Open main menu

Wiktionary β

User talk:JohnC5

Archive – 201420152016

*dʰegʷʰ- reversionEdit

You told me to comment on your talk page if I think this rollback in in error. Well, Sanskrit happens to be my second language and the word dagdha means burned. If you doubt this go to this link. This word, as obvious to any student of PIE, fits much closer the reconstructed dʰegʷʰ. Personally I think only people who know Sanskrit should get to decide what are the Sanskrit descendants of a reconstructed PIE root. —This unsigned comment was added by Idielive (talkcontribs).

Why did you remove the existing word that was listed there, and replace it with a word that does not descend from the verb *dʰégʷʰeti but is listed as if it were? —CodeCat 21:04, 6 January 2017 (UTC)
@Idielive: Sorry for the delay. I'm not sure whether you noticed, but in the diff after the one in which you reverted, I put दग्ध (dagdhá), which is a verbal adjective, under its correct PIE etymon *dʰₔgʷʰ-tós. When you made your edit, you placed this verbal adjective meaning “burned” under the section for *dʰégʷʰ-e-ti, which is the verb meaning “to be burning.” The correct descendant of this verbal form is दहति (dáhati). Furthermore, you addition was ill-formatted, which is what lead to the immediate reversion. This was an understandable mistake you made, but, because PIE and Sanskrit did not possess unified verbal systems, on this project we split out the specific verbal formations and their descendants. As such, the *o-grade causative दाहयति (dāhayati) has a separate but cognate lineage from the *e-grade thematic root present दहति (dáhati). Does that make sense? —JohnC5 13:52, 7 January 2017 (UTC)
@JohnC5: I understand, but dagdha has to have some link to the root dʰegʷʰ right? The loss of the initial breathy consonant dʰ would have served as an a coloring sound shift, that being the only way to explain the a vowel in दग्ध. दग्ध should be placed in a section of its own, since it would not descend from dʰégʷʰ-e-ti (or else it would have a final ti like dáhati). Idielive 7 January 2017
@Idielive: As mentioned before, दग्ध (dagdhá) comes from *dʰₔgʷʰ-tós, where it is currently listed in the entry, *dʰegʷʰ-. The change from *dʰegʷʰ- to *degʷʰ- is called “Grassmann's law” and the change from *dagʰ-ta to *dag-dʰa is called “Bartholomae's law.” —JohnC5 00:02, 8 January 2017 (UTC)

Share your experience and feedback as a Wikimedian in this global surveyEdit

  1. ^ This survey is primarily meant to get feedback on the Wikimedia Foundation's current work, not long-term strategy.
  2. ^ Legal stuff: No purchase necessary. Must be the age of majority to participate. Sponsored by the Wikimedia Foundation located at 149 New Montgomery, San Francisco, CA, USA, 94105. Ends January 31, 2017. Void where prohibited. Click here for contest rules.


X-SAMPA to IPA functionsEdit

Your changes to Module:IPA technically work, but they don't do quite the same thing. When the current version of {{X2IPA}} is substituted, it outputs the HTML that {{IPA}} would create, rather than template code. For example, this diff. — Eru·tuon 02:51, 23 January 2017 (UTC)

@Erutuon: We should have to use subst:. We should just be calling the template directly. Right? —JohnC5 02:56, 23 January 2017 (UTC)
Well, the goal was to make it possible for me to type in X-SAMPA but have it saved as an instance of {{IPA}} with the equivalent IPA. Right now, you get (more or less) the same thing you would if you put an instance of {{IPA}} into Special:ExpandTemplates, for instance IPA(key): [ɛɡˈzɑmpɫ̩]IPA(key): [ɛɡˈzɑmpɫ̩]. You get the HTML that {{IPA}} outputs. (The only difference is that Special:ExpandTemplates removes HTML comments, whereas substing apparently doesn't.) — Eru·tuon 03:03, 23 January 2017 (UTC)
@Erutuon: Oh, oh, oh. I get it. You're saying you don't want the XSAMPA input to appear but instead be replaced with the correct IPA input. I'm dumb. Gimme a minute, and I'll revert some parts of it. —JohnC5 03:09, 23 January 2017 (UTC)
@Erutuon: Actually, I see you were trying to do something. Would you like me to change it back or what? —JohnC5 03:15, 23 January 2017 (UTC)
Oh, I was just goofing off, seeing if I could simplify your code. No, go ahead with what you were doing. — Eru·tuon 03:17, 23 January 2017 (UTC)
@Erutuon: I put it back. I'm not suuuuper crazy about this template substitution system you've got, but out of sight—out of mind, eh? —JohnC5 03:25, 23 January 2017 (UTC)
It's pretty inelegant, but it is so much more convenient than using {{x2i}} in each parameter. — Eru·tuon 03:51, 23 January 2017 (UTC)

