Wiktionary Request pages (edit) see also: discussions
Requests for cleanup add new \| history \| archives Cleanup requests, questions and discussions.	Requests for verification/English add new English request \| history \| archives Requests for verification in the form of durably-archived attestations conveying the meaning of the term in question.	Requests for verification/CJK add new CJK request \| history Requests for verification of entries in Chinese, Japanese, Korean or any other language using an East Asian script.	Requests for verification/Italic add new Italic request \| history Requests for verification of Italic-language entries.	Requests for verification/Non-English add new non-English request \| history \| archives Requests for verification of any other non-English entries.	Requests for deletion/Others add new \| history Requests for deletion and undeletion of pages in other (not the main) namespaces, such as categories, appendices and templates.
Requests for moves, mergers and splits add new \| history \| archives Moves, mergers and splits; requests listings, questions and discussions.	Requests for deletion/English add new English request \| history \| archives Requests for deletion of pages in the main namespace due to policy violations; also for undeletion requests.	Requests for deletion/CJK add new CJK request \| history Requests for deletion and undeletion of entries in Chinese, Japanese, Korean or any other language using an East Asian script.	Requests for deletion/Italic add new Italic request \| history Requests for deletion and undeletion of Italic-language entries.	Requests for deletion/Non-English add new non-English request \| history \| archives Requests for deletion and undeletion of any other non-English entries.	Requests for deletion/Reconstruction add new reconstruction request \| history Requests for deletion and undeletion of reconstructed entries.
{{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

Templates:

{{rfd}}
{{rfd-sense}}
{{rfd-redundant}}
{{archive-top|rfd}} + {{archive-bottom}}

Shortcuts:

2019 edit

Reconstruction:Proto-Germanic/erþaburgz edit

I have no problem with the word or whenever it existed or not. My issue is in regards to whenever the word should be reconstructed as Proto-Germanic *erþaburgz (“earthen mound, earthwork”) or *erþōburgz. This example is one of many PGmc where the first noun of the reconstructed compound ends with "ō" but the reconstructed compound has medial "a". I would normally check the descendent to see if I can deduce more information, however, most have no medial compound vowel e.g. Old English eorþburh, Old High German erdburg, Old Norse jarðborh. So now, I'm left wondering what form it should be. 𐌷𐌻𐌿𐌳𐌰𐍅𐌹𐌲𐍃 𐌰𐌻𐌰𐍂𐌴𐌹𐌺𐌹𐌲𐌲𐍃 (talk) 03:19, 2 November 2019 (UTC)Reply[reply]

The default medial vowel in pre-Germanic had become -o- for the thematic classes, as in Celtic and Latin. PGmc medial*-ō- would presumably have left some trace in OHG. Burgundaz (talk) 08:54, 28 April 2020 (UTC)Reply[reply]

Kept. The discussion both here and on the talk page really seems to be whether the page should be moved to a different title, not deleted outright. —Mahāgaja · talk 14:00, 21 August 2023 (UTC)Reply[reply]

May 2020 edit

Reconstruction:Proto-Germanic/gallô edit

Since all of the descendants from this have been moved over to *gallǭ, I think this can be deleted. DJ K-Çel (talk) 02:34, 30 May 2020 (UTC)Reply[reply]

No. *gallô is the ancestor of the OE form, and *gallǭ the rest. --{{victar|talk}} 02:49, 30 May 2020 (UTC)Reply[reply]

Well, at this discussion @Leasnam: had said: "I've moved *gallō to *gallǭ, since the West Germanic descendants are weak. I've also added the descendants of *gallô to *gallǭ. I think we can delete *gallô."

But it looks like English gall and its ancestors were deleted about a week ago from *gallǭ. DJ K-Çel (talk) 03:01, 30 May 2020 (UTC)Reply[reply]

It all depends on whether we want to keep the *gallô page solely for the lone Old English galla. Or we could consider the OE term a gender change from Proto-West Germanic *gallā f from Proto-Germanic *gallǭ and place it there. Leasnam (talk) 04:05, 22 July 2020 (UTC)Reply[reply]

keep - this is a different word to *gallǭ. I've made updates, and removed the tag. Leasnam (talk) 19:50, 3 August 2023 (UTC)Reply[reply]

RFD-kept —Caoimhin ceallach (talk) 14:26, 16 October 2023 (UTC)Reply[reply]

Reconstruction:Proto-West Germanic/būti edit

Only has reflexes in one descendant, and an uncertain borrowing. This could easily have been formed within the separate history of Dutch. —Rua (mew) 10:52, 10 March 2020 (UTC)Reply[reply]

Keep: The etymology is so widely circulated that even if it is wrong, which is hard to say (though I do prefer a direct Gaulish etymology for the Latin), it should just have an entry anyhow. --{{victar|talk}} 19:58, 10 March 2020 (UTC)Reply[reply]

But should that be a Proto-West Germanic entry? The term is literally has only one descendant, that's not enough evidence to claim it's of PWG date. —Rua (mew) 20:15, 10 March 2020 (UTC)Reply[reply]

Well, it's mostly reconstructed as PG, so PWG is even safer, no? --{{victar|talk}} 20:46, 10 March 2020 (UTC)Reply[reply]

A term with only a Dutch descendant (which is doubtful, as Etymologiebank says the term is Low German in origin) can't even be reconstructed for PWG, let alone PG. —Rua (mew) 10:38, 11 March 2020 (UTC)Reply[reply]

And others the opposite, and others still both inherited. --{{victar|talk}} 20:10, 14 March 2020 (UTC)Reply[reply]

Related to this a Frankish label could be handy for PWG with only Dutch and Latin descendants. --{{victar|talk}} 20:58, 10 March 2020 (UTC)Reply[reply]

Reconstruction:Proto-West Germanic/fellō edit

Same as *dubbjan above. —Rua (mew) 12:37, 10 March 2020 (UTC)Reply[reply]

Keep: OHG added. --{{victar|talk}} 19:52, 10 March 2020 (UTC)Reply[reply]

You're aware that this is RFV, right? There's no keep/delete votes. —Rua (mew) 20:16, 10 March 2020 (UTC)Reply[reply]

Keep: --{{victar|talk}} 20:25, 10 March 2020 (UTC)Reply[reply]

...kay. —Rua (mew) 10:38, 11 March 2020 (UTC)Reply[reply]

@Rua: So does that resolve this? --{{victar|talk}} 20:08, 14 March 2020 (UTC)Reply[reply]

I'm not entirely sure if having only an OHG descendant is enough either. But I'll leave that to third parties to decide. @Mnemosientje, Mahagaja, DerRudymeister —Rua (mew) 20:18, 14 March 2020 (UTC)Reply[reply]

