutely recognized as a node in a family tree and your insistence that Wiktionary stick to the completely inaccurate term "Proto-Brythonic" for what is properly (Archaic) Neo-Brittonic/Brythonic (the preferred term among all Celticists today for early medieval, immediate ancestor of Welsh, Cornish, and Breton) is just bizarre. I don't even understand the motivation behind it! The term is wrong and is used by no one outside of Wiktionary to refer to Neo-Brittonic. M.Aurelius.Viator (talk) 22:06, 15 January 2023 (UTC)Reply

Dumping ground of an entry, full of unsourced reconstructions. -- Sokkjō (talk) 07:30, 19 January 2023 (UTC)Reply

what do you mean by "unsourced"? there are sources on the page --Ioe bidome (talk) 22:02, 17 February 2023 (UTC)Reply

Keep. Source or not, most of the etymologies are demonstrable with the comparative method. Saph668 (talk) 16:27, 23 November 2023 (UTC)Reply

Forged PBS word and PIE word made up on it. There and no Baltic descendants and PBS word is based on false etymology of PS *moriti. Sławobóg (talk) 13:25, 19 January 2023 (UTC)Reply

Keep. Not sure what's wrong with the etymology. Checks out at the PIE and PBS levels, though I'm not too familiar with Slavic sound laws. Claiming that it doesn't exist in Proto-Indo-European is also disregarding the Indo-Iranian descendants. -saph 🍏 13:51, 8 January 2024 (UTC)Reply

Comment. Actually, if the Slavic cognates really are false (and to be clear, I don't necessarily think they are), then the Indo-Iranian cognates would be insufficient for pIE reonstruction, as the particular verb form could have been formed in that branch alone from inherited *mer-, 'die,' + *-(o)-éye-, ',' resulting in the attested meaning 'kill' (< 'cause to die,' a meaning attested also in Sanskrit though apparently not in Slavic). The Sanskrit reflexes of *(o)-éye-, (ā)-áya- and (a)-áya- (depending on the shape of the root; refer to Brugmann's law for the details, which are irrelevant here), are highly productive, and so an independent formation is highly plausible (I assume the same is true for other Indo-Iranian languages, although my familiarity with them is somewhat limited). In fact, it is plausible even if the Slavic cognates are sound, as the suffix was apparently productive there as well (although again, a Slavicist's input on that point would be very helpful, as it is one of the IE branches I am less familar with). As for the semantic development, this is also easily repeatable; compare, e.g., sinas-, 'kill,' a word widespread in many varieties of Japonic other than standard Japanese (with essentially the same segmental shape in all of them), formed from sin-, 'die,' + -as-, ',' with exactly the same development.

My inclination is towards a Weak Keep, assuming the Slavic cognates are not proven false. In general, it seems to me that Wiktionary errs on the side of keeping reconstructions that are at least plausible, which seems fitting for a dictionary whose main purpose is reporting what scholars have posited and not deciding which of those scholars is correct. Jln Dlphk (talk) 19:06, 29 March 2025 (UTC)Reply

@Sławobóg LIV does reconstruct the PIE term, although with a question mark (I assume this means it might also be a later formation). Exarchus (talk) 10:07, 13 November 2024 (UTC)Reply

Proto-Slavic. Tagged by 2A00:23C7:9C97:8201:A977:F253:F051:826F, but not listed. 70.172.194.25 06:47, 10 February 2023 (UTC)Reply

Unusual approach for Proto-Slavic, but I see no flaw nor obvious alternative. Fay Freak (talk) 05:59, 31 December 2023 (UTC)Reply

Discussion moved from WT:RFVNE.

Both created by User:Sławobóg, who displays apparent tendencies to shove Polish etymologies (often augmented with his personal interpretations) on otherwise controversial terms:

• Old East Slavic Стрибогъ (Stribogŭ) is described (in The Tale of Prince Igor) to have association with the winds and is viewed in this way by plenty of Slavists. Phonetic evidence also points towards *-i- (including for the related verb Proto-Slavic *strybati; it's reconstructed as *stribati by Melnychuk, O. S., editor (1982–2012), “стрибати”, in Етимологічний словник української мови (in Ukrainian), Kyiv: Naukova Dumka).
• Old East Slavic Хърсъ (Xŭrsŭ) has an alleged connection with the Sun. For example, in some regions of Bulgaria, it is associated with summer solstice^{Remark: Dubious sources.}. It is viewed as a solar deity by a sufficiently large group of Slavists in order to disregard this hypothesis entirely. Also, there is not enough evidence to assert Slavic origin despite what Sławobóg claims. The theonym is attested so sporadically that there is always going to be a shadow of doubt.

Безименен (talk) 08:44, 8 March 2023 (UTC)Reply

TL;DR: User:Sławobóg explained in (unrequested) depth why the view he supports is correct. And I still don't get how this justifies ignoring all points of disagreement with past theories. Безименен (talk) 01:37, 9 March 2023 (UTC)Reply

• In *Strybogъ there is mention about that other (wind) etymology, but it is also explained why it is unlikely. I respect Melnychuk, but he is wrong on this one, it is definitely *strybati, Smoczyński reconstructs *stryti (“to flow; to roll”) and derives it from Proto-Balto-Slavic *srūˀtei (y < ū cf. *ū́ˀdrāˀ, *tūˀ, *sū́ˀnus). -t- between and in Slavic is normal (cf. *ostrъ < *aśrás). Mixation of -i- and -y- is mentioned (per Trubachov, but there are others), and it is confirmed since 11th century, I'm glad you ignored that. Root *stryb- is widely attested hydronyms, especially rivers (also in Poland and Polabian), that makes derivation from *sterti (“to extend”) problematic on semantic ground. Another semantic problems face Ukrainian words with root stryb- that mean "to run (away)"; "to jump" - that has semantic shift cognates for example in Latvian deja, diet (“to run; to jump; to cavort; to dance”). Is there any example for "to extend" > "to flow"/"to jump"? There is also dialectal Russian стриба́ть (stribátʹ, “to flash (about lightning)”) which has almost exact semantic cognate in Serbo-Croatian strujiti (“to flow (about electricity)”) from the same root. It should be mentioned, that the context in text also allows aquatic etymology (arrows are brought by the wind from the sea). TLDR: wind is correct on phonetic side, but the supposed root of *sterti doesn't explain semantics of Ukrainian/Russian words and doesn't explain why that root is used in so many hydronyms. Plus, there is kinda-morphological problem - *bogъ (“god”) in the theonym is attested for *Daďьbogъ only, but that theonym has very unique etymology, from interjection *daďь bože/*bogъ daďь (“God willing”).
• This is simply a lie. Scholars are divided on this one, some connect Him to the sun, others to the moon. Researchers who support the solar interpretation have no evidence for it. The Tale of Igor's Campaign:

Prince Vseslav was a judge for his subjects, he distributed cities among princes, but by night he ran like a wolf, from Kiev he ran to Tmutarakan, before the cock crowed, as a wolf he ran along the road of the great Khors.

Borissoff's Proto-Indo-European *ǵʰers- is completly wrong, PIE ǵʰ gives Slavic g, z or ž, never x/k, he also can't tell which sources are credible. Iranian etymology is also rejected on phonetic grounds. That etymology also assumes we had like 3 Sun Gods in Slavic mythology (*Daďьbogъ, *Sъvarožiťь), but not Moon God, which was basic god for Indo-Europeans (and all humans actually). Etymology is pretty simple here: there are mixations of k : x in Slavic languages (x comes from earlier k/s), I gave examples for that. That makes earlier form *Kъrsъ completly correct. In theonymy/onomastics there is simple rule, that theonyms come from homophonous appellative, and we have that appellative: *kъrsъ; compare also Proto-Germanic *Þunraz from *þunraz, or Proto-Indo-European *dyḗws. Bezimenem also forged some relation of Khors with Bulgarian solstice, there no such thing, Khors is known only from Old East Slavic texts and toponyms.

• There is general problem with Iranian etymologies (loved so much by East and South Slavs). In the past some scholars believe there were dozens of Iranian words borrowed into Slavic, including theonyms, today it is just few. Scholars pushed Iranian etymology even more for words that had in them, but for many years now we know thats native sound. All gods have native etymology (Simargl is only weird one). I work on Slavic mythology a lot, meanwhile Bezimenem believes that pagan Slavs knew concept of the sin, probably because he refuses to understand that concept of morality and concept of sin or two different concepts, even our dictionary explains it, just like most pagan peoples. He said that I have no basic knowledge and that I'm ignorant, but he reversed my correct (kinda) change, because he doesn't know about ę : ě alternation, even tho it is mentioned and explained in our both dictionaries of Proto-Slavic. I also explained some of theonyms stuff to him on my talk page. Sławobóg (talk) 12:05, 8 March 2023 (UTC)Reply

  @Sławobóg: You are wasting everyone's time with your explanations. The fundamental issue with your etymologies is not that they lack justification, but that they don't acknowledge alternative opinions made by experts. It's not your business to decide which etymology is the most accurate.

  Also, where exactly have I claimed that sin and morality are the same? Stop twisting my words. I've claimed the two categories are related and that pagans had some comprehension of categories that are related to the concept of sin. You, on the other hand, have asserted that proto-Slavs had "no such thing like" a concept of sin. This is an absolute statement. It implies absolute lack of comprehension of anything related to sin, including morale and ethics. If you want me to stop criticizing you, stop talking in absolutes.

  PS The account of Bulgarian folklore traditions related to Xъrsъ are attested by Анчо Калоянов and mentioned in w:bg:Xърс (ref: СбНУ). Go argue with them what they have "forged", I'm not their advocate. Even without that, it's undeniable that the lunar theory is not universally excepted, hence, it's unjust to impose it as such. That's the whole point of my complaint! Безименен (talk) 16:50, 8 March 2023 (UTC)Reply

  So thank you for wasting my time on writting stuff you don't even care to read. As for that alleged Bulgarian thing, it his just his opinion, not a fact, there is no historical information about Khors' feast. You are completly ignorant on the topic. You can't even distinguish opinion from historical fact and you don't understand what you are reading lol. I even downloaded these books and I can't confirm Калоянов saying anything like that, there different topic on page 163 (first and second part) vol 2, vol 3 talks "Jewish" epithet used on some occassion. It doesn't matter, it is 19th century book, we have books from 21th century that understand that topic better. Sławobóg (talk) 17:51, 8 March 2023 (UTC)Reply

  Wow hahaha is it another one source which discuss the "Jewish Angel Khors" from the book "The discussion between the three priests" from the 14-15 centuries. You can find very weird mythology stuff in the related Russian apocryphal texts, like a Great Solar Chicken and the Anti-Satan, so i'm not sure if any quote from "The Discussion" can be used for the research on etymology of any word. Any apocryphal text of this kind has rough translated borrowings and anachronist conceptions. By the way, Kaloyanov doesn't mention it as a Russian text unrelated to Bulgarian geography, probably because the legend about a Bulgarian author.

  PS Slawobog, do you have more sources on k/ch sound change in Slavic languages? Seems like a very fresh researches. Are you sure they are legit? Tollef Salemann (talk) 18:34, 8 March 2023 (UTC)Reply

Well, religion is barmy. If you get too much into it you can make a nefarious creature but still be on something. Not bad to make mention of the alternative versions you consider less likely. I don’t like the tendency of Bezimenen giving all versions in a randomly ordered list as if no theory had advantages over the other, but this does not mean that we can’t at least add the references of the theories you consider “pseudo-scientific” (science is a big word when it’s about assessment of probabilities by experience with conjectural material) and note nigh the reference (in a further reading or references header, depending on what you deem their purposes) why you don’t even let it into the main article text. Fay Freak (talk) 18:35, 8 March 2023 (UTC)Reply

Haven't read the whole Kaloyanov yet, so im not saying he is a pseudo-science-dude making a mummy collection. I just know from my experience that the topic "Names of Slavic Pagan Gods" has tendency to get a lot of doubtful sources with shady agenda. Tollef Salemann (talk) 18:52, 8 March 2023 (UTC)Reply

@Tollef Salemann: I wouldn't take Kaloyanov's views uncritically, however, that's not the point here. It seems he bases his observations on ethnological research made by God knows who and on the following sources: "Ходене на Богородица по мъките", "Беседа на тримата светители", "Слово на Христолюбеца" и "Слово на тълкователя". I apologise that I can't reflect more on the problem in hand. My goal is not to refute Slawobog's etymology, but to make him acknowledge the dissonance among the scientific community on the topic of "Pagan Gods". Безименен (talk) 19:23, 8 March 2023 (UTC)Reply

It can be a Bulgarian text behind it, but the passage about "Jewish Angel Khors" may come from another not-Bulgarian source, but anyway is kinda weird to use it in etymology research. He is not the one who did it. I hope that the origins of this angel-passage gonna be revealed some day, but it seems like some Greek-Slavic translation localization, compare e.g. to translations of Jupiter in the old Norse Trojumanna saga as Krita-Thor, very similar to that one which was made with Khors in this old Russian texts. Im still not sure if it is anything to do with the etymology tho :) Tollef Salemann (talk) 19:34, 8 March 2023 (UTC)Reply

@Tollef Salemann: Admittedly, it is not directly related. I just don't see why we should leave an option for this discussion to arise in the first place. One could simply address common past controversies under Wiktionary's etymology and to forget about them once and for all. If this was done beforehand, I wouldn't have dug these dubious/confusing/forged views up. Безименен (talk) 23:15, 8 March 2023 (UTC)Reply

@Tollef Salemann It is not related at all. It is used by some supporters of fringe theories. "Jewish" thing is indeed very weird here. My book say it just meant "Pagan", just like "Greek", but it doesn't give further explanation or other examples of this use. Sławobóg (talk) 10:41, 9 March 2023 (UTC)Reply

Yeah there is used the word "Ellin" as "Greek". It is usually used in the old Russian sources as a synonym of the pagan Greeks and their beliefs (except maybe one or two examples). Tollef Salemann (talk) 10:58, 9 March 2023 (UTC)Reply

@Sławobóg: Again, where in your etymologies it is implied that 21th cent. theories have superseded old ones? If an etymon in question is so apparent as you make it appear, there wouldn't have been so many theories and fault opinions in the past.

@Fay Freak: I'm fine with trimming down some of the sources that I add, as long as it is apparent that given contested question bears uncertainty. I often struggle to represent faithfully the complexity of a given etymology without going overboard. Безименен (talk) 18:55, 8 March 2023 (UTC)Reply

Discussion moved from WT:RFVN.

Proto-Slavic. There are no sources, none of the given descendants match the reconstructed verb, the proposed PIE root does not exactly match the semantics of the alleged Slavic verb. A related term Slovene réža, which the creator advices us to see, is given by Slovene linguists as a derivative of Slovene režáti (“to gape”) < Proto-Slavic *ręža̋ti.

Short Google search points towards Simargl (Slavic deity), where at least there is quotation of the Polish researcher Łuczyński. It looks to me, this whole lemma was reconstructed by User:Sławobóg in his zealous attempt to demonstrate that all Slavic deities bear Slavic etymology (rather than borrowed from Iranian or another source).

PS Ignore for now the inconsistency with the established Wiktionary convention. The lemma should be given as *reťi (like Proto-Slavic *moťь = *mogtь), but this can be fixed easily. Безименен (talk) 10:18, 11 March 2023 (UTC)Reply

I don't remember why I put these words there, but these are etymologically unrelated. So none of you "criticism" is related to that lemma. You, again, didn't do any research about this word and you know nothing about it. Descendants of that lemma are usually linked to Proto-Slavic *rǫgati/*ręgati (“to mock, taunt; to scold, chide”) with alleged original meaning “to burst, break” but (1) meaning shift looks very weird, (2) nasals are unexplained, *ręgati gave Croatian rúgati, meanwhile Croatian words mentioned in my lemma have -e- in the anlaut, Polish *rzęga is expected (Boris gives an explanation here, but the Croatian word allows this to be contradicted). My only factual error is that this lexeme should be called *regati. Good job dude, keep trolling. Sławobóg (talk) 12:52, 11 March 2023 (UTC)Reply

I struggle to understand what you're saying. Can you use explicit language? What terms are you referring to with "these" and "that"? Why are you linking *regti (“to cut”) and *rǫgati (“to scold”)? According to Snoj's dictionary, the later is assumed to be related with *regati (“to croak”). Are you implying that Slovene regati (“to crack open”) and Slovene regati (“to croak”) are identical? The only mention of the former that I could find is in Pleteršnik's "Sloveno-Nemški slovar" and he makes distinction between the two. Is *ręgati gave Croatian rúgati a typo or you seriously think pSl. *-ę- is supposed to yield SCr. -u-? How am I supposed to know anything about *regti when no source is provided, there is no mention of it in standard dictionaries, and none of the proposed descendants actually match the reconstruction (you've listed participles, a deverbial noun in Polish and a barely attested secondary verb)? Безименен (talk) 14:56, 11 March 2023 (UTC)Reply

No, *-ǫ- can give Scr. -u- and there was alternation between *-ǫ- and *-ę-. And no, I don't think these 2 are identical, I said some suggested that *regati (“to cut”) is identical with *rǫgati. I didn't do any research on *rǫgati/*ręgati because these words not related to lemma I created and I don't know if Slovene regate (“to croak”) is related to them or not. I'm going to fix my lemma now. Sławobóg (talk) 19:28, 11 March 2023 (UTC)Reply

Actually it should be *reťi... Sławobóg (talk) 20:15, 11 March 2023 (UTC)Reply

@Sławobóg: Comment after the improvements made in *reťi: As far as I get it, based on the provided sources, all the descendants that you list except dial. Polish rzega can be derived from Proto-Slavic *ręgati. I wouldn't base a whole pSl. reconstruction on a single dialectal derivative. Not directly related to the reconstruction of *regti, but still relevant to the topic is this *Rьglъ that Łuczyński reconstructs. The expected l-participle of *regti is *reglъ, not **rьglъ. The later could hypothetically follow from *rьgnǫti, but still this doesn't justify reconstructing *regti. Безименен (talk) 23:56, 11 March 2023 (UTC)Reply

@Bezimenen it can't be from *ręgati because problematic semantic shift, I mentioned that. Polish word is semantically close to South Slavic ones. I think maybe rẹ́gniti might derive from something else. The way this lemma is made has nothing to do here with Rьglъ. Sławobóg (talk) 17:30, 14 March 2023 (UTC)Reply

Talking about a "problematic semantic shift", going from 'to be dark' to 'to cut' doesn't seem unproblematic to me. Exarchus (talk) 09:28, 9 December 2024 (UTC)Reply

Etymology is rather semantically dubious, “to swallow” > “deep place”, and should be left to the etymology section of Albanian grellë. Entry also created by block-evade user. Pinging @ArbDardh -- Sokkjō 01:29, 18 April 2023 (UTC)Reply

Latin. See español. Earlier, speakers of such languages would have called themselves ladino or cristiano when Old Occitan began to innovate the word. Kwékwlos (talk) 10:08, 3 May 2023 (UTC)Reply

Delete. Nicodene (talk) 13:48, 2 May 2024 (UTC)Reply

Delete, contradicts information already on Wiktionary. Fay Freak (talk) 14:37, 2 May 2024 (UTC)Reply

@Fay Freak this is not a reason for deletion. If it were, one could equally make the same argument for deleting whichever pages contradict this page. You need to reference actual reliable sources in order to establish the reliability or non-reliability of this reconstruction. Brusquedandelion (talk) 00:13, 12 May 2024 (UTC)Reply

@Brusquedandelion: It is: the nominated-for-deletion page contradicts the information, the facts on the other page which I assume as true due to detail and references, not simply the expression of it. Indeed I care whether pages contradict each other but don’t put this forward as a reason. You made this up in your head from my rhetorics. Fay Freak (talk) 00:22, 12 May 2024 (UTC)Reply

Comment: @Romanophile, as the page creator, can you comment on this RfD? Thank you. Also, @Kwékwlos, can you provide reliable sources arguing in favor of your view here? Likewise for Romanophile. Brusquedandelion (talk) 00:11, 12 May 2024 (UTC)Reply

https://www.etymonline.com/word/spaniel (((Romanophile))) ♞ (contributions) 23:37, 13 May 2024 (UTC)Reply

That site provides no factual basis for such a reconstruction. Nicodene (talk) 01:33, 14 May 2024 (UTC)Reply

Check The Barnhart Dictionary of Etymology. (((Romanophile))) ♞ (contributions) 01:51, 27 June 2024 (UTC)Reply

Ymir of Norse mythology is obscure and its etymology is purely conjecture, not warranting a full entry. @DerRudymeister -- Gowanw (talk) 20:05, 8 May 2023 (UTC)Reply

All modern specialists in Germanic and Norse language and mythology reconstruct *jumja- which is generally connected to Indo-European *yémHos, cf. Böldl (2013: 112), which can be reconciled, as Kroonen (2013) does, by reconstructing a zero-grade pre-Proto-Germanic form *i̯m̥H-i̯o-. Interestingly, as observed by Kroonen (2013), Old Irish emon can actually go back either to a zero-grade *imH-o- or a full grade *iemH-o- (the traditional reconstruction). This doesn't mean we have to reconstruct a zero-grade but it does leave open the option.

Positing a long stem vowel **Ýmir as per Huld apud Mallory & Adams (1997) on the other hand is entirely unwarranted and unnecessary. There doesn't appear to be any reason to outright reject the connection posited by the majority of experts in the field.

Böldl, Klaus (2013). Götter und Mythen des Nordens: Ein Handbuch . Munich: C.H.Beck. Hörntje (talk) 13:38, 10 June 2024 (UTC)Reply

"... and its etymology is purely conjecture" is ignoring just how much we know about Germanic sound changes, and also ignores the (rather clear) cognates in other Indo-European languages. Haimariks Wandilaz (talk) 16:52, 1 October 2024 (UTC)Reply

1. ^ Mallory, J. P., Adams, D. Q., editors (1997), Encyclopedia of Indo-European Culture, London, Chicago: Fitzroy Dearborn Publishers, page 608

Proto-Indo-European. Reconstruction by @Gnosandes (which although straightforward, it must be noted it is unsourced). I don't see why we can't handle this in the derived terms of *bʰardʰéh₂, as we usually do. Catonif (talk) 19:12, 23 May 2023 (UTC)Reply

Delete: The PIE word for beard has always been contentious. No sense making it worse with derivatives that are clearly secondary. --{{victar|talk}} 23:52, 23 May 2023 (UTC)Reply

Delete. Derivatives of this extension are already listed on *bʰardʰeh₂. -saph 🍏 16:34, 7 December 2023 (UTC)Reply

Not attested in Old Norse. History should be merged into Proto-West Germanic *parruk. @Nicodene --{{victar|talk}} 00:29, 25 May 2023 (UTC)Reply

What a twist.

Pinging @Leasnam as well, who added the mention of an Old Norse parrak. Do you recall where that may have been from? Nicodene (talk) 05:47, 25 May 2023 (UTC)Reply

Koebler has it in Old Norse, as parrak and parak (B&T also lists parak and parraka (“to keep pent in or under restraint”)), but Koebler postulates it as a loanword from Old English or Middle Low German. However, I want to say I had gotten it from somewhere else long before I started using his works, but I cannot remember. parrak wouldn't be right anyway, it would have to be *pǫrkr or something like that and the form parrak always bothered me. Leasnam (talk) 06:21, 25 May 2023 (UTC)Reply

The circular reference in the etymologies of Proto-West Germanic *parruk and Latin parricus also might need to be looked at. Leasnam (talk) 06:50, 25 May 2023 (UTC)Reply

Indeed, that's what prompted this investigation (Talk:parricus).

So did both Danish and Swedish inherit some form of this par(r)ak? The meaning 'fish-pond' does seem consistent with that. On the other hand, the meaning 'park' suggests a modern loan directly or indirectly from French parc. Nicodene (talk) 07:01, 25 May 2023 (UTC)Reply

As far as I can tell, parrak in only attested in medieval Icelandic (Landnámabók) as a nickname of unknown meaning: “Eystein rejste fra Norge til Island; han var søn af Rane, der var søn af Hildir Parrak.” Hellquist is the one that cites parraka as a “laan fra ags. pearroc”, but the -ak is a more typical Saxon vocalisation, fwiw. --{{victar|talk}} 07:22, 25 May 2023 (UTC)Reply

Hellquist does not give the form "parraka", either in the first edition (which is on Runeberg) or in the third (which is in my lap). Instead he says "En gammal biform föreligger i ags. pearroc, inhägnad (eng. parrock) = fhty. pferrich, (får-)fålla o.d. (ty. pferch), osv., mlat. parricus." For the main form of the word (park, y. fsv. parker) he says "Väl ytterst från fra. (enl. somliga i sin tur av iberiskt urspr.)" 217.210.158.171 20:53, 9 July 2023 (UTC)Reply

Swedish parak exists but is defined as “lamb, kid”. It conceivably derives in meaning from “penned animal”, but even so, it would be more than likely a Low German borrowing. I can't find any source that claims Danish park also means “fish pond”, though it wouldn't be too hard of a semantic shift. @Leasnam, it seems you added that back in 2012 (have we really been working on these entry for 11 years? 0_0). --{{victar|talk}} 08:01, 25 May 2023 (UTC)Reply

I added the Scandinavian descendants in 2016 ]. Rua (CodeCat) and I (and a few others) made many of the early ones...and we didn't yet distinguish a node at PWG. Yeah, it's been that long. Leasnam (talk) 18:56, 25 May 2023 (UTC)Reply

Yeah, I created the PWG back in 2012 when we still called it Frankish. :p --{{victar|talk}} 19:14, 25 May 2023 (UTC)Reply

Yep, and at that time the two were still considered distinct :) Leasnam (talk) 15:01, 26 May 2023 (UTC)Reply

From where is the claim that parak exists as a Swedish word? Wiktionary doesn't have it, it's not in SAOB, not in Hellquist, and Korp gives no relevant hit, only a spurious one arising from inserting a space into parakit. 217.210.158.171 21:01, 9 July 2023 (UTC)Reply

This is the main form for parrkaz. If we delete this, would we delete parrikaz as well? —Soap— 08:33, 15 March 2024 (UTC)Reply

There has been an RFV (now auto-converted into an RFD) on this page since April 2019. Nicodene (talk) 22:12, 1 June 2023 (UTC)Reply

I deemed this RFV passed a year and a half ago, but apparently forgot to remove the tag. I'm sorry. Unless anyone brings up further objections I can remove it. —Caoimhin ceallach (talk) 23:22, 1 June 2023 (UTC)Reply

Can we move it do *ḱad- though? I see it was proposed in the past RFD as well, but never done. Catonif (talk) 13:33, 15 June 2023 (UTC)Reply

Delete. Not everything starred somewhere needs a page. Fay Freak (talk) 14:41, 2 May 2024 (UTC)Reply

Duplicate entry of Reconstruction:Proto-Germanic/skagją. -- Sokkjō 04:04, 21 June 2023 (UTC)Reply

For context: *leyd- § History, ludo § History and User talk:Catonif § Proto-Indo-European *leyd-. I invite editors to check these links and the entries' revistions, and of course to check the sources firsthand, before stating an opinion.

Proto-Indo-European, in the sense "to play", added by Victar. This revision shows the references which endorse a single root meaning "to let, allow; to let go, release". The semantic explaination can be found here. The current revision puzzlingly cites the same sources while claiming something wildly different, which is supported, as the PIE entry currently stands, by the two additional references of Pokorny and Adams (of course potentially many more works of the same and earlier time-period can be cited). Their works respectively are undeniably of enormous value, yet science is moving forward, and I've yet to find an authoritative source explicitely rejecting Rix's view in favour of the simpler traditional Baltic-less "to play" one. Catonif (talk) 20:29, 29 June 2023 (UTC)Reply

To echo what I wrote earlier, most recent does not equal most right, and just because something is published, doesn't mean it shouldn't be scrutinized. Rix posits the theory that the Balto-Slavic terms meaning “to let (go)”, which are traditionally found under *leh₁d- (“to let go, become tired”), should instead be reconstructed as from *leyd- (“to let (go)”). He then goes one step further and suggests the root is one the same as the well-established root *leyd- (“to play”).