Etymology of "blueEdit

Hello, i wanted to let you know that there are certain contradictions with the proto-indo-european-etymology of the word blue, it says that it comes from *blēwaz (blue, dark blue) and this one from *bʰlēw- (yellow, blond, grey). But blewaz estymology states that it comes from *bʰleh₁- and that flavus is a congnate. But *bʰleh₁- page doesn't show any related meaning. And flavus states that it comes from *bʰel- and so other cognates that that page points. Tom 144 (talk) 16:11, 23 January 2017 (UTC)

There are a whole bunch of related roots of the general shape *bʰel-, *bʰle-R?-, etc. I'll try to sort this out later somewhat. If it doesn't change for a few days, pester me. —JohnC5 16:49, 23 January 2017 (UTC)

Sinistro/dextro reciprocal interchiral agnosis?Edit

  • 23:29, January 22, 2017 JohnC5 (talk | contribs) deleted page Template:IPAchar/old (Obsolete and orphaned template / unused template subpage: content: '<!-- --><span class="IPA" lang="">{{{1}}}</span><!-- -->{{#if:{{{2|}}}|, <span class="IPA" lang="">{{{2}}}</span>}}<!-- -->{{#i...' (last edit: [[Special:Contributions/ZxxZ...)

Chuck Entz (talk) 10:41, 24 January 2017 (UTC)

@Chuck Entz: Fixèd! Sorry for the hassle. —JohnC5 16:41, 24 January 2017 (UTC)

Why is "erreur" feminine?Edit

It comes from Latin error which is masculine, and its Spanish descendant error is also masculine. --kc_kennylau (talk) 18:28, 28 January 2017 (UTC)

I've looked around a bit and don't have a good answer. If fr.Wikt is to be believed, then Old and Modern Occitan also have feminine error. The best I have at the moment, is that sometimes words just change gender. —JohnC5 19:12, 28 January 2017 (UTC)
It's not very helpful, but Latin third declension nouns in -or, which are nearly all masculine (exceptions being due to natural gender) do become feminine in many Romance languages, but generally very few become fixed thus in the standard (another one similar to French erreur is Portuguese cor f, from Latin color m). However, if you look into more rural dialects, like those of western and northern Spain, you can find that many more -or nouns have become fixed as feminine. —Μετάknowledgediscuss/deeds 02:41, 29 January 2017 (UTC)


Why did you delete this? It means "sixth" and is a separate etymon from Reconstruction:Proto-Slavic/šestь "six". Benwing2 (talk) 21:14, 4 February 2017 (UTC)

@Benwing2: I've restored it, but as before it has no usable content. Please add the appropriate material. —JohnC5 07:48, 6 February 2017 (UTC)
I see, that explains the deletion. In any case, thanks for restoring it, and it's fixed. Benwing2 (talk) 10:26, 6 February 2017 (UTC)
@Benwing2: Sorry for the delay. I was at a juggling convention all weekend. —JohnC5 15:42, 6 February 2017 (UTC)

Keep the Turkish etymologies on (don't delete them)Edit

@JohnC5 I was going to contribute to the etymology of Turkish bulut and please the etymology on (with some other Turkish terms I have contributed to). Please don't delete it thank you! Helolo1 (talk) 03:13, 7 February 2017 (UTC)

None Helolo1 (talk) 03:14, 7 February 2017 (UTC)