If there really aren't any other West Germanic reflexes, then I'd be inclined to delete and just say the Latin is a loanword from OHG. It's not clear where OHG fello comes from, though, since Proto-Germanic *faluz doesn't have an OHG reflex. —Mahāgaja · talk 21:14, 14 March 2020 (UTC)Reply[reply]

Based on the context of the Latin attestations, it looks to have originated from Frankish, not OHG, and if we were to say it didn't exist in PWG, we have to somehow explain how it was novelly constructed in OHG. --{{victar|talk}} 08:00, 1 April 2020 (UTC)Reply[reply]

Could it possibly be from Proto-Germanic *fal(l)jô, a derivative of *faluz, or perhaps a derivative of Old Dutch fellen (“to destroy, ruin”)/Old High German fellen (“to throw down, defeat, demean, devalue”) ? Leasnam (talk) 02:20, 24 August 2020 (UTC)Reply[reply]

I think it makes most sense to move the entry to *falljō, from Proto-Germanic *faljô, a derivative of *faluz. Leasnam (talk) 22:27, 28 January 2023 (UTC)Reply[reply]

PWG *-jō was plenty productive, so *falu +‎ *-jō is also possible. --Sokkjō (talk) 09:44, 29 January 2023 (UTC)Reply[reply]

I've moved the entry to RC:Proto-West Germanic/falljō. @Mahagaja, Leasnam if you believe this addresses the RFD, please resolve. -- Sokkjō 19:44, 16 October 2023 (UTC)Reply[reply]

Reconstruction:Proto-Germanic/daliþō edit

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[reply]

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

April 2020 edit

Reconstruction:Proto-Slavic/vьśь edit

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

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

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

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

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

Care to share the recordings of native late Proto-Slavic speakers proving that “ś didn't exist”? Something like *Ix → *Iç → *Iɕ replace I with ɪ, invalid IPA characters (I→*I→*I) 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[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[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[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[reply]

Reconstruction:Proto-Germanic/wirtiz edit

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[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[reply]

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

July 2020 edit

Reconstruction:Proto-Indo-European/bʰeh₂ǵos, Reconstruction:Proto-Slavic/bazъ edit

“Transformed” Pokorny stuff, ominously sourced by the Leiden school.

The beech isn’t in the range (!) of the Proto-Indo-European homeland.
The Slavic page is properly *bъzъ. The Serbo-Croatian descendant does not count for *bazъ because Proto-Slavic *ъ generally gives a in Serbo-Croatian, the Russian and Ukrainian given are obscure dialectal forms, as well as the Bulgarian, which are unstressed while Bulgarian has suffered vowel reduction and Bulgarian а (a) and ъ (ǎ) are very close; ominously one gives an Old Church Slavonic only for *bъzъ. The current Slovak form which I added, apart from being anomalous as a feminine, can also be from ъ, this can be seen *dъždžь → dážď and the variation for *čexъlъ. Against the evidence from all Slavic languages one cannot posit such a byform, more easily *bazъ is an etymologist’s fabrication to shoehorn all into an Indo-European-etymology. Which does not work anyhow because the Slavic words mean elder, not beech. These plants are not confusable.
- The page is in ESSJa, ’tis true, but apart from the entry’s age as I have noticed often, they do not take a stand for every entry in their Proto-Slavic dictionary, which is but hypothetical. They apparently create some index files, here motivated by Pokorny, and look what they can find to support the form, then they publish all anyway if the result is negative. See the RFD already filed for the adjective *bazovъ in WT:RFDO, Useigor did not understand this and created bare objectionable entries this way.
Proto-Germanic *bōks means “book” but there is yet no proof the Germanic peoples used beechbark writing or anyone else as opposed to birchbark writing. And how can *bōkō (“beech”), different paradigms, be from the same Proto-Indo-European form? There is something unaccounted. The existence of that word also conflicts with *bʰeh₂ǵʰús (“arm”) giving *bōguz, as the consonant outcome differs and because “the slot is filled” i.e. the alleged word for a tree is too similar to a word for the arm for both having existed.
- Also the sound correspondence is not necessariy, I see that PIE eh₂ can give (with and without acute) ō, there is *bʰréh₂tēr → *brōþēr, *wréh₂ds → *wrōts, but it depends on ablaut and more typical is the result ā or a, e.g. *séh₂ls → *saltą, *stéh₂tis → *stadiz. Also considering the actual sound value in PIE [a] given at *bʰréh₂tēr. This additionally decreases the likelihood of PIE origin.
Albanian bung is very tentative and random as always.
Armenian բոխի (boxi) has been thrown out of the equation meticulously after the creation of the PIE, much reasoned at its entry.
Where is the Gaulish word attested? Probably fishy if it is claimed to be only Gaulish but not retained in other Celtic languages. What do the other Celtic languages have? With such things I am accustomed to have the suspicion that it is somehow conjectured from unfathomable placenames.
The Latin word may be an early borrowing from Northwest Greek φᾱγός (phāgós), like even mālum (“apple”); as Italy was Greek-settled and the beech is found in Italy only at some places and not right at Rome, only somewhat outwards. Whereas the beech is very frequent in the Proto-Hellenic area. In Latin likely a foreign word. I say this also from general impressions about substratum origins of Latin plant names, after having dealt with many Latin plant names and their origins.
This is well a loanword after Proto-Indo-European when Germans, Italians/Romans and Greeks took new settlements judging by analogy. Remarkably the Slavic words *bukъ and *buky are Germanic borrowings for some reason, apparently because the Slavs settled right at the Northeast of the distribution of the beech, of course also Hungarian bükk (“beech”) is loaned. So if not even the Slavs before expansion (3rd century CE) had a word for the beech, the Proto-Indo-Europeans hadn’t either; if the Slavs borrowed this word, the Germans and Greeks and Romans did it likewise earlier. The correct etymologies for the German and Greek words are “borrowed from an unknown source common to [Greek|Proto-Germanic]”. Fay Freak (talk) 15:37, 27 July 2020 (UTC)Reply[reply]

@Fay Freak: agreed. This has always been a dubious reconstruction, made worse by shoehorning more descendants to it, and further comical by reconstructing it with *-eh₂-. Also see {{R:ine:HCHIEL|86}} --{{victar|talk}} 18:28, 30 July 2020 (UTC)Reply[reply]

I have read it. So I have found it is actually a debunked canard since half a century ago, called beech argument. It might have went past the Soviet theorists. In Krogmann, Willy (1954), “Das Buchenargument”, in Zeitschrift für vergleichende Sprachforschung auf dem Gebiete der Indogermanischen Sprachen (in German), volume 72 1./2, →DOI, page 13 it is expounded how the Gaulish name is derived by reconstruction, from placenames. It is to be added that the literature finds it problematic that the Greek word means an oak and not a beech. Fay Freak (talk) 20:44, 31 July 2020 (UTC)Reply[reply]

I disagree. Where else would "book" come from — This unsigned comment was added by 72.76.95.136 (talk) at 19:57, 6 December 2020 (UTC).Reply[reply]

It says where it possibly comes from. Often explained as in Germanic from the word for beech, which last is a word borrowed from somewhere. I do not need to have an explanation for or know everything to disprove an etymology. Your argument is none. Otherwise aliens built the pyramids because “how else”. Fay Freak (talk) 23:32, 16 December 2020 (UTC)Reply[reply]

Several invalid arguments here.