The semantics of this merge are dubious, which Rix himself claims tentatively, marking each and every reconstruction as uncertain. The safer bet is to continue to keep this as two separate roots, whilst noting Rix's theory on the entry. --{{victar|talk}} 00:03, 30 June 2023 (UTC)Reply

@Victar So we're removing references now? (diff) The Leiden books are just citing LIV and we don't need to recite it every time it's mentioned, just the original source. Scholars citing each other is exactly the point, it means the idea is widespread and makes sense. I understand the concept of reference clutter whenever there's an undisputed claim, but here you're claiming that Rix is the only one that tries to merge these roots (diff), and it seems like you are trying to make it seem like that's the case, while it undeniably isn't. hey aren't studies on PIE. Bonkers, they're Indo-Europeanists! Do I need to wait for Lubotsky's project to be finished for the ref to be usable? ost recent does not equal most right. Certainly can't be applied to every single case, but given your expertise in the field I don't think I need to be the one explaining to you that the vast majority of times ewer works that are more up-to-date with modern scholarship have precedence over the old ones (from WT:REF). We can place your question marks wherever you will, but I don't think we should split the root in two for them. I personally find Rix's semantical explaination convincing, considering the same shift is observable notably in Lithuanian léisti (“to take a break (from work), pass time, rest”) and relax : cf. Italian lasciare (“to let”), as Latin lūd- was initially the opposite of work, for which compare also Ancient Greek σχολή (skholḗ, “leisure, free time, rest”) → school, with the Latin parallel lūdus (“school”). Catonif (talk) 20:05, 3 July 2023 (UTC)Reply

No sources were removed, but attributing them on the entry as a primary source to the reconstruction is incorrect. If we cited every secondary source of a reconstruction, there would be dozens upon dozens. In fact, I added a source for what might be the oldest reconstruction for *leyd- (“to let go”), theorizing the connection between the Balto-Slavic and the Albanian, but with no mention of the Latin and Greek words for “to play”. --{{victar|talk}} 21:30, 3 July 2023 (UTC)Reply

Only found in West Germanic and of unknown origin. Please please merge edit history into Reconstruction:Proto-West Germanic/bast and delete. -- Sokkjō 17:25, 17 July 2023 (UTC)Reply

Delete. Catonif (talk) 10:12, 11 September 2023 (UTC)Reply

Delete. Actually leaning keep because this should be WT:RFE. Often said to be akin to band, I say Bändsel (“string”). Meid argues that no trace of loss of n exists . The ending of береста needs to be illuminated before deletion can proceed. 2A00:20:6091:625E:BD00:D4C:7C1E:9B5C 09:19, 4 August 2024 (UTC)Reply

Sanskrit राम (dark) is often cited to be related to Old English rōmig (“sooty”), with unknown further etymology, but no one connects to it *h₁reh₁- (“to separate; sparse”). I presume the user was confused. --{{victar|talk}} 22:43, 19 July 2023 (UTC)Reply

Delete. Fay Freak (talk) 05:09, 20 September 2023 (UTC)Reply

Proto-Tungusic. Created by TheSilverWolf98. And tagged by Ardahan Karabağ. -- Dubukimchi (talk) 16:36, 13 August 2023 (UTC)Reply

No entra-WG descendants. Edit history should be merged into PWG *marþr and deleted. -- Sokkjō 07:54, 1 September 2023 (UTC)Reply

Delete. Catonif (talk) 10:09, 11 September 2023 (UTC)Reply

Latin. Tagged but not listed by Nicodene. Ultimateria (talk) 03:31, 4 September 2023 (UTC)Reply

*x̱änɦï is an alternate (Starostin 1994) reconstruction of *ɬ:ɔn (reconstruction from Schrijver 2021), I don't see why there is a separate Wiktionary page for *x̱änɦï, when the alternate reconstruction is already in the page *ɬ:ɔn --Qmbhiseykwos (talk) 00:20, 8 September 2023 (UTC)Reply

@Qmbhiseykwos: You're going about it backwards; *ɬ:ɔn should be deleted, and then *x̱änɦï would be moved to Reconstruction:Proto-Northeast Caucasian/ɬ:ɔn, assuming that's what people want. You should start a discussion first on making your point on why this is a preferable reconstruction, because as is, all our PNC reconstructions so far have been based on Starostin and we don't want two competing systems. --{{victar|talk}} 01:21, 8 September 2023 (UTC)Reply

Here is the case I will make for why Schrijver's reconstructions are preferable reconstructions to Starostin's PNEC reconstructions. I quote Johanna Nichols ("The Nakh-Dagestanian Consonant Correspondences," 2003:208) at length

"Given this agreement, the obvious research priority would seem to be reconstructing Proto-Nakh-Daghestanian (PND). But, oddly enough, this is the one thing that has not been done. The groundwork for Nakh-Daghestanian (ND) comparative studies was laid by Trubetzkoy (1930), who established some notoriously non-transparent but quite regular correspondences; however, his cognate sets are for 'North Caucasian' and include Northwest Caucasian (NWC) as well as Nakh-Daghestanian words. Gigineishvili (1977) reconstructs Daghestanian only. Bokarev (1981, and the earlier 1961 work on which it is based) deals with ND, but unequally, comparing Daghestanian protoforms and cognate sets to Nakh translation equivalents without determining whether the Nakh words are actually cognate to their Daghestanian translations (often they are not). Diakonoff and Starostin 1986 assume that Hurro-Urartean is related to Nakh-Daghestanian, and assemble putative cognate sets so as to maximize similarities between the two families. Nikolayev and Starostin (1994) assume that Nakh-Daghestanian and Northwest Caucasian form a family and offer reconstructions for that putative family, again assembling putative cognate sets so as to maximize matches and similarities between ND and NWC. This method offers no guarantee that the ND subpart of an ND-NWC putative cognate set is a proper ND cognate set, and in approximately a third of the cases they are not (Smeets 1989, Nichols 1997). As the putative cognate sets determine the correspondences and hence the reconstructions, there is no certainty that a proper PND has actually been reconstructed. Neither Diakonoff and Starostin, nor Nikolayev and Starostin, take on the burden of proof and discuss whether the incidence of resemblances exceeds chance expectation, nor do they present examples of the kind of shared morphological paradigmaticity that would strongly support genetic relatedness. Accordingly, the possibility of external relations to NWC and/or Hurro-Urartean must be regarded as an opinion for which no support has been offered."

Nichols further explains that "Nikolayev & Starostin 1994 frequently split Nakh-Daghestanian cognate sets up among two or more putative North Caucasian cognate sets. Nichols 1997 found that over one-third of the then-known secure Nakh-Daghestanian cognate sets were split up and redistributed by Nikolayev and Starostin."

The reconstruction provided here is based on regular (Nichols 2003) consonant correspondences that Nichols identifies in the same paper, and regular vowel correspondences worked out in Schrijver (2021 "A history of the vowel systems of the Nakh languages (East Caucasian) with special reference to umlaut in Chechen and Ingush", 2018 "The Origin of Vowel Alteration in Avar-Andi-Dido (North-East Caucasian)")- it is limited because not every vowel phoneme of PNEC has been reconstructed in Schrijver 2021, nor has data from Khinalug, Dargwa, Lak, or Lezghian languages been incorporated into the reconstruction. However, I still think this is better than Starostin's reconstruction, because while Starostin's reconstruction can claim inspiration from Khinalug, Dargwa, Lak, or Lezghian languages in addition to Avar-Andi-Tsezic (or Avar-Andi-Dido) languages, it has the significant disadvantage of rather irregular correspondences everywhere. And those irregular correspondences are used to explain a set of data which contains 2/3 cognates, 1/3 false cognates. Nichols and Schrijver instead proceed from established cognate sets in Gigineishvili and build sound laws off of those, rather than building sound laws off of a wider data set that includes false cognates. This is why, although their reconstruction is less "complete" than Starostin's, their reconstruction is still significantly better methodologically than Starostin's, and unlike Starostin's reconstruction, seems to actually explain a good deal about ablaut and vocalism in Nakh and Avar-Andi-Tsezic languages.

Based on this case I think that this Schrijver (2021) based reconstruction should be used instead of Starostin's. Qmbhiseykwos (talk) 02:33, 8 September 2023 (UTC)Reply

There is already an Uzbek entry at emoq so this is already an attested term, not a reconstructed one? Unless this is a reconstruction for midevil/pre-modern Uzbek, in which case it should be in the Arabic script not Latin. سَمِیر | Sameer (^{مشارکت‌ها} • ^{کتی من گپ بزن}) 17:11, 23 September 2023 (UTC)Reply

Also, Uzbek is still written in the Arabic script in Afghanistan and in China. Rodrigo5260 (talk) 18:04, 23 September 2023 (UTC)Reply

Keep and delete emoq instead. emoq is completely unattested (as far as I know) in Modern Uzbek and is only the hypothetical infinitive form of its conjugations. Note that the page emoq is of very low quality also, sporting an impossible to confirm pronunciation, an erroneous conjugation table (the verb is defective and takes a very limited number of conjugations) and a poor gloss. Newgrass 82 (talk) 04:18, 5 June 2024 (UTC)Reply

Comment Also, Sameer, how did you decide which article was the correct one to delete? Why are you sticking your nose in languages you don't even remotely understand? Newgrass 82 (talk) 04:25, 5 June 2024 (UTC)Reply

The following Proto-Basque reconstructions lack sources, don't conform to Proto-Basque phonology or are otherwise dubious.

• Reconstruction:Proto-Basque/arhe unsourced
• Reconstruction:Proto-Basque/beh- unsourced, probably based on the hypothesis that behi (“cow”) and behor (“mare”) are related (according to Trask " cannot be evaluated")
• Reconstruction:Proto-Basque/caheca unsourced (it doesn't show up in the linked sources), Proto-Basque doesn't allow word-initial /k/. According to the General Basque Dictionary it is onomatopoeic.
• Reconstruction:Proto-Basque/caio same as above (unsourced, initial /k/). Mitxelena links it to French geai (“jay”) and Corominas to Spanish gavia (“seagull”).
• Reconstruction:Proto-Basque/cedaŕ same as above. Possibly a derivative of ke (“smoke”), whose conservative form is probably eke.
• Reconstruction:Proto-Basque/concoŕ initial /k/. Trask lists it as one of several kV(n)kV- expressive formations.
• Reconstruction:Proto-Basque/eɫhauso Trask lists modern Basque elhauso (standard spelling elauso) as a variant of the straightforward compound el(h)ur (“snow”) + auso (“blizzard”)
• Reconstruction:Proto-Basque/gernu contains the unusual cluster /rn/, according to Trask "Probably native, with an unrecoverable phonological history".
• Reconstruction:Proto-Basque/izarr- the only descendants are Romance borrowings. There's no indication that Romance borrowed specifically from Proto-Basque and not from some other substrate language.
• Reconstruction:Proto-Basque/katena initial /k/. While modern Basque kate is clearly a (later) Latin/Romance borrowing, Proto-Basque borrowed Latin c- as *g- in general.
• Reconstruction:Proto-Basque/kaka initial /k/. Probably just a borrowing from Spanish.
• Reconstruction:Proto-Basque/-kunbe no indication in the given source that the suffix -kume (from ume (“child”), itself from Proto-Basque *unbe) dates from Proto-Basque.
• Reconstruction:Proto-Basque/sapar Trask lists modern Basque sapar as one of many dialectal variants of lahar.
• Reconstruction:Proto-Basque/xainx́ unsourced, the palatal fricative and affricate (assuming that's what x and x́ stand for) were not phonemic in Proto-Basque.
• Reconstruction:Proto-Basque/ziNaguRi unsourced, Proto-Basque nouns with more than two syllables are uncommon.

Santi2222 (talk) 16:32, 10 October 2023 (UTC)Reply

These all look like sound reasons for deletion to me, though I wont vote yet as I know little about the language. In fact I have a question ... do we know at what point the deletion of -n- occurred? Was this before or after the language began to comfortably loan words with initial voiceless stops? Because katea, while a redlink, is so far as I know a modern Basque word for chain, and if /n/ was lost at a time when we say that the language still only allowed voiced stops in word-initial position, perhaps there were exceptions to this rule. Best wishes, —Soap— 17:07, 10 October 2023 (UTC)Reply

A note: it seems the word for chain may have irregularly changed to kate. —Soap— 17:39, 10 October 2023 (UTC)Reply

I don't know at what point Basque acquired initial voiceless stops, but according to Trask, the loss of intervocalic /n/ happened during the early medieval period and was complete by the early 11th century. Again according to Trask, all instances of Romance /p/, /t/, /k/ are borrowed as /p/, /t/, /k/ (with many exceptions in the initial position). Regarding the loss of final -a, modern Basque kate is a reanalysis of *katea as kate +‎ -a (definite article). Actually, katea is attested as an indefinite form, but nowadays it is the singular (definite) absolutive form of kate (I don't know if any varieties still have the form with final -a). Santi2222 (talk) 18:09, 10 October 2023 (UTC)Reply

Also, while Im here, though Im probably not the first to notice, I want to point out that the Basque Etymology link we use on so many pages currently redirects to the website of a mountain resort in Bulgaria. Since the original link was also from Bulgaria, my guess is that it's the same company and that someone at what is now a resort had an interesting side hobby. Nonetheless the link isn't much use if we can't get to the original document, and I wonder if it might have been compiled from a different original source. Thoughts? —Soap— 17:56, 10 October 2023 (UTC)Reply

The "Basque Etymology" link used to point to an etymological dictionary by John Bengtson. The main point of that dictionary seems to be to connect Basque with "Proto-Sino-Caucasian" so I seriously doubt we can find anything useful in there.Santi2222 (talk) 18:24, 10 October 2023 (UTC)Reply

Likely a West-Germanic construction. See RC:Proto-West Germanic/lapil. -- Sokkjō 19:33, 14 October 2023 (UTC)Reply

There's only one descendant. In addition, *iz meant "he" in Proto-Germanic, not "self, same". —Caoimhin ceallach (talk) 23:18, 20 October 2023 (UTC)Reply

@Leasnam -- Sokkjō 00:38, 21 October 2023 (UTC)Reply

Deleted. Leasnam (talk) 00:51, 21 October 2023 (UTC)Reply

@Leasnam What's the etymology of Old English ilca? I don't think it should have been deleted so quickly. -- Sokkjō 00:54, 21 October 2023 (UTC)Reply

Restored, and moved to PWG. Not sure if the RFD applies to this as well. The Old English etymology is equivalent to ī- +‎ līċ. Old English ī- has a partial cognate in Middle Low German īdich (“the same”), Middle High German iedic (“the same”). Not sure at the moment what the best way to handle this is. Leasnam (talk) 01:14, 21 October 2023 (UTC)Reply

@Leasnam If it's from PWG *iʀlīk, wouldn't the outcome in OE be *īlīċ (or *irlīċ?) -- Sokkjō 02:47, 21 October 2023 (UTC)Reply

The Nordic terms are borrowed from Middle Low German, according to {{R:no:NEO}}, which is usually the case with so few terms. @DerRudymeister -- Sokkjō 21:32, 22 October 2023 (UTC)Reply

Icelandic is probably the crux for this. Is the known to appear there late or show other signs of being secondarily from Norwegian? --Tropylium (talk) 18:08, 23 July 2024 (UTC)Reply

Reconstruction:Proto-West Germanic/plakkju is dicey enough, no need to make it worse with a Proto-Germanic entry. Merge history into aforementioned PWG entry. @Leasnam -- Sokkjō 04:19, 23 October 2023 (UTC)Reply

Proto-Indo-European. A term with descendants only in Latin and Hellenic, of further unknown etymology, not without phonological problems for which some ad hoc solutions have been proposed, often marked with question marks even in relatively early works such as Pokorny's. Two modern sources, square brackets mine:

• Mallory, J. P., Adams, D. Q., editors (1997), “?*dl̥kus ~ *glukus”, in Encyclopedia of Indo-European Culture, London, Chicago: Fitzroy Dearborn Publishers, page 560b

  Even accepting this proposal though , there is little reason for proposing an IE form given that the distribution is limited to neighbors with a long history of contact.

• De Vaan, Michiel (2008) “dulcis”, in Etymological Dictionary of Latin and the other Italic Languages (Leiden Indo-European Etymological Dictionary Series; 7), Leiden, Boston: Brill, →ISBN, page 182

  Even if it were correct, we would still not have an etymology for Graeco-Latin *dl(u)ku-. It is likely that we are dealing with a common borrowing from an unknown source.

The Armenian քաղցր (kʻałcʻr) is occasionally mentioned by some sources, for which I ping @Vahagn Petrosyan for input, although judging from the etymology we provide for it, I doubt it's what is going to save the PIE entry. Catonif (talk) 15:57, 24 October 2023 (UTC)Reply

The Armenian is super-irregular and is not helpful for your root. I have a more economical explanation for Armenian as my original research, which I have not put in our entry for քաղցր (kʻałcʻr) yet. Vahag (talk) 17:39, 24 October 2023 (UTC)Reply

Keep: It's used as a big case study for #dl in PIE and is well sourced because of it. I can think of dozens of PIE entries that should be deleted before this one. -- Sokkjō 16:01, 22 May 2024 (UTC)Reply

I moved the entry content for RC:Proto-West Germanic/kwellan to RC:Proto-West Germanic/kwalljan#Etymology 2, as it's less gymnastics explaining the germination with a strong j-present then a wonky secondary full-grade.

Thus RC:Proto-West Germanic/kwellan and RC:Proto-Germanic/kwellaną should me history merged/deleted into RC:Proto-West Germanic/kwalljan. @Holodwig21, DerRudymeister -- Sokkjō 07:26, 31 October 2023 (UTC)Reply

Delete I think its better if it is a j-present. I guessing the germinated entry has to done with the source giving the verb as *kwellaną. Without a non-west Germanic descent there is no telling whenever this verb is a j-present or formed through a germinate. 𐌷𐌻𐌿𐌳𐌰𐍅𐌹𐌲𐍃 𐌰𐌻𐌰𐍂𐌴𐌹𐌺𐌹𐌲𐌲𐍃 (talk) 20:06, 2 November 2023 (UTC)Reply

I'm not opposed to remove the Proto-Germanic entry, but I'm not in favor of reconstructing a class 6 j-present as there is no evidence for that at all. All the descendants show a class 3b inflection. Not only Orel but also Seebold reconstructs this verb as such. None of them bother to explain the geminate, but if I were to do an educated guess where it originates from I would say the verb was influenced by semantically related *wellaną (“to well up, spring out, bubble forth”). --DerRudymeister (talk) 20:07, 6 November 2023 (UTC)Reply

@DerRudymeister: That's a good point, though j-presents have a way of being rebuilt. Kroonen believes in full-grade iteratives, so I rewrote the etymology using that on RC:Proto-West Germanic/kwellan and reverted RC:Proto-West Germanic/kwalljan. RC:Proto-Germanic/kwellaną should still be deleted, given how unsure this verb is. -- Sokkjō 05:16, 11 November 2023 (UTC)Reply

Proto-West Germanic. I removed the Old English losian as that comes from a root meaning "loss" shared with Old Norse. The GOH is incorrect, losōn means to "hear, listen"; and lōsōn (“to loose, free”) goes back to Proto-West Germanic *lausōn. The Low German and Dutch look derived from the adjective. Leasnam (talk) 16:52, 10 November 2023 (UTC)Reply

I am grappling with a couple of issues related to Proto-West Germanic *narrjō. If this existed in PWGmc, shouldn't we at least see i/j-mutation in German ? German is pretty reliably affected in this regard. Also, it is generally accepted that the Middle Dutch and Middle Low German words are borrowings, the latter especially so since it is first attested in a translation of Luther's Bible.

If Old High German narro is native, and related to Middle High German narren, nerren (“to growl, snarl”), then the entry would still have the wrong form to give rise to Old High German narro and Middle High German narre. Leasnam (talk) 23:16, 20 November 2023 (UTC)Reply

Moving the entry to *narrō with an unexplained germination is probably the way to go. The term could go back to a frequentative, i.e. *nar/ʀ-(a)rōn. I don't see any strong reason to assume the MDut. and MLG forms are borrowings from MHG. It's first attested in MLG by Agricola, not in Bugenhagen's Bible translation. -- Sokkjō 05:31, 12 December 2023 (UTC)Reply

I moved it to RC:Proto-West Germanic/narrō and added Lloyd as a source, who seems to be convinced in a native formation. They also note some Frisian forms. -- Sokkjō 02:07, 13 December 2023 (UTC)Reply

Proto-West Germanic. Tagged by @Victar but not listed. This, that and the other (talk) 06:59, 30 December 2023 (UTC)Reply

RFD not created by me, but an IP. I just commented out the long rational and asked them to create a RFD discussion.

Keep: The IP's rational is that, "the compound is so obvious, it could even have developed independently in various languages." That can be said of countless Germanic terms, and the simplicity of a compound has never been grounds to reject a reconstruction. They also claim that it should also be deleted because it is limited to Continental West Germanic, but again, regionality is also not valid reason for disclusion, see CAT:Regional Proto-West Germanic. -- Sokkjō 10:27, 2 January 2024 (UTC)Reply

Proto-West Germanic. Tagged by @Sokkjo (Victar) but not listed. The edit summary was "not the place from a huge rational". This, that and the other (talk) 07:00, 30 December 2023 (UTC)Reply

@This, that and the other: look again. It was tagged by a (generally knowledgable) IP, who laid out their rationale- chapter and verse- in the template itself. Sokkyo merely commented out the verbiage with the explanation (if I may paraphrase) that it would be better to present it in an RFD discussion than on the page. Chuck Entz (talk) 07:44, 30 December 2023 (UTC)Reply

@Chuck Entz Thanks Chuck, and apologies to Victar. I understand what was meant by that edit summary now. Here's the IP's rationale:

There is no evidence that this old enough to qualify as "Proto-West Germanic". In fact there is strong evidence to the contrary:

1.) Not present in English; 2.) First attested in OHG in the 11th century, not attested in any other old language; 3.) The underlying form outside High German is clearly "cipolla" as no b-forms exist (Middle Dutch "chibole" is from French ; MLG forms with "zw-" are obvious High German loans); 4.) Even in High German it's probably "cipolla", p-forms are attested in MHG. The -b- comes from Proto-West Germanic *bollā. If so, it lacks the first phase of the High German consonant shift, which would automatically exclude it. -- But either way, there's nothing: No English, no unity in form, no early attestation.

This, that and the other (talk) 08:50, 30 December 2023 (UTC)Reply

1. There are plenty of valid PWG reconstructions without an Old English form, see CAT:Regional Proto-West Germanic for some examples. 2. Many PWG reconstructions are only attested in OHG simply because OHG is the most well attested. 3. What often happens is Middle English and Dutch are reenforced by Old French, so you'll find quite a few ME terms deriving both from OE and OF. For the discussion on ML /c/ > PWG, see #Reconstruction:Proto-West_Germanic/krūci 4. Not sure the point they're trying to make here, as there is no "p-form" in HG. -- Sokkjō 04:03, 31 December 2023 (UTC)Reply

Keep. The IP arguments are kind of non sequitur. “-b- comes from Proto-West Germanic *bollā.” – could have been even more so in PWG? Dialectal variation /p/ ~ /b/ in early times for such a lovely smol plant can be expected. MLG onsets can also be pseudo-High-Germanisms, like when a word was borrowed from Romance but perceived as borrowed from High German, because the High-German-speaking Franconians ruled France. All arguments are much less pressing than IP thinks.

It is also not comfortable to assume that in the 5th century CE, after Germanic invasions in Rome but before the High German consonant shift, the only word for onions was *lauk. Wasn’t Albanian qepë also borrowed before even /k/ in Latin was affricatized? Don’t know what the “uncertain” part is. At the same time, for comparison, Arabic كُرَّاث (kurrāṯ) long was borrowed in addition to بَصَل (baṣal), such important plants are internationalisms in earliest possible antiquity. In Africa south of the desert when Arabs came they spread the onion. The Slavs I think only knew ramsons before borrowing *lukъ to name the onion, yes, in Proto-Germanic times European nutrition was depressing, in Late Antiquity less so. Fay Freak (talk) 04:34, 31 December 2023 (UTC)Reply

As for the ultimate origin, the comparisons to etymologize Latin cēpa are of course garbo. With Welsh cennin going back to *kasninā compared to Proto-Slavic *česnъ (“garlic”), the equation with its singulative is only motivated to claim PIE inheritance; the Albanian one is contradictory when at the same time it is in the descendants section. I might have chosen to leave it that, some contradictions cannot be resolved, but I ween that it is an Anatolian borrowing cognate to قُبَّعَة (qubbaʕa), hence the variation with caepa and κάπια (kápia, “onions”) only used in Ceryneia, and weird indeclinable cēpe n in addition to caepum; anno urbis conditae hence the food with the Italics was also depressing. Fay Freak (talk) 04:54, 31 December 2023 (UTC)Reply

If kept, rename to *tsibollā. Using a single letter implies that this is a distinct phoneme, which I previously expressed doubts about. It's more likely that native speakers treated it as a consonant cluster, albeit one that doesn't exist natively. Similar to how non-Slavic speakers will treat Slavic "c". —Rua (mew) 10:16, 9 February 2024 (UTC)Reply

Proto-West Germanic. A creation of Victar/Sokkjo, tagged by @Leasnam with the edit summary "one descendant with unknown gender". I have to say I tend to agree. This, that and the other (talk) 07:01, 30 December 2023 (UTC)Reply

You're overlooking the derived term. -- Sokkjō 04:03, 31 December 2023 (UTC)Reply

Proto-Germanic. Tagged by @Sokkjo with the reason "PWG entry *duli already exists and is on unknown origin", but not listed. This, that and the other (talk) 07:09, 30 December 2023 (UTC)Reply

Tagging the entry creator @Leasnam.

Not sure this can really be taken back to Proto-Germanic. If it helps, I can comment on the Romance series. The latter don't really suggest much of a time depth either, as their distribution mostly follows follows the zones of contact with Germanic. The exception is one 'prong' that has penetrated through Occitan and into Basque and Catalan, but that doesn't do much to change the overall picture. Overall we are well within the range of typical for Frankish borrowings (probably with some degree of lexical "piggy-backing" via French). Mind, I've not yet sorted out the details of the various vowel outcomes. Nicodene (talk) 14:25, 19 February 2024 (UTC)Reply

I don't have a great deal of objection to removing this entry; however, there is also the Old High German dola (“tube, duct, channel”) which suggests a Proto-Germanic *dulō, with Proto-Germanic *dulją being a diminutive/derivative of it. I tend to think that PWG terms with stem-final j-gemination likely occurred in PGmc and were inherited as fossilised irregular nouns rather than gemination being part of a productive paradigm in PWG. Is it fair to say this or am I missing something ? Leasnam (talk) 17:36, 19 February 2024 (UTC)Reply

There are no listed descendants in the entry. Also, the thesis linked in "Further reading" itself does not support this reconstruction (p. 150). Ελίας (talk) 11:55, 1 January 2024 (UTC)Reply

Move to Proto-Philippines. There have been some descendants added in the meantime. The only regular reflexes are found in Philippine languages and in a Malayic variety of eastern Borneo; the latter might be a borrowing from a yet unknown local source. Blust reconstructs *qaRutay for Proto-Philippines, which makes sense.

All other "reflexes" rest on some kind of linguistic hocus-pocus by Donohue and Denham based on the rationale "anything goes when words wander". I mean, just look at the chart on p. 304. Most of these assumptions are ad hoc and only serve to squeeze the various forms into the procrustean wanderwort bed. –Austronesier (talk) 18:26, 29 July 2024 (UTC)Reply

Proto-Slavic. One branch does not Proto-Slavic make, especially since Czech and (second) Upper Sorbian don't work, the header is broken, and four descendants (of which two are Sorbian) can easily innovate this later. Thadh (talk) 15:54, 16 January 2024 (UTC)Reply

Keep. Fixed errors and added more material. One branch can make Proto-Slavic and that view is popular in Slavistics. Sławobóg (talk) 16:27, 16 January 2024 (UTC)Reply

1 descendant. Same policy from Italic should be applied to Hellenic. -saph 🍏 06:53, 17 January 2024 (UTC)Reply

What constitutes a valid Proto-Hellenic entry, for the record? One where the Greek descendant is found in sufficiently divergent forms in multiple dialects? Nicodene (talk) 13:49, 19 February 2024 (UTC)Reply

Presumably that or a Mycenaean cognate. -saph 🍏 14:05, 19 February 2024 (UTC)Reply

Right.

Delete, in any case. Nicodene (talk) 21:34, 25 February 2024 (UTC)Reply

In the Nihon Ryōiki, it's attested as 乎乃 according to JDB p. 837. Apparently the Nihon Ryōiki perserves ko otsu distinction of no. @Eirikr Chuterix (talk) 19:10, 18 January 2024 (UTC)Reply

Considering that the ko-otsu vowel differences were merged in the 800s, and that the Nihon Ryōiki probably wasn't completed until 824, and that all manuscripts are apparently incomplete with the oldest dating to 904, this is all potentially late enough that I'd want to be certain that the spellings there actually reflect ko-otsu vowel values, and not post-merger spellings where the older distinctions were already lost.