You should know that the source you linked to in the etymology tends to focus a lot on speculative, long-range etymologies that aren't widely accepted in the linguistic community. For instance, the linguistic mainstream doesn't consider the evidence for Altaic strong enough to justify using it in etymologies. We only allow mentioning it if it's accompanied with a warning that it's disputed and controversial. Chuck Entz (talk) 03:55, 7 February 2017 (UTC)

Lithuanian ModuleEdit

Hi John C5, at User:Awesomemeeos/lithuanian (you may correctly edit if you want), I'm creating data found on the Wikipedia phonology article. Let's put both phonetic and phonemic transcriptions, and deal with the standard accent first. Thanks foꝛ reading!AWESOME meeos * (chōmtī hao /t͡ɕoːm˩˧.tiː˩˧ haw˦˥/) 02:19, 10 February 2017 (UTC)

@Awesomemeeos: Cool. It may take me a few days to get started on this. Also, do we know the environments in which the ⟨ą⟩ variations /æː/ ~ /ɛː/ occur, or do we have to input them manually? —JohnC5 06:03, 10 February 2017 (UTC)
They are predictable, where /æː/ is advanced from /ɛː/, when after palatalised consonants, e.g. gilią (/ˈɡʲɪ.lʲæː/). Take a look at [w:Lithuanian phonology]. I'm already writing the template, and the module is lt-pronAWESOME meeos * (chōmtī hao /t͡ɕoːm˩˧.tiː˩˧ haw˦˥/) 06:08, 10 February 2017 (UTC)
I have a friend who is Lithuanian in ethnicity, and he will help me write some of the phonology and help with some of the testcases XD – AWESOME meeos * (chōmtī hao /t͡ɕoːm˩˧.tiː˩˧ haw˦˥/) 06:09, 10 February 2017 (UTC)
@Awesomemeeos: I think you have misread the article. ⟨Cą⟩ → /aː/, whereas ⟨Cʲą⟩ = ⟨Cʲę⟩ = ⟨Cę⟩ → /æː/ or /ɛː/, but it doesn't provide any more detail about the variation. Also, I don't see where it says that ⟨a⟩ → /aː/. —JohnC5 06:21, 10 February 2017 (UTC)
For ⟨a⟩ → /aː/; see wikipedia Help:IPA for LithuanianAWESOME meeos * (chōmtī hao /t͡ɕoːm˩˧.tiː˩˧ haw˦˥/) 06:22, 10 February 2017 (UTC)


What's the reason for the revert?
L&S for example has "anceps (once ancipes, Plaut. Rud. 4, 4, 114; cf. Charis, pp. 67 and 96 P.; Prisc. p. 754 P ... )" and "bĭceps, cĭpĭtis (old form bĭcĭpes, like ancipes for anceps, acc. to Prisc. p. 754 P. ... )". For triceps and centiceps it doesn't mention a form in -cipes, and quadriceps and multiceps are missing in L&S.
That clearly shows that -cipes was uncommon in Classical and Late Latin because else L&S or other dictionaries like Gaffiot would have had it and it shows that it was Old Latin as it was used by Plautus. - 21:18, 17 February 2017 (UTC)

Wiktionary considers Old Latin to be a separate language, itc-ola, and so it cannot act as an alternative form. I'll update is appropriately. —JohnC5 21:59, 17 February 2017 (UTC)

Unlock Template:senseEdit

Can you unlock this please? I need to make a few changes. —CodeCat 23:30, 17 February 2017 (UTC)

@CodeCat: Done. Please tell me when you are done? —JohnC5 00:30, 18 February 2017 (UTC)
Done, though I see no particular reason to protect it fully. —CodeCat 00:32, 18 February 2017 (UTC)


You reverted this edit, which was the first time i've tried to edit a reconstruction lemma. Please explain why or how or where one should add that info. Thanks, Espoo (talk) 04:59, 23 February 2017 (UTC)