1) “Transformed” Pokorny stuff, ominously sourced by the Leiden school." -- This is frankly just rude, not reasoned. Kroonen's dictionary is extremely respectable (even if one disagrees with it) and tested by peer-review, unlike this nomination for deletion. The reconstruction is cited by philologists in other "schools" than Leiden. Check out e.g. Ringe (Pennsylvania/Oxford). Wiktionary should be reflecting the general scholarly consensus, not novel, non-peer-reviewed proposals of independent-minded contributors.

2) "beech not in homeland" -- irrelevant, as many words change in meaning over time, and with different environments in different geographical locations

3) "yet no proof the Germanic peoples used beechbark writing". No, but Germanic peoples' first contact with "books" would probably be Roman writing-tablets, which were often made of beech wood. (See e.g. https://en.wikipedia.org/wiki/Vindolanda_tablets)

4) "Latin word may be an early borrowing from Northwest Greek φᾱγός" -- Even if that is true, that still leaves Germanic (NW Indo-European) and Greek (S Central Indo-European) as cognates, which is generally regarded as sufficient to support the hypothesis that it is ancestral to both of those branches. But what on earth is "an unknown source common to [Greek|Proto-Germanic]" other than the common ancestor of the European side of PIE?

Signed: an anonymous academic peer-reviewer, who is a tenured Professor in a Philology Faculty (no, not Leiden). But the decision about whether to delete the page or not should be taken on the merits of the arguments alone. 82.132.228.243 11:19, 9 December 2021 (UTC)Reply[reply]

Furthermore, the argument that “the slot is filled” i.e. the alleged word for a tree is too similar to a word for the arm for both having existed" is unreasonable, because homophony and doublets are actually perfectly common phenomena cross-linguistically. 94.196.220.242 14:57, 9 December 2021 (UTC)Reply[reply]

You are rude and not reasoned, okay?

Bare editors are stricter about meaning differences than I am, e.g. I presume @Metaknowledge mildly not amused about this lumping beeches and elders and what not. To reconstruct we need to pin down a more or less vague meaning, which these equations do not meet, and formally it is a scarecrow greater than many reconstruction pages we decided to delete, just not on first glance, but after a review of the possibilities (possibilities are hard to assess for the casual observer by magnitude, hence all those antivaxxers; our judgement needs specific training for the assessment of specific possibilities, so even if you are a professor in one area you may stay without insight in a closely related area and ignore its possibilities even though these should influence the decision).

We all have read very odd things that are peer-reviewed, as some academics have built parallel universes to make a living. And the beech argument is one of it, not a respected theory any more (if I understood respect correctly as being more than being constantly repeated out of courtesy and the university habit of citing everything that is available) but a fringe view, certainly not adding, in the traditional meaning of science, to our knowledge, but you are right that the decision about whether to delete the page or not should be taken on the merits of the arguments alone, since you yourself know your colleagues enough to distrust them.

It is symptomatic though that a tenured professor in philology fails to consider the presence of unknown language groups before Indo-European; that’s how one regularly comes up with reconstructions that should never be made: if all you have is a hammer, everything looks like a nail, if you only know Indo-European, “everything” is from Indo-European. It is partially not your fault, given that the other language groups of Europe are scatteredly or not at all attested, and partially it is your fault in so far as you never deal with language areas where well-documented languages have stood in manifest contact (the usual case with the bulk of scholarship about Greek and Germanic and Germanic languages, you know those who live in a Germanic country and study Germanic do kind of a cheap thing, and if a European classical language is added this only opens the horizon a little).

Different semantic fields of a language have various propensities to contain borrowed terms, plant-names are especially notorious in it. If I was only an Indo-Europeanist I would hardly know but in the Semitic, Turkic, Iranian languages half of them (I exaggerate but little) are certainly loans from each other or other less-known or completely unknown language groups; e.g. another Wiktionary professor saw that خُلَّر‎ (ḵullar) is surely borrowed and likely Hurrian but for the Greek ὄλυρα (ólura) the mainstreamers fail to do anything but speculate Indo-European (native or “pre-Greek”) origins although with the Near East data they should have classified it as wanderwort. However about every second time I open a Greek etymology Beekes claims a Greek word to be pre-Greek: while the intrinsic value of this label is close to zero due to the multifariousness of the frequent Pre-Greek claim, the idea of unknown sources of borrowings has been defended very well and is being concretized while we edit Wiktionary etymologies more and more.

So we pray you, Professor, to register and solve words occasionally, and especially if to disprove people as rude and uneducated as me. The more you learn of this dictionary business the more you realize that there is a thin line between daring comparisons—adding to our knowledge by maverickism—and academic dishonesty. And IPs are evil. After all you already do not rely on the majority of the comparisons on which that PIE “reconstruction” is made, if Germanic and Greek are left: In our experience the farer away a reconstructed historical language the more descendants one needs, and for PIE two are regularly (without very good reasons) not enough, while for Proto-Slavic not rarely one is enough—if a term must have been formed in Proto-Slavic, e.g. *mězgyrь, while for PIE there are too many millennia in between of what could have happened and we do not know that *bʰeh₂ǵos must have been internally derived in PIE (usually between Arabic, Iranian and Turkic and often inside their language groups themselves we know where a term was formed and hence whence borrowed by our understanding the internal morphologies of the languages: all things you do not know for this term).

This is all to say that, in comparison to more certain etymologies, here you know absolutely nothing. But you should somehow be confident about a reconstruction rather than many mismatches and coincidences and alternative scenarios (and I have engaged in shaky reconstructions out of excitement, but this is so shaky that it crumbles apart the more you think about it—if it were better I would come to maintain this PIE term: obviously I come correct in thinking about reconstructions, you will hardly deny this experience in having a consistent and carefully weighed approach about reconstruction entries). Fay Freak (talk) 17:12, 9 December 2021 (UTC)Reply[reply]

According to w:ru:Буковый аргумент, the modern range of w:Fagus orientalis falls inside the Proto-Indo-European homeland. And its range several millenia ago might have been broader than today. --Ghirlandajo (talk) 18:38, 26 June 2022 (UTC)Reply[reply]
I'm surprised there haven't been any updates to this request for deletion thread since last year. Is anyone willing to contribute journal articles discussing the presence of this constructed word for beech outside of the late Northwest Proto-Indo-European dialect continuum? In the meantime, I will provide citations from the EIEC, Calvert Watkins: "The American Heritage Dictionary of Indo-European Roots", and Beekes Etymological Dictionary of Greek. Calvert Watkins was not part of the Leiden school as far as I'm aware; nor is Douglas Q. Adams, who was one of the main editors of EIEC.Rigognos Molinarios (talk) 12:23, 11 February 2023 (UTC)Reply[reply]

This entire question is a bit of a mess, especially as some of the associated reconstructions are dubious, but I believe that the original posting user has made several critical errors, although the associated page should be updated.