Given also the possibility that wo might have only represented ⟨wo₁⟩, for which Arisaka's law would necessitate that the following o would have to also be ⟨o₁⟩, we should make sure to confirm. ‑‑ Eiríkr Útlendi │^{Tala við mig} 21:05, 18 January 2024 (UTC)Reply

Proto-Balto-Slavic. No sources, added by IP who has (re)added unsourced material that did not survive RFD previously. Vininn126 (talk) 20:14, 20 January 2024 (UTC)Reply

I've added sources for this one at least. --{{victar|talk}} 02:47, 21 January 2024 (UTC)Reply

Same as above, made presumably by the same user. --{{victar|talk}} 03:07, 24 January 2024 (UTC)Reply

Keep. Etymology section on here probably would not fit in on the main page. It's also not really an alternative reconstruction, more like an alternative form. -saph 🍏 19:19, 25 January 2024 (UTC)Reply

• Convert to a hard redirect. —Mahāgaja · talk 19:50, 25 January 2024 (UTC)Reply

Proto-Samic. Entries with no information other than derived terms. This, that and the other (talk) 00:17, 5 February 2024 (UTC)Reply

Delete. Should be kept in etymology sections. -saph 🍏 18:21, 7 February 2024 (UTC)Reply

Keep: We've never deleted reconstructions just because they only have derived terms. --{{victar|talk}} 21:20, 7 February 2024 (UTC)Reply

They don't even have meanings... what is the point of such an entry? This, that and the other (talk) 09:46, 8 February 2024 (UTC)Reply

I've fixed the templating. It's a noun with two denominative verbs, but yes, it lacks a meaning. I could venture to guess “error, mistake”, but I'll let the Sami editors, ex. User:Rua, figure that out. --{{victar|talk}} 17:47, 8 February 2024 (UTC)Reply

Okay, that's better. I'll close this discussion as far as it concerns *čājē. This, that and the other (talk) 00:13, 10 February 2024 (UTC)Reply

Added sources for the reconstruction of the root, also as a verb rather than noun (thus now moved to *čājētēk). Is this one closed yet or not? Seems to be still listed on the page itself. --Tropylium (talk) 16:50, 23 July 2024 (UTC)Reply

Keep - two derived terms with productive suffixes is usually enough to reconstruct a base noun at some stage of the language. Thadh (talk) 18:17, 8 February 2024 (UTC)Reply

@Thadh It seems to me that in the case of *čëkë there is only one derived term, as the two terms listed are alternative reconstructions of the same term. This, that and the other (talk) 00:11, 10 February 2024 (UTC)Reply

@This, that and the other: In that case we need to fix the entries for the derived term and move everything there. Thadh (talk) 00:53, 10 February 2024 (UTC)Reply

Keep, so long as the derived terms are valid. Nicodene (talk) 11:26, 9 February 2024 (UTC)Reply

Proto-Hellenic. One descendant, same policy should be applied to Hellenic as to Italic. Bundling this in with all the other one-descendent Proto-Hellenic pages. -saph 🍏 18:55, 6 February 2024 (UTC)Reply

See also Wiktionary:Requests for deletion/Reconstruction#Reconstruction:Proto-Hellenic/pʰílos. -saph 🍏 18:57, 6 February 2024 (UTC)Reply

Proto-Slavic. Tagged for speedy (!) deletion in September 2023 with the reason "misspelling of *smьjati". This, that and the other (talk) 09:45, 8 February 2024 (UTC)Reply

Not sure what the issue is (or how this constitutes a 'misspelling'?). From what I can tell, OCS reflects this form. -saph 🍏 16:42, 8 February 2024 (UTC)Reply

Gaulish reconstruction with no descendants or derived terms. —Mahāgaja · talk 21:28, 18 February 2024 (UTC)Reply

Not a reconstruction, this is attested as avallo in Endlicher's glossary. Keep and move out of reconstruction namespace. -saph 🍏 13:11, 19 February 2024 (UTC)Reply

I have added the quote. User @Treehill moved this page and *aballo- formerly claimed an alternative spelling while admitting the fruit is attested. Against the cognitive dissonance I point out a recent discussion about lemmatization of Trümmersprachen attested in transcriptive mentions. @Mahagaja has to move back; avallo or avallon are likely options for me, but inconsistent with the tree name kept at *aballo-, which should perhaps be moved across the reconstruction namespace to *avallos. On the other hand one can normalize at b with the argument that the u in the manuscript stands for /β/ already, depending a bit on what Mahagaja thinks the Gaulish sound. Fay Freak (talk) 15:13, 19 February 2024 (UTC)Reply

If it's attested as avallo, then obviously that's what the entry needs to be named. The gloss poma (which seems to be plural, unless Medieval Latin has made a feminine noun out of it) isn't sufficient to know whether the fruit or the tree is being referred to. —Mahāgaja · talk 16:51, 19 February 2024 (UTC)Reply

@Mahagaja: I mean hyphenation has not developed in the 5th century CE to designate a stem or root to be expanded by endings. It is possible that the author of the glossary abstracted the endings away but we should add something for it to be comprehensible to modern readers; we do so in the inflection table anyway, but if it is the tree then the table is unfounded. Actually I tend to assume that it is the tree, due to the topographic content of the glossary overall, so move to avallos (where s is the inflectional ending that we don’t separate in titles by hyphens, and add the URL which I forgot when formatting the quote)! Otherwise we don’t know which is to be had in the mainspace and which a reconstruction! Fay Freak (talk) 17:57, 19 February 2024 (UTC)Reply

Whatever this is, it's unlikely to be a masculine ending in -os, since the Proto-Celtic word for the fruit is *abalom n and the word for the tree is *abalnā f. The -ll- (which can come from *-ln-) makes the tree more likely, while the -o makes the fruit more likely. But I see no reason at all to put the entry at anything other than the attested spelling avallo. —Mahāgaja · talk 18:05, 19 February 2024 (UTC)Reply

What about avallo-? Would be artificial though, nothing compellingly reasonable I see. -ll- is a weak argument for the tree, and even in spite of the topographic content of the glossary a Roman might have heard the name of the fruit and assumed it to equal the name of the tree, as in Latin usually only the ending differs for fruit and tree, just like I did it right now. And given the Proto-Celtic tree word both reconstruction pages are unreasonable and should be deleted. Due to the Romance meanings of poma I assumed an apple fruit first so we have it attested and the tree would be different. Fay Freak (talk) 19:04, 19 February 2024 (UTC)Reply

Dear all, So in French studies of the Gaulish language, there is a distinction between :

• the attested form of Late Gaulish avallo which is translated as poma in Endlicher's wordlist. avallo is a late form with spirantisation of b into a v, and the loss of the final -s at the nominative singular seen in Late Gaulish.
• *abalo-, for apple : is the reconstructed based on the above term and many descendents in Latin and onomastic studies of Celtic base toponyms -> acerabulus (leading to the French érable), the many Alon, Avalon, Valuéjols and Valeuil in France or known toponyms like Αβουλα (given by Ptolemee in Spain), Abulobrica or personal names like Abulus, Abalanis, Abbula, etc. Most probably neuter.
• *aballo-, for apple tree : most likely one of the few -o- words being feminine in Gaulish. Like in Brittonic, there was likely a distinction aballo- / aballo- for apple / apple tree in Gaulish, probably lost in the late form of the language. Attested in current placenames like Ollon, Avalleur, etc. and in former proper names like Abullius, Abelus, Abullus, etc.

The final "-" is part of the reconstruction of the Gaulish form as it indicates that we don’t know for sure whether it was -on and -os.

Kind regards, Treehill (talk) 20:04, 19 February 2024 (UTC)Reply

I hardly think the compiler of Endlicher's glossary listed a noun stem in his word list rather than a whole word. It's probably avallom with the final m in the process of being absorbed into the preceding vowel (as happened in Vulgar Latin) on the way to the final syllable being lost (as happened throughout Insular Celtic). (There was no ‑s because the word was neuter, not masculine.) The fortis ll is phonologically regular in Old Irish ubull, but not (AFAIK) in Gaulish. Perhaps there was influence from the word for the tree *aballā on the word for the fruit *abalom, making the latter pick up an unetymological fortis/geminate ll. —Mahāgaja · talk 20:10, 19 February 2024 (UTC)Reply

So according to Treehill there was actually an aballo in Late Gaulish writen avallo by a Roman due to Romance spirantization but the for earlier stages we can reconstruct two more pages? Thanks both for not shirking to add complications. 😣 Fay Freak (talk) 20:48, 19 February 2024 (UTC)Reply

Dear User:Mahagaja, I think you misunderstood me, avallo, is not a noun stem. It is the Gaulish word as attested by the compiler in a very late form of Gaulish. The "o" is the final letter. The two others are Classical Gaulish forms reconstructed from an earlier form of Gaulish whose ending is uncertain *abalo- (was likely an -on word) or *aballo- (likely ended in -os). So the three entries as they exist now are the only one attested. Writing avallon (or worse, avallom) would be an incorrect anachronism putting a late spirantised letter with an earlier form of the nominative singular.

Same with aballo User:Fay Freak as it would do the opposite -> a pre-spirantisation "b" of earlier forms of Gaulish, with a late nominative singular.

The three entries should remain as they are, as stated in the sources indicated : avallo, Reconstruction:Gaulish/abalo- an Reconstruction:Gaulish/aballo- are the only correct forms. Treehill (talk) 21:52, 19 February 2024 (UTC)Reply

My comment about the noun stem was in response to Fay Freak's suggestion of using avallo- with a hyphen as the page name. What is the evidence that Gaulish had an o-stem noun aballos? RC:Gaulish/aballo- also lists no descendants or derived terms and should therefore also be deleted. —Mahāgaja · talk 08:21, 20 February 2024 (UTC)Reply

Descendants were added as per your request. Treehill (talk) 10:14, 20 February 2024 (UTC)Reply

Same as below. —Caoimhin ceallach (talk) 13:19, 20 February 2024 (UTC)Reply

@Treehill, Mahagaja, Saph668, Fay Freak David Stifter wrote an article (in German) about the Celtic word for apple: An apple a day.... The Gaulish forms are discussed in the first section. About Endlicher's glossary he says:

Zu verschiedenen Anlässen habe ich argumentiert, dass Endlichers Glossar kein Zeugnis eines noch gesprochenen Spätgallischen ist, sondern dass es sich um eine gelehrte Wörtersammlung nach dem Sprachtod des Gallischen handelt, die zum größeren Teil auf Exzerpten aus spätantikem lateinischen Schrifttum beruht. Sein Zweck war die antiquarische Erklärung gallischer Ortsnamen bzw. Ortsnamenelemente (so auch Blom 2011: 177–181).

On various occasions I have argued that Endlicher's glossary is not evidence of a still spoken Late Gaulish, but that it is an erudite collection of words compiled after Gaulish had gone extinct, based largely on excerpts from late Latin literature. Its purpose was the antiquarian explanation of Gaulish place names or place name elements (see also Blom 2011: 177-181).

So while it's a valueble source and pertinent to the entry, I don't think we can use Endlicher's glossary as "attestations" in the strict sense. Therefore both the Gaulish words for apple and apple tree should be considered reconstructed. —Caoimhin ceallach (talk) 14:08, 20 February 2024 (UTC)Reply

No, what follows is that the attestation is not an additional Late Gaulish in addition to two earlier Gaulish reconstructions. A mention excerpted from an earlier mention. Hence one of the pages has to be moved to the mainspace. Unless we assume the part was reconstructed from toponyms in the beginning, hence the ending at a morpheme boundary. Fay Freak (talk) 15:02, 20 February 2024 (UTC)Reply

Delamarre writes in his Dictionnaire gaulois français :

*abalo-, *aballo- : pomme, pommier
Mot qui apparait dans le glossaire de Vienne sous la forme avallo, glosé poma avec déjà spirantisation du b en v.

*abalo-, *aballo- : apple, apple tree
This word appears in the Vienna glossary in the form avallo, glossed poma with the b already spiralized into v .

He then goes on listing descendents and cognate in other IE languages.

Savignac writes :

pomme nf, abalo- (neutre)
Mot donné par le Glossaire de Vienne sous la forme avallo, glosé poma « pomme », avec spirantisation du -b- en -v-, et déduit du nom propre Abal(l)us, comparable au viel irlandais ubull (de *ablu-), gallois afal, breton aval « pomme ». Remonte à une forme *ablu-/*abol- « pomme », d'où viennent le lituanien óbuolas, vieux slave abluko, vieux haut allemand apful, allemand Apfel, anglais apple « pomme », cf la glose thrace dinupula (de *kun-abola « pomme à chien »)

pomme nf, abalo- (neuter)
Word given by the Vienna Glossary in the form avallo, glossed poma apple, with spirantisation of the -b- into -v-, and deduced from the proper name Abal(l)us, comparable to Old Irish ubull (from *ablu), Welsh afal, Breton aval 'apple'. Goes back to a form *ablu-/*abol- "apple", from which come Lithuanian óbuolas, Old Slavonic abluko, Old High German apful, German Apfel, English apple "apple", cf the Thracian gloss dinupula (from *kun-abola "dog apple").

Best Treehill (talk) 17:32, 20 February 2024 (UTC)Reply

I read all of the above again and the Beer Parlour thread you linked to, but I'm still not following. Which page has to be moved to the mainspace and why? —Caoimhin ceallach (talk) 22:48, 20 February 2024 (UTC)Reply

Adding this term as it also has no descendants or derived terms. —Mahāgaja · talk 08:23, 20 February 2024 (UTC)Reply

As per your request, attested descendants and derived terms have been added. Treehill (talk) 10:15, 20 February 2024 (UTC)Reply

I reordered them and added Wikipedia links for placenames. Is this a good way of doing that? —Caoimhin ceallach (talk) 11:50, 20 February 2024 (UTC)Reply

Thanks, but since this form is attested in the form avallo it should be moved there (main space, not Reconstruction space). —Mahāgaja · talk 13:28, 20 February 2024 (UTC)Reply

Savignac writes :

pommier nm, aballo-
Mot déduit de NP Abellus, Abullius ... , du nom de lieu Abella, en Campanie, qualifiée de malifera « porte-pomme » par Virgile, d'où vient le terme aveline « grosse noisette », Aballo, devenu Avallon (Yonne) ... , comparable au vieil irlandais aball, gallois afall, breton avallenn « pommier ». Clairière-des-Pommiers *Aballo-ialon, Avaloiolum, Valuéjols (Cantal) et Valeuil (Dordogne, Eure), La Pommeraie Aballo, Avallon, Ollon (Drôme, Avalono, en 1 252), Marché-du-Pommier *Aballo-duron, Avalleur (Aube). Remonte à une forme celtique *aballo-, issue d'un plus ancien *abalnosl*abalna « pommier », dérivé en *-no- du nom de la pomme *ablu-/*abol-.).

apple tree nm, aballo-
Word deduced from NP Abellus, Abullius ... from the place name Abella, in Campania, qualified as malifera 'apple-bearer' by Virgil, from which comes the term aveline 'large hazelnut', Aballo, which became Avallon (Yonne) ... comparable with Old Irish aball, Welsh afall, Breton avallenn "apple tree". Clairière-des-Pommiers *Aballo-ialon, Avaloiolum, Valuéjols (Cantal) and Valeuil (Dordogne, Eure), La Pommeraie Aballo, Avallon, Ollon (Drôme, Avalono, in 1 252), Marché-du-Pommier *Aballo-duron, Avalleur (Aube). Traces back to a Celtic form *aballo-, from an older *abalnosl*abalna "apple tree", derived in *-no- from the name of the apple *ablu-/*abol-).

Best Treehill (talk) 17:32, 20 February 2024 (UTC)Reply

While the minds are focussed on Celtic apples, maybe anyone is interested in this: Wiktionary:Requests_for_moves,_mergers_and_splits#Reconstruction:Proto-Celtic/abalom_→_Reconstruction:Proto-Celtic/abūl. —Caoimhin ceallach (talk) 00:37, 21 February 2024 (UTC)Reply

Keep both. In the reconstruction namespace. —Caoimhin ceallach (talk) 02:12, 1 June 2024 (UTC)Reply

@AleksiB 1945, Illustrious Lock Now that there are codes for CAT:Proto-South Dravidian language and CAT:Proto-South Dravidian I language, can this entry be undeleted or recreated? Kutchkutch (talk) 03:50, 24 February 2024 (UTC)Reply

But it only has two cognates. Wait for now. Illustrious Lock (talk) 13:53, 24 February 2024 (UTC)Reply

Redundant entry of Old English ōs, the expected outcome in OE as a result of the Ingvaeonic nasal spirant law, inherited from Proto-North Sea Germanic *ǭs. @Æzelf89, Leasnam -- Sokkjō 03:44, 27 February 2024 (UTC)Reply

Counterpoint: Onswini is attested on the Collingham Stone(s), by a runic inscription reading "æfter, onswini cu(ning)", showing that an older varient of the name Ōswine was used during the early OE period, thus allowing an older form of the term ōs to be fossilized. Æzelf89 (talk) 04:10, 27 February 2024 (UTC)Reply

Archaic feature are sometimes retained in proper nouns, but regardless, it isn't evidence of it surviving in this term. -- Sokkjō 05:23, 27 February 2024 (UTC)Reply

That's fair. However, I don't feel comfortable simply deleting the page all together. Same with *ans. I think it would be better to note both of them as archaic, and state in both pages that both words would've only survived within given names and not in normal usage. Æzelf89 (talk) 15:05, 29 February 2024 (UTC)Reply

You're also duplicating content, which, even if it was a valid alternative, is not how to deal with alternative forms, see Old English gǣst. -- Sokkjō 19:57, 29 February 2024 (UTC)Reply

Fair, fair. I'll take care of that and edit them accordingly. Æzelf89 (talk) 20:00, 29 February 2024 (UTC)Reply

@Æzelf89, I moved to Onswini to ᚩᚾᛋᚹᛁᚾᛁ (ónswini), as that is how it is attested. It also alludes to that possibility of this simply being an intentional archaism, as is quite common with late inscriptions, cf. the Old Persian Artaxerxes inscriptions.

Also, I want to point out again that given and place names are dubious sources for term attestations. Names like Ansigar and Aslac are likely to be foreign names of immigrants to England, see Migrants in Medieval England, c. 500-c. 1500. The moneyer Aslac, which you cite as a source for *ās, was most certainly a Dane, as the coin which names him was minted under Danelaw for King Cnut. And the memorial coinage the moneyer Ansigar appears on was also likely a foreigner from continental Europe, see Edmund the Martyr#Memorial coinage. -- Sokkjō 19:38, 1 March 2024 (UTC)Reply

@Sokkjo, I would agree with your point about names such as Ansigar being almost certainly foreign, especially on account of their late attestation as indicated by their listings in the PASE Database.

Additionally, I did some research into the Collingham Stone inscription, and I found Multiple Sources (Hempl., George (1897), The Collingham Runic Inscription, & Stevens, William O. (1904), The Cross in the Life and Literature of the Anglo-Saxons) discrediting the reading of ᚩᚾᛋᚹᛁᚾᛁ; which were from approximately half a century after it was cited in Sievers An Old English Grammar, as translated by Cook (for which LeornendeEaldEnglisc supplied the wrong page number; it is mentioned on p86, not p128), and it would appear that no consensus now exists as to the reading of this inscription. However, having viewed a photograph of it myself here, I can say with confidence that the rune that the originally cited Stephens proposed to be the second ᚾ is most certainly ᚦ, and the first rune(s), proposed to be ᚩᚾ, are completely illegible due to weathering, save for part of a diagonal in the top left.

Therefore I submit that the two reconstruction entries, as well as that of ᚩᚾᛋᚹᛁᚾᛁ be deleted. Yeldred Gengo (talk) 09:55, 26 March 2025 (UTC)Reply

@Æzelf89, pinging you again. -- Sokkjō 17:01, 5 March 2024 (UTC)Reply

@Æzelf89 still hasn't responded. @Mahagaja, can you read the above and take action? -- Sokkjō 06:50, 31 May 2024 (UTC)Reply

@Sokkjo: I'm not sure why I didn't do anything about this 10 months ago when you asked, but I have now deleted RC:Old English/ons, RC:Old English/ans and RC:Old English/as. I didn't delete ᚩᚾᛋᚹᛁᚾᛁ (ónswini), though. —Mahāgaja · talk 14:37, 26 March 2025 (UTC)Reply

Only the names Giddinge and Gyddingas are attested, which might derive from a personal name *Gydda, but it's not certain. Best left to an etymology section. -- Sokkjō 05:49, 27 February 2024 (UTC)Reply

@Leasnam, pinging you. -- Sokkjō 06:52, 31 May 2024 (UTC)Reply

Like above, not strong enough evidence to reconstruct this term. -- Sokkjō 05:51, 27 February 2024 (UTC)Reply

This should be merged into Proto-Germanic *wulkną. -- Sokkjō 03:08, 28 February 2024 (UTC)Reply

Same as Reconstruction:Old English/ons above. -- Sokkjō 19:39, 29 February 2024 (UTC)Reply

Again, per above. -- Sokkjō 22:19, 29 February 2024 (UTC)Reply

A Proto-Ugric word with a solid Proto-Uralic etymology with no morphological change - we don't have these, but only terms which are not inherited from Proto-Uralic. — SURJECTION ^{/ T / C / L /} 08:49, 2 March 2024 (UTC)Reply

"to wade" > "to rise" does not seem solid. Aikio can rely on Khanty kül- "arise, go to shore". Okay. The different meaning still justifies an entry, whether as separate definition, ety or lemma, still. 2A02:3033:206:9573:F921:E292:BEAC:54D1 23:12, 2 April 2024 (UTC)Reply

It doesn't. The semantic shift can be recorded in the Proto-Uralic entry by giving glosses. — SURJECTION ^{/ T / C / L /} 06:13, 3 April 2024 (UTC)Reply

Dubious. See rational at Reconstruction:Proto-Germanic/sneganą#Etymology. --{{victar|talk}} 09:07, 2 March 2024 (UTC)Reply

And how does that invalidate the other proposed descendants...? -saph 🍏 00:49, 30 March 2024 (UTC)Reply

Delete. Much too uncertain. Sanskrit नाग (nāga, “snake”) is probably not related. —Caoimhin ceallach (talk) 11:00, 24 April 2024 (UTC)Reply

On what basis do you deny its cognacy? -saph 🍏 11:01, 24 April 2024 (UTC)Reply

See the etymology in the entry. —Caoimhin ceallach (talk) 11:48, 24 April 2024 (UTC)Reply

Again, *sneganą is not the only proposed relation here. I don't see a reason why *(s)nogHō > *snakô "snake" and *(s)nogós > *nāgás "snake" is so unreasonable. -saph 🍏 13:57, 24 April 2024 (UTC)Reply

It is not unreasonable in and of itself, but the other etymology is better because a) it accounts for the meaning "elephant", which is old, and b) it has a precise Lithuanian cognate. With that, the PIE basis for *sneg- is gone. —Caoimhin ceallach (talk) 14:37, 24 April 2024 (UTC)Reply

"naked" > "snake, elephant" seems like a bit of a stretch. -saph 🍏 15:56, 24 April 2024 (UTC)Reply

I wouldn't know why. Words meaning "naked" are often used to describe hairless animals. In English too, e.g. naked mole-rat. —Caoimhin ceallach (talk) 08:45, 25 April 2024 (UTC)Reply

As a descriptor. You wouldn't just say "naked," no? -saph 🍏 12:00, 25 April 2024 (UTC)Reply

In English, adjectives aren't freely substantivised. But in other languages this is normal and common. Take for instance Proto-Germanic *berô (“bear”, literally “brown”), *hasô (“hare”, literally “grey”), Latin cervus (“deer”, literally “horned”). —Caoimhin ceallach (talk) 12:36, 25 April 2024 (UTC)Reply

Has this ever happened in Sanskrit? Why would it develop into two only vaguely similar concepts, anyway? -saph 🍏 15:59, 25 April 2024 (UTC)Reply

Substantivization of adjectives is extremely common in Sanskrit, aided by the fact that adjectives and nouns aren't distinguished at all in inflection. I would even go as far as saying that most adjectives also have a substantival meaning. Eg. मार्ग (mārga, “road < wild (one)”), भैरव (bhairava, “name of Śiva < gruesome (one)”), ब्राह्मण (brāhmaṇa, “Brahmin < pertaining to Brahman”).

Anyway, नाग (nāga) originally meant "elephant". It was later used to mean "snake", but originally only in contexts where its mythological closeness to the elephant was to be emphasised. For example in the Mahābhārata the king of the snakes (nāga) is Airāvata, the divine elephant. (See P. Thieme, Zeitschrift für vergleichende Sprachforschung auf dem Gebiete der Indogermanischen Sprachen, vol. 78, no. 3/4, p. 178, fn. 1) —Caoimhin ceallach (talk) 10:30, 26 April 2024 (UTC)Reply

I wonder if the semantic shift 'elephant' > 'snake' really comes from them both being hairless, or if it has to do with the similarity of an elephant's trunk to a snake. —Mahāgaja · talk 10:53, 26 April 2024 (UTC)Reply

That's Wackernagel's theory too, from *nāgahasta lit. "snake-arm" or so, like the Latin anguimanus. But in Vedic nāga (almost?) always means "elephant", not "snake". —Caoimhin ceallach (talk) 11:19, 26 April 2024 (UTC)Reply

Delete, then. -saph 🍏 12:50, 26 April 2024 (UTC)Reply

@Caoimhin ceallach you can't just call for pages to be deleted based on your own personal speculation that it's "much too uncertain". This reconstruction is attested in reliable sources. At a minimum you have to provide other reliable sources that dispute this reconstruction, but even then that would not necessarily be enough to delete the reconstruction, only to problematize it. Brusquedandelion (talk) 00:25, 12 May 2024 (UTC)Reply

I added a reliable reference at the entry for नाग (nāga), contesting this reconstruction. Contrary to what you say, this article has no references. If you have references for Proto-Indo-European *(s)neg-, please add them. If you want more doubt that this reconstruction is correct, check Kroonen, Guus (2011) “*snego, *snakkaz 'snake'?”, in The Proto-Germanic n-stems: A study in diachronic morphophonology, Amsterdam, New York: Rodopi, →ISBN, page 333f., which also connects German Schnake (“mosquito”), Dutch snoek (“pike”), and Icelandic snagi (“pin”) with Proto-Germanic *snego ~ *snakkaz, which are all irreconsilable with a root meaning “crawl”. Some root meaning “pointed” or “stinging” is more likely. —Caoimhin ceallach (talk) 15:26, 12 May 2024 (UTC)Reply

Delete. I already changed some outdated etymologies referring to this root. Exarchus (talk) 10:00, 16 March 2025 (UTC)Reply

Being an ecclesiastical term, I don't think this can go back to PWG. OE and OHG were independent borrowings. OHG nunna is not attested till the latter half of the 9th century (?) Leasnam (talk) 20:24, 4 March 2024 (UTC)Reply

Ecclesiastical Latin still falls in the same timeframe as Proto-West Germanic. -- Sokkjō 20:51, 4 March 2024 (UTC)Reply

Parallel to Old High German priest (“priest”) from Old English prēost, the Old High German nunna appears to be a borrowing of Old English nunne, the Old English itself being an early borrowing from the Latin. Both OHG terms nunna and priest appear Late 9th c. after the Anglo-Saxon missions to Germany :\ Leasnam (talk) 22:19, 4 March 2024 (UTC)Reply

That's certainly possible. Hard to say morphologically. -- Sokkjō 00:00, 5 March 2024 (UTC)Reply

This is why it's also important to know the history :] Leasnam (talk) 00:55, 5 March 2024 (UTC)Reply

You could make the same argument for any Christian term, but that's certainly not the case. -- Sokkjō 00:59, 5 March 2024 (UTC)Reply

Do you think this may be the reason why we have so few reconstructions for religious terms ? We didn't overlook these...they're not candidates. Leasnam (talk) 01:03, 5 March 2024 (UTC)Reply

We? These entries weren't created previously because we did not have a Proto-West Germanic language on the project before. Why not just mention on the entry that the OHG attestation is late, and may instead be a borrowing from Old English? -- Sokkjō 01:27, 5 March 2024 (UTC)Reply

The idea seems to be that reconstruction should be backed up by evidence. EWAhd says that o > u is a High German internal development from ecclesiastical Latin, from a Late Latin Lallwort of rather different meaning, as also noted on nonnus and nun. That's not a reliable basis for any kind of reconstruction. If it were, you would have to consider the ecclesiastical usage possibly borrowed from (pseudo) PWG. DurdyWendy (talk) 16:00, 10 March 2024 (UTC)Reply

That change u > o is a general West Germanic a-umlaut feature, so likely it was borrowed from attested Medieval Latin variant nunna. That, or the underlying WG form is actually *nunnjā, from nonna + *-jā. -- Sokkjō 19:40, 14 March 2024 (UTC)Reply

Comment: Christianity existed in the West-Germanic Sprachraum since Late Antiquity, particularly the areas bordering the historical Roman Empire. The existence of words pertaining to Christianity in PWG is certain (Category:gmw-pro:Christianity). Simply being an ecclesiastical term does not preclude its existence in PWG. — Mnemosientje (t · c) 12:09, 19 March 2024 (UTC)Reply

Thought about this a bit and now explicitly voting delete. The case for this reconstruction seems too weak, even though I am sure at least some Christian terminology was borrowed into (late) PWGmc. — Mnemosientje (t · c) 11:41, 8 January 2025 (UTC)Reply

*b and *a are irregular reconstructions. Other problems: same west-central distribution as *bʰeh₂ǵos (in RfD), an Afro-Semitic comparison (not cited), three out of the six references do not support this root, two of the other three are agnostic about grouping, outdated in many parts, and the combination of meaning and sound sounds like you really serious with the comparative methadone. The Reconstruction note is a red-flag: "Various points suggest a post-Indo-European borrowing".