@Espoo: Yeah, sorry about that! In reconstruction pages (especially roots), we try to terminate the descendant chain as soon as we arrive at a suitable descendant. This can mean anything from a descended protolanguage (Proto-Germanic, Proto-Italic) or the oldest attested languages in a particular branch (Sanskrit, Latin, Ancient Greek, Gothic). If an entire set of descendants is know for a protolanguage article, but that article doesn't exist, the descendants are sometimes listed (say you have a bunch of Slavic descendants but the Proto-Slavic article doesn't exist yet, you can store those descendants on the PIE page until the PS page is created). It's all fairly confusing, but the objective is to terminate as quickly as possible.
I'll admit, the process is somewhat arbitrary and case-by-case. For instance, we almost never list descendants of Latin since the Latin articles tend to be good and will likely have a list of descendants there. Sanskrit or Old Persian articles, on the other hand, tend to be someone scanty, so lists of their full descendants may often be found in the PIE articles. —JohnC5 06:04, 23 February 2017 (UTC)
OK, i get it, sort of. But doesn't that mean we should first check whether descendants are indeed listed on the suitable cutoff page before deleting them from the reconstruction lemma? And how should they be added to vox; using "descendants include"? --Espoo (talk) 17:13, 23 February 2017 (UTC)
@Espoo: The descendants are under vox#Descendants. It could certainly be expanded there. —JohnC5 17:15, 23 February 2017 (UTC)
oops --Espoo (talk) 17:37, 23 February 2017 (UTC)

Why is "lēx" feminine?Edit

The Latin word lēx is feminine, yet the Latin word rēx which has the exact same declension (save the initial consonant) is masculine. I understand that rēx has to be masculine because it is referring to a male (the king), but why is lēx feminine? Thank you. --kc_kennylau (talk) 14:08, 7 March 2017 (UTC)

@Kc kennylau: Athematic nouns in early PIE used to have only animate or inanimate genders and desinences, the former becoming masculine and feminine in later PIE and the latter, neuter (A table may be found here). Once this tripartitie distinction developed, non-neuter animate nouns without a clear physical gender were divided in a more-or-less arbitrary manner (*wréh₂ds f (root), *h₂óḱris f (top, protrusion), *h₂énts m (forehead; front)). So athematic non-neuter nouns may share the exact same declension but opposing genders for no reason. —JohnC5 14:38, 7 March 2017 (UTC)
Thank you. I corrected the link for you. --kc_kennylau (talk) 14:50, 7 March 2017 (UTC)

Why did you revert my revisions?Edit

I’ve asked exactly this question at least twenty-five times the past year, be it on Wikipedia, Wiktionary, or Wikia:
Why did you revert this and this revision of mine?
PapíDimmi (talk | contribs) 17:37, 7 March 2017 (UTC)

@PapiDimmi: You ask very good questions. Each of these edits contained different errors:
  • In this edit you moved the Alternative forms header below the headword, which is not the correct order. Also, you provided synonyms not broken out by {{sense}} which certainly do not apply to all the sense and are debatable in the first place.
  • For this edit, you removed the call to {{also}} and gave a prescriptive analysis of the word's usage.
  • In the case of choffer, you've only cited a recent video game that is set in a fictional universe and omitted actual, attested meanings of the word.
I hope this clarifies my edits somewhat. —JohnC5 18:05, 7 March 2017 (UTC)
“Youre” is not an alternative form of “you’re”; it’s a misspelling. “You’re” is a contraction of “you are,” which is why it has an apostrophe. Removing the apostrophe makes it a typo, not an alternative way to spell it, so why should the entry not mention that it is a misspelling?
Also, the also template is unnecessary when the definition is “misspelling of you're.”
PapíDimmi (talk | contribs) 18:11, 7 March 2017 (UTC)
@PapiDimmi: Just because an entry links to another does not mean we do not use the {{also}} template. This template is entirely agnostic of the content on the page. As to “you're” vs “youre”, this topic has been debated many times on Wiktionary. If a spelling has been well enough attested, it doesn't matter that formal grammars proscribe its use. —JohnC5 18:19, 7 March 2017 (UTC)
"Alternative forms" is allowed as a level 4 header below the headword, following a recent vote. —CodeCat 18:21, 7 March 2017 (UTC)
@CodeCat: Oh, I totally missed that vote. Thanks, Code. My comment about the synonyms remains. —JohnC5 18:24, 7 March 2017 (UTC)
I have another question: why did you delete Template:italic?
PapíDimmi (talk | contribs) 20:44, 7 March 2017 (UTC)
Because it is wasteful of Wiktionary's transclusion time and can be done on the page in fewer characters than it takes to invoke that template. —JohnC5 21:06, 7 March 2017 (UTC)


Just because people often misspell a word doesn’t mean that the word is suddenly an official alternative form of the actual word.