The range of the beech tree not being within the hypothetical homeland of the Proto-Indo-Europeans, as mentioned earler, is irrelevant to whether the word occurs in PIE. Since the descendent languages cannot seem to agree to what tree it refers, which is not that unusual for tree or animal words that are not highly general, it could also originally have referred to a tree other than a beech, but seemingly one which is usable as a food source (per the different meanings for hypothesised descendents being 'elder', 'beech', and 'oak').

The Proto-Slavic word being properly *bъzъ does nothing to detract from the reconstruction, as Proto-Slavic *bъ- as a cluster regularly comes from PIE *bʰe-. The real question is whether *bъzъ can result from bʰeh₂ǵ-. *pazъ from *peh₂ǵ- suggests no, however, *bogъ being derived from *bʰeh₂g- suggests that the vowel a derivative of the word *bʰeh₂ǵos would reflect in Proto-Slavic is uncertain (admittedly the following consonant here is velar and not a palato-velar, but point is still relevant), so this is still a possible derivative. If another aspirate-eh₂-palatovelar cluster derivative could be found in Proto-Slavic anywhere, that would be helpful.

The Germanic reconstruction is an absolute ludicrous mess, agreed, and the attempted connection between 'beech' and 'book' is completely silly, plus one would expect a Proto-Germanic *bōkaz to come from *bʰeh₂ǵos and not *bōkō. However, the Proto-Germanic word for beech is apparently more appropriately reconstructible to *bōkijā, or more credibly neuter *bōkiją which is easily derivative of *bʰeh₂ǵos. Also *bʰeh₂- appears to regularly become *bō- in Proto-Germanic from a quick check of Proto-Germanic lemmas, so the additional attempted phonology points from the original poster are also spurious.

Agreed, the Gaulish word is not credible as it is derived purely from onomastics. However the Albanian word is only listed as possibly related, as it may possibly be, and it means 'oak' identically to Greek φᾱγός, so it is fair as a tentative or hypothetical relation given the difficulties inherent in Albanian reconstructions. Dismissing it out-of-hand is childish.

If the Latin word fāgus were an EARLY Greek loan from the more conservative Doric dialects, one would expect pāgus or phāgus in Latin and NOT fāgus as the Greek aspirate did NOT sound like a fricative in Latin even into the reign of Augustus, therefore it is near-impossible that the Latin word is a loan from Greek, and it is more reasonable to conclude genuine common descent. The beech not being immediately in the area of Rome but in the mountains a few dozen miles away is also completely spurious as an argument as this implies Latin speakers never left home, nor had need of a word for things not immediately in their environment but which were plentiful close by. Rome is also not on the sea coast, yet Latin also magically has more than one indigenous word to describe seas and seaborne fish! Plus, the Greek φᾱγός means 'oak', so magically the Latins would also have to inherit a Greek loanword with a completely different meaning to the Greek word for no discernable reason.

So to conclude, a credible word reconstruction in at least Latin and Greek, which is more than enough to posit a PIE origination, and additionally with plausible Germanic and Slavic derivatives, although the Slavic is on shaky ground. I'm inclined towards allowing the page for *bʰeh₂ǵos to stand, but the *bazъ page is spurious and should be deleted, and the Proto-Germanic 'beech' and 'book' derivations need serious attention by someone more capable than me at Proto-Germanic research, as they are in a completely silly state currently. I somewhat doubt any Indo-Iranian derivatives will be found given that beech and oak trees are not exactly plentiful on the Iranian plateau or in northern India, but otherwise at least four major families are covered. Someone could additionally look into Armenian as its word for beech, hačari, seems at face value to be similar-looking, but I am also completely unskilled at Armenian phonology and it would have to involve some manner of special cases as 'h' normally results from PIE *p in Armenian as I understand. This is still less sketchy than the still-existent page for PIE *h₂éyos, which is almost certainly an accidental fabrication, but nonetheless mistaken. Andecombogios (talk) 04:47, 22 February 2023 (UTC)Reply[reply]

Mallory and Adams also accept the existence of a proto-indo-european term for beech, in their Encyclopedia of indo-european culture, (its on pages 58-60), so this isn't even limited to leiden school Ioe bidome (talk) 21:02, 7 August 2023 (UTC)Reply[reply]

August 2020 edit

Reconstruction:Proto-Semitic/gamal- edit

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[reply]

Delete. How will we avoid the lengthy cognate lists? Fay Freak (talk) 12:49, 8 August 2020 (UTC)Reply[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[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[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[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 [of this proto-Arabic word] is a Semitic wanderwort which was also the source of [... ... ...]" ? - -sche (discuss) 06:10, 13 September 2020 (UTC)Reply[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[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[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[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[reply]

November 2020 edit

Reconstruction:Proto-Indo-European/h₁óytos edit

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[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[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[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[reply]

Delete. Two words don't warrant a PIE reconstruction. Highly speculative. Ghirlandajo (talk) 21:17, 7 June 2022 (UTC)Reply[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[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[reply]
- @Fay Freak You already voted above. Chuck Entz (talk) 05:16, 23 May 2023 (UTC)Reply[reply]

December 2020 edit

Reconstruction:Proto-Indo-European/dʰews- edit

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

Reconstruction:Proto-Indo-European/gl̥h₁éys edit

Not enough evidence, most likely did not exist; one used *múh₂s instead for all kinds of mice. The nominated word, somewhat varying by form in literature, is also claimed on the basis of some Iranian forms, which however have two much more likely explanations I mentioned at Arabic جُرَذ‎ (juraḏ, “rat”) (an Iranian stem related to biting and stinging etc. (semantically Russian куса́ть (kusátʹ), if it is not clear from an English horizon) and a Semitic borrowing from a widespread Semitic stem related to gnawing). Looking at the edit history @Victar has removed an alleged Indo-Iranian reconstruction allegedly descending from this Proto-Indo-European, likely without explanation because it is obviously baseless, before someone re-added the Sanskrit words गिरि (giri) and गिरिका (girikā, “making hills, a mouse?”) again, which apart from a new formation from the word for a hill may also be a borrowing like other words (from which we have Bengali ইঁদুর (ĩdur) etc., “from a lost Vedic substrate language”). Munda forms ɡoɖo ~ guɖu are not farther than the alleged Indo-European cognates. g + either r/l are too common consonants. Then, how has the paradigm given for the Indo-European page any relation to the paradigm of the Sanskrit word or the paradigm of the Latin word glīs, which has the stem glīs- or glīr-. Ancient Greek γαλέη (galéē) means a mustelid, which is an animal not that similar or not all confusable or confoundable. I realize that Latin mūstēla (“weasel”) contains the word for mouse, but it does so because the weasel eats mice, but the weasel is no mouse by utmost historical phantasy, they had and have to be kept apart. So the glossing of the Indo-European page “mouse, dormouse; weasel” is impossible – no language can use one and the same word for mice and weasels. The Thracian άργιλος (árgilos, “mouse”) is not similar at all and needs other explanations, which it has, mentioned in the linked article Studies in Thracian vocabulary I–VII. Fay Freak (talk) 23:03, 16 December 2020 (UTC)Reply[reply]