Meilet takes -culum for un suffixe de nom d'instrument comme en grec, but this is not clear cut because it either is as in Greek or it is not as in Greek and so irrelevant to the comparison. The suffix is supposed to be -culum, which would leave ba to be explained, so *bakklom has to be addressed. acclamo, acclaro, acclino, acclivis, accolo etc. are not clear counter examples. -ulum is a dumb diminutive. DurdyWendy (talk) 16:57, 14 March 2024 (UTC)Reply

Should be history merged into RC:Proto-West Germanic/wrīban and deleted. -- Sokkjō 02:29, 16 March 2024 (UTC)Reply

Delete. — Mnemosientje (t · c) 12:09, 19 March 2024 (UTC)Reply

Entry created for Gothic *𐌱𐌴𐍂𐌹𐌻𐍃 (*bērils) as a possible borrowed form to Old French baril, but if the Gothic existed at all, it could just be an intra-Gothic formation, like Proto-West Germanic *biril and Old Norse berill. @Mnemosientje -- Sokkjō 20:29, 17 March 2024 (UTC)Reply

Delete. — Mnemosientje (t · c) 12:09, 19 March 2024 (UTC)Reply

Uncertain etymology. Should be history merged into RC:Proto-West Germanic/slīban and deleted. -- Sokkjō 07:04, 18 March 2024 (UTC)Reply

Delete. — Mnemosientje (t · c) 12:09, 19 March 2024 (UTC)Reply

The etymology for this term isn't super convincing and I think should be consolidated to RC:Proto-West Germanic/fak, with no Germanic cognate outside West Germanic. -- Sokkjō 04:32, 20 March 2024 (UTC)Reply

Delete. —Caoimhin ceallach (talk) 11:43, 24 April 2024 (UTC)Reply

Unknown origin with no cognates outside West Germanic. History merge into RC:Proto-West Germanic/finkan and delete. Same for:

• RC:Proto-Germanic/funkô → RC:Proto-West Germanic/funkō
• RC:Proto-Germanic/fangijaną, RC:Proto-Germanic/fankijaną → RC:Proto-West Germanic/fangijan
• delete unneeded alternative RC:Proto-West Germanic/fankijan

-- Sokkjō 06:23, 21 March 2024 (UTC)Reply

• Delete. No reason to assume proto-Germanic antiquity (as the lemma is attested only in West Germanic languages). --Ghirlandajo (talk) 16:05, 6 January 2025 (UTC)Reply

Etymology of RC:Proto-West Germanic/dāhwā is uncertain. Should be history merged and deleted. -- Sokkjō 03:30, 23 March 2024 (UTC)Reply

Etymology of RC:Proto-West Germanic/hagatussjā is uncertain. Should be history merged and deleted. -- Sokkjō 06:09, 23 March 2024 (UTC)Reply

From what I can tell this reconstruction is valid when it comes to the form, it's what would be expected. However, we have no idea how far back it goes, thus I can neither concur nor argue against you. Haimariks (talk) 09:00, 23 May 2024 (UTC)Reply

Etymology of RC:Proto-West Germanic/skāk is uncertain. Should be history merged and deleted. -- Sokkjō 22:21, 23 March 2024 (UTC)Reply

If it's possible to history merge this with Latin fluo, I think that would be better than having a separate page. As described on the pages, there is not a consensus on the pre-Latin form of this verb. I also don't see any cognates within Italic. Furthermore, I am skeptical of the perfect *flūzai and past participle *flūssos. The redirect at Reconstruction:Proto-Italic/flowō should also be deleted. Urszag (talk) 23:48, 26 March 2024 (UTC)Reply

Reconstructed as an o-grade to already existing *h₁ep-. @ElkandAcquerne --{{victar|talk}} 04:46, 5 April 2024 (UTC)Reply

The *h₁op- entry could be folded into the *h₁ep- entry although Mallory and Adams only state that *h₁op- may derive from *h₁ep-. As well derived roots often have their own entries on Wiktionary. ElkandAcquerne (talk) 13:33, 5 April 2024 (UTC)Reply

Before you attempt to create any other Proto-Indo-European entries, please read WT:AINE. You'll see there that we reconstruct entries in e-grade. -- Sokkjō 02:34, 6 April 2024 (UTC)Reply

Delete and merge into *h₁ep-. This isn't significant enough that we have to make an exception to policy. -saph 🍏 03:07, 13 April 2024 (UTC)Reply

'Delete and merge' is also my opinion as an IP passing by, FWIW.--62.73.72.3 03:25, 19 January 2025 (UTC)Reply

Proto-Italic. One Latin descendant. Made up on a dubious etymology of Pokorny's. -saph 🍏 12:04, 9 April 2024 (UTC)Reply

I don't see any evidence of this form surviving into Old Dutch. @Preupellor -- Sokkjō 17:41, 13 April 2024 (UTC)Reply

I made this entry because its genitive plural (frōno) did survive (attested) as an adjective, and *frō or its variant *frōn were found back many times in toponyms. Therefore, my conclusion was that it must have been around in Old Dutch. Preupellor (talk) 17:57, 13 April 2024 (UTC)Reply

@Preupellor: The Old Dutch noun frōno descends from PWG *frauwjanō (“lordship”), itself, yes, built from the genitive plural of *frauwjō. -- Sokkjō 20:02, 13 April 2024 (UTC)Reply

@Sokkjo: Well, the noun frōno was never attested, but the adjective was attested in the Mfr. Reimb. A, and if that does in fact stem from PWG too, that still doesn't explain the various toponyms that use frōn- rather than frōno- ("fronland", "fronakre", "vronlo"). Well, unless *frōn was also directly inherited from PWG, rather than a form of *frō. Preupellor (talk) 20:11, 13 April 2024 (UTC)Reply

@Preupellor: PWG *frauwjanō (“lordship”) simply had more productivity than *frauwjō (“lord”), which was displaced by *hairiʀō (“lord, master”) in West Germanic, and why it didn't survive into Dutch. -- Sokkjō 20:40, 13 April 2024 (UTC)Reply

@Sokkjo: Sorry for the late reply, but I think we should just let a moderator decide over this. Preupellor (talk) 16:23, 30 April 2024 (UTC)Reply

@Preupellor: There are no moderators, but if there is anyone you'd like to comment, you can tag them using {{ping}}. -- Sokkjō 17:04, 30 April 2024 (UTC)Reply

@Sokkjo: Thank you. Honestly the lack of people who check other people's edit, and even though I don't completely agree with you in this case, I'm at least happy that you're at least trying to filter out the errors... Preupellor (talk) 17:39, 30 April 2024 (UTC)Reply

Maybe I'm just bored and meddling, but this appears to have solid descendents only in West-Germanic. The hypothetical Gothic term with Vulgar Latin *skīna is unreliable as the etymology is undecided between Frankish, Gothic or Lombardic. The Gothic reconstruction (@Fakename 2016, @Villager 2020) is not sourced. Apease a Zulu (talk) 16:47, 19 April 2024 (UTC)Reply

Agreed and that's probably the case for a lot of alleged Gothic borrowings. Looks like @Leasnam was the first to add them to the page. -- Sokkjō 16:39, 29 April 2024 (UTC)Reply

Move, this is indeed a secondary formation with only West Germanic evidence. In Proto-Germanic this would be a root-ablauting n-stem *skijô (weak stem *skin-), whence also Old English scīa (“shin”) from the strong stem. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 06:00, 18 July 2024 (UTC)Reply

No basis for assuming a Proto-Gallo-Romance form considering the absence of a cognate in Occitan (or Catalan). In all likelihood e(d)age was formed within Old French from e(d)et + -age and borrowed/calqued into Old Franco-Provençal. Nicodene (talk) 22:35, 19 April 2024 (UTC)Reply

Proto-Ryukyuan. Rationale was "not attested until Early modern japanese". Ultimateria (talk) 20:47, 26 May 2024 (UTC)Reply

I'm not sure what that's got to do with the price of tea in Okinawa, since (1) Japanese isn't among the descendants of Proto-Ryukyuan and (2) the whole point of reconstructed words is that they aren't attested. —Mahāgaja · talk 21:39, 26 May 2024 (UTC)Reply

It's that hanas- in Japanese is not attested until early modern Japanese, which means all Ryukyuan terms containing this word were all borrowed from Japanese. Chuterix (talk) 00:39, 6 June 2024 (UTC)Reply

I believe the differing reflexes of the initial p- in the Ryukyuan words suggest that it was separately inherited. —Soap— 16:15, 23 June 2025 (UTC)Reply

The descendants could have be phonetically modified to be like its inherited counterparts, though. Chuterix (talk) 22:14, 23 June 2025 (UTC)Reply

@Hazarasp: Is there any grounds for reconstructing a form that basically is a ghost? There's no process fn- > sn- anywhere I know of. Chuterix (talk) 00:36, 6 June 2024 (UTC)Reply

@Chuterix: see the etymologies for sneeze, snore and snort. This isn't isolated at all. Chuck Entz (talk) 05:30, 6 June 2024 (UTC)Reply

RFV passed. See the comment I agree with below. Chuterix (talk) 05:36, 6 June 2024 (UTC)Reply

• Delete, even if this variant is likely to have existed alongside the verb fnoren, there's no benefit in our having a Reconstruction entry for it when it has no descendants. —Mahāgaja · talk 06:04, 6 June 2024 (UTC)Reply

Delete per above. —Caoimhin ceallach (talk) 17:27, 6 June 2024 (UTC)Reply

Keep - but it does have a descendant, Middle English snore. If you look at the etymology of English snore, it's nice to have this logical step between Middle English snore and Old English fnora. Leasnam (talk) 14:20, 7 August 2024 (UTC)Reply

I think there's a good argument to be made for Old Norse Rín being borrowed from Proto-West Germanic *Rīn. @Mahagaja -- Sokkjō 19:42, 18 June 2024 (UTC)Reply

there's literally Latin borrowings from before 1 AD 😭 Haimariks Wandilaz (talk) 02:45, 10 October 2024 (UTC)Reply

Keep. You say "there's a good argument to be made"—has anyone made it? We don't delete entries based on one user's speculation. Where is the linguistic scholarship on the etymology of this word? Do linguists trace it back to Proto-Germanic or Proto-West Germanic? Cite some sources before you call for entries to be deleted. -Literally Satan (talk) 18:37, 3 January 2025 (UTC)Reply

RFD kept.

Proto-West Germanic. The Old English tryddian (< trieddian ?) is a variant of treddian (if not a sole survival of Proto-West Germanic *truddjōn). The West Frisian and Dutch forms are ultimately from the French. The French is a borrowing of a High German term parallel to Old High German trotōn. Leasnam (talk) 16:51, 21 June 2024 (UTC)Reply

Proto-West Germanic. Similar to above, this should be moved to Proto-West Germanic *treddōn, or maybe *traddjōn. Leasnam (talk) 16:55, 21 June 2024 (UTC)Reply

(Notifying TAKASUGI Shinji, Atitarev, HappyMidnight, Tibidibi, Quadmix77, Kaepoong, The Editor's Apprenice, Saranamd): additional pings: @Chom.kwoy, @Surjection, @Mahagaja, @Theknightwho Requesting a speedy deletion for all of them. Reconstruction:Proto-Koreanic/tàLá & Reconstruction:Proto-Koreanic/hyeng were deleted for this same reason. It's been brought up to this user directly (on Discord) since the start of this year that their reconstructions simply do not make sense in terms of Koreanic family history. The sources also do not support the exact constructions, except for *seki, but that one is shaky as well. This user has openly stated that they do not have that much experience with Koreanic languages, "I don't know much about Korean linguistics" at Wiktionary:Etymology scriptorium/2024/January § ᄣᅢ, let alone to create their own Reconstruction. They've also since been told to stop working on Koreanic entries for this reason. Said user has also been blocked before for similar issues. The entries in question include:

• Reconstruction:Proto-Koreanic/cenki: If anything it should be put at Reconstruction:Proto-Koreanic/ceNki, the current reconstruction doesn't make sense logically, and the source does not support it.
• Reconstruction:Proto-Koreanic/matV: This reconstruction solely exists to posit a Japanese borrowing, completely ignoring how Koreanic forms 맡 (math) & 마당 (madang) would point to an earlier *matVk, per basic historical Koreanic linguistic principles. The source also does not support it. Edit while writing: Since me bringing this problem up, it's been changed from *mati to *matV, but the same issues apply.
• Reconstruction:Proto-Koreanic/kwòmá: Having tones in the reconstruction is the first blatant problem. We're not even sure if Old Korean had tones, per Lee Ki-Moon and S. Roberts Ramsey in A History of the Korean Language (2011), let alone Proto-Koreanic. Additionally, the source cited does not have any PK reconstruction listed, solely talking about the initial vowel likely being /o/ due to vowel raising in Baekje, with Western Old Japanese likely being a borrowing from it. There's nothing that shows that the second vowel is /a/.
• Reconstruction:Proto-Koreanic/kopoLi: What is the <L> supposed to represent? Also, according to the source listed, Western Old Japanese borrowed from Baekje and not Proto-Koreanic. Vovin also states that the MK form ᄀᆞ옳 (kowolh) comes from *kopor but even that is shaky because it's missing the explanation for the /h/.
• Reconstruction:Proto-Koreanic/i-: There's no reason for this reconstruction to exist. Why is it a prefix or verb stem when it's neither of those in Koreanic languages? We can show a borrowing to Jurchen -> Manchu without a poor reconstruction; the reconstruction is also not found in the source cited.
• Reconstruction:Proto-Koreanic/seki: This is the only one that may be passable since it's directly cited by Vovin, but as @Chom.kwoy has noted, "It does seem like one possibility of a internal reconstruction but theres not enough evidence to be sure other than the tentative loanword in Japanese, so same situation as *mati. It could have been different suffixes added to the same root, like 셓〯 (sěyh) < *se-ih and 석〯 (sěk) < *se-k. Compare 서〯 (sě, “three”), 사ᄋᆞᆯ〮 (sàól, “three days”), 사릅 (salup) "three years old (animal)"

This has caused a severe problem with the quality of our entries, leading to more misinformation spread, as I've openly talked about beforehand. See: this Reddit thread for an example. This is precisely why I'm getting tired of cleaning up Koreanic entries, as stuff like this gets to stay up after creation even though they're blatantly incorrect, while we wait for the slow wheels to Wiktionary to get to them. I'm very annoyed to have to put this work in now, and this is why we have good editors leave. AG202 (talk) 03:31, 27 June 2024 (UTC)Reply

All Proto-Koreanic etymologies have been removed from their attested entries. :) Chuterix (talk) 03:48, 27 June 2024 (UTC)Reply

p.s. Not just that I don't know (much) about Koreanic linguistics, but also that I don't know much Korean at all. :/ Chuterix (talk) 03:51, 27 June 2024 (UTC)Reply

For the mistaken pK *matV(k), I was fooled by the mistaken assumption that -h < (*)-k was some sort of (place or locative) suffix (e.g. AFR's hypothesis that it is somehow cognate to OJ ko2 "place"). For the taLa that was deleted, it was based on some assumption that Vovin cited, that Old Korean did not have vowel harmony, although a look at Ramsey's History of Korean Language said: "For as we move back in time from Contemporary Korean to Early Modern Korean to Middle Korean, we find that the system of vowel harmony becomes more and more regular, with fewer and fewer exceptions to its rules."

The syllable <L> in *koporVk supposes it was either */r/ or */l/. Chuterix (talk) 03:57, 27 June 2024 (UTC)Reply

(Notifying TAKASUGI Shinji, Atitarev, HappyMidnight, Tibidibi, Quadmix77, Kaepoong, The Editor's Apprentice, Saranamd): : Re-pinging in case the initial ping did not go through. AG202 (talk) 15:23, 22 August 2024 (UTC)Reply

Delete all per proposer.--Saranamd (talk) 00:34, 23 August 2024 (UTC)Reply

Comment: I've deleted some of these which the creator nominated for speedy deletion with the rationale "created in error". Ultimateria (talk) 18:06, 27 August 2024 (UTC)Reply

The etymology for this terms is contested. Should me history merged into RC:Proto-West Germanic/hūdijan. -- Sokkjō 06:26, 1 July 2024 (UTC)Reply

Delete per proposer. —Caoimhin ceallach (talk) 23:19, 6 August 2024 (UTC)Reply

Keep, unclear rationale. Many things are contested, that is not enough reason to delete. Orel does reconstruct the lexeme for PG.--Anonymous44 (talk) 00:01, 30 March 2025 (UTC)Reply

Created by banned IP-user 24.108.18.81.

Cited sources:

• Michel Ferlus, "Formation of Ethnonyms in Southeast Asia" pdf;
  • Result of my attempt at source-verification: Ferlus reconsctructs the "etymon *k.raːw", yet he does not classify it as Austroasiatic.
• Sidwell, Paul and Felix Rau (2015). "Austroasiatic Comparative-Historical Reconstruction: An Overview." In Jenny, Mathias and Paul Sidwell, eds (2015). The Handbook of Austroasiatic Languages. Leiden: Brill.
  • Result of my attempt at source-verification: Sidwell & Rau (2015:340-365) reconstruct no such item as PAA *k.raːw "man"; relevant pages are 350 & 351

I found *k(ə)ra:w "human being" on page 646 of Frédéric Pain (2008). "An Introduction to Thai Ethnonymy: Examples from Shan and Northern Thai". Journal of the American Oriental Society. 128 (4): 641–662. Even, Pain (2008) is old & may not reflect current scholarly consensus in Austroasiatic linguistics. Erminwin (talk) 17:14, 3 July 2024 (UTC)Reply

I question this as a PIE suffix. "*(o)-neh₂" nouns are simply derived from r/n-stems + *-h₂ or *-eh₂, which are themselves late. --{{victar|talk}} 03:03, 7 July 2024 (UTC)Reply

What do you make of the fact that none of the examples listed on the page have associated r/n-stems? —Caoimhin ceallach (talk) 22:57, 6 August 2024 (UTC)Reply

As you and I have discussed in the past, the rebuilding of r/n-stems is a very common occurrence, and at times, they don't survive all, see the RC:Proto-Indo-European/dóh₃r̥. By the logic of the entry RC:Proto-Indo-European/-néh₂, we should have *(é)-nom, *(é)-nus, *(ó)-nis, *(é)-rom, etc. suffixes. --{{victar|talk}} 05:20, 8 August 2024 (UTC)Reply

I get that part and we can argue about that again some other day. No matter the origin, what's to say it didn't become a suffix in its own right? —Caoimhin ceallach (talk) 08:11, 8 August 2024 (UTC)Reply

The academic literature. The source the IP used on the entry, agrees, and cites their examples as secondary substantives of earlier *(∅)-nós adjectives, thus not their own suffix, cf. etymology of RC:Proto-Indo-European/déh₃nom. --{{victar|talk}} 09:39, 8 August 2024 (UTC)Reply

I had already checked that and no, that's not what it says... "Substantives have -no- and -neh₂-". Nothing about how they're derived. —Caoimhin ceallach (talk) 12:05, 8 August 2024 (UTC)Reply

The examples are under the bullet, "-no-‚ forms adjectives:". If you want to read more about no-stem substantives, see Ozoliņš and Fortson, as two examples. --{{victar|talk}} 19:51, 8 August 2024 (UTC)Reply

@Caoimhin ceallach: You misunderstand: Your question was: "... the fact that none of the examples listed on the page have associated r/n-stems?" It is answered by Victar's "... substantives of earlier *(∅)-nós adjectives". Beekes had aduced the adjectives and related nouns under the heading "Derived nouns" (2011:180). You doubt that we may read Beekes' statement as "Substantives have -no- and -neh₂-", but you are incorrectly stating that this is definitely not what was meant. For context: the chapter is followed by a discussion of Caland System and Vrddhi. The latter is topical right now in WT:ES#काम्बोज. By the way, one ("1") gives a *(∅)-nós derivation but there is no zero-grade to go on. Beekes and de Vaan agree on a possible *-no- suffix but Greek οἶος does not. What do you make of that? To me it is an isogloss sine qua non. Alisheva (talk) 20:12, 8 August 2024 (UTC)Reply

Thanks, now I get it. (As for "one", you're right, on the face of it it's not the same suffix. But this touches on something that's been bothering me for a while: in our entries we inseparably associate a suffix with a gender and an accent-ablaut paradigm. But all works on IE linguistics that I'm familiar with don't do that. They regard suffixes, genders, and accent-ablaut paradigms as more-or-less distinct phenomena which interact, giving rise to certain patterns. If we did that we'd say *óy-no-s is derived from root *ey- with a suffix *-no- and a static R(ó)-S(e/o) paradigm, declining in all three genders.) —Caoimhin ceallach (talk) 22:25, 8 August 2024 (UTC)Reply

Delete. Not for the reason given but because they're secondary to *-no-stems. —Caoimhin ceallach (talk) 22:28, 8 August 2024 (UTC)Reply

1. ^ Beekes, Robert S. P. (2011) Comparative Indo-European Linguistics: An Introduction, 2nd edition, revised and corrected by Michiel de Vaan, Amsterdam, Philadelphia: John Benjamins Publishing Company, §13.1.2, page 181
2. ^ Ozoliņš, Kaspars (2015) Revisiting Proto-Indo-European Schwebeablaut (PhD doctorate), Los Angeles: University of California, page 140
3. ^ Fortson, Benjamin W. (2004, 2010) Indo-European Language and Culture: An Introduction, Oxford: Blackwell

Should be merged into RC:Proto-West Germanic/failī. --{{victar|talk}} 01:50, 22 July 2024 (UTC)Reply

Delete. —Caoimhin ceallach (talk) 16:05, 7 October 2024 (UTC)Reply

Coptic. Tagged but not listed back in May 2021 by Absudar (talk • contribs) with the comment "ⲟⲩⲥⲉⲣϩⲁⲡⲓ isn't attested anywhere and it's a speculative reconstruction only used on Coptic Wiki Incubator". —Mahāgaja · talk 20:54, 27 July 2024 (UTC)Reply

• Deleted. —Mahāgaja · talk 17:29, 20 June 2025 (UTC)Reply

Seemingly late secondary formation around the Latin era. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 22:41, 9 August 2024 (UTC)Reply

Delete. --kc_kennylau (talk) 21:43, 10 August 2024 (UTC)Reply

Delete.--Urszag (talk) 04:55, 26 February 2025 (UTC)Reply

Obvious late formation in Latin of unclear origin, all possible cognates have the wrong velar so De Vaan says "etymology unclear". De Vaan doesn't even specify the PIt. verb formation for this one. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 22:41, 9 August 2024 (UTC)Reply

Delete. --kc_kennylau (talk) 21:42, 10 August 2024 (UTC)Reply

Delete.--Urszag (talk) 04:55, 26 February 2025 (UTC)Reply

-men could have been added at any time to a base apparent in Proto-Germanic *grasą. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 22:44, 9 August 2024 (UTC)Reply

Delete. --kc_kennylau (talk) 21:46, 10 August 2024 (UTC)Reply

Proto-Italic. Created by IP in 2021; linked from frons#Latin in 2020 (edit) by User:Kwékwlos. De Vaan is uncertain on both "frōns, frondis" and "frōns, frontis", so I think both etymologies should be removed entirely. --kc_kennylau (talk) 12:29, 10 August 2024 (UTC)Reply

Delete (there shouldn't be Proto-Italic entries with one descendant anyway). I edited the etymology at frons. Exarchus (talk) 10:45, 16 March 2025 (UTC)Reply

This *lian is a double entry. The verb is actually attested and also has a non-reconstructed entry lian. The reconstructed entry can be deleted. --DerRudymeister (talk) 10:56, 25 August 2024 (UTC)Reply

The verb is mentioned, but not used, in a Latin text/word list. Does that count enough for being attested? —Rua (mew) 14:14, 8 January 2025 (UTC)Reply

User:RwiTexx is creating a slew of questionable and poorly put-together PIE entries that should be deleted.

• Reconstruction:Proto-Indo-European/lewd-
• Reconstruction:Proto-Indo-European/sríHgos
• Reconstruction:Proto-Indo-European/gʰeldʰ-
• Reconstruction:Proto-Indo-European/(s)kweh₁t-

--{{victar|talk}} 01:27, 27 August 2024 (UTC)Reply

While I agree that some of the roots and lemmas I have created a page for are questionable (as noted in some of these pages), the “poorly put-together” part of your request stems from the fact that I do not have an advanced knowledge of Proto-Indo-European, rather laying the foundation for other users to improve and refine these pages.

RwiTexx (talk) 00:51, 28 August 2024 (UTC)Reply

Please don't do that. We don't need a bunch of incomplete junk entries. --{{victar|talk}} 05:58, 28 August 2024 (UTC)Reply

Keep *sríHgos, no opinion on the others. The correct answer to an incomplete entry is to fix the entry, not delete it, on principle. I would personally only nominate PIE entries for deletion if the validity of a reconstruction is doubtful (i.e. when we shouldn't be reconstructing a word in the first place). I cleaned up *sríHgos just now. @RwiTexx, despite Victar's irritations I see potential in you. Backing the reconstruction pages you make with reliable sources is a good baby step (LIV is fine, but there's much more out there, including the sources in Category:Proto-Indo-European reference templates). — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 23:23, 28 August 2024 (UTC)Reply

Added to *gʰeldʰ-, but it still could use more sources and discussion, though admittedly there is not much out there as this is a highly marginal ‘root’. Its main issue is a lack of research. Anyway I see no reason not to keep it now as long as the page clearly expresses the issues; we have plenty of questionable PIE reconstructions that are nonetheless academically important. Now that the other entries have been added to, there is no real motivation for deleting any of these. Use {{needsources}} etc. to mark them for cleanup instead.
That said, @RwiTexx, Victar is right at least in that you shouldn't be creating such minimal-effort PIE entries, especially not ones with no listed derivatives/descendants. Incomplete is OK; un-evidenced isn't. Also, a user should not have to visit the cited source to see the evidence for a reconstruction, and a single Pokorny reference is never enough (always use more recent sources in combination and follow WT:AINE). Note that edits to the reconstruction space are slow, with few active editors (and even fewer who know what they're doing), and protolanguages are notorious for attracting sloppy, amateurish linguistics, so it looks bad for Wiktionary when we do what you did. — 2600:4808:9C31:4800:A1B8:3B7D:966F:AF34 03:42, 31 August 2024 (UTC)Reply

Also expanded *lewd-, but it's still sloppy and needs help. — 2600:4808:9C31:4800:A1B8:3B7D:966F:AF34 04:37, 31 August 2024 (UTC)Reply

Keep *ḱwes-. I haven't looked at the others. I would like to echo @Mellohi!'s sentiments. @RwiTexx, it's probably better to do one root well be checking a number of references than to do five with little detail.

Does anyone understand why most sources don't reconstruct a final laryngeal? —Caoimhin ceallach (talk) 21:55, 1 September 2024 (UTC)Reply

LIV assumes *श्वस्ति (śvásti) was rebuilt to श्वसिति (śvasiti) by analogy with अनिति (aniti), but beyond that I'm not sure. — 2600:4808:9C31:4800:EC22:7EB3:B87D:9BD9 20:26, 7 September 2024 (UTC)Reply

@Caoimhin ceallach Can the RFD on this page be removed? Seems fairly well-sourced now. Dragonoid76 (talk) 18:49, 21 March 2025 (UTC)Reply

Keep *(s)kweh₁t-. —Caoimhin ceallach (talk) 20:21, 7 October 2024 (UTC)Reply

Keep *(s)kweh₁t-, now that the entry has been fixed. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 19:54, 19 February 2025 (UTC)Reply

RFD kept *ḱwes- due to being satisfactory verification with sources. Is *(s)kweh₁t- good to go too? — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 04:49, 22 March 2025 (UTC)Reply

Outside of LIV, there doesn't seem to be much enthusiasm about *(s)kweh₁t-. So if we keep it, there should be a note on this (Matasovic says "the reconstructed PIE root shape is highly unusual"). For Proto-West Germanic *skuddjan, Kroonen gives "no clear etymology", Orel gives "probably related to σκύζομαι (skúzomai, “to be angry”)" and EWN is "very uncertain" about further cognates. De Vaan tentatively suggests "*ku(o)t-" at 'quatio'. Exarchus (talk) 10:31, 22 March 2025 (UTC)Reply

@User:Antiquistik, you're creating Old Persian and Old Median entries for highly conjectural etymologies of Ancient Greek (etc.) entries. Terms with singular alleged borrowing should only exist in the etymology of the borrowing itself. --{{victar|talk}} 08:44, 4 September 2024 (UTC)Reply

@Victar That's fair. I'll avoid doing this in the future. Antiquistik (talk) 09:02, 4 September 2024 (UTC)Reply

@Victar Was nominating Reconstruction:Old Persian/Arbarāivāh a mistake though? Because it is attested in Akkadian too, and it seems that only the contraction should be deleted. Antiquistik (talk) 09:24, 4 September 2024 (UTC)Reply

No. The Babylonian could very well be from this, but the Ancient Greek looks like it's derived elsewise, perhaps from *arbah +‎ 𐎠𐎼𐎡𐎹 (a-r-i-y), but who knows. Either way, too conjectural for an OP entry. --{{victar|talk}} 09:45, 4 September 2024 (UTC)Reply

@Victar Regarding the other nominations for deletion, you did so after I had split the contractions of *Bayaçāh, *Raucakah and *Spiθradātah into separate pages, or had created them with separate pages for their contractions. And it appears that I had by accident omitted the Elamite and Aramaic forms of *Bagavahyah.

I have no objections to deleting the other pages, though I have corrected the abovementioned errors on these four pages specifically. Can you look at those four again and see if any of them can be salvaged? Antiquistik (talk) 10:02, 4 September 2024 (UTC)Reply

RC:Old Persian/Bayaçāh is highly dubious, both in the connection of the alleged borrowings, and reconstruction. I've removed the RfDs from RC:Old Persian/Raucakah and RC:Old Persian/Spiθradātah, with those being fixed. --{{victar|talk}} 22:52, 4 September 2024 (UTC)Reply

@Victar And RC:Old Persian/Bagavahyah? Antiquistik (talk) 03:00, 5 September 2024 (UTC)Reply

The second part of the Elamite borrowing looks like a different word entirely, for which Schmitt suggests Old Persian *vanyah (“victorious”), and Ancient Greek Βαγώας (Bagṓas) could just was well be from Old Persian *Bagava(h)uš, or some derivative of it. The Imperial Aramaic is probably the best evidence of the Old Persian, but that being the case, it should live in its etymology section. --{{victar|talk}} 23:18, 6 September 2024 (UTC)Reply

All Proto-Afroasiatic reconstructions with no descendants, or only one descendant.