E.g., lots of people write “its” instead of “it’s,” “there” instead of “their,” “could of” instead of “could have,” “who’s” instead of “whose,” “isnt” instead of “isn’t,” and so on, but that doesn’t mean that these misspellings are alternative forms of the words. The are merely misspellings.
PapíDimmi (talk | contribs) 11:03, 13 March 2017 (UTC)

Regarding your revision, what is the point of {{also|you're}} when the definition of the entry links to you're?
By the way, are you paid to follow my contributions page and undo all my revisions?
PapíDimmi (talk | contribs) 16:40, 13 March 2017 (UTC)
@PapiDimmi: As I've previously stated, the {{also}} template is always used regardless of the content on the page. It may seem silly, and sometimes is, but if there is a another page to which {{also}} applies, then it is used.
I am not paid to do anything. As an administrator, I am charged with ensuring that additions to Wiktionary are both correct and formatted within our guidelines, especially when that concerns the edits of new or unknown editors. You're getting better, which I appreciate, but you have not yet mastered all of our norms. —JohnC5 16:47, 13 March 2017 (UTC)
Oh, you’re an admin? TIL. Anyway, I think that the use of {{also}} is very unnecessary when the same link/links is/are used in the definition of the entry, but that’s none of my business.
PapíDimmi (talk | contribs) 16:49, 13 March 2017 (UTC)
@PapiDimmi: Yeah, as I say, the usage of {{also}} is often silly, but then again, there are bots that would add it if I did not. —JohnC5 16:51, 13 March 2017 (UTC)

Why did you revert my revision?Edit

Link. PapíDimmi (talk | contribs) 11:00, 18 March 2017 (UTC)

It's an adjective, and "does not" is a verb. —CodeCat 13:19, 18 March 2017 (UTC)
@PapiDimmi: Also, the two attested meanings are not in possession of and not containing any, both of which are represented by without. This might be a little deceptive, as the meanings are different, and maybe the senses should be split. —JohnC5 17:33, 18 March 2017 (UTC)
The reason I made that edit is that I want to clarify the meaning of the word “niggerless.” Do you use the adjective to describe a person who does not own an African American person, or do you use it to describe a place that does not contain any African Americans (or perhaps both; and if that’s the case, maybe it should have two definitions).
PapíDimmi (talk | contribs) 16:23, 22 March 2017 (UTC)
@PapiDimmi: I don't understand you. You asked a question whose answer I already gave in this same discussion. —JohnC5 16:53, 22 March 2017 (UTC)
You said that the definitions should be split, and I asked what the actual definition of the word is. Does it mean A, B, or both?
PapíDimmi (talk | contribs) 16:55, 22 March 2017 (UTC)

Again, why did you revert my revision?Edit


What’s wrong with just explaining it in the edit summary? Hastily reverting a revision is fine if it’s vandalism, but this is not vandalism.
PapíDimmi (talk | contribs) 17:03, 22 March 2017 (UTC)