Re the claim that a weasel is too different from a mouse, so the two have to be kept apart - the Norwegian word for a weasel is snømus, literally 'snow-mouse'. De Vaan, as cited in the Wiktionary entry, interprets mūstēla as an original diminutive *mūstr-elā 'a small mouse-like animal', not as 'mouse-eater'. Perhaps a taboo replacement (cf. the conflation of worm and snake, arguably an even more important difference)? --79.100.144.23 22:51, 10 July 2022 (UTC)Reply[reply]

You're right that it might be a conflation of substrate words. But suppose it didn't exist in PIE―then why are the supposed cognates so similar? Sheer coincidence doesn't seem to be an option, so they would have come from related substrate/adstrate terms that had diverged semantically in their own distant past. It simply pushes back the question further into an unreconstructible era. Just because two animals don't seem "confusable" doesn't invalidate an etymological connection between words for them. Far stranger shifts have happened in far shorter time than the whole history of Indo-European. Also, the original term might have been an adjective or some other unspecific label that would have been applicable to various animals (cf. *bʰerH- (“brown”), the supposed root of both bear and beaver; similarly semantically divergent substantizations are par for Indo-European linguistics). So your semantic arguments don't really hold up. If there were no formal problems with the phonetics, the cognate set would hardly be questionable. Likewise, some questionable items being formerly part of an etymology have no bearing on the present state of the etymology, nor do tentative, less-supported connections to other language families like Dravidian or Semitic, unless of course the evidence for those etymologies is stronger (in this case, it clearly isn't). The differing paradigms are an important question, but nothing unresolvable; most can be regarded as different ablaut variants of an i-stem, and the question of the exact form of the original noun (or other nominal) is at that point almost irrelevant to the validity of the etymology. Thus the only really worthwhile criticism here is on the phonetic grounds. ― 69.120.66.131 01:54, 16 January 2022 (UTC)Reply[reply]

“Clearly” the evidence for Semitic and Munda origin of the Indo-Aryan terms are stronger—what does “supported” even mean? No, but coincidence is a stronger option, due to the ubiquity of the consonants in the game, and on the Arabic page I have offered a better derivation for the Iranian part. Anyway you admit that a noun is formally not reconstructible nor even any root of any defined meaning. The original might have been this, might have been that, or: there were multiple unrelated originals. You have a Greek and a Latin and Sanskrit word of bare different meaning. The Thracian is a meme speculation.

Now I also see the Greek and Latin derived from *gelH-, on Proto-Slavic *golъ, underline how arbitrary the assumptions underlying the Indo-European reconstructions of these words are. Fay Freak (talk) 20:42, 8 April 2022 (UTC)Reply[reply]

This reconstruction looks highly dubitable and is not supported by mainstream linguists. Delete. --Ghirlandajo (talk) 21:25, 7 June 2022 (UTC)Reply[reply]

Delete: I removed the Proto-Indo-Iranian because the Sanskrit is a ghost word, only attested in dictionaries^[EWAiaI:448], and to quote Schrijver, "as it stands, the comparison [between Latin glīs and Ancient Greek γαλέη (galéē)] is neither formally nor semantically compelling."^{[Schrijver:1991:242-243]} --{{victar|talk}} 08:38, 23 May 2023 (UTC)Reply[reply]

Delete -- 𝘗𝘶𝘭𝘪 𝘮𝘢𝘪𝘺𝘪^{(𝘵𝘢𝘭𝘬)} 07:01, 6 June 2023 (UTC)Reply[reply]

January 2021 edit

Reconstruction:Proto-Halmahera-Cenderawasih/wakaʀ edit

No descendants listed. Page creator is no longer active at Wiktionary. —Mahāgaja · talk 19:30, 30 January 2021 (UTC)Reply[reply]

The reference gives three reflexes of Proto-Malayo-Polynesian *wakaR for "SHWNG", which is another name for Halmahera-Cenderawasih: Babuza wā, Windesi "war" (Wikipedia treats Windesi as a dialect of the w:Wamesa language with the language code wad, while we treat wad as the language code for Wandamen, which Wikipedia says is the name of another dialect of Wamesa) and Ansus woa. Of course, we don't have entries for any of those and Blust is known to have regularized orthographies elsewhere in the same work to make comparison easier, so I wouldn't use it as a source for the terms themselves.

I would note that the Proto-South Halmahera-West New Guinea index lists wakaR, and a suffixed form *wakaR-i (with four more reflexes). Of course, it has similar indexes for proto-languages that probably don't exist, such as Proto-Western Malayo-Polynesian, so it would be nice to have some source to back this up. Chuck Entz (talk) 23:40, 30 January 2021 (UTC)Reply[reply]

February 2021 edit

Reconstruction:Latin/consutura edit

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

Indeed: [1], [2], [3]. 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[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[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[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[reply]

March 2021 edit

Reconstruction:Proto-Indo-European/ǵeh₂r- edit

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

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

April 2021 edit

Reconstruction:Proto-Germanic/aganą edit

The conjugation table does not agree with the title, however I am unsure what it should be changed to. Pinging @Rua Mårtensås (talk) 15:17, 23 April 2021 (UTC)Reply[reply]

Since the only direct descendant is Gothic 𐍉𐌲𐌰𐌽 (ōgan), I suspect Reconstruction:Proto-Germanic/aganą should be moved to Reconstruction:Proto-Germanic/ōganą. —Mahāgaja · talk 15:41, 23 April 2021 (UTC)Reply[reply]

And looking through the page history, I see it was at Reconstruction:Proto-Germanic/ōganą until Rua moved it to aganą in January 2019. —Mahāgaja · talk 15:47, 23 April 2021 (UTC)Reply[reply]

The Proto-Germanic page title is fine. The only thing wrong with the conjugation is that the present singulars should be *ōg, *ōht and *ōg, respectively. The infinitive, not being inflected for number or person, is properly *aganą which follows the stem of the non-singular present, just like the present participle *agandz does, as we see in the fossilized Gothic 𐌿𐌽𐌰𐌲𐌰𐌽𐌳𐍃 (unagands). The Gothic form of the verb as *ōgan, with *ōg spread throughout the paradigm, including the secondary present participle 𐍉𐌲𐌰𐌽𐌳𐍃 (ōgands), shows high degrees of regularization as is pretty typical for Gothic, which supports the theory that it was a superstratum language with many non-native speakers.