(While we’re at it, this is not entirely a serious proposal, but—can we just ban Proto-Afroasiatic entry pages already? At present there’s absolutely no academic consensus on Proto-Afroasiatic reconstructions at all, or even on what principles to reconstruct from, or what sound correspondences are valid. Ehret reconstructs totally differently from Orel and Stolbova, and both of their works are riddled with basic errors regarding the comparanda they cite. Why do we have pages for Proto-Afroasiatic terms when there is no agreement on the most basic principles of how to reconstruct them?) — Vorziblix (talk · contribs) 06:58, 5 September 2024 (UTC)Reply

I kind of agree with just banning PA reconstructions. We already did do that informally for Proto-Cushitic. Thadh (talk) 12:11, 5 September 2024 (UTC)Reply

Delete the lazy pages. Fay Freak (talk) 16:05, 5 September 2024 (UTC)Reply

I can only speak to the last entry, which I created, but I see no reason to delete it. It's sourced appropriately. If we want to delete all PA reconstructions, fine, but I don't see why this is any different from other pages citing Ehret. It's also worth nothing that there are two other entries in the book that I didn't add because I wasn't sure how to handle them. Vergencescattered (talk) 19:29, 5 September 2024 (UTC)Reply

The entries should be convincing by themselves, not by the references. Formally banning any accepted proto-language won’t happen since Wiktionary could not even agree on formally banning Proto-Albanian, so we can only delete pages as unsupported, frequently though it be, and this is far from being supported. The reference and the particular entry in it is laughable, it cites a random misspelt/as spelt grammatically impossible fringe word from an Arabic dictionary that can have thousand other derivations from the root د ب ب (d b b) (not created on Wiktionary yet because there is so much in it), and a proto-Eastern Cushitic without discernible reflection. Fay Freak (talk) 00:28, 6 September 2024 (UTC)Reply

It's from an academic source, so I don't see how it's "laughable," and frankly I don't think you should be so rude to your fellow editors. Why should the site make decisions based on your word rather than on actual academic sources? Vergencescattered (talk) 03:29, 1 October 2024 (UTC)Reply

The problem with Afro-Asiatic is the astronomical time depth: attestation includes some of the earliest writing on the planet, but even back then the branches were already quite distant from each other. When you're looking at something so far away, the details get blurry. Chuck Entz (talk) 13:19, 1 October 2024 (UTC)Reply

Based on what I've read about the state of Proto-Afroasiatic reconstruction, I'd support removing all Proto-Afroasiatic entries from Wiktionary, though I'd be open to having that overruled if someone with a more informed opinion wants to argue for why they should be included.--Urszag (talk) 21:22, 7 September 2024 (UTC)Reply

No source for this reconstruction and with many errors. See etymology scriptorium. Exarchus (talk) 09:25, 6 October 2024 (UTC)Reply

While you can disagree on whether the root is palatalized or the inclusion of certain descendants, the reconstruction *Gʰrem- is very well documented, and an Rfd is an overreaction. Keep. --{{victar|talk}} 09:50, 6 October 2024 (UTC)Reply

The root is already at Reconstruction:Proto-Indo-European/gʰrem-. Exarchus (talk) 11:43, 6 October 2024 (UTC)Reply

Then what you should have created was an WT:RFM, not an Rfd. --{{victar|talk}} 19:36, 6 October 2024 (UTC)Reply

Ok then...

Topic moved to Requests for moves, mergers and splits Exarchus (talk) 08:05, 7 October 2024 (UTC)Reply

Proto-Italic. No non-Latin terms. De Vaan does list three Italic cognates, but they're loanwords from Latin. —Caoimhin ceallach (talk) 14:05, 7 October 2024 (UTC)Reply

Keep tentatively. De Vaan actually says that both vinciō (“to bind”) and vincō (“to win, defeat”) are from the exact same nasal-infix thematic verb. vinciō is backed up by direct derivative *winkelom, which alongside Latin vinculum exists a non-borrowed Umbrian derivative previślatu. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 21:40, 8 October 2024 (UTC)Reply

I already removed some forms (see edit comment) and the remaining roots are now at *dʰar-. So I think this can go as I see no basis for reconstructing this for PII. The thing to keep in mind is that *dʰāráyati was used as a basic present verb. Exarchus (talk) 22:03, 18 October 2024 (UTC)Reply

An "o-reduplicated athematic root aorist" is a complete fiction. I already placed the Sanskrit participle at *ǵʰers-, so this can go. Exarchus (talk) 10:44, 19 October 2024 (UTC)Reply

Move: Exarchus, you do this thing where you push to delete entries that instead should be moved -- in this case to RC:Proto-Indo-Iranian/ȷ́ʰaȷ́ʰársa. The entry is well sources, so you would have to explain why it's better suited there in the etymology. --{{victar|talk}} 09:02, 24 October 2024 (UTC)Reply

How do you reconcile the long vowel in Sanskrit: जाहृषाण (jāhṛṣāṇá)? --{{victar|talk}} 20:04, 24 October 2024 (UTC)Reply

It probably doesn't have a phonological explanation. There's an entire book written about these forms: Krisch (1996) Zur Genese und Funktion der altindischen Perfekta mit langem Reduplikationsvokal. —Caoimhin ceallach (talk) 23:01, 24 October 2024 (UTC)Reply

Not a fan of a random long vowel. If not an o-reduplicated form, which do exist (ex. poposcī), maybe a secondary reduplication from a lengthened-grade aorist. --{{victar|talk}} 01:51, 25 October 2024 (UTC)Reply

It's not random. It just doesn't have a clean, simple explanation (yet). Unless you can properly motivate it I can't take your idea seriously. —Caoimhin ceallach (talk) 12:52, 25 October 2024 (UTC)Reply

Than come up with an explanation if it isn't random. --{{victar|talk}} 10:14, 28 October 2024 (UTC)Reply

I had missed this discussion completely. Firstly I would guess the o-reduplication in Latin 'poposci' was secondarily formed from the present (the -sc- also indicates this origin).

About the long vowel in Sanskrit perfects: some are phonologically regular, caused by initial root laryngeal. This analogically spread to some others, with 1) rythmic tendencies (long root vowel avoids long reduplication) and 2) apparently also a semantic tendency with long reduplication occurring more often in perfect verbs with a present meaning. Source: {{R:iir:Kummel:2000|page=21-22}}. In the case of जाहृषाण (jāhṛṣāṇa), this is indeed used with the meaning of a present (see page 604). Exarchus (talk) 17:50, 29 October 2024 (UTC)Reply

You're (intentionally?) misunderstanding. As Exarchus said, some principles are known, but the exact distribution of long syllable-reduplication is not well understood. That doesn't mean it's random. It also doesn't mean we should plump for the first ad-hoc explanation that we come across. —Caoimhin ceallach (talk) 22:20, 29 October 2024 (UTC)Reply

"a secondary reduplication from a lengthened-grade aorist" Is there any example of that? Such a form would rather have short reduplication, because of the rythmic tendency I mentioned above.

I'll add that reduplication with long vowel is particularly frequent in roots with form 'CarC' or 'va(R)C', so also हृष् (hṛṣ). In Avestan, reduplication with 'ā' occurs only in forms with 'r' in the next syllable (remarkably including the 3pl. ending). Exarchus (talk) 19:51, 29 October 2024 (UTC)Reply

Delete. Only one descendant and this is one of these cases. But why call it reduplicated stative. Shouldn't we decide between stative and perfect and then stick to it? —Caoimhin ceallach (talk) 12:10, 24 October 2024 (UTC)Reply

We don't have policy against single descendants entries in PII. Also, where would you like me to put the sources that support descent from *ǵʰers-, if not directly after it? --{{victar|talk}} 20:04, 24 October 2024 (UTC)Reply

I think it's better not to make up entire stem formations based on single forms. But if you insist, put the standard explanation first with references and then your own ideas, so as not to create the false impression that it's sourced. —Caoimhin ceallach (talk) 23:07, 24 October 2024 (UTC)Reply

You didn't answer my question. Where should I have put the source citing the root, if not directly after the root? --{{victar|talk}} 01:51, 25 October 2024 (UTC)Reply

You don't need me to help you with this.

From Proto-Indo-European *ǵʰe-gʰórs-e, perfect of *ǵʰers- (“to stand on end”). Alternatively, from Proto-Indo-European *h₁é-ǵʰo-ǵʰers-t ~ *h₁é-ǵʰo-ǵʰr̥s-ént, o-reduplicated athematic root aorist of the same root. —Caoimhin ceallach (talk) 12:48, 25 October 2024 (UTC)Reply

Cool, so exactly where I had it. --{{victar|talk}} 08:00, 28 October 2024 (UTC)Reply

What do you mean? Putting references at the end of the sentence "From Proto-Indo-European *h₁é-ǵʰo-ǵʰers-t ~ *h₁é-ǵʰo-ǵʰr̥s-ént, o-reduplicated athematic root aorist of *ǵʰers- (“to stand on end”)" implies that the references support all of the statements in the sentence: not just the part about "*ǵʰers- ", but also the part about "*h₁é-ǵʰo-ǵʰers-t ~ *h₁é-ǵʰo-ǵʰr̥s-ént, o-reduplicated athematic root aorist". Putting references at the end of the sentence "From Proto-Indo-European *ǵʰe-gʰórs-e, perfect of *ǵʰers- (“to stand on end”)" is therefore different. Apologies if I've misunderstood the edit history, but if I've understood correctly, it's concerning that you don't understand or won't acknowledge the significant difference between Caoimhin ceallach's suggested placement of references and the way that you had placed the references.--Urszag (talk) 08:58, 28 October 2024 (UTC)Reply

"Putting references at the end of the sentence implies that the references support all of the statements in the sentence". No it doesn't. There are plenty of cases, if not most cases, where a source only cites a root, without reconstructing a lemma, and if the root is at the end of the etymology, as it usually is, that's where you would put the reference. This is common practice. --{{victar|talk}} 09:28, 28 October 2024 (UTC)Reply

Thank you for clarifying your viewpoint. It appears to me that Caoimhin ceallach doesn't agree with that position (and I don't either). On Wikipedia, at least, the content guidelines on citations specify that material should not be positioned closely before an in-line citation if it doesn't come from the cited source (since it could mislead a reader into thinking it is supported by that source). I see no reason Wiktionary should follow a different policy in this regard: while it's true that Wiktionary is different in allowing "original research" to some degree, that doesn't make it any less important to use formatting that clearly identifies what material ought to be attributed to the cited sources.--Urszag (talk) 09:44, 28 October 2024 (UTC)Reply

What you're referring is more along the lines of citation bundling -- which this also isn't. Simply the part that is being sourced is at the end of the sentence. --{{victar|talk}} 10:09, 28 October 2024 (UTC)Reply

How does "If a sentence or paragraph is footnoted with a source, adding new material that is not supported by the existing source to the sentence/paragraph, without a source for the new text, is highly misleading if placed to appear that the cited source supports it" not apply? (While this talks about adding new material, the same concern applies whenever a sentence has material that isn't supported by the in-line citation that follows it.) If for some reason you are unwilling to restructure your sentences in cases like this, I would consider it preferable to list the sources as general references rather than in-line references. While the heading title "Further reading" is not ideal as a means of clearly attributing sources, I think it is adequate: also, I feel like giving sufficient attribution is generally not as important a concern as avoiding misleading attribution in this context: "from root X"-type information often is more or less common knowledge among scholars of a language group, anyways, so there is less risk of falling into plagiarism than when writing etymology sections based on scholarly work that deals with less widely repeated details of etymologies.--Urszag (talk) 10:27, 28 October 2024 (UTC)Reply

Applying Wikipedia rules to Wiktionary isn't always feasible, as you alluded to earlier. We're not building a narrative with multiple paragraphs -- these are mechanical chains of descent, starting with the most recent form and ending with the earliest, which is often a root. If we adhered to the strict interpretation you're suggesting, we'd either have to restructure the etymology to be ass-first, or as you suggest, forgo inline sources altogether. Neither are particularly attractive solutions.

Maybe there is a technical solutions, with mouseover highlighting of portions of text that support each element of the chain. --47.23.164.194 20:03, 30 October 2024 (UTC)Reply

On pages for reconstructed words like this, I see no pressing need for the etymology to be presented in a single short sentence: the etymology is one of the major things most visitors to the page will likely be interested in. If it requires more than one sentence to make it clear which part of the etymology is supported by academic sources, and which part is uncited speculation by Wiktionary editors, that seems well worth it.--Urszag (talk) 20:55, 30 October 2024 (UTC)Reply

In most cases, leaving out the uncited speculation by Wiktionary editors is a clear improvement: shorter, and more accurate. Very much in the case of *Háȷ́ʰāȷ́ʰaršt. Exarchus (talk) 21:27, 30 October 2024 (UTC)Reply

As for the objection about presenting the etymology backwards, I can see why sticking to a consistent formula might seem superior. But my viewpoint is that ultimately, it is baked into the design of Wiktionary that this is a dictionary aimed at being used by humans reading it, not at being used as a formally structured database, so using an unconventional but understandable format for reasons of clarity seems fine to me. Humans can understand "From X, via Y" or similar structures. Having our entries include more database-like content would be a good thing, and that can be included in the back end via templates, but I don't think it's necessary to be so inflexible about how the free text portions of our etymology sections are formatted.--Urszag (talk) 21:15, 30 October 2024 (UTC)Reply

A citation following a statement should support that statement, not contradict it. It is ludicrous that we are arguing about this. —Caoimhin ceallach (talk) 00:19, 29 October 2024 (UTC)Reply

@Victar This is precisely the same issue that was raised in Wiktionary:Beer parlour/2024/October#User:Victar and false citations, where you place references at the end of a sentence which also contains your own, unsourced views, thereby creating the misleading impression your original research is sourced. You were offered a solution there, too, and yet here you are still refusing to get the point. I am blocking you for a week, and you are lucky it is not the longer block Benwing2 argued for (). If this happens again, the block length will be for a month, and after that indefinite; we cannot tolerate this kind of academic dishonesty here. Theknightwho (talk) 00:27, 30 October 2024 (UTC)Reply

@Caoimhin ceallach "Shouldn't we decide between stative and perfect and then stick to it?" Good idea, I'm fine with 'perfect', which is probably the correct term to complement 'present' and 'aorist'. Exarchus (talk) 17:54, 29 October 2024 (UTC)Reply

I would go with that too, but I admit that I'm not familiar with the arguments for calling it 'stative'. —Caoimhin ceallach (talk) 21:56, 29 October 2024 (UTC)Reply

You can use 'imperfective'/'perfective'/'stative' or 'present'/'aorist'/'perfect'. But 'stative' is also used for other things, like "statives in *-éh₁-" or some verbs reconstructed by LIV with 'stative endings' (*-or instead of *-tor). Exarchus (talk) 22:12, 29 October 2024 (UTC)Reply

All descendants turned out to be something else (aorist subjunctive, or causative). Exarchus (talk) 12:34, 8 November 2024 (UTC)Reply

I moved the Iranian to RC:Proto-Iranian/ǰāmáyati. Where did you move Vedic Sanskrit गमति (gámati) to? --{{victar|talk}} 08:37, 12 November 2024 (UTC)Reply

I've added it at *gʷémt for clarification. Exarchus (talk) 09:08, 12 November 2024 (UTC)Reply

Incidentally, {{R:iir:Lipp:2009}} cites Old Avestan 𐬘𐬀𐬨𐬀𐬌𐬙𐬍 (jamaⁱtī) as a 3sg.conj.act., which supports a thematic verb reconstruction, as does {{R:ira:ESIJa}}, who reconstructs a thematic present. --{{victar|talk}} 03:17, 13 November 2024 (UTC)Reply

Ehm, German 'Konjunktiv' and English 'subjunctive' are the same thing. The Russian dictionary does appear to give a thematic present, but that seems to be an outdated analysis (from Pokorny p.464?). Exarchus (talk) 08:36, 13 November 2024 (UTC)Reply

Whoops, I knew that. --{{victar|talk}} 19:12, 13 November 2024 (UTC)Reply

If it would be a present subjunctive, then that would exactly indicate an athematic verb.

And now I see that Lipp actually gives "Konj.Aor." so really no discussion here. Exarchus (talk) 19:41, 13 November 2024 (UTC)Reply

As final reference on this topic, see the article "Vedisch gámati" in Karl Hoffmann, Aufsätze zur Indoiranistik, p.384-6 Exarchus (talk) 20:46, 14 November 2024 (UTC)Reply

I can only guess why you thought a subjunctive 'jamaitī' would point to a thematic verb, but maybe you were mixing up with Proto-Germanic. The 'i' in Avestan is just an epenthetic vowel.Exarchus (talk) 13:33, 15 November 2024 (UTC)Reply

Actually, ESIJa's use of references is quite dubious (or at least sloppy): at page 118 of vol.3, they give Bartholomae's analysis of the Avestan root gam-, giving (apparently) a root present (class 1) and a thematic present (class 3). Then they give as reference "", but if you look up what Kellens says (Liste du verbe avestique, p.19, and also in Le verbe avestique, p.353), you'll see he gives those forms (like 'jamaitī') as aorist. Exarchus (talk) 16:41, 13 November 2024 (UTC)Reply

Opinions on ESIJa notwithstanding, a 3sg.pres.subj. is still a viable possibly. Why omit it? --{{victar|talk}} 19:12, 13 November 2024 (UTC)Reply

Delete. —Caoimhin ceallach (talk) 08:50, 13 November 2024 (UTC)Reply

Proto-Malayo-Polynesian.

There are several things about this that make me very nervous:

1. Blust's Austronesian Comparative Dictionary doesn't reconstruct this for Proto-Malayo-Polynesian, but for Proto-Phillipine. The existence of that branch is disputed, but this suggests that the term may have arisen locally after Proto-Malayo-Polynesian had split up.
2. It was added by AleksiB 1945, who mostly edits in Dravidian languages, and has been blocked for editing in languages they don't know.
3. The etymologies of Proto-Tai *kluəjꟲ and Proto-Mon-Khmer *t₁luəjʔ were changed to make them direct borrowings from this term
4. It's apparently not used in the Malayo-Polynesian languages to refer to the edible fruit of the true banana Musa × paradisiaca, but to the plant of the inedible abaca or Manila hemp, which is used for the fibers in its stalks.

While I don't dispute the possibility of some contact between some Austronesian or pre-Austronesian language with ancient proto-languages in Southeast Asia, I wonder about the relative timing of all of these proto-languages that would make direct borrowing possible between them. Each of these proto-languages only existed for a period of time before splitting into daughter languages.

I should also mention that the existence of Proto-Mon-Khmer is disputed, but Wiktionary has kept it for now since there hasn't been enough work yet on Proto-Austroasiatic to replace the large body of published reconstructions for Proto-Mon-Khmer. That doesn't have any effect on the matter at hand, so we can ignore it in this discussion.

Pinging @Austronesier, who knows a lot more about this than I do. Chuck Entz (talk) 17:33, 15 November 2024 (UTC)Reply

The whole thing is based on Donohue & Denham (2009), which is cited with a bare URL in the "Further reading" section. Donohue & Denham's wild speculations that connect various entirely unrelatable forms to *qaRutay can be found on pp.301ff. This is really, really bad amateur etymologizng; even for borrowings, such an anything-goes approach is below every standard of historical-comparative linguistics. Moreover, the list contains spurious form like "Batak" (which Batak variety?) "galo". Probably they meant Toba Batak gaol, but that doesn't fit their pseudo-reconstruction "kaloy". Looks quite Greenbergish, no? To quote Blust's comment in his ACD entry: "The highly speculative ethnobotanical hypothesis they propose based on this is simply not supported by scientifically reliable linguistic evidence.".

*qaʀutay is a good reconstruction for Proto-Phillipine. The cognate set listed by Blust has two non-trivial sound correspondences (Tboli k vs. non-phonemic initial glottal stop in the rest; and l ~ g ~ h, all regular for *R). This, together with the non-connected areas of attestation point to inheritance rather than borrowing. If one accepts Proto-Philippine, then *qaʀutay must be reconstructed at this level, since Casiguran Dumagat, Hanunoo, and Tboli/Blaan belong to three different subbranches (North Luzon, Greater Central Philippine, South Mindanao). If one doesn't, it would be assigned to the level of the immediate proto-language ancestral to North Luzon, Greater Central Philippine, South Mindanao. According to Laurie Reid, who rejects Proto-Philippine and considers the branches on the Philippines to be first-order subgroups of Malayo-Polynesian, this should be Proto-Malayo-Polynesian.

Proto-Philippine is treated as a valid clade Wiktionary, which I think is good practice in spite of some criticism out there. So let's move *qaʀutay to Proto-Philippine and throw all the other junk from the entry. –18:39, 15 November 2024 (UTC)

User @2.207.102.112 suggested this to be removed at the etymology scriptorium. Exarchus (talk) 20:19, 15 November 2024 (UTC)Reply

Keep: {{R:nl:EWN|vals}} goes into the etymology from Latin falsus +‎ *-isk, and *falskōn +‎ *-ī is a pretty straightforward formation. --{{victar|talk}} 07:03, 16 November 2024 (UTC)Reply

They don't suggest a PWG date of borrowing, as they give Old French as possible intermediary. Exarchus (talk) 22:33, 19 November 2024 (UTC)Reply

Very few published works address PWG at all. Certainly, the word was at least reinforced by Old French, as were most Latin borrowing. I find it highly unlikely the every Old Germanic branch independently borrowed Old French fals while also merging it with suffix *-isk. --{{victar|talk}} 22:01, 20 November 2024 (UTC)Reply

The -sch in German is said to be from Middle Dutch. Exarchus (talk) 23:10, 20 November 2024 (UTC)Reply

That's a pretty rare direction of borrowing. {{R:EWddS}} lists only two examples Middle High German terms deriving from Middle Dutch, hübsch and Klippe. Kluge claims German falsch is borrowed from Old Northern French falske, but I can't find any evidence of such a form.

By all accounts, the verb appears to be older (see *falskōn), which would point to a deverbal adjective, or at the very least, an adjective corrupted by the verb.

--{{victar|talk}} 00:42, 21 November 2024 (UTC)Reply

As has been said at the etymology scriptorium: "We can even in see how it spread from Dutch and West Central German eastward (namely through Veldeke, cf. Pfeifer)." Exarchus (talk) 09:37, 21 November 2024 (UTC)Reply

The relevant forms are rather reconstructed to be from a reduplicated present, see *h₁wegʷʰ-. Exarchus (talk) 22:11, 19 November 2024 (UTC)Reply

Delete: I support deleting all alternative form entries for PIE roots. --{{victar|talk}} 22:04, 20 November 2024 (UTC)Reply

Redirect. In general, I support making alternative form entries in Reconstruction space into hard redirects. Since Reconstruction space entries are language-specific (unlike main space entries), hard redirects do no harm. —Mahāgaja · talk 08:44, 21 November 2024 (UTC)Reply

The point here is that *h₁ewgʰ- might simply not have existed. Exarchus (talk) 13:09, 21 November 2024 (UTC)Reply

Many notable reconstructions "might not have existed". We delete them if they are not justified, not if they have a justification that isn't absolutely watertight. — 2600:4808:9C31:4800:9CE6:E253:38E7:60AE 23:07, 26 November 2024 (UTC)Reply

To clarify: Pokorny reconstructs this as eug^u̯h- (p.348), but nowadays everyone gives this is as reduplicated *h₁eh₁ug⁽ʷ⁾ʰ-, so no need for any redirect I'd think. Exarchus (talk) 16:01, 30 November 2024 (UTC)Reply

Exists solely to explain Latin ānser, despite at least two problems: the (presumed analogical) disappearance of Proto-Italic *h- and the addition of a suffix, either Latin/Italic *-eros or *-ēr or thematicization of s-stem as *-ezos. We cannot know at which stage either of these modifications took place; for all we know it was any of the following in Proto-Italic: *hānseros, *ānseros, *hānsezos, *ānsezos, *hānsēr, *ānsēr, *āns; or none of the above. — 2600:4808:9C31:4800:9CE6:E253:38E7:60AE 23:03, 26 November 2024 (UTC)Reply
Edit: I have changed my mind about the suffix, as the s-stem explanation seems most likely, but then this needs to be explained on the entry. The problem of chronology for the initial consonant and thematicization persists though. — 2600:4808:9C31:4800:9CE6:E253:38E7:60AE 20:04, 27 November 2024 (UTC)Reply
Edit 2: Is the lengthening of *-ans- to *-āns- of Proto-Italic age? If not, the page should be moved to Reconstruction:Proto-Italic/hans even if my above concerns are disregarded. — 2600:4808:9C31:4800:9CE6:E253:38E7:60AE 20:12, 27 November 2024 (UTC)Reply

Similar issue to the above: there is little reason to assume this suffixed form existed at the Proto-Italic stage, as it could have been *hesteros or *hestēr, or something unexpected like *hestern. Having *hesternos as an entry wrongly implies that we know (or linguists are in agreement) about the particular morphological and chronological development of the Latin word, and perhaps worse implies that any Italic cognates if they existed should be expected to fit this form. — 2600:4808:9C31:4800:9CE6:E253:38E7:60AE 23:54, 26 November 2024 (UTC)Reply

Duplicate of Reconstruction:Proto-Germanic/skaubaz that is left inappropriately as a stem. Merge histories I guess. — 2600:4808:9C31:4800:C586:BBB5:F031:8DCF 07:14, 8 December 2024 (UTC)Reply

I propose to delete the Proto-Eastern Caucasian etymologies due to the lack of general reconstruction and the lack of elaboration of Wiktionary:About Proto-Northeast Caucasian.

• Reconstruction:Proto-Northeast Caucasian/bɔrc̣(V)
• Reconstruction:Proto-Northeast Caucasian/marq̄ʷa
• Reconstruction:Proto-Northeast Caucasian/mɔč̣č̣(V)
• Reconstruction:Proto-Northeast Caucasian/mɔRc̣c̣i ~ mɨRc̣c̣i
• Reconstruction:Proto-Northeast Caucasian/ruƛ̣(V) ~ *ruƛ̣ƛ̣(V)
• Reconstruction:Proto-Northeast Caucasian/rɔḳʷ(ə)
• Reconstruction:Proto-Northeast Caucasian/weršē
• Reconstruction:Proto-Northeast Caucasian/x̱änɦï
• Reconstruction:Proto-Northeast Caucasian/ḳwēmṭī
• Reconstruction:Proto-Northeast Caucasian/ɬ:ɔn
• Reconstruction:Proto-Northeast Caucasian/ɢ̄ə

I also very much doubt that the reconstructions of Johanna Nichols, Wolfgang Schulze, Evgeny Bokarev and Bokar Gigineishvili were used. For example, they use the reconstructed phoneme PNEC or PND *ʒ to produce different reflexes in the descendants: PAA *c̣c̣, Lak z, PN *ṭṭ, PL & PD *cc, Khinalug c̣ and PT *c. It is strange that this phoneme gives such "heavy" reflexes in descendants. Because most of all it resembles the reflexes of the Chamalal dialects, for example, *bac̣a > Gigatl (1) bazza ‹базза́›; Chamalal (2) bac̣a ‹ба̀цӏа›, Gadyri (3) badda ‹бадда› (“wolf”) - User:ɶLerman/Chamalal dialects.

Reflex (1) more closely resembles (see above) the Lak reflex z, reflex (3) resembles the Proto-Nakh reflex *ṭṭ, and reflex (2) resembles all other reflexes.

ɶLerman (talk) 06:39, 19 December 2024 (UTC)Reply

Lack of an About page is not a general reason to delete reconstructions, we have quite a lot of proto-languages without one. We might want to adjust the transcription here in several ways of course (at least the inconsistency between "ɬ:", "č̣č̣" and "ɢ̄" for various geminates probably should be resolved).

Voiced > geminate is a common recurring development across East Caucasian: the "plain tenuis" series (notated *p *t *c…) is uniformly in fact aspirated, which allows voiced consonants to devoice, producing pure tenues; these then have an areal tendency to be enhanced by gemination to distinguish them from the aspirates. Such correspondences are abundantly demonstrated e.g. just within Dargwic or Lezgic.

A position of "there is no reconstruction, therefore this actually proposed and sourced reconstruction is in fact not a reconstruction" seems, uh, circular to me. Can you elaborate better what are your actual objections? If you just want to integrate more sources than just NCED, you are free to do so, including moving any of the entries from their NCED reconstructions to any forms closer to Nichols', Schulze's etc. — as in your first-mentioned example seems to have actually been done, even. But why this would have to involve deleting anything is unclear to me. Surely you are not insisting treating e.g. the various East Caucasian words for 'moon' as not even cognate? --Tropylium (talk) 14:56, 28 December 2024 (UTC)Reply

The etymology is uncertain beyond West Germanic. Should either deleted or, preferably, merged into RC:Proto-West Germanic/skirmi. --{{victar|talk}} 19:41, 19 December 2024 (UTC)Reply

Latin. Duplicate of ranceō.

It also doesn't make sense for a reconstruction entry to say "very rare": either it's attested or it isn't. If "a few other forms are marginally attested in Late Latin", that means it's not a reconstruction; it's just a rare term. Theknightwho (talk) 04:07, 28 December 2024 (UTC)Reply

I agree that it should be deleted, but giving the confusing edit history, I suggested that a history merge with "ranceo" might be appropriate to make sure no attribution is lost.--Urszag (talk) 04:13, 28 December 2024 (UTC)Reply

Seconded. Nicodene (talk) 05:12, 28 December 2024 (UTC)Reply

Duplicate of RC:Proto-Germanic/kōz. See Reconstruction talk:Proto-Germanic/kūz. @Gudamunduz --{{victar|talk}} 05:19, 28 December 2024 (UTC)Reply

• Redirect (for the time being) or speedy delete and salt, a straight-out accidental mistake like this (*kōz was originally erroneously at *kūz for a long time) is not worth spending time in the backlogged RFD queue for. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 08:00, 2 January 2025 (UTC)Reply

Discuss why the entry was moved to *kōz in the first place. —Rua (mew) 14:21, 8 January 2025 (UTC)Reply

Simply put, *kōz is the straightforward descendant of nom. sg. *gʷṓws/*gʷéh₃-u-s, and must be reconstructed to account for Old Saxon, High German and East Norse which derive from a stem *kō-. Leveling to *kūz cannot have happened at the Proto-Germanic level since both North Germanic and West Germanic did not consistently generalize the same stem within themselves, and having zero grade nom. sg. vs. full/ō-grade oblique stem makes zero sense as a root/u-stem noun ablaut pattern. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 04:34, 16 January 2025 (UTC)Reply

Latin (specifically, Proto-Romance). Clearly redundant to Late Latin odiō, which is widely attested. It also arguably even existed in Classical Latin, too (e.g. odītur can be found in Tertullian, though I guess he could be seen as very early Late Latin). Theknightwho (talk) 17:45, 30 December 2024 (UTC)Reply

• Delete as attested term that does not belong in Reconstruction: space. —Mahāgaja · talk 22:09, 30 December 2024 (UTC)Reply

"Odītur" is a form of odīre, not odiāre: there's a difference in conjugation that is obscured if you just refer to the first-person singular present active form.--Urszag (talk) 22:45, 30 December 2024 (UTC)Reply

As Urszag notes there is a conjugation difference: odire versus *odiare.