@PapiDimmi: Two things: I have provided a discussion on the talk page, and the rollback tool which I need to use to rollback multiple edits has a set text which I cannot change. —JohnC5 17:05, 22 March 2017 (UTC)
Then don’t use the rollback tool; use the undo button. PapíDimmi (talk | contribs) 17:09, 22 March 2017 (UTC)
...No. —JohnC5 17:11, 22 March 2017 (UTC)
Just FYI, the you can undo multiple changes with the undo button as well. Go a diff of range of revisions you want to undo and there will be an undo button on the top right of the page. If nothing else, this is useful to undo multiple revisions by different editors, only a subset of revisions by the same editor, or a range of revisions that does not include the latest revision. --WikiTiki89 18:08, 22 March 2017 (UTC)
@Wikitiki89: Yeah, I know, but I wasn't in the mood to explain the difference between the rollback and undo features. Thanks though. —JohnC5 18:10, 22 March 2017 (UTC)
Why not? It saves me from the trouble of continuously asking me why you reverted my revision.
PapíDimmi (talk | contribs) 22:51, 22 March 2017 (UTC)
You wouldn't have to ask if you started by reading John's input to relevant discussions and assessing whether there is a possibility that you might be in the wrong, based in your observations of other comparable entries. Going through that process every time would probably help you figure out why your edits were reverted, and hopefully help you to become a better, more conscientious editor. —Μετάknowledgediscuss/deeds 22:57, 22 March 2017 (UTC)
…No. PapíDimmi (talk | contribs) 22:58, 22 March 2017 (UTC)
Okay. Bear in mind that to the extent that John is in the right (as he consistently has been, which is unsurprising given his extensive experience) and you are unwilling to listen to him, your edits will continue to be reverted. Edit warring, even when explanations are not immediately given, may be grounds for blocking. —Μετάknowledgediscuss/deeds 00:34, 23 March 2017 (UTC)
I'm such a kindly person, giving out warnings like that. Equinox isn't quite as forgiving, it seems. But he is right that your Twitter is bizarre. —Μετάknowledgediscuss/deeds 00:38, 23 March 2017 (UTC)
@Metaknowledge If you're such a kindly person, why don't you at least shorten the block? — AWESOME meeos * ([nʲɪ‿nəʐɨˈmajtʲe sʲʊˈda]) 02:24, 23 March 2017 (UTC)
Questioning Equinox is not what I'm here to do. —Μετάknowledgediscuss/deeds 03:28, 23 March 2017 (UTC)
Wow, that Twitter is really something. I also tend to be skeptical of users who primarily edit offensive words. Regardless, I'm sorry to have caused all this argument. —JohnC5 03:40, 23 March 2017 (UTC)
Reminds me a bit of Skankhunt42 in the latest season of South Park. It's funny how real life often mirrors South Park episodes. --WikiTiki89 14:08, 23 March 2017 (UTC)


I don't think splitting it into two senses was a good idea. Compare catless, dogless, or almost any other -less. They really just mean "without X", whether it's a person who is without X (a moneyless beggar) or a place (a moneyless bank) or a system (a moneyless economy)... no need to split hairs... Equinox 02:15, 23 March 2017 (UTC)

I have to agree on this. —CodeCat 14:16, 23 March 2017 (UTC)
I already changed it back. —JohnC5 14:45, 23 March 2017 (UTC)
Oh, you realized that I wasn’t the one who did it? Good job, mate.
PapíDimmi (talk | contribs) 10:05, 10 April 2017 (UTC)

Unprotect Module:headwordEdit

I would like to make some changes to this module, could you unprotect it please? —CodeCat 22:59, 24 March 2017 (UTC)

@CodeCat: Done. Please inform me when you are finished. —JohnC5 23:05, 24 March 2017 (UTC)
@CodeCat: but I thought that you were an admin so that means you get the rights to edit it! — AWESOME meeos * ([nʲɪ‿nəʐɨˈmajtʲe sʲʊˈda]) 23:17, 24 March 2017 (UTC)

Fix Dbachmann's editsEdit

Hi John! I wonder if you can do a favour for me; to reformat his edits, especially on etymologies. I tried to do some, but I gave up and wasted time. Would you mind if you check on them please? Ta! — AWESOME meeos * ([nʲɪ‿nəʐɨˈmajtʲe sʲʊˈda]) 06:56, 28 March 2017 (UTC)

*pekʷ- reversionEdit

Why were the descendant Slavic/Russian reverted (and removed)? In the respective page etymologies, *pekʷ- is given as the ancestor.