Likewise for aorist-presents, the infinitive stem follows the non-singular present stem of the verb, as seen in *wiganą, although here too the present singulars should be *wīhō, *wīhizi, *wīhidi respectively; this is proved by the fact that otherwise -h- would have been unrecoverable to later Germanic speakers, and Gothic could not have regularized the -h- throughout... Which again in Gothic, we see regularization as the singular stem of the verb *wīh-, is spread to the other present formations: 𐍅𐌴𐌹𐌷𐌰𐌼 (weiham) instead of original *wigam, and again in a derived form of the present participle, the original short vowel is kept, but the velar is regularized to -h- instead of original -g-: 𐌰𐌽𐌳𐍅𐌰𐌹𐌷𐌰𐌽𐌳𐍉 (andwaihandō) instead of *andwigandō. Burgundaz (talk) 22:05, 18 May 2021 (UTC)Reply[reply]

Seconding that the title agrees perfectly well with the IPA and conjugation table, even if perhaps the vocalisation of the present singulars is incorrect as Burgundaz says. Tristanjlroberts (talk) 23:48, 7 February 2022 (UTC)Reply[reply]

May 2021 edit

Reconstruction:Proto-West Germanic/krūci edit

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[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 [t͡s]. 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[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[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[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[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[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[reply]

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

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

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

Reconstruction:Proto-Germanic/ainijaz edit

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[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[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.

Reconstruction:Latin/retina edit

Contra the provided Grandgent (1907) reference, which states, "A few that must have existed are not attested at all: *refusare, Substrate V, 234; *retĭna = ‘rein’, Substrate V, 237", the term is actually attested in a 10th-century example given by the Du Cange dictionary (which I assume it did not occur Grandgent to consult in this case). I suggest deleting the Reconstruction:Latin/retina entry, using the actual Early Medieval Latin entry retina instead as the etymology for all the relevant Romance language entries (e.g. Spanish rienda, French rêne). While it is true the attestation is fairly late, from the 10th century, such a late attestation doesn't mean the word didn't have an earlier existence.--Ser be être 是^talk/stalk 00:14, 16 May 2021 (UTC)Reply[reply]

Separate question: the etymology we give at "retina", "An abbreviation of the Classical retināculum n", looks suspicious to me: do any other words ending in -culum/-cula have "abbreviated" forms that are inflected like non-abbreviated first declension words? I'm not sure, but could it instead be an alternative formation directly from the the verb retineo?--Urszag (talk) 02:51, 16 May 2021 (UTC)Reply[reply]

@Urszag: Yeah, very much. In fact, both the Trésor de la langue française informatisé and the Coromines & Pascual dictionary explain rêne/rienda as a deverbal derivation of retineō, not some kind of "abbreviation" of retinācula. I'll change the entry accordingly. I imagine User:I'm so meta even this acronym, who added the etymology, was thinking of "back-formation" from the apparent -cula diminutive.

(That said, I find the explanation vaguely, mildly attractive, considering retināculum is attested in classical written Latin with a similar meaning, even though I can't think another example of back-formation applied on a diminutive...)--Ser be être 是^talk/stalk 05:12, 16 May 2021 (UTC)Reply[reply]

Incidentally, I just looked at the etymology of English "retina", and it looks quite wrong, considering "tunica retina" does appear attested in medieval Latin. The Corpus Corporum website gives examples from Gullielmus the abbot's De natura corporis et animae (11th century) and Roger Bacon. I don't know why it's called Vulgar or why it has an asterisk... I wonder what the OED has?--Ser be être 是^talk/stalk 05:15, 16 May 2021 (UTC)Reply[reply]

OED says “post-classical Latin retina (13th cent. in British and continental sources) < classical Latin rēte net (see rete n.) + -īna -ine suffix4, so called on account of its finely fibrillar texture resembling a net”. — SGconlaw (talk) 07:20, 16 May 2021 (UTC)Reply[reply]

@Sgconlaw: Thank you. That was helpful to edit the current English etymology and add the relevant Latin entries (which were completely absent until now).--Ser be être 是^talk/stalk 17:15, 16 May 2021 (UTC)Reply[reply]

@Ser be etre shi: My thoughts: 1) if it was formed from the verb, it must have been both before the shift of accent on the root and before the connection with the verb was no longer felt, otherwise it would have followed the shift of accent; 2) this makes the backformation proposal quite appealing; 3) as a matter of principle, I don't believe it's possible to list clearly inherited Romance terms as inherited from Medieval Latin, which is a parallel language to these varieties and not their parent. Which recalls the problem of lack of definition of Medieval Latin on the website. Brutal Russian (talk) 10:17, 18 May 2021 (UTC)Reply[reply]

@Brutal Russian: Those are pretty good points. The lack of accent shift is pretty noticeable. And I agree it depends on how we define Medieval Latin, and the model of what Romance languages come from what... I imagine you're familiar with Roger Wright and the related "communication verticale" literature that tries to model (or explores the question of) early medieval Latin and early "pre-literary" Romance as registers of a single language, and for a longer period of time at that if it's Hispano-Romance and Italo-Romance too. I tend to personally prefer that view to the other of calling post-classical words attested in early medieval Latin "Vulgar", even marking them with an asterisk while also quoting them, all because Romance is taken to be parallel varieties to early medieval Latin.

At any rate, you've basically convinced me about the back-formation, and made me more conscious that this complaint has more to do with the model in my head than anything else (and the model in my head is definitely not the only one; early Romance as parallel to medieval Latin is a commonly-used view). I suppose it's okay to leave the Reconstruction:Latin entry there.--Ser be être 是^talk/stalk 00:40, 23 May 2021 (UTC)Reply[reply]

Reconstruction:Old English/ynne edit

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[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[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[reply]

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

Reconstruction:Proto-West Germanic/maginōn edit

@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[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[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[reply]

Reconstruction:Proto-Indo-European/bʰruHg- edit

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

June 2021 edit

Reconstruction:Proto-West Germanic/kauli edit

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[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[reply]

Reconstruction:Proto-West Germanic/buterā edit

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[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[reply]

Reconstructed Latin edit

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[reply]

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

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

August 2021 edit

Reconstruction:Proto-Germanic/hagastaldaz edit

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

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

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

Reconstruction:Proto-Semitic/šawmān- edit

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[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[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[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[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[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[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[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[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[reply]

April 2022 edit

Reconstruction:Old Portuguese/africão edit

Reconstructed entry with no derived terms. - Sarilho1 (talk) 12:29, 5 April 2022 (UTC)Reply[reply]