However, the basis for reconstructing *odiare is flimsy at best. The Romance forms are neither common nor widespread, and both are easily explained otherwise:

1) The Old Catalan ujar can simply be backformed (with subtraction of the common en- prefix) from enu(t)jar < inodiare, a verb which unlike *odiare is securely attested and survives widely in at least one branch of Romance.

2) The Italian uggire (and also uggiare) can simply be formed from the noun uggia.

This isn’t enough for me to vote for deleting *odiare (though I’d never have made the entry myself) but I think it merits a note on the entry. Nicodene (talk) 23:57, 30 December 2024 (UTC)Reply

@Nicodene Note that second conjugation odeō is also attested in Late Latin. I'm not sure how this interacts with either Old Catalan or Italian, but it's worth considering. It seems to have had an unstable conjugation for quite some time. Theknightwho (talk) 04:22, 31 December 2024 (UTC)Reply

@Theknightwho What is the quote? Nicodene (talk) 04:41, 31 December 2024 (UTC)Reply

@Nicodene I remember seeing something more concrete (that I now can't find), but in any event, Marius Plotius Sacerdos apparently mentions odendus: Participia haec duo futuri temporis, hic osurus, hic odendus et praeteriti hic osus. That only works by interpreting the underlying verb as second conjugation odeō or third conjugation odō, which neither entry supports. Apparently odentes diem lexicalised as an expression, too. My source here, in the footnote. Theknightwho (talk) 21:02, 17 January 2025 (UTC)Reply

@Theknightwho Thank you.

Having looked further into all this, I have some points to add:

• The sense of Old Catalan ujar is best explained by deriving it from enujar < Late Latin inodiāre, whose descendants very often have, or can have, the sense “to tire, make weary”.
• The Italian word-family of uggia (“thick shade produced by trees, preventing the growth of anything below; harmful or destructive influence over something; situation inducing a feeling of disquiet; the feeling of disquiet itself”) and uzza (“damp, cool breeze of the morning or evening; thick shade produced by trees”), as well as the associated verbs, is neither phonologically nor very semantically consistent with Latin ŏdium (“hatred”). Most modern sources favour the etymology *ūdia < ūdus (“damp”).
• The rest of Romance seems to offer no verb at all that would be consistent with inheritance from *odiāre (“to hate”).

Accordingly, I now vote delete. Nicodene (talk) 05:13, 19 January 2025 (UTC)Reply

Alleged descendants have irreconcilable semantic differences, created by User:RwiTexx. --{{victar|talk}} 19:03, 30 December 2024 (UTC)Reply

• Keep, I do not agree that there are "irreconcilable semantic differences". Seething > mad > clumsy gets us the Slavic term, the Greek word can be seen as "product of seething > emission > breath/scent" or the like, and I also found a Baltic "rage" verb anyhow. The root has good acceptance of at least Baltic and Germanic descendants, from {{R:bsl:EDBIL|page=396}} (which also accepts cognacy of the Greek term), {{R:gem:EDPG|page=435}}, {{R:ine:EIEC|page=76}}, LIV, Orel, etc... will ask @Caoimhin ceallach, Exarchus for further input. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 15:41, 31 December 2024 (UTC)Reply

  In any case, the meaning of Proto-Slavic *šutъ as "clumsy" seems original research to me. Maybe it's rather: seething > mad > fool Exarchus (talk) 16:21, 31 December 2024 (UTC)Reply

Keep, the meanings are divergent, but the semantic developments are well-motivated. I think the PIE status is best shown by the Narten-present, which must be archaic. —Caoimhin ceallach (talk) 18:41, 31 December 2024 (UTC)Reply

Fritz (starting at p.293) says the basic meaning of Greek ἀϋτμή (aütmḗ) is "vapour" ("Dampf"), as it frequently occurs with fire. Beekes references Fritz, but still gives "breath; scent" as meaning, probably an oversight.

So I see no reason not to keep this. Exarchus (talk) 20:13, 31 December 2024 (UTC)Reply

A gigantic mess.

• You have Nowicki and De Vaan arguing that the Latin word is a late -īnus derivative;
• You have the other camp struggling to get straight from *h₃rḗǵnih₂ to here via implausible means;
• And the Marrucinian comparandum regen- is severely unclear since the word ending does not survive; it could be and has been read regenei, which does not allow for *rēgīnā.

In the wake of this, it is dubious whether Latin rēgīnā is traceable to PIt. in the first place. We'd be better off to move all etymological information to Latin rēgīna and then delete this entry. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 14:23, 31 December 2024 (UTC)Reply

Delete. Completely agree. Difficulties of reconstruction can also be mentioned at Proto-Indo-European *h₃rḗǵnih₂. —Caoimhin ceallach (talk) 18:54, 31 December 2024 (UTC)Reply

Nonsense unsupported reconstruction. The descendants listed, which anyway are West Germanic–only, reflect *funhtaz, *fūhtaz and *funhtigaz. (The derived terms are easily suffixed forms of *funhtaz.) Move etymology to Proto-West Germanic *fų̄ht. — 2600:4808:9C31:4800:982D:2083:BB23:FD71 08:52, 1 January 2025 (UTC)Reply

I tend to agree. Why does Kroonen reconstruct a u-stem? —Caoimhin ceallach (talk) 14:54, 6 January 2025 (UTC)Reply

Proto-West Germanic.
The Dutch word is attested from 1599 as far as I can see, despite several sources citing a "Middle Dutch gloren". No Middle Low German, No Old Frisian, only attested in Middle English from 1390, possibly as an alteration of glaren. Leasnam (talk) 15:20, 3 January 2025 (UTC)Reply

I'll just mention that etymologiebank gives a reconstruction *gluzan- for this, also connecting Swedish glora, glosa. Orel also gives Norwegian glora, "to stare". Exarchus (talk) 18:53, 3 January 2025 (UTC)Reply

I know. I mentioned that . Leasnam (talk) 22:19, 3 January 2025 (UTC)Reply

So you think Norwegian and Swedish borrowed glora somehow? There's also Icelandic glóra (“glimmer, faint light”). Exarchus (talk) 22:50, 3 January 2025 (UTC)Reply

The Scandinavian terms are probably a frequentative of words for "glow" (< Old Norse *glóra (“to glow continually, keep glowing”)). This may also be true of some of the West Germanic words, like Saterland Frisian glorje (“to glow faintly, smoulder”), and some of the Middle English senses of gloren. Leasnam (talk) 23:02, 3 January 2025 (UTC)Reply

Furthermore, all the other similar-looking terms are very recent. There is no Old Norse or Middle Low German forms, and none in German. Leasnam (talk) 23:04, 3 January 2025 (UTC)Reply

Given the long vowel, I am not sure Dutch gloren belongs here. I would expect *glōrōn to have yielded a Dutch form *gloeren. — Mnemosientje (t · c) 14:14, 4 January 2025 (UTC)Reply

The Middle Dutch might be a Low Saxon borrowing, especially given how late it is. The alternative is reconstructing a short vowel byform:

• PIE *ǵʰl̥h₁-ró-s > PG *gluraz > PWG *glor ⇒ *glorōn or
• PG *glēsaną > *gluzaz > PWG *gloʀ ⇒ *gloʀōn.

Either pathways presume an original adjective and a possible nominalization. --{{victar|talk}} 20:53, 4 January 2025 (UTC)Reply

If related, Dutch gluren (“to peek”), Middle Low German gluren (“to wink, leer”), West Frisian glurje (“to peek”) would point to PWG *glurjan (synchronically PIE *ǵʰl̥h₁-r-ye-ti) or *gluʀjan (*ǵʰl̥h₁-sye-ti). --{{victar|talk}} 22:18, 4 January 2025 (UTC)Reply

Proto-West Germanic *glurjan would produce Old Frisian *glera > West Frisian *glarre, therefore West Frisian gluorje, glûrje and gloere (“to peep, peek, blink”) cannot be inherited from that word. Middle Low German gluren is a corruption of geluren, gelûren, from ge- + lûren (“to lie in wait, lurk, peek”), so that word is ruled out. Dutch gluren is a modern Dutch word. It's not inherited from Middle Dutch. It's more likely that the Dutch and Frisian words are borrowings of the Low German. Leasnam (talk) 02:38, 5 January 2025 (UTC)Reply

<<The Middle Dutch might be a Low Saxon borrowing>> - and once again, there is no Middle Dutch *gloren (nor *glooren, *gloeren, *glouren) in the MNW (Middel­nederlandsch Woordenboek). Leasnam (talk) 18:42, 5 January 2025 (UTC)Reply

You're correct on *glurjan not working for West Frisian glurje, and {{R:nl:EWNT}} instead reconstructs *glūrjan to glûrje, but at no point was I saying the forms under RC:Proto-West Germanic/glōrōn, like gluorje, could descend from it. Also, I wouldn't "rule out" the Middle Low German so definitively. {{R:nl:EWN}} cites Dutch gluren as influenced by loeren, but not derived from it. --{{victar|talk}} 00:09, 6 January 2025 (UTC)Reply

Dutch gluren is probably a borrowing from Low German. Middle Low German gluren, glûren (“to lurk, peep, squint, blink”) is probably from Middle Low German ge- + lûren (“to lurk”) based on the meaning, and not related to Dutch gloren. Back at the main issue: there is no evidence for Proto-West Germanic *glōrōn. Leasnam (talk) 02:58, 6 January 2025 (UTC)Reply

So just so I'm on the same page, you think West Frisian gluorje, gloarje, German Low German gloren, glören, and Old Norse *glóra (Norwegian glore, Swedish glora, Icelandic glóra) are all independent frequentatives from *glōaną, and Middle English gloren, glouren, glooren is a unexpected variant of glaren? --{{victar|talk}} 04:15, 6 January 2025 (UTC)Reply

No. I'm not certain about those mentioned above, but I am certain that they cannot be explained by Proto-West Germanic *glōrōn. Just state that each origin is unknown, and you can compare them to one another, that's it. You cannot make up a theoretical ancestor term to explain them that shows up after 1390. That's unscientific. Leasnam (talk) 05:08, 6 January 2025 (UTC)Reply

If you wanted to, Old Norse *glóra (“to glow, smoulder”) would be an appropriate candidate for reconstruction, yet that won't do anything to help with the PWGmc issue at *glōrōn. Leasnam (talk) 05:26, 6 January 2025 (UTC)Reply

Old Norse glóra actually looks to be attested, according to Cleasby & Vigfusson, see . --{{victar|talk}} 05:32, 6 January 2025 (UTC)Reply

Excellent. That's where I would put the Middle English gloren coming from. If the Low German is also from a Scandinavian source, then that could explain the Dutch and Frisian. Pokorny has a Proto-Germanic *glōraz (“shining”), but lists no descendants from what I can see. Otherwise, Old Norse -ra was a productive verbal suffix, so it could have been created in Old Norse. Leasnam (talk) 14:15, 6 January 2025 (UTC)Reply

Are we sure that glóra is attested as Old Norse? It would be weird for Orel to reconstruct *ʒlōrōjanan and then give glóra as Norwegian instead of Old Norse (if the latter were attested). De Vries (AnEW, p.175) gives glóra as a feminine nickname. Exarchus (talk) 18:43, 6 January 2025 (UTC)Reply

The Íslensk stafsetningarorðabók lists both a verb and feminine noun. I don't know the dates of these attestations, and if they're Old Icelandic or Old West Norse. --{{victar|talk}} 22:58, 6 January 2025 (UTC)Reply

Scientific enough for Orel to reconstruct it. --{{victar|talk}} 05:30, 6 January 2025 (UTC)Reply

Was Orel not human ? Leasnam (talk) 17:23, 7 January 2025 (UTC)Reply

<<and Middle English gloren, glouren, glooren is a unexpected variant of glaren?>> - To be clear, Old English *glārian > Middle English glaren, gloren is a normal expectancy, parallelled by Old English slāpan > Middle English slapen, slopen; Old English sārian > Middle English saren, soren; Old English rārian > Middle English raren, roren; Old English grāpian > Middle English grapen, gropen; etc. Leasnam (talk) 21:59, 9 January 2025 (UTC)Reply

I don't think the evidence for *gʰleh₂dʰ- is strong enough to warrant an entry. The Germanic is better explained as a secondary past ptc. to *glōaną, per Kroonen, and the remaining possible Latin and Balto-Slavic forms can live at RC:Proto-Indo-European/ǵʰleh₁-. --{{victar|talk}} 21:52, 3 January 2025 (UTC)Reply

Merge and redirect or keep as root extension, but absolutely do not delete. Deletion hides the original contribution history of whatever would be added to *ǵʰleh₁-, and *gʰleh₂dʰ is still a decently-often-cited root in scholarship, so at least a redirect is warranted. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 04:31, 4 January 2025 (UTC)Reply

Most of the edits are from me, but sure, history merge it into RC:Proto-Indo-European/ǵʰleh₁-, if others think that's for the best. --{{victar|talk}} 20:25, 4 January 2025 (UTC)Reply

Proto-West Germanic.
One descendant, Modern Saterland Frisian Glor. Most likely a derivative of Saterland Frisian glorje (“to glow”). Further, the expected descendant of Old Frisian *glōr would be Glour. Leasnam (talk) 22:25, 3 January 2025 (UTC)Reply

Compare also West Frisian gloarre, gloar (“glow, embers”); German Low German Gloor (“glow, embers, coals”); and Dutch gloor (“glow, radiance”), each derived from their respective verbs. Leasnam (talk) 08:45, 4 January 2025 (UTC)Reply

Why consider them all separate nominalizations? --{{victar|talk}} 20:59, 4 January 2025 (UTC)Reply

Another Proto-Italic entry with a sole descendant. While it looks inherited, and so something like it can a priori be assumed to have existed in Proto-Italic (though to me it seems indistinguishable from a substratum word), reconstructing it with *a nonetheless requires some major assumptions about “unrounding”, namely its chronology and the reasons for it. De Vaan's reconstruction as *laku- is not the only possibility, even assuming inheritance. Not all such unroundings from presumed original *o are reconstructed to Proto-Italic time depth—for example Proto-Italic *boþjo- is given for Latin badius, an arbitrary choice over *baþjo- (or similar; this is disregarding the equally puzzling choice of consonants)—and so the assumption that the velarized *l caused this effect already in Proto-Italic rather than at a later stage seems arbitrary. At the very least these uncertainties should be noted on the page with alternative possible reconstructions listed. — 2600:4808:9C31:4800:FDE2:3540:DE4C:D241 21:29, 7 January 2025 (UTC)Reply

Okinawan. Rfd-sense: warabi na

Do not delete; the names were taken from Okinawa-go Jiten (1963), NINJAL.

Proto-West Germanic. Only Modern English and Middle High German (+ some modern High German descendants, some of which may derive from Italian tata). Middle High German tate does not appear to be attested; Köbler (who claims its origin as being onomatopoeic, i.e. baby talk) refers to Lexer's mid-19th century MHG dictionary as his source for this lemma, whereas Lexer claims that tate is "zu folgern aus tetel", that is, that its existence is to be deduced from the existence of tetel. I think this hypothetical link is best kept to a mention in the etymology sections of the actual mainspace entries affected, particularly dad, as it seems rather too weak to dedicate an entire PWGmc entry to it. — Mnemosientje (t · c) 12:24, 13 January 2025 (UTC)Reply

Delete, blatant baby-ish language that could have been separately created. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 20:22, 13 January 2025 (UTC)Reply

Delete...I've been adding some additional terms, but this shouldn't be regarded as an attempt to save. I am good with removing this. Leasnam (talk) 15:29, 14 January 2025 (UTC)Reply

Proto-Sino-Tibetan. Connecting these "descendants" via any regular sound correspondences seems flat out impossible, especially between Chinese 鼬 (yòu) and anything else (how can a *b-initial become an *l initial...). {{R:sit:Hill 2019}} doesn't even try to touch these words. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 00:26, 20 January 2025 (UTC)Reply

Delete. Agree with reasons provided, and there seems to be no attempt to even apply the comparative method here. Vampyricon (talk) 00:44, 20 January 2025 (UTC)Reply

Proto-Sino-Tibetan. Apparently post-proto-ST wanderwort according to Sagart and also Hill & Fellner, and thus unreconstructible. Fails the comparative method smell test very hard (where did the final nasal go in Chinese?). — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 04:02, 20 January 2025 (UTC)Reply

Despite the fact that the correspondence between TB ones with Chinese ones for horse still need to be further examined, it's not impossible for an etymon to have both forms with or without a coda in different languages.

Cf.

Ch. 日 *C.nik : Tib. ཉི་མ nyi-ma : Bur. နေ ne "sun"

Ch. 犬 *ʷʰˤʔ : Tib. ཁྱི khyi : Bur. ‌ေ‌‌ခွး khwe” "dog"

Janinga Chang (talk) 10:03, 15 February 2025 (UTC)Reply

The situation for "horse" is incomparable: for "sun" and "dog", Chinese has a final lost in other Sino-Tibetan languages, but here it's Chinese lacking a final that appears elsewhere. Furthermore, those other two words have replicable sound correspondences that appear in other words: OC -k + Tibetan zero + Burmese zero for "sun" (which Hill denotes with *-kə) and OC -r + Tibetan zero + Burmese -j for "dog". In contrast, no such ŋ-deleting correspondence exists. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 04:46, 20 February 2025 (UTC)Reply

Proto-Sino-Tibetan. A reconstruction at the Proto-Sino-Tibetan level really should not be supported by just Sinitic. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 06:14, 20 January 2025 (UTC)Reply

Proto-Sino-Tibetan. Back-projection of a Lolo-Burmese word way too far back. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 02:08, 21 January 2025 (UTC)Reply

ReRR- isn't a valid root shape, as far as I'm aware. Yet, even if it were, the two descendants (the o- and i-stems) either belong from an older single nominal paradigm, or are two separate m-stems. I also question merging the words for worm and ant. --{{victar|talk}} 07:19, 23 January 2025 (UTC)Reply

Delete, looks like an irrational montage from other pages. Fay Freak (talk) 15:55, 30 June 2025 (UTC)Reply

Proto-Malayic. Rationale was "wrong reconstruction, since Adelaar's paper doesn't state this alternative form". Ultimateria (talk) 03:12, 26 January 2025 (UTC)Reply

Proto-Germanic. There is no apparent grounds for reconstruction. Linked to from *gardaz and *gardô without further clarification, and from Latin Ingardis, which seems to me like it would not derive from an i-stem *gardiz due to the absence of umlaut. (The -is suffix is probably Latinate.) — Mnemosientje (t · c) 10:41, 18 February 2025 (UTC)Reply

It is presumably based on feminine names like Old High German Hildegard and Old Norse Gerðr. The latter, at least, must derive from Proto-Germanic *gardiz. For Ingardis compare Old Norse Ingigerðr. ᛙᛆᚱᛐᛁᚿᛌᛆᛌ ᛭ Proto-Norsing ᛭ Ask me anything 16:22, 19 February 2025 (UTC)Reply

Latin borrowings into Welsh whose proto-pages were created by an IP. I do not see how they necessitate a Proto-Brittonic stage (plenty of Irish borrowings of Latin words have lenition and secondary treatments of long vowels). Middle Breton has no trace of inheriting any of these borrowings from a Proto-Brittonic stage (having instead arm, Egi(p)t, aer etc.). @Mahagaja @Caoimhin ceallach for further advice. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 23:16, 26 February 2025 (UTC)Reply

Are all the Welsh forms in question explicable as borrowings via Irish? Nicodene (talk) 23:58, 26 February 2025 (UTC)Reply

I did not intend to imply such a thing; I am presuming that the Welsh borrowed straight from Latin mutatis mutandis at a later point of time than a putative "Proto-Brittonic" stage (maybe in unattested Old Welsh?), because Breton clearly borrowed directly and independently instead of inheriting the Latin loans from a proto-Brittonic ancestor. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 01:32, 27 February 2025 (UTC)Reply

*arβ̃ also has a Cornish descendant, which I've added. GPC mentions an Old Cornish apui as a cognate of awyr, but I don't know what to make of that p. I'm also not sure we know enough about Old Welsh to be sure when sound changes like ē ō → uɨ ʉ and m g d → β̃ ɣ ð stopped happening, i.e. the latest possible date for the borrowings. —Mahāgaja · talk 08:00, 27 February 2025 (UTC)Reply

Are there ‘post-Proto-Brythonic’ borrowings with the relevant sound-changes?

Couldn’t Breton have had older forms displaced by borrowings from French? Nicodene (talk) 08:35, 27 February 2025 (UTC)Reply

(1) Well, there's the weird case of dogfen, which was apparently coined in the 16th century but given all the sound changes as if it been borrowed into Proto-Brythonic during the Roman Britain era and then inherited into Welsh. (2) Yes, it could, but without the evidence of Old or Middle Breton forms, we can't make that assumption. —Mahāgaja · talk 11:27, 27 February 2025 (UTC)Reply

(1) Ideally not neologisms by a scholar mimicking sound-changes. (2) Considering the time span between the Breton arrival on the mainland and the time of the earliest substantial texts, I don’t see how making the opposite assumption would be any safer. Nicodene (talk) 18:31, 27 February 2025 (UTC)Reply

Proto-Indo-European. Etymology not phonetically possible, several "cognates" (like the Celtic ones) do not actually exist or make no phonetic sense (like the Indo-Aryan words), and one of the sources (Kroonen) cited in the entry explicitly calls it a Western European Wanderwort and thus cannot be of Indo-European stock. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 16:53, 28 February 2025 (UTC)Reply

Delete, but I think the origin of this entry is that some pages do link to this. This reconstruction seems to go all the way back to Rua's 2011 version of Proto-Germanic *samdaz. Exarchus (talk) 17:18, 1 March 2025 (UTC)Reply

I've now noticed that this reconstruction is also at Requested entries for PIE. I suppose that list needs to be cleaned up. Exarchus (talk) 15:02, 2 March 2025 (UTC)Reply

Proto-Sino-Tibetan. Based off this STEDT entry, but the descendants are impossible to relate to each other by sound laws, and the random smattering of Burmese "descendants" does not instill any confidence. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 21:46, 2 March 2025 (UTC)Reply

Delete since STEDT already has methodological issues. Add to that the fact that "mother" frequently starts with /m/ cross-linguistically and the lack of correspondences elsewhere, this ends up looking like a Nostratic-level error. Vampyricon (talk) 22:05, 2 March 2025 (UTC)Reply

Firstly PIE participles generally come from a verb and not directly from the root.
Then when you look at the descendants, Sanskrit and Greek are active aorist participles, Latin is a present participle with the verb coming from *wid-eh₁- per de Vaan, and Germanic is the participle of *witaną, reconstructed to come from the PIE perfect. So they aren't fully cognates. *widónts can be reconstructed as the participle of the thematic aorist *widét, but we normally don't have pages for that, and I think the gen.sg. would be *widóntos and not *widn̥tés. Exarchus (talk) 16:45, 3 March 2025 (UTC)Reply

Delete since the "descendants" cannot be inherited from a common ancestor. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 19:59, 3 March 2025 (UTC)Reply

Proto-Indo-Iranian. Of unknown origin, and the Avestan and Sanskrit forms cannot share a direct common ancestor (the Sanskrit is a feminine aH-stem, the Avestan is a neuter s-stem; also Sanskrit has a long root vowel while Avestan has a short one). Any information is better handled in separate Sanskrit and Avestan entries. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 02:00, 18 March 2025 (UTC)Reply

As already proposed at the Etymology scriptorium. Exarchus (talk) 13:26, 19 March 2025 (UTC)Reply

Old English. Supposedly based on a Middle English adjective, yet no such adjective exists. Leasnam (talk) 22:59, 19 March 2025 (UTC)Reply

I think we need to sort this out and leave one thing. We can also simply indicate an alternative reconstruction or write reconstruction notes. I have to abandon Matasović's rule because I can't find the source I read several years ago. ɶLerman (talk) 22:15, 20 March 2025 (UTC)Reply

Okay, it was all on Wikipedia: https://en.wikipedia.orghttps://dictious.com/en/Proto-Balto-Slavic_language#Satemization ɶLerman (talk) 22:21, 20 March 2025 (UTC)Reply

Proto-Indo-Iranian. Descendants are of dubious provenance: see Wiktionary:Etymology scriptorium/2025/March#Reconstruction:Proto-Indo-Iranian/ȷ́aluš. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 22:13, 21 March 2025 (UTC)Reply

Old French.
The English word is directly from Middle Latin. The Old French word derived from Frankish *laubijā is loge. *lobie would not be possible phonetically under normal sound changes. Leasnam (talk) 21:58, 27 March 2025 (UTC)Reply

This entry is presumably meant to represent a pronunciation like , which would have been used before the French sound change that turned into . The spelling -ie was sometimes used in Old French to represent , generally in borrowed words, as in palie, monie, chanonie, apostolie.--Urszag (talk) 22:42, 28 March 2025 (UTC)Reply

Given that we label as Old French the entirety of early medieval Oïl, there may well have been varieties where did not yield an affricate result. I’ll have a look. Nicodene (talk) 09:50, 29 March 2025 (UTC)Reply

The Germanic term is an adverb, not an adjective, and Orel links it to Latin citrā and citrō, not citer. For the latter, de Vaan indeed reconstructs Proto-Italic *ki-tero-, but there don't seem to be cognates for this specific formation, so I wouldn't reconstruct it for PIE (definitely not have a separate page for it). Exarchus (talk) 20:55, 28 March 2025 (UTC)Reply

Only 'descendants' would be the adjectives Proto-Germanic *samafadrijaz and Ancient Greek ὁμοπάτριος (homopátrios). Exarchus (talk) 11:47, 23 April 2025 (UTC)Reply

We could reconstruct something based on those two, but certainly *somoph₂tḗr is not correct (you can't just prefix *somo- to *ph₂tḗr and get it to mean 'sibling'). ᛙᛆᚱᛐᛁᚿᛌᛆᛌ ᛭ Proto-Norsing ᛭ Ask me anything 18:17, 28 April 2025 (UTC)Reply

For that matter, how do we know that the Old Norse and Ancient Greek reflexes aren't just independent innovations? And are there any other descendants of the Proto-Germanic form? We could really use some references here all around. Though it would be unfair to mention this and the Proto-Germanic entry in the same breath- the creator of this one is so totally clueless that I've undone a number of their edits and blocked them for a week. Chuck Entz (talk) 06:07, 29 April 2025 (UTC)Reply

The Old Norse reflects a nominalization < *samafadrijô. This type of contruction was hardly productive in ON so it probably does go back to Proto-Germanic, but both Ancient Greek and PGmc have the same productive elements in this case, so it's impossible to say whether it's inherited or coined in both. It's like stony, stenig. -y and -ig are cognate but both are still productive, so in a sense they are both inherited from Proto-Germanic *stainagaz and independent. ᛙᛆᚱᛐᛁᚿᛌᛆᛌ ᛭ Proto-Norsing ᛭ Ask me anything 18:41, 11 May 2025 (UTC)Reply

Proto-Finnic, etymology 2. Only found in Livonian; I suppose this is based on an assumption of apocope from *walketa ~ *waleta > *valëta > *valta. I think we can explain the Livonian more directly from *valëda (cf. *hopëda > õ’bdõ) or even *valkëda (cf. *korkëda > kuordõ), although granted, my knowledge of Livonian sound changes is rather poor. — SURJECTION ^{/ T / C / L /} 14:45, 4 May 2025 (UTC)Reply

Hmm, the final vowel being a might require a bisyllabic form though. I don't know enough about the chronology here to decide whether Proto-Finnic is required to explain that. — SURJECTION ^{/ T / C / L /} 16:09, 4 May 2025 (UTC)Reply

Dubious. English chaw is first attested in the 16th century (both the noun and the verb, supposed to be an anomalous by-form of chew). @Leasnam has put together some Dutch and Old/Middle High German forms (uncited now cited) to claim this reconstruction. Does it hold water? (@Victar and other resident Germanicists.) Nicodene (talk) 00:57, 7 May 2025 (UTC)Reply

Keep - the requestor is requesting deletion of the entry due to a conflict with an etymology he is pushing at English jaw. However, this reconstruction is completely sound based on comparative methods used across Proto-West Germanic reconstructions. Leasnam (talk) 01:24, 7 May 2025 (UTC)Reply

The Middle Dutch can be found couwe f "jaw", (pl) "mouth, snout". The Old High German can be found as kouuua "oral cavity, pharynx". Old High German ou answers to PWG au. Leasnam (talk) 01:52, 7 May 2025 (UTC)Reply

Leasnam invented this reconstruction about 6‒7 hours ago because of a discussion, which has been ongoing for days now, in which he refuses to acknowledge overwhelming evidence against a Proto-West-Germanic etymology for English jaw.

I will repeat here what the statistics show about Leasnam's assumption that Middle English joue ("jaw") had an unattested variant *chawe and came from an also-unattested Old English *ċēawe. The question is: if a word that began with /tʃ/ in Old English really had coexisting variants with /tʃ/ and /dʒ/ in Middle English, how likely is it that out of 30+ attestations in Middle English, not a single one would actually show the /tʃ/ variant? For comparison let's take another Middle English word chā̆vel, which came from an (attested) Old English ceafl.

For chā̆vel I count 33 occurrences of forms with /tʃ/ and 3 with /dʒ/ (source: MED quotations). For the aforementioned joue I count 0 occurrences of forms with /tʃ/ and 33 with /dʒ/ (source: MED quotations).