‘no derived terms’— I think you meant descendants. But unfortunately, seeing as the term used in Modern Portuguese is africano, a cultismo, the unattested Old Portuguese word has no descendants. The entry looks legit, tho’ a reference would have been better. Keep for now. ·~ dictátor·mundꟾ 13:30, 5 April 2022 (UTC)Reply[reply]

Question: If aflicão is attested (it has a quotation), why does it claim to be an alternative form of a merely reconstructed word? 70.172.194.25 00:46, 9 April 2022 (UTC)Reply[reply]

Theoretically, a rarer form of a word assumed to be oftener in another but unattested form can be attested and against direct corpus evidence assumed the alternative form of an unattested form. But the same question has occurred to me and it is still left: the actual attestation situation against this speculation. Fay Freak (talk) 15:17, 9 April 2022 (UTC)Reply[reply]

The issue here is that the word is only attested as a glaring hypercorrection. — Ungoliant ^(falai) 23:28, 11 April 2022 (UTC)Reply[reply]

@Inqilābī I do agree that africano is a cultism, but it's not clear to me why *africão is the expected native form. Wouldn't *afrigão be a more obvious reconstruction? - Sarilho1 (talk) 14:10, 19 August 2022 (UTC)Reply[reply]

Delete, I reformatted aflicão so this entry is no longer needed. This, that and the other (talk) 02:32, 27 February 2023 (UTC)Reply[reply]

The intervocalic /k/ is still problematic for an inheritance. Nicodene (talk) 03:55, 27 February 2023 (UTC)Reply[reply]

It gets weirder, though. For some reason I haven't understood yet, the source normalizes the word as "afriçãos" with the "r" and a "ç" that has corresponding pronunciation of /t͡s/ (implying a Latin form of *afritianus?). - Sarilho1 (talk) 00:08, 10 March 2023 (UTC)Reply[reply]

Delete, this entry seems unnecessary if it is not attested in this form. I don't see why it matters that aflicão is a "glaring hypercorrection". And shouldn't /f/ in this context be subject to voicing (compare ábrego), making the actual hypothetical inherited form *avrigão?--Urszag (talk) 04:17, 27 February 2023 (UTC)Reply[reply]

May 2022 edit

Reconstruction:Proto-Indo-European/Hrugʰís edit

Szemerényi, Oswald (1967), “Славянская этимология на индоевропейском фоне”, in , В. А. Меркулова, transl., Вопросы языкознания (in Russian), issue 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[reply]

Reconstruction:Proto-Germanic/pīpaną edit

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[reply]

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

June 2022 edit

Reconstruction:Old English/giestranmorgen edit

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[reply]

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

September 2022 edit

Reconstruction:Proto-Slavic/židъ, Reconstruction:Proto-Slavic/Rimъ edit

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[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[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[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[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[reply]

"*Rọ̄ma" is simply wrong, as it implies that the form is unattested. Needless to say, Roma [ˈroːma] 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[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[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[reply]

Reconstructed Latin noun and verb forms edit

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[reply]

They're there for etymology reasons. Delete, and substitute (eg. 1st conjugation) every From [[*~āre]], infitive of [[*~ō]] with From [[*~o|*~ō, *~āre]] 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[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[reply]

Reconstruction:Proto-Indo-European/h₂ewlo- edit

Strewn together entry of semantically impossible words. @GabeMoore --Sokkjō (talk) 04:45, 30 September 2022 (UTC)Reply[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[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[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[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^[Cicero]; beehive^[Varro+]; hull (of a ship)^[Tac.]”), 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[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[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[reply]

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

Viredaz has connected Old Armenian աւղ (awł, “ring”) with ագ-ուցանեմ (ag-ucʿanem, “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[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[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[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[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[reply]

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

October 2022 edit

Reconstruction:Proto-Indo-European/énu edit

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[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[reply]

Reconstruction:Proto-Germanic/feþlą edit

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[reply]

Reconstruction:Proto-Indo-European/ḱlinéh₂ti edit

Discussion moved from WT:RFDO. 11:38, 19 October 2022 (UTC)

None of the formations in the descendants actually match the PIE form, so why does this exist? There was a Latin form listed before, but per De Vaan, it doesn't belong there. —CodeCa t 14:04, 31 July 2017 (UTC)Reply[reply]

Beekes clearly cites *klin-je/o- as the root of the the Greek form, and that itself comes from an older nasal present. Kroonen cites the older original form as either *ḱli-neh₂- or *ḱli-neu-. That is what was reconstructed on this page. de Vann concurs and explains the long -ī- as being "introduced from the root aorist *klei- i *kli- (cf. cliēns)." --Victar (talk) 14:12, 31 July 2017 (UTC)Reply[reply]

This doesn't address the main point at all, but merely confirms it. Why do we reconstruct this if there are zero forms which actually descend from it? —CodeCa t 14:15, 31 July 2017 (UTC)Reply[reply]

The Latin form is cited in sources as an example of this original form. Source add. --Victar (talk) 14:24, 31 July 2017 (UTC)Reply[reply]

Perhaps, but the entry is now in conflict with the (also sourced) etymology given at clīnō and *ḱley-. We clearly cannot follow the sources here, as they contradict each other. This is an issue with a lot of your editing, Victar. You blindly go with sources which often posit very bold hypotheses that don't have widespread acceptance, you don't critically examine them. You also do not seek out consensus; whenever someone reverts your questionable additions, you ignore the fact that reverting an addition means no consensus, and repeatedly reinstate. When discussions finally start, you also refuse to wait for a consensus, but reinstate your edits as soon as you think you have proven your point, as on Reconstruction talk:Proto-Indo-European/ḱlitós. You need to stop hiding behind sources and start listening to editors. —CodeCa t 14:27, 31 July 2017 (UTC)Reply[reply]

If we're going to bring this to personal attacks simply because you disagree with the sources, let me say that your actions are very unbecoming of an editor. I attempt to start a dialog with you, as exampled here, and your solution every time is to simply assert you are correct and delete the entry. You were stripped of your adminship because of your uncooperative behavior which you continue to this day. --Victar (talk) 14:34, 31 July 2017 (UTC)Reply[reply]

You attempted to start a dialog, then completely ignored it and put the content back anyway. Multiple times. Without consensus. And you still haven't given a good argument for keeping this entry, and you still continue to add content without consensus. —CodeCa t 15:02, 31 July 2017 (UTC)Reply[reply]

Ignored it? I put forth my thoughts, and your response was you're wrong, I'm deleting it again. That is not a dialog on your part. You are not one to talk at all about consensus. For example, there was a clear consensus that laryngeals existed in PII and many of its descendants, and despite that, you systematically deleted them, at which point @JohnC5 had to insist that you stop. This project is not a dictatorship for your ruling. --Victar (talk) 15:45, 31 July 2017 (UTC)Reply[reply]

No person is correct in all decisions. CodeCat has made mistakes, as have I. Recently, Victar, you're been running roughshod across these entries, frequently ignoring language specific-considerations in favor of utopian reconstructionism. I've tried to be fairly conservative in the way I've edited on here, and CodeCat is very useful in reining in my reconstructive excesses. Similarly, sometimes she has personal opinions which need curtailing, but this is done in discussion with other editors. If there is no consensus, it is better to not add it all then to add highly speculative material. Again, I've made these mistakes many times, and indeed at one point CodeCat told me to go read a bunch of literature since I was adding so many bad reconstructions.