Now, according to Fisher's exact test, the odds that the discrepancy in outcomes between the two words occurred purely by chance is 1.5721960560778892e-16 (source: vassarstats, two-tailed p). In decimal form that is 0.00000000000000015721960560778892 (source:wolfram alpha) which comes out to about 1 out of 6.36 quadrillion. So, judging by comparison with chā̆vel, the odds that joue had forms with /tʃ/ that just—by coincidence—happened never to be recorded in the Middle English corpus are indistinguishable from zero.

Note that the MED explains Middle English joue as simply borrowed from Old French jöe~joue (with /dʒ/). Nicodene (talk) 06:57, 7 May 2025 (UTC)Reply

Not everybody believes what you and Hazarasp believe. The origin of Middle English jowe is uncertain at best, otherwise disputed. You need to stop trying to force everyone to see things YOUR way. You're absolutely not always right. I've proven that MULTIPLE times over the past few days. Leasnam (talk) 02:17, 7 May 2025 (UTC)Reply

Your comment above proves that this has nothing to do with Proto-West Germanic *kauwā. Your concerned instead with how the English descendant of this term might pose problems with your made-up etymology at jaw. I wasn't even challenging your assertions there. I simply added an alternative origin at the end. I don't see why you're flipping out over it. It was all calm until you brought this here. Leasnam (talk) 02:17, 7 May 2025 (UTC)Reply

Brass tacks: the data supports Proto-West Germanic *kauwā. It would have been created eventually anyway. Now is as good a time as any. Leasnam (talk) 02:02, 7 May 2025 (UTC)Reply

Perhaps? I'll leave commenting on the Dutch and German to others. As far as English is concerned, I'll point out that *hauwan yields hew and, per one of your own reconstructions, *gakauw > Middle English icheu > English chew (“act of chewing”).

As for all the shouting and such - I'm not replying to that. Nicodene (talk) 03:39, 7 May 2025 (UTC)Reply

ēa develops differently after palatised consonants ċ, ġ, and sċ and cannot be reliably predicted (e.g. Old English sċēawian > Middle English schewen, but also schawen & schowen; ċēas (“chose”, past of ċēosan) > Middle English chees, chase, or chos-; Old English ċēapian > Middle English chepen, but also chapen/chappen; Old English ġēa (“yes”) > Middle English ȝe, ȝa, or ȝo; Old English ċēast (“strife”) > Middle English chest, chast, or chost) Leasnam (talk) 04:02, 7 May 2025 (UTC)Reply

I'll begin by stating that several examples you give are nonprobative:

• chos and chas are both Late Middle English analogical formations: the former from the past participle chosen, ychose (itself analogical for earlier (i)coren), and the latter from class 4 strong verbs; compare Middle English bar, spak, Early Modern English bare, spake "bore, spoke".
• Forms of chepen with /a/ are due to trisyllabic shortening, which regularly occurred across most of the verb's paradigm; this would've been reinforced by the vocalism of compounds such as chaffare and chapman (with shortening before a consonant cluster). The predominance of forms with /ɛː/ is therefore presumably due to the influence of the noun chep.
• Besides ġēa, ġǣ, ġē, OE also possessed a variant ġā for "yes" which ME ȝa, ȝo could easily derive from.

However, forms of chest and schewen appear to examples of OE ēa developing to early ME /ɑː/ after a palatal which are difficult to deny (see Jordan 1974 §81); the vocalism in -o- is due to the rounding of that vowel south of the Humber-Lune line. However, both in the two words under consideration and the additional examples Jordan adduces, examples of ēa developing into /ɑː/ are much rarer than the regular development (into /æː/ > /ɛː/). The only exception is yoman, which is possibly spurious since it may not have etymological êa: a derivation from *ġēamann competes with that from yongman; even if the former etymology holds, the following /m/ could've exerted a special backing influence.

Contrastingly, forms of joue with -e- are markedly rarer than those of -o- or -a-, especially if the forms geawe, geowe represent /ˈd͡ʒau̯(ə)/, /ˈd͡ʒɔu̯(ə)/, with the the digraph <ge> representing /d͡ʒ/, as in e.g. pygeoun, rendering a derivation from *ċẽawe unlikely. However, I believe the clinching evidence for deriving joue from Old French joe rather than OE *ċēawe to be the consonantism rather than the vocalism; it doesn't help that only one proposed etymon is actually attested. Hazarasp (parlement · werkis) 10:41, 7 May 2025 (UTC)Reply

So it ends up about the same. Your theory assumes:

1) An English vowel outcome that is ‘difficult to account for’

2) That a Middle English *chawe existed alongside Middle English joue without ever being documented. Per the above calculations—and if you prefer, I can run other statistical tests as well—the odds of that are ≈1 out of 6.36 quadrillion. This is without adding the consideration that Old English *ċēawe is also unattested.

Nicodene (talk) 04:47, 7 May 2025 (UTC)Reply

It's not difficult. It's unpredictable. It will invariably be e, a, or o. Leasnam (talk) 04:50, 7 May 2025 (UTC)Reply

See, I was right. This discussion is not about why *kauwā should be deleted. I'm leaving it. Leasnam (talk) 13:31, 7 May 2025 (UTC)Reply

My general position is that if other reliable etymological reference works reconstruct a particular term, I prefer to have our own entry for that reconstructed term, an entry which can then mention any potential issues; conversely, if we are reconstructing a term all by ourselves that no-one else has reconstructed, and there is doubt about whether we're right, I view having an entry with scepticism. So if other reference works reconstruct this term, I am inclined to keep it; if they don't, I'm inclined to delete it and mention anything like "alternatively, jaw might be cognate with foobar" (ideally "...according to XYZ") in the entry for jaw. - -sche (discuss) 15:59, 7 May 2025 (UTC)Reply

Sources that I've seen do not go into detail when it comes to variants of reconstructions. If you're lucky, you might get a mention of a variant in passing. Proto-West Germanic *keuwā and *kauwā are extremely similar in form and meaning and derive from the same immediate source: Proto-West Germanic from *keuwan, so it is highly unlikely that anyone in print will devote the time (and ink) to detail them individually. You're more likely to see the descendants of them lumped together, as Koebler does , but those familiar with Germanic etymology and phonology know that OHG kiuwa and kouwa cannot possibly derive from the same immediate source. It is impossible. Leasnam (talk) 17:39, 7 May 2025 (UTC)Reply

My determination for reconstructing *kauwā is to explain not only OHG kouwa, MLG couwe, cauwe, DUM couwe, but also EME chaw. Some on this discussion are convinced that chaw poses some sort of threat to their work. That's for them to work out. All I know is that EME chaw didn't just appear out of nowhere, nor was it derived from the verb chaw="chew", as a derivation from the verb would mean "that which is chewed" and not "that which does the chewing", i.e. "jaw", and which is evidenced by Etymology 1 Noun at chaw (“that which is chewed”). A word derived from "chew" in Middle or Modern English meaning "chewer, jaw" would require an agent suffix, like -er. The absence of such a suffix on chaw="jaw" indicates that the word is much older and was created either in Old English or PWG, as the -e of Old English and the -ā of PWG could serve as agent suffixes. Keep. Leasnam (talk) 17:39, 7 May 2025 (UTC)Reply

All I know is that EME chaw didn't just appear out of nowhere

Chaw, as a noun meaning jaw, is ‛apparently a by-form of jaw n.1, modified by association with the verb chew or its by-form chaw’. Nicodene (talk) 22:29, 7 May 2025 (UTC)Reply

From the same source jaw is ‛Of uncertain origin. Perhaps a word inherited from Germanic.’. Leasnam (talk) 22:44, 7 May 2025 (UTC)Reply

The question is whether there exists a credible objection to chaw^noun developing as a by-form of jaw influenced by chew or chaw^verb. So far, there are none.

The OED makes a single objection to the Old French etymology, based on their assumption that the Middle English joue~jaue does not have the vowel outcome that would be expected of a borrowing of Old French jöe~joue. This assumption turned out to be incorrect, as noted by Foster 1970 and Woledge 1970 and as demonstrated by the following examples:

• Middle English moue~maue < Old French möe~moue
• Middle English poue~paue < Old French pöe~poue
• Middle English scroue~scraue < Old French escröe~escroue

Nicodene (talk) 23:52, 7 May 2025 (UTC)Reply

Yeah yeah yeah it's the same old line with you. I'm not convinced. You can stop now. Repeating what doesn't work expecting a different result is the definition of insanity. Leasnam (talk) 00:03, 8 May 2025 (UTC)Reply

Since this discussion isn't really about deleting Proto-West Germanic *kauwā, but is instead (as was always the intent of Nicodene) to be another inroad for him to angle his little magic mirrors and smokescreens at me - does anyone object to closing this discussion and removing the tag ? Leasnam (talk) 00:12, 8 May 2025 (UTC)Reply

@-sche Is Leasnam's behaviour in this thread considered acceptable conduct for an administrator on this site? Nicodene (talk) 10:18, 9 May 2025 (UTC)Reply

Having done some reading on this, my question is whether it's abnormal for the ablaut in the preterite of strong verbs like *keuwan (“to chew”) to restructure the verb overall (~*kauwān?) or carry over into derivatives like *keuwā (“jaw”, ~*kauwā?) at the level of Proto-West-Germanic already.

Assuming it's not, one can just put *kauwā under *keuwā, the way that *kauwān is under *keuwan, along with a tree like:

• >? Old English: *ċēowe (or *ċēawe if < *kauwā; see above)
  • >? Middle English: jowe, jawe
    • English: jaw
    • Scots: jow
  • >? Middle English: *chowe, *chawe
    • English: chaw
    • Scots: chow

I don't see the use in separate pages (*kauwā, *ċēowe, *ċēawe, *chawe) and, per the below, others may not either. Nicodene (talk) 22:44, 10 May 2025 (UTC)Reply

@Nicodene Your recurrent vehemence on seeing your views implemented is great, but RFDing a perfectly sourceable reconstruction based on your views and a naïve self-conducted statistical test (across a sparse corpus) comes across as silly—almost as silly as attempting to hide the corresponding etymology, again perfectly sourceable, from the main entry. Why? Keep unless further justification supports deletion. 🌙🐇 ^{⠀talk⠀ ⠀contribs⠀} 03:34, 12 May 2025 (UTC)Reply

@Lunabunn: With the personal attacks such as naïve and silly, this comes across as voting out of personal hatred.

across a sparse corpus

There are over 30 examples of each word, which is plenty for Fisher's exact test.

perfectly sourceable reconstruction

Which source gives a Proto-West Germanic *kauwā?

attempting to hide the corresponding etymology

How?

Nicodene (talk) 03:51, 12 May 2025 (UTC)Reply

@Nicodene The Old English corpus is sparse. This isn't to say that words aren't attested, but rather that their distribution is biased, skewed, and limited. The test is naïve in the neutral sense; it assumes random, representative sampling, which was not established. Silly is not a personal attack, just the impression I get from this RFD—unless you somehow consider the inclusion of these terms (or the lack thereof) to be a token of your character?

Re: sourcing, see @Leasnam's response to sche above. Re: attempting to hide the corresponding etymology, Talk:jaw. 🌙🐇 ^{⠀talk⠀ ⠀contribs⠀} 04:00, 12 May 2025 (UTC)Reply

@Lunabunn

The Old English corpus is sparse.

My test did not have anything to do with the Old English corpus.

This isn't to say that words aren't attested, but rather that their distribution is biased, skewed, and limited. The test is naïve in the neutral sense; it assumes random, representative sampling, which was not established.

Fisher's exact test is designed to test exactly that: how likely it is that accidental happenstance would explain the discrepancy in the data.

Re: sourcing, see @Leasnam's response to sche above

The comment in which he points out that it's unlikely for any source to give this reconstruction?

Re: attempting to hide the corresponding etymology, Talk:jaw.

Maybe I've misread you, but to me attempting would mean actually taking some kind of action.

Nicodene (talk) 05:20, 12 May 2025 (UTC)Reply

@Nicodene Re: points 1 and 2, that is precisely what I am taking issue with. You cannot use a statistical test that depends on a representative sampling without first examining the corpus for possible bias. I am not claiming your results are necessarily invalid, but rather pointing out your methodology (or at least presentation thereof) is insufficient. Re: point 3, did you not see the rest of the sentence? Should we not have a reconstruction entry for any form not explicitly given verbatim in literature? Re: point 4, this is a wiki. Pushing a point over and over again to, in effect, "win" a discussion by exhaustion of all other involved parties might as well be direct action. 🌙🐇 ^{⠀talk⠀ ⠀contribs⠀} 14:29, 13 May 2025 (UTC)Reply

@Lunabunn

You cannot use a statistical test that depends on a representative sampling without first examining the corpus for possible bias.

Could you supply evidence for the bias that you suppose would explain the massive discrepancy we observe between those two words (n > 30)?

Re: point 3, did you not see the rest of the sentence? Should we not have a reconstruction entry for any form not explicitly given verbatim in literature?

You claim that this is a perfectly sourceable reconstruction. When asked to source it, you point me to a comment where Leasnam says this reconstruction is unlikely to be sourceable. I don't think that supports the claim.

Re: point 4, this is a wiki. Pushing a point over and over again...

I, like Leasnam, proposed using a collapsible box once, not over and over again. We then both forgot about the idea over a week ago, so I'm not sure why you're pushing point over and over again now. Nicodene (talk) 19:30, 13 May 2025 (UTC)Reply

@Nicodene It's on the part of the person conducting the test to assess bias, not on mine. Frankly I don't care; I just believe in inclusionism over unconstructive whataboutism. Re: sourcing, does any understanding of either my claim or Leasnam's explanation beyond the most literal reading escape you? Leasnam explained that no such reconstruction is likely to be found in literature, but it is nonetheless implied through prose. You have a way with making every discussion completely unbearable to participate in, so please excuse my refusal to engage further. My stance was stated and you have provided nothing in your series of responses in direct address. Cheers, 🌙🐇 ^{⠀talk⠀ ⠀contribs⠀} 00:31, 14 May 2025 (UTC)Reply

@Lunabunn:

It's on the part of the person conducting the test to assess bias, not on mine.

I've tested for random chance, and I made sure to select the most comparable word I could imagine (semantically similar, attested with similar frequency in Middle English, and derived from an Old English word beginning with /tʃ/).

If there is some form of bias that I've not considered, which may have borne drastically different results for chavel and joue, then please - tell me, so I can actually see whether that plays out elsewhere in the corpus data. If you won't specify what form of bias you have in mind, then your objection remains unfalsifiable.

Frankly I don't care

You don't care about evidence?

Re: sourcing, does any understanding of either my claim or Leasnam's explanation beyond the most literal reading escape you? Leasnam explained that no such reconstruction is likely to be found in literature, but it is nonetheless implied through prose.

Listing a mixed bag of descendants of *keuwan or *keuwā is not the same as saying that Proto-West Germanic already had forms like *kauwā and *kauwān, restructured around the ablauted past singular of the original verb. The question of whether such a phenomenon is known to have existed at the level of Proto-West Germanic was touched on in an earlier comment already. I'm curious what specialists in Germanic have to say.

You have a way with making every discussion completely unbearable to participate in

Thanks. I'll refrain from returning the personal attack.

Nicodene (talk) 01:29, 14 May 2025 (UTC)Reply

Expanding a bit on this - if it's the case that the forms grouped under *kauwā would need the English chaw to be (clearly) projected back to PWG, as opposed to amounting to some later areal phenomenon, I'm not sure said projection is tenable with chaw first being attested in the 16th century (and readily explainable through other another etymology). Input from specialists in Germanic would be greatly appreciated, as ever. Nicodene (talk) 11:07, 14 May 2025 (UTC)Reply

Per the above. Nicodene (talk) 18:40, 7 May 2025 (UTC)Reply

Keep, per the above. Leasnam (talk) 18:48, 7 May 2025 (UTC)Reply

@Nicodene so now that you realise you cannot sweep *kauwā under the carpet, you're going after *ċēawe. <smdh> Leasnam (talk) 19:04, 7 May 2025 (UTC)Reply

Delete: Leasnam is attempting to prove that jaw (Middle English jowe) can derive from this form rather than disprove the competing derivation from Old French joe. Therefore his success would render his proposed etymology a possibility rather than a certainty, and I don't believe a mere possibility is enough to warrant the creation and maintenance of a reconstruction page, especially given the certitude its existence implies. Hazarasp (parlement · werkis) 01:28, 8 May 2025 (UTC)Reply

That's not true, sir. I have not seriously attempted to derive jaw from this, only chaw. Nor have I interfered with the etymology at jaw that points to joe. Keep - it reasonably explains chaw in light of English and other cognates in other Germanic languages. Leasnam (talk) 01:50, 8 May 2025 (UTC)Reply

I have not seriously attempted to derive jaw from this

Diff.

Nicodene (talk) 01:57, 8 May 2025 (UTC)Reply

Alternatively, perhaps from Middle English jawe, a variant of *chawe (whence modern English chaw (“jaw”)) - this is not what I call a serious attempt. According to the OED (and you) Chaw, as a noun meaning jaw, is ‛apparently a by-form of jaw n.1 but their jaw is not your jaw. Their jaw is uncertain, possibly from Germanic. Leasnam (talk) 02:07, 8 May 2025 (UTC)Reply

We need to show BOTH, without prejudice. Leasnam (talk) 02:08, 8 May 2025 (UTC)Reply

The OED does not say anything about a *chawe. Nicodene (talk) 02:16, 8 May 2025 (UTC)Reply

I somehow mistyped that. Right, they didn't. Thanks. I fixed it. Leasnam (talk) 02:31, 8 May 2025 (UTC)Reply

For reference, Leasnam’s quote here cuts short right before the part where he added {{inh|en|ang|*ċēawe|t=jaw}}. Nicodene (talk) 07:17, 8 May 2025 (UTC)Reply

Additionally, I agree that an entry is not necessary. It can be listed at Proto-West Germanic *kauwā without one. Same is true for *ċēowe below - it's already at *keuwā. Leasnam (talk) 02:03, 8 May 2025 (UTC)Reply

Per the above. Nicodene (talk) 00:10, 8 May 2025 (UTC)Reply

Keep per the above.

The real issue here is that we have a user (Nicodene) bent on relentlessly pushing their own agenda (which is fine by me, I have no qualms about that) - yet when an agenda involves curtailing and trying to silence other competing views, then I have a problem. Then it is a serious offence. Leasnam (talk) 00:23, 8 May 2025 (UTC)Reply

@Nicodene, are those the only ones ? I've created many more Old English and Proto-West Germanic reconstructions, do you not wish to delete those too ? I'm guessing not, because you only care about the ones that prove you WRONG. Leasnam (talk) 00:25, 8 May 2025 (UTC)Reply

Delete: Even if one insists on deriving jaw from a Old English etymon, having two seperate entries is unnecessary and redundant. Hazarasp (parlement · werkis) 01:41, 8 May 2025 (UTC)Reply

I agree. This is the weaker of the two, but it is the one posited by the OED, hence it was created first. We can still cite it in etymologies, but it isn't necessary to have an additional entry. Leasnam (talk) 01:43, 8 May 2025 (UTC)Reply

Per the above. Nicodene (talk) 02:17, 8 May 2025 (UTC)Reply

I am scared to follow the long argument you had about jaws with Leasnam. Fay Freak (talk) 15:52, 30 June 2025 (UTC)Reply

While deriving English shoal from Old English sċolu (“troop, host”) through such a form is not in principle impossible, its semantic agreement of with Dutch school (“shoal, school (of fishes)”) raises suspicions that it is its source; Bosworth-Toller provide a sense "shoal, school" for sċolu, but this can be ignored as they neglect to provide any pertinent examples. The only issue with such a derivation is the adaptation of Dutch /sx/ as /ʃ/, but this could either be a nativisation after the example of correspondences involving Germanic *sk such as ship:schip or reflect a borrowing from West Flemish rather than Dutch, specifically French Flemish, where *sk becomes ; perhaps shoal's borrowing was precipitated by the loss of the Pale of Calais. In any case, the existence of a viable alternative etymology leaves *schole's existence too uncertain to warrant a entry. Hazarasp (parlement · werkis) 03:46, 8 May 2025 (UTC)Reply

Latin-only, uncertain etymology. Even if the root were well-attested and uncontroversial (which it is far from), we could not be certain that it was an i-stem in Proto-Italic (its only cognate in Indo-Iranian is a u-stem, and Latin frequently remodeled things to i-stems), so giving anything more than the root is dishonest. As it stands, the etymology is implied as a secure one.
Moreover, it is highly suspicious that there is also Latin pestis without etymology, which is basically identical to this PIt. form save for the fricative middle consonants being unvoiced. Maybe these are two dialectal forms or parallel cluster simplifications of the same original word, but since we can't know this I would strongly suggest framing the IE etymology as less certain; it's only reasonable to take any etymology with a grain of salt given the smallest hint of substrate influence, as at the very least it's a sign to regard the situation as not straightforward. — Ganjabarah (talk) 03:32, 14 May 2025 (UTC)Reply

We voted to delete Proto-Italic terms with one descendant a while ago per @Catonif. The vote applies here unless, perchance, some Italic cognate of pēdis has surfaced in the years since de Vaan 2008. Nicodene (talk) 04:13, 14 May 2025 (UTC)Reply

Yeah. I realized shortly after sending this that it was needlessly verbose. It would have been enough to say “Latin-only”. — Ganjabarah (talk) 01:26, 15 May 2025 (UTC)Reply

Delete per above.--Urszag (talk) 03:34, 15 May 2025 (UTC)Reply

No information is lost by subjecting it to deletion. Fay Freak (talk) 15:50, 30 June 2025 (UTC)Reply

Both only have Latin descendants. The verb also has a disputed etymology. According to De Vaan, others suspect that the form Latin ignoro may have emerged from *əngnārāō if the -a- shifted to -o- on the model of ignotus. However, De Vaan also explicitly states that “it seems highly unlikely that *en-gnarare was replaced by *en-gnorare, while the adj. gnarus ‘knowing’ itself was left unchanged.” Instead, De Vaan—citing Nussbaum—explains Latin ignoro as a derivation from a verb *gnō-sāō (“to know”), itself a denominative to *gnō-sā- (“knowledge”). He also argues that the term may have directly derived from a PIE verb due to the presence of Hittite kane/išš-zi.

This PIE verb is reconstructed by De Vaan as *ǵnéh₃-s-ti. I do not know enough about PIE to reconstruct the verb myself, but I will provide a link to this research paper in case anyone reading this wishes to add such information. Graearms (talk) 13:24, 16 May 2025 (UTC)Reply

Delete both; the verb due to being formed on a dubious etymology, and the adjective because the negative adjective prefix is too productive to ascertain if ignarus is old. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 05:35, 17 May 2025 (UTC)Reply

Easy delete. — Ganjabarah (talk) 21:50, 21 May 2025 (UTC)Reply

These are all Proto-Italic terms with one descendant. I feel we should probably delete them considering the vote.

• Reconstruction:Proto-Italic/aimelos
• Reconstruction:Proto-Italic/anats
• Reconstruction:Proto-Italic/arðōs
• Reconstruction:Proto-Italic/atta
• Reconstruction:Proto-Italic/attā
• Reconstruction:Proto-Italic/auzis
• Reconstruction:Proto-Italic/awikaps
• Reconstruction:Proto-Italic/bakklom
• Reconstruction:Proto-Italic/deknōnā
• Reconstruction:Proto-Italic/dekos
• Reconstruction:Proto-Italic/diskō (and its redirect)
• Reconstruction:Proto-Italic/dlongos
• Reconstruction:Proto-Italic/dokeō
• Reconstruction:Proto-Italic/dolaō
• Reconstruction:Proto-Italic/domanos
• Reconstruction:Proto-Italic/domaō
• Reconstruction:Proto-Italic/domos
• Reconstruction:Proto-Italic/dormjō
• Reconstruction:Proto-Italic/dweos
• Reconstruction:Proto-Italic/ensis
• Reconstruction:Proto-Italic/erzāō
• Reconstruction:Proto-Italic/ēɣʷrjos
• Reconstruction:Proto-Italic/farβātos
• Reconstruction:Proto-Italic/feiðos
• Reconstruction:Proto-Italic/feros
• Reconstruction:Proto-Italic/feɣʷris
• Reconstruction:Proto-Italic/feβros
• Reconstruction:Proto-Italic/flagmā
• Reconstruction:Proto-Italic/flāgmen
• Reconstruction:Proto-Italic/flāwos
• Reconstruction:Proto-Italic/flēō
• Reconstruction:Proto-Italic/folgos
• Reconstruction:Proto-Italic/folkjō
• Reconstruction:Proto-Italic/formokaps
• Reconstruction:Proto-Italic/formos
• Reconstruction:Proto-Italic/foðjō
• Reconstruction:Proto-Italic/foɣʷeō
• Reconstruction:Proto-Italic/fronts (this term seems to also just be a fabrication as De Vaan doesn't reconstruct it)
• Reconstruction:Proto-Italic/fugā
• Reconstruction:Proto-Italic/fēzālis
• Reconstruction:Proto-Italic/fōr
• Reconstruction:Proto-Italic/fūmos
• Reconstruction:Proto-Italic/glouβō
• Reconstruction:Proto-Italic/gnāskōr
• Reconstruction:Proto-Italic/grāmen
• Reconstruction:Proto-Italic/grānom
• Reconstruction:Proto-Italic/gustus
• Reconstruction:Proto-Italic/gāwidēō
• Reconstruction:Proto-Italic/gʷelāō
• Reconstruction:Proto-Italic/gʷrūtos
• Reconstruction:Proto-Italic/gʷīwō
• Reconstruction:Proto-Italic/haizēō
• Reconstruction:Proto-Italic/heliwos

Graearms (talk) 21:52, 16 May 2025 (UTC)Reply

Also *auzōs, *awizdjō, *breɣʷisemos, *breɣʷjōs, *dormiō, *doukō, *endo, *engents, *enðernos, *ezos, *fakinos, *flagros, *flūō, *folgēō, *hāns, *hesternos, *hostis, *iowaō, *jakēō, *jakjō, *jekʷor, *jougos, *jowestos, *jowezāō, *jowos, *jugom, *jungō, *juwenis, *kadō, *kaikos, *kalowā, *kankros, *kaprā, *karkros, *kaudā, *kedate, *kedō, *kentom, *kerazrom, *kerwos, *klāwos, *klounis, *klāwis, *knuks, *koksā, *kommoinis, *komsmis, *kord, *kornū, *korpos, *korteks, *kortos, *korzō, *krāzrō, *krezðō, *krinō, *kriznis, *krūros, *kūdō, *kūlos, *kʷekʷō, *kʷelōnos, *kʷolpā, *lakus, *latēō, *lawāðrom, *legō (ety 1), *leɣʷis, *linō, *loukeō, *magjōs, *magnos, *maktos, *mari, *medēōr, *memonai, *mentis, *mentjōr, *mikskō, *molgeō, *moneō, *monēō, *mordeō, *morjōr, *mormīkā, *mortis, *mūs, *neks, *nepōts, *neptis, *nēō, *nizdos, *nogʷedos, *nokeō, *nokts, *nosteros, *nōs, *nounākentā, *ōdai, *onos, *ōwom, *ōs, *paukos, *pawēō, *pāzēō, *pelnis, *pendēō, *pendō, *perītlom, *petnā, *pezdō, *piskis, *plēō, *plēōs, *poomos, *portus, *pōtos, *praiɣodā, *priōs, *pugnos, *pūgjō, *pungō, *rapjō, *ratos, *reipā, *rēōr, *rowestikos, *rowos, *saljō, *sangwens, *sekʷontinos, *sekʷōr, *sēmən, *sentjō, *serpents, *serwāō, *seɣwēros, *sizō, *skindō, *smīɣeslī, *sniks, *sorβeō, *spēs, *stērlā, *stlātos, *stoltos, *supoleigaklom, *sūō, *swekrus, *swekuros, *swezrīnos, *swezōr, *taksō, *tempos, *tepēō, *tlokʷāks, *tlokʷōr, *tolazāō, *torzeō, *treiðlom, *troudō, *tundō, *walēō, *wāðō, *waðom, *wegeō, *wegēō, *weizos, *wenezāōr, *wenjā, *wentos, *werēōr, *westis, *wetos, *wīkentī, *wizgā, *wizēō, *wokā, *wolpis, *wormis, *wrādīks, *əngʷnis, *ənratos, *ɣlaðros, *ɣrondā, *-de, *-fakos, *-faks, *-foros, *-jos, *-jōs, *-kaps (plus everything suffixed with it), *-kaputis, *-kʷelā. Nicodene (talk) 02:43, 17 May 2025 (UTC)Reply

Yes. There's a lot. Catonif has a list started at User:Catonif/great-pit-cleanup. A problem I think is that the vote didn't really establish a clear process for removing these. Also @Graearms I'm not sure I understand why you have been editing to add more Proto-Italic forms into Latin entries such as glūbō (adding *glouβō) and auceps (adding *-kaps) if you also feel like the actual Proto-Italic entries should be deleted. Is your position that we should reconstruct these forms, but just not have separate pages for them? Some of the issues Catonif mentioned aren't resolved by that approach.--Urszag (talk) 02:09, 17 May 2025 (UTC)Reply

@Urszag I added a link back to *awis on auceps because I figured that a link back to the Proto-Italic entry would be convenient once *awikaps was deleted. I was trying to avoid generally breaking the dictionary once all these Proto-italic forms were deleted and the links dissipated. This is also why I've added the Latin descendants to PIE roots such as *dʰegʷʰ- instead of just linking the Proto-Italic term. I added terms such as *glouβō to glūbō as I figured the consensus on Wiktionary was already to show the Proto-Italic reconstruction on the Latin term as plenty of Latin terms do already mention the possible Proto-Italic form. Graearms (talk) 03:20, 17 May 2025 (UTC)Reply

Thanks for clarifying! I feel like as a general rule, it is best to avoid referring to forms constructed only on the basis of Latin as "Proto-Italic" in Latin etymology sections, since they might be pre-Latin but post-Proto-Italic (e.g. *-kaps seems like an example where that could conceivably be the case). There might be special cases where mentioning such a form is necessary, e.g. where it isn't trivial to see how the Proto-Indo-European sources developed to the Latin form.--Urszag (talk) 05:33, 17 May 2025 (UTC)Reply

The guiding principle in whether to reconstruct a proto-form should be that any proto-lemma that we are reasonably sure existed in a proto-language (and not innovated in, in this case, Old Latin or proto-Latin or proto-Latino-Faliscan) should get a page. Giving a proto-lemma a page helps us to see what terms, in aggregate, are known to exist at a proto-stage when browsing through lemma and topic (sub)categories.