In this matter, I would agree with CodeCat that there seem to be several competing explanations or maybe several competing forms. This entry is at this point too speculative to merit its own entry. —John C5 17:57, 31 July 2017 (UTC)Reply[reply]

I am completely open to being wrong and absolutely crave discussions so that I might learn from them. I do ping you and CodeCat on entries that I hope either of you can look over. I probably would have done the same once I finished working on the entries, like I did here.

My objection is in CodeCat's method of simply reverting someone with no explanation other than "you're wrong", if that much, which I find in bad form. We should be encouraging editors to go through the sources and add this information to the project, and, in turn, correcting mistakes, not disregarding them with reverts and throwing the baby out with the bath water.

You both know, I'm a clean and methodical sourcer (I added |passage= to many reference templates to improve them further), and I want to get these entries done by the books as much as anyone else. Three sources, Kroonen, de Van, and Beekes all suggest a nasal verb root, but if that's too speculative, lets have a conversation and come to some less speculative alternatives. I'm happy to see that CodeCat has now added some to the root entry, but this was only after my objections to the deletion of the entry. --Victar (talk) 18:27, 31 July 2017 (UTC)Reply[reply]

I usually err on the side of caution, and I wouldn't even have created a PIE page in this instance. The descendants are too different to really pin them down onto a single form. Kroonen etc. may be right about the original -neh₂- suffix, but this is very speculative considering that it doesn't appear in that form in any language. It appears to me like they pulled it out of thin air. The nasal itself is plentifully attested, but its exact nature is too unclear. What I find especially telling is that Greek added an additional -ye- present suffix to it, as if it wasn't "present enough" in its old form. The Germanic form, with a stative formed, seemingly, from a characterised present, is even more puzzling. So my preferred option would be to just say "We don't know" and only list the forms without trying to pin a particular underlying formation on them, as Kroonen and De Vaan do. Note that Kroonen and De Vaan disagree on what they think the original form was, and I find neither of their proposals particularly compelling. De Vaan's proposal might work if the root had an additional final laryngeal, but that's ad hoc and only solves the puzzle for Latin, it makes it harder for all the others. —CodeCa t 18:37, 31 July 2017 (UTC)Reply[reply]

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘ Haha, "present enough" made me laugh. I can certainly see where you're coming from and Kroonen definitely phones it in at times. I think we can agree though that three independent innovations of *ḱli-n- seems highly unlikely. I've moved over the sources for to *ḱley-, so please feel free to delete *ḱlinéh₂ti. Also, please copy your comment regarding Germanic weak class 3 over, which I think is helpful. Thanks. --Victar (talk) 03:06, 1 August 2017 (UTC)Reply[reply]

Re CodeCat and "...personal attacks simply because you disagree with the sources...": +1 on the Uther-meter. Chuck Entz (talk) 02:59, 1 August 2017 (UTC)Reply[reply]

@Chuck Entz: That is not a helpful comment for dialog. --Victar (talk) 03:06, 1 August 2017 (UTC)Reply[reply]

Keep - again, if it's properly sourced to credible material, we keep it. if there are conflicting opinions in sourced material, we include information about the conflict. we shouldn't be making the call on which is "right", that is original research (though debating/arguing about it is fun! ^__^). our job on wiktionary, as with wikipedia, is to provide a compilation of the existing information. it's not our job (on here, in the article-space) to express our own ideas/opinions about the etymologies (or definitions, etc.) Lx 121 (talk) 13:18, 29 April 2019 (UTC)Reply[reply]

Hoping to get some more input on this at the proper venue. This, that and the other (talk) 11:38, 19 October 2022 (UTC)Reply[reply]

Delete. I think that it doesn't make complete sense to keep it as it's own entry, but rather to move it to *ḱley-, and then show it as forms descending from the derived form *ḱlinéh₂ti, Qwed117 (talk) 23:31, 9 November 2022 (UTC)Reply[reply]

Reconstruction:Latin/werra edit

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[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) [4]: "[The Frankish Germanic idiom's] 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[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[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[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[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[reply]

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

November 2022 edit

Reconstruction:Proto-Semitic/ʔabay- edit

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

Reconstruction:Proto-West Germanic/biggō edit

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[reply]

December 2022 edit

Reconstruction:Proto-Indo-European/ḱóymos edit

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[reply]

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

Reconstruction:Proto-Germanic/in walą fallaną edit

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[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[reply]

Delete all three. --Sokkjō (talk) 04:08, 19 December 2022 (UTC)

Wiktionary Request pages (edit) see also: discussions
Requests for cleanup add new \| history \| archives Cleanup requests, questions and discussions.	Requests for verification/English add new English request \| history \| archives Requests for verification in the form of durably-archived attestations conveying the meaning of the term in question.	Requests for verification/CJK add new CJK request \| history Requests for verification of entries in Chinese, Japanese, Korean or any other language using an East Asian script.	Requests for verification/Italic add new Italic request \| history Requests for verification of Italic-language entries.	Requests for verification/Non-English add new non-English request \| history \| archives Requests for verification of any other non-English entries.	Requests for deletion/Others add new \| history Requests for deletion and undeletion of pages in other (not the main) namespaces, such as categories, appendices and templates.
Requests for moves, mergers and splits add new \| history \| archives Moves, mergers and splits; requests listings, questions and discussions.	Requests for deletion/English add new English request \| history \| archives Requests for deletion of pages in the main namespace due to policy violations; also for undeletion requests.	Requests for deletion/CJK add new CJK request \| history Requests for deletion and undeletion of entries in Chinese, Japanese, Korean or any other language using an East Asian script.	Requests for deletion/Italic add new Italic request \| history Requests for deletion and undeletion of Italic-language entries.	Requests for deletion/Non-English add new non-English request \| history \| archives Requests for deletion and undeletion of any other non-English entries.	Requests for deletion/Reconstruction add new reconstruction request \| history Requests for deletion and undeletion of reconstructed entries.
{{attention}} • {{rfap}} • {{rfdate}} • {{rfquote}} • {{rfdef}} • {{rfeq}} • {{rfe}} • {{rfex}} • {{rfi}} • {{rfp}}

Wiktionary:Requests for deletion/Reconstruction