Overall my thoughts and principles on these Proto-Italic nominations are case-by-case:

• Keep the following:
  • Anything (even one-descendant entries) inherited from a PIE non-root lemma. Thus I think *kord, *sniks, *mortis, *nokts, etc. should stay since PIE reconstructions of their ancestral lemmas are supported by non-Italic cognates. The PIE form must pass through Proto-Italic in order to be inherited into Latin, ergo the word must have existed in PIt. and thus should get a page.
  • Any one-descendant entry that forms a derived term attested in at least a separate branch to the single descendant on the base entry. For instance, *-kelos has a derived term that appears in Sabellic, and thus should be kept on this basis.
• Merge and redirect things that more neatly fit together as a word family, e.g. *-kaps should be merged into a mention in the related terms at *kapjō. It would be neat to be able to easily find all cross-Italic members of a word family in a single page, without having to wade through a bunch of non-Italic reflexes on a PIE page (which can get quite large).
• Delete the following:
  • Obvious dubiously back-projected Latin derivatives we have no reason to think were coined earlier than Latin, like *supoleigaklom.
  • Verbs formed with *-āō without cross-branch cognates; the suffix was productive enough that such a derivative verb could be formed at any time.

— Ceso femmuin mbolgaig mbung, mellohi! (投稿) 05:16, 17 May 2025 (UTC)Reply

nything (even one-descendant entries) inherited from a PIE non-root lemma. Thus I think *kord, *sniks, *mortis, *nokts, etc. should stay since PIE reconstructions of their ancestral lemmas are supported by non-Italic cognates.

I suppose I don't have any conceptual objection to such entries, but I do wonder what purpose they would serve?

ny one-descendant entry that forms a derived term attested in at least a separate branch to the single descendant on the base entry.

Agreed. These are de facto multi-descendant entries anyway, via their derived terms.

Merge and redirect things that more neatly fit together as a word family, e.g. *-kaps should be merged into a mention in the related terms at *kapjō.

If there are no suffixes cognate to Latin -ceps in other Italic languages, I'd honestly just consider that a derivative of Latin capiō to the extent that recipiō and such are. One could try setting up pages for Old Latin *-caps, *recapiō, etc. I suppose. Nicodene (talk) 06:25, 17 May 2025 (UTC)Reply

"but I do wonder what purpose they would serve" To populate topic and grammatical categories like Category:itc-pro:Body parts and Category:Proto-Italic consonant stem nouns? Optimally those categories should have all such topical terms reconstructible to Proto-Italic in them, and to put a word in such a category, the proto-word needs to have a page. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 14:03, 17 May 2025 (UTC)Reply

I'd honestly just consider that a derivative of Latin capiō to the extent that recipiō and such are.

By what morphological process? I'm aware that some authors think so, but there aren't really any parallels. If there's even the slightest chance that this suffix reflects a PIE root noun then it's etymologically important and interesting enough to warrant a page. The debate on whether it existed in Proto-Italic can be explained there; we have many reconstructions that probably didn't exist but nonetheless warrant an entry with the right disclaimers. The derived terms should be deleted, but then show the Latin descendants at *-kaps (or move it to *kaps). And *(-)keps should probably be listed as an alternative reconstruction on principle. — Ganjabarah (talk) 22:10, 17 May 2025 (UTC)Reply

FWIW, overall I agree with all of @Mellohi!'s stipulations (and I've always been a bit uncomfortable with blanket deletions of Proto-Foo entries with a single descendant). Benwing2 (talk) 07:10, 17 May 2025 (UTC)Reply

I would like to note one suggestion I have that I was just reminded of in this discussion: I think we should probably remove all the prefixed forms on *kapjō and *fakjō. Since all of those forms seem to only have the Latin descendant, then I don't see why we can conclude that the prefixation necessarily occurred at the Proto-Italic stage. I have some concerns as well about the case of *staēō. I assume the reconstructions *praistaēō and *eɣstaēō are based on the South Picene forms 𐌀𐌃𐌔𐌕𐌀𐌉𐌞𐌇 (adstaiúh) and 𐌐𐌓𐌀𐌉𐌔𐌕𐌀𐌝𐌍𐌕 (praistaínt). However, De Vaan categorizes all of the Sabellic forms under a PSab tem *sta-iē-, so I don't know if we can conclusively state that the prefixation occurred on a Proto-Italic term and not just the Proto-Sabellic or South Picene term.

I suppose a problem with all of these affixed forms is that you can't technically know when the affixation occurred. Even if we have multiple forms in separate Italic languages containing the affix, they still could have been borrowed or influenced by each other or just separate innovations. All these languages were capable of adding such morpehemes. I think Mellohi! was also touhing on a similar concern with his proposal to remove terms suffixed in -āō without other IE cognates. I don't necessarily know what to do about all of this though. Perhaps we could follow the model of asserō and merely state that *adserō can be reconstructed off the basis of the cognate in Oscan aserum but not necessarily that the Proto-Itailic form absolutely must have existed. I still feel like this solution leaves some problems though. If we start mentioning possible Proto-Italic reconstructions just to be comprehensive, then this principle could be extended to pages like afficiō and we could mention the existence of a possible reconstruction in *adfakjō. However, this seems way too close to what Catonif called "daydreaming." Overall, I don't really know what to do about it. I don't actually support adding dubious PIt reconstructions like *adfakjō to pages just to be slightly more comprehensive. But aren't all affixed Proto-Italic terms somewhat dubious in that you can't prove when the affixation occurred? Something also irks me about leaving out any information at all. I also don't support deleting every affixed Proto-Italic entry or really any at all currently. I was just thinking about the prefixed forms of *kapjō and *fakjō and then fell down a thought spiral tangentially related to the topic at hand. I feel as if a large part of the problem is trying to communicate to the reader that not all reconstructions are created equal: Some are far more reliable the others. When a reader looks at *adserō, they'll see it treated the same way *h₁éḱwos and thus assume that both reconstructions are equally valid. I don't really know what to do about any of this though.

I would support Mellohi!'s proposal to merge PIt terms that are part of word families. This is already basically what De Vaan does: He broadly groups all PIt terms related to each other in a single entry. See how De Vaan mentions all of the terms related to sacer on one page. This feels like a good way of ensuring that the remaining Proto-Italic entries are far more useful to the reader. You can see how a page like *kastrom becomes far more useful by centralizing a bunch of terms in one place. It also allows us to preserve the useful bits of these PIt terms with one descendant whilst avoiding all the downsides. Consider a page like *agʷnos, which recounts De Vaan's proposed connection to avillus without needing a whole separate page.

We could also do this we derived terms. De Vaan does this as well: For instance, he includes *sakrātrīks alongside his discussion of other terms related to sacer. Basically, this would entail merging PIt terms derived from other PIt terms onto the page for the original term. I can foresee such merges further centralizing information, allowing for the remaining PIt entries to be even more comprehensive and more useful. Terms such as *serwos and *opos become much more useful by including all the Italic terms ultimately derived from it. If we had a separate page for *opezāō and then moved all the Sabellic descendants to that page, then the original entry for *opos becomes somewhat redundant. Such merges could also allow us to get rid of entries like *adserō, which contains only information also easily includable on *serō. I am willing to concede that this proposal might be more a product of my own daydreaming and my own personal preference for big PIt entries that contain a lot of information rather than any major benefit this provides to the dictionary. Graearms (talk) 15:08, 17 May 2025 (UTC)Reply

I would support Mellohi!'s proposal to merge PIt terms that are part of word families.

This isn't so much a proposal as a standard practice on Wiktionary. There's no world in which we cease to mention all of these on some entry.

Such merges could also allow us to get rid of entries like *adserō

What? There's absolutely no reason to do that. If it's a separate word in Proto-Italic, it's at the level of a separate entry (whether we host the entry on its own page or not—in this case we do and have no reason not to). That's like saying we should have no entries for English unhappy, happily, happier, happiest, happiness or unhappiness, because they “contain only information also easily includable on happy.” Who needs categories like Proto-Italic terms prefixed with *ad- English terms prefixed with un-, right? Except your case is even more drastic, because there are also languages in the descendants sections that would be lumped too. — Ganjabarah (talk) 22:22, 17 May 2025 (UTC)Reply

I'm not definitively committed to this idea myself and I do understand that its a bit (perhaps a lot) fantastical. There seems to be some precedent for this notion in the existing Proto-Italic entries that include all the information of derived terms (like *weɣō or *speks) and also the PIE terms that also include the derived terms and their descendants. I would also like to note that those English entries do contain information not includable within the unaffixed forms: pronunciation, quotes, example sentences, and translations. Also, the grammatical paradigms and inflection table templates that can be included on these pages are far more useful in attested languages that people actually use, as some people might use them to study the language or they might search for a non-lemma form on Wiktionary to reach the lemma form. I still am not willing to die on this hill and I understand why someone find it idiotic. Graearms (talk) 23:12, 17 May 2025 (UTC)Reply

I think you misunderstand the motivation. Those are all not permanently dependent on the larger entry; the ones with reflexes in multiple languages just haven't been created yet. Once created the descendants will be moved but the headword will remain, as a bluelinked (how ironic) derived term. No Wiktionary rule, guideline or common-sense principle is stopping me from going ahead and creating both Proto-Italic *adweɣō and Proto-Italic *komweɣō right now.

On that note, thanks for reminding me to create *adferō. — Ganjabarah (talk) 23:24, 17 May 2025 (UTC)Reply

Okay, I'm willing to concede this point. I probably know a lot less about linguistics than you do, so I am willing to take your word for it. Graearms (talk) 23:31, 17 May 2025 (UTC)Reply

By the way:

When a reader looks at *adserō, they'll see it treated the same way *h₁éḱwos and thus assume that both reconstructions are equally valid.

I could argue that *adserō is more valid in some ways than is *h₁éḱwos, and not just based on the different time depths. For instance, it was very likely athematic *h₁éḱus in an earlier stage of PIE and therefore the page should probably be moved to that. And *adserō is less likely to be a wanderwort, whereas words for horses spread all the time. Not to mention the regularity of descendants—did Proto-Hellenic *íkkʷos actually come from it regularly?
This is a bit tongue-in-cheek; I'm not saying I seriously doubt *h₁éḱwos is accurate to one form of the word that existed at some time and place, just that we can be a bit more sure about the phonetics of *adserō if it existed than we can about *h₁éḱus/*h₁éḱwos, even if the latter is more likely to have existed in some form. My point is that all of these protolanguage reconstructions have asterisks for a reason, and a simple rule like "it's more uncertain if it had a prefix" doesn't carry all that much weight in the scheme of things. The only way to convey all the information you desire about levels of certainty would be piling up endless notes and disclaimers—but we can do something even better: point readers to sources, which ideally they will read until they reach an expert level of understanding of the subject. That would finally solve it. ;) — Ganjabarah (talk) 23:10, 17 May 2025 (UTC)Reply

I assume the emoticon at the end is a subtle insult direct towards me. I get it; you won the argument. I don't think I'm as knowledgeable as you are, so I won't try and assert that my opinion is more valuable than yours in this discussion. I think I may have taken this entire discussion in a direction it was not originally intended to go. I was mostly thinking about 𐌀𐌃𐌔𐌕𐌀𐌉𐌞𐌇 (adstaiúh) and 𐌐𐌓𐌀𐌉𐌔𐌕𐌀𐌝𐌍𐌕 (praistaínt) and got sidetracked. I'll concede all my points. I do get that they may have been generally ignorant and stupid. Graearms (talk) 23:56, 17 May 2025 (UTC)Reply

It's not meant to convey superiority or malice of any kind lol. Actually the opposite: it means I'm being jovial with my sarcasm. I was saying it's an unsolvable problem, as one would have to be an expert to understand all the assumptions, nuances, biases and so on that go into any reconstruction (and even then likely miss a few). There is a threshold of relevance below which Wiktionarians must agree not to worry about the issues, and I think the concerns you raised about *adserō fall just short of that threshold. It's a valid concern and fine that you brought it up; this is not a battle and no one is keeping score. — Ganjabarah (talk) 00:49, 18 May 2025 (UTC)Reply

┌┘

Proposed plan, in the interest of building consensus:

1. Handle derived terms the way we do for PIE, with e.g. *méh₂tēr housing *meh₂ter-yḗh₁-s and descendants thereof.
2. Keep PI reconstructions that are supported by descendants of secure etymology, including descendants of derived terms, in more than one Italic language.
3. Keep PI reconstructions for any term inherited in a straight line from a secure PIE reconstruction, with only trivial morphological changes. E.g. Mellohi's *kord, *sniks, *mortis, *nokts.
4. Remove PI reconstructions that simultaneously fail the conditions for 2 and the conditions for 3.

Thoughts? Nicodene (talk) 22:01, 17 May 2025 (UTC)Reply

Support tentatively. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 23:54, 17 May 2025 (UTC)Reply

I agree almost verbatim with Mellohi!'s explanation above: The guiding principle in whether to reconstruct a proto-form should be that any proto-lemma that we are reasonably sure existed in a proto-language should get a page, on whatever consensus basis, and so I too prefer to consider them case by case rather than based on a rigid set of rules, which will tend to miss something. To your list I would add that many terms in Latin are known to have been formed earlier than Latin based on our understanding of that language, but that doesn't necessarily make them of Proto-Italic age, and so that's where the logic gets messy and subjective. Perhaps more "Old Latin" reconstructions would work? But the issue there is that we don't know as much about Old Latin morphology to know what was productive in it. For the same reason we have many North Germanic–only entries in Proto-Germanic rather than reconstructing them in Proto-Norse (an attested language). The established protolanguages are more convenient and we understand them better, resulting in a tradeoff between convenience + morphophonemic accuracy and time-depth accuracy. Note also that there is some dispute about what counts as a unique Italic language: Lanuvian and Praenestine are too poorly attested to know whether they were sister languages or dialects of Old Latin (Wiktionary assumes the latter, but see Latino-Faliscan languages on Wikipedia). Since we don't currently recognize Proto-Latino-Faliscan, their most recent recognized common ancestor would have to be Proto-Italic if they are even slightly separated from Old Latin. This should only be an issue in a couple of cases. — Ganjabarah (talk) 22:15, 21 May 2025 (UTC)Reply

We can provisionally accept ‘Proto-Latino-Faliscan’ reconstructions as Proto-Italic, pending a possible split in the future. I'm not sure we have any as of yet.

To your list I would add that many terms in Latin are known to have been formed earlier than Latin based on our understanding of that language

Without meeting the criteria for either 2 or 3 above?

Bearing in mind that unstressed vowel reduction of the *-faks > -fex type ‘took place within VOL , probably between 500–300 BC’ (LaFond 2024: 164). A reconstruction based solely on undoing that process takes us no further back in time than for instance the mamarcom attested in Satricum in the ‘mid-6th century BCE’ (Rigobianco 2024: §4.1). Nicodene (talk) 11:45, 22 May 2025 (UTC)Reply

Proto-Austronesian. Not listed in ACD as a Proto-Austronesian word. Alfarizi M (talk) 19:59, 22 May 2025 (UTC)Reply

Proto-Sino-Tibetan. Tibetan form shows no sound correspondence whatsoever to the Old Chinese reconstructions. It does resemble the Middle Chinese form more closely, indicating that the word is a post-PST Wanderwort, and thus we can't reconstruct a PST form. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 15:15, 24 May 2025 (UTC)Reply

Delete. — justin(r)leung _{{ (t...) | c=› }} 03:08, 26 May 2025 (UTC)Reply

Comment: This may also be relevant. — justin(r)leung _{{ (t...) | c=› }} 17:17, 26 May 2025 (UTC)Reply

Proto-Sino-Tibetan. Yet another sound correspondence disaster, what more can be said here? — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 19:03, 24 May 2025 (UTC)Reply

Keep but significantly revise it. Jacques (2022) (admittedly not yet published) proposes that: One term for a crustacean is possibly reconstructible, attested in Chinese (蟹 ɣɛɨX ← *m-kˁreʔ ‘crab’) and Kiranti (Khaling ghrɛ̂ː ‘crayfish’, from pre-Khaling *gʰrawa, Jacques 2016). (p. 24). Erminwin (talk) 05:23, 26 May 2025 (UTC)Reply

Proto-Sino-Tibetan. Another sound correspondence disaster. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 02:48, 26 May 2025 (UTC)Reply

I'm not well-versed enough in ST sound-correspondence (e.g. "ablaut of *u ~ *a in the proto-language" (source: Reconstruction:Proto-Sino-Tibetan/kV-sum)) to vote keep or delete. Even so, I recommend removing 莽 and 茻 from list of descendants, as they both belong to division I (一等), so their Old Chinese ancestors likely had no medial *-r-, thus making them unlikely descendants of PST *m-r(a/u)k or *m-ljak. Erminwin (talk) 20:23, 26 May 2025 (UTC)Reply

Old High German.
This word is attested, as Old High German lenc (“left”, adjective) and also found in the derivative lenka (“left hand”) (—the page shows a linka with the same meaning). Middle High German shows both lenc~linc (“left”), as does Middle Low German (lenk~link), so it's conceivable that OHG had an unattested *link form (?), but if so, it should point to lenc, lenk. Leasnam (talk) 17:20, 29 May 2025 (UTC)Reply

I've created an entry at lenk. Leasnam (talk) 17:51, 29 May 2025 (UTC)Reply

Resolved. Leasnam (talk) 03:46, 1 June 2025 (UTC)Reply

Proto-Finnic. Obsolete and probably also incorrect reconstruction, which is also irregular (one would explain *neici). (The supposed PU *nejde also looks a bit suspect.) — SURJECTION ^{/ T / C / L /} 18:51, 29 May 2025 (UTC)Reply

EVE suggests Proto-Finnic *neici might have existed and inherited from Proto-Uralic *näjte. — SURJECTION ^{/ T / C / L /} 19:00, 29 May 2025 (UTC)Reply

@Surjection: *neici wouldn't give any of the attested descendants regularly though, would it? One could posit an analogical leveling *neici : *neiẟen to *neiti : neiẟen, but this didn't have to happen in the descendants individually, it could have been in (a late stage of) the proto-language already. Thadh (talk) 20:37, 29 May 2025 (UTC)Reply

You're correct, *neici would not have given any of the descendants, and I don't think any descendant would be attested in any language either. But I suppose the argument is that *nici, *nite- > *nitoi / *nitei. e-stem inflection (neiti > neiden) is found in older/dialectal Finnish, though; that would have to be analogous. — SURJECTION ^{/ T / C / L /} 20:40, 29 May 2025 (UTC)Reply

@Surjection: actually, now I think of it, neiti-neijen is also attested in Ingrian runos. A quick search gives hits in Venakontsa, Kolkanpää and Säätinä, so both Ingrian Finnish and Soikkola Ingrian (Izhorian). I wonder if Karelian and Votic runos show such a form as well. Thadh (talk) 21:44, 29 May 2025 (UTC)Reply

Okay, definitely (Northern) Karelian and a lot more Ingrian Finnish in Central and Northern Ingria. Not sure about Votic, I can find one occurrence of neide (Neide päivä valgeuttasa), but I can't gather from context whether this is from *neiti or need. Thadh (talk) 22:04, 29 May 2025 (UTC)Reply

I believe all entries for Proto-Nuristani should be deleted, as they do not meet Wiktionary’s standards for reconstructed languages — a concern I have raised previously.

The primary source for these reconstructions, Richard Strand’s website, is unpublished and outdated, falling short of WT:CFI#Reconstructed languages requirements. The reconstructions themselves are little more than slightly altered forms of Proto-Indo-Iranian, without a systematic account of sound changes across Kamkata-vari, Ashkun, Waigali, and other Nuristani varieties.

In short, these entries are ad hoc, insufficiently sourced, and noncompliant with WT:RECONS. I therefore petition they be deleted pending the publication of citable, comparative research.

--{{victar|talk}} 22:00, 3 June 2025 (UTC)Reply

Note the existence of D. N. Nelson's thesis "THE HISTORICAL DEVELOPMENT OF THE NURISTANI LANGUAGES (KAFIRI, INDO-IRANIAN)". Table of Contents advertises a section with changes from Proto-Nuristani to the Nuristani lects. Can anyone with institutional access evaluate this? — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 07:14, 7 June 2025 (UTC)Reply

Pinging @Kwékwlos, the creator of the nominated entries, to ask for further information/sources on sound correspondences/changes in Nuristani. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 07:28, 7 June 2025 (UTC)Reply

There's https://www.researchgate.net/publication/298750373_On_the_position_of_Nuristani_within_Indo-Iranian by Irén Hegedűs and Vaclav Blazek which provides a set of PNur. correspondences, but it seems to be very tentative and contains some errors (eg. PNur. *c /ts/ > Ash. c /ts/ instead of /s/). Jakob Halfmann is working on a Proto-Nuristani etymological dictionary, but he hasn't revealed much. But we can examine Halfmann's articles (https://www.researchgate.net/publication/368760201_Advances_in_the_historical_phonology_of_the_Nuristani_languages, https://www.researchgate.net/publication/380269159_Nuristani_Theonyms_in_Light_of_Historical_Phonology, and https://www.researchgate.net/publication/376709559_Lad_law_a_Bactrian_loanword_in_the_Nuristani_languages), which further elucidates some aspects of Nuristani historical phonology.

I suppose we'll have to wait for Halfmann's output if we want to start adding Proto-Nuristani reconstructions. Kwékwlos (talk) 12:53, 7 June 2025 (UTC)Reply

With no direct descendants, this is too speculative for its own page. --{{victar|talk}} 06:28, 7 June 2025 (UTC)Reply

Delete, but only because Proto-Brythonic *kaɨr has an equally good if not better alternative etymology, and so there may be no form in *-r-. — Ganjabarah (talk) 06:21, 21 June 2025 (UTC)Reply

To be clear, I'm not objecting to having the descendants therein listed on the root entry, just that this medial form is too speculative for its own page. --{{victar|talk}} 15:18, 30 June 2025 (UTC)Reply

Delete. Probably did not exist. Fay Freak (talk) 15:38, 30 June 2025 (UTC)Reply

No Iranian descendants after removal of Avestan term. I suppose we can have a page for a Proto-Indo-Aryan reconstruction, but the connection with PIE *h₁wer- (“broad”) is in any case very speculative and doesn't seem to be even mentioned by Mayrhofer. Exarchus (talk) 16:12, 7 June 2025 (UTC)Reply

Of the few PIA entries that should exist, this is one of them. I say it should be moved to RC:Proto-Indo-Aryan/Hwárunas. --{{victar|talk}} 06:24, 8 June 2025 (UTC)Reply

I'd say move to PIA *Wárunas, as the laryngeal was based on a shaky connection (is there even a source for this?) to PIE *h₁wer-. Exarchus (talk) 08:05, 8 June 2025 (UTC)Reply

The loss of laryngeals in PIA is a can of worms probably better left to its own separate discussion. --{{victar|talk}} 10:21, 8 June 2025 (UTC)Reply

Loss of laryngeals might be a can of worms, but there's no indication there was a laryngeal here to start with. Exarchus (talk) 10:48, 8 June 2025 (UTC)Reply

It's hardly definitive proof though, and could be explained multiple ways, i.e. transmission loss into Hittite, or a feature late-PIA or dialectal Mitanni. --{{victar|talk}} 20:53, 8 June 2025 (UTC)Reply

If you want to indicate where there might have been a laryngeal, then you'd need to give *(H)wár(H)unas Exarchus (talk) 07:45, 9 June 2025 (UTC)Reply

Sure, but we should be consistent -- either reconstruct Proto-Indo-Aryan as retaining laryngeals in all positions, or not. Incidentally, not all scholars agree that Mitanni was even PIA, i.e. Parpola (1999). --{{victar|talk}} 10:14, 11 June 2025 (UTC)Reply

Redirect. It never existed in PBS, useless reconstruction for Slavic word that is Slavic innovation. Sławobóg (talk) 15:18, 18 June 2025 (UTC)Reply

Delete, you are right. It is unlikely to be used, too, though maybe-maybe occurring in literature. Fay Freak (talk) 15:40, 30 June 2025 (UTC)Reply

Proto-West Germanic. Reconstruction is very tenuous: this word is generally regarded as being a late medieval German product of dissimilation; note the etymology (and sources) at sammeln - I cannot find sources that contradict this and endorse a common (West-)Germanic origin. The word first pops up in the 14th century in High and Low German and shortly thereafter spreads elsewhere, appearing first in northeastern Middle Dutch (as evidenced by the derivation samelinge, early 15th c.) and much later still in West Frisian (sammelje is listed as mid-19th century by the WFT). I find no evidence prior to the 14th century. — Mnemosientje (t · c) 14:55, 20 June 2025 (UTC)Reply

I'll move the descendants back to Proto-West Germanic *samnōn Leasnam (talk) 02:55, 22 June 2025 (UTC)Reply

Assuming this true, it is conlanging and should be deleted, and put under a descendants section in Middle High or Low German (somewhere around the isoglosses the difference between High German and Low German is not real anyway). Fay Freak (talk) 15:43, 30 June 2025 (UTC)Reply

Uncertain etymology, so the "Proto-Albanian" form cannot be reconstructed. At least two alternatives are possible, which see at gjalpë. — Ganjabarah (talk) 06:18, 21 June 2025 (UTC)Reply

Never mind, it's probably fine with the updated etymology. But probably still doesn't warrant an entry. — Ganjabarah (talk) 21:03, 21 June 2025 (UTC)Reply

Delete: Same issue found with many Proto-Albanian entries. -- {{victar|talk}} 16:06, 1 July 2025 (UTC)Reply

Changing to keep. It's a pretty well agreed upon etymology, and those alternatives are just different reconstruction conventions. --{{victar|talk}} 05:59, 4 July 2025 (UTC)Reply

There are no descendants in the entry. Lerman (talk) 20:05, 27 June 2025 (UTC)Reply

There are atleast 7 derived terms which is enough to have it. Sławobóg (talk) 20:22, 27 June 2025 (UTC)Reply

Right, this term is generally agreed upon, having formed endearing terms in the later proto-period. Keep. Fay Freak (talk) 15:45, 30 June 2025 (UTC)Reply

Slovene actually has an o-stem sòłn m, translated as "Lichtreflex im Auge" in Pleteršnik's dictionary. It's not exactly a descendant, but still close enough. Безименен (talk) 17:33, 30 June 2025 (UTC)Reply

Proto-Slavic.

Created prematurely. Descendants are attested barely around 15-16 century + etymology is uncertain. Not worthy of Proto-Slavic treatment. I already moved etymological details and refs under Bulgarian дори (dori), so nothing of value is going to be lost when the pSl. entry is deleted. — This unsigned comment was added by Bezimenen (talk • contribs) at 09:48, 28 June 2025 (UTC).Reply

Note: The above editor motioned to delete his own entry after four years. Fay Freak (talk) 15:46, 30 June 2025 (UTC)Reply

I have removed Old High German Saturnus as a descendant as it defies the High German consonant shift. It goes without saying that the Old English is also much more likely a direct borrowing from Latin which occurred after West Germanic ceased to be a single dialect continuum. The reconstruction also does not even match the first vowel of *Sāturnas dag, which it is supposed to support. There is no purpose to this reconstruction (which exists only on Wiktionary and more specifically in Victar's mind). Note also that the Ripuarian descendant at *Sāturnas dag should have undergone the sound shift (no intervovalic /t/ is safe) if it is inherited. Gōdanā Sā̆turnas dag iwwi! — 2600:4808:9C30:C500:A054:39DA:F18C:3FC9 14:59, 28 June 2025 (UTC)Reply

The ad hominem attack is petty. --{{victar|talk}} 16:38, 30 June 2025 (UTC)Reply

• Delete —Mahāgaja · talk 12:34, 29 June 2025 (UTC)Reply

Delete. Old English Saturnus cannot possibly be inherited from PWGmc. ᛙᛆᚱᛐᛁᚿᛌᛆᛌ ᛭ Proto-Norsing ᛭ Ask me anything 14:09, 30 June 2025 (UTC)Reply

Delete: Roman gods were certainly known to WG peoples, but there probably isn't much point to having this entry just for *Sāturnas dag. @Mnemosientje --{{victar|talk}} 16:38, 30 June 2025 (UTC)Reply

RFD-(speedy-)deleted. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 20:31, 8 July 2025 (UTC)Reply

Proto-Germanic *ī gives six derived terms, but all of these seem to occur only in Gothic, so I'd rather assume 𐌴𐌹 (ei) became a productive enclitic particle in Gothic. This would imply deletion of the following pages:
- Reconstruction:Proto-Germanic/izī
- Reconstruction:Proto-Germanic/saī
- Reconstruction:Proto-Germanic/sunizī Exarchus (talk) 14:01, 29 June 2025 (UTC)Reply

Entry based on dubious etymology and created by since permablocked user User:Dardhanos. Should be deleted and left to the etymology on Albanian urtë. --{{victar|talk}} 00:32, 1 July 2025 (UTC)Reply