Giacintoc (talk) 18:42, 1 April 2017 (UTC)

@Giacintoc: They are already listed under *pékʷeti, which is a subpage of *pekʷ-. —JohnC5 18:57, 1 April 2017 (UTC)
@JohnC5: Thank you, I understand now. Giacintoc (talk) 05:07, 2 April 2017 (UTC)
I'd suggest using {{see desc}} (or perhaps a new, more appropriate template) in such cases, to clarify that further descendants are listed on the subpage. Benwing2 (talk) 16:19, 2 April 2017 (UTC)


I notice that you cited a whole bunch of sources, but from the references it doesn't appear that any of them actually cite the verb itself, just the root. I would prefer that citations for verbs be explicitly for the verb, so that would mean they should be for a thematic root present in this case. Do any of the sources mention that formation specifically? —CodeCat 13:18, 2 April 2017 (UTC)

Rix does and I imagine Ringe does as well, although I don't have the latter handy right now. Rix writes it as *bʰéu̯dʰ-e- (described as "Präsens") and gives Vedic bhódati, Gothic ana-biudan, Greek πεύθομαι, OCS bljudǫ, etc. as descendants. Benwing2 (talk) 16:15, 2 April 2017 (UTC)
Ok, that's fine then. What about the stative and causative? —CodeCat 16:27, 2 April 2017 (UTC)
@CodeCat: So, for the imperftective Pokorny mentions "Themat Präs.", Rix shows *bʰéu̯dʰ-e-, Ringe and Adams I copied by accident, and Beekes, Cheung, Derksen all directly mention a thematic present. As for the stative and causative, they are both directly mentioned in Pokorny and Rix, and their descendants are mentioned in Derksen, Beekes, and Matasović. —JohnC5 17:34, 2 April 2017 (UTC)
Thanks. You do excellent work and I don't normally have any reason to question it, but I hope you don't mind if I stay vigilant anyway. —CodeCat 17:54, 2 April 2017 (UTC)
@CodeCat: No worries. I'm just happy someone is reading these articles to make sure I've done everything correctly! —JohnC5 18:00, 2 April 2017 (UTC)


I don't see why adding that would be bad, just a small side note worthy to mention, and also it's true. 23:50, 6 April 2017 (UTC)

@ You usage note was ill-formatted, unsourced, and inaccurate. —JohnC5 00:10, 7 April 2017 (UTC)

Why did you undo my revision?Edit

Link. PapíDimmi (talk | contribs) 05:50, 7 April 2017 (UTC)

We link words the first time they show up in the definitions, and generally do not do so again later in the definitions unless necessary. —Μετάknowledgediscuss/deeds 06:17, 7 April 2017 (UTC)
Really? Can you give me the link to the policy that say that? I often see links repeated in multiple definitions in entries.
By the way, I removed the link to “situation” because “situation” is not related to the definition, and it’s a common word that most people understand and thus does not need to be linked to.
PapíDimmi (talk | contribs) 06:19, 7 April 2017 (UTC)
It's not policy, but it's standard. John knows this and you don't, so you will have to follow his lead. As for "situation", I actually agree, but a little extra linking is harmless. —Μετάknowledgediscuss/deeds 06:25, 7 April 2017 (UTC)
Why is there no policy about it?
I’ve seen many Wiktionary entries with repeated links.
PapíDimmi (talk | contribs) 09:24, 10 April 2017 (UTC)
There are a fair number of unwritten rules of Wiktionary layout. It is now moot, given that you are blocked indefinitely, but whatever. —JohnC5 15:19, 10 April 2017 (UTC)
JohnC5, but I've wondered if it's possible (not necessarily PapíDimmi), but to have another admin (like you) to unblock or shorten the block of somebody, if they think the block is unfair? — AWESOME meeos * ([nʲɪ‿bʲɪ.spɐˈko.ɪtʲ]) 21:17, 10 April 2017 (UTC)
Yes, admins can and do modify blocks that are abusive, which are generally long blocks by an involved party. However, admins are respectful of each other's decisions, and I don't question other admins' judgement on blocks more than once every couple years, I'd guess. In this block, Equinox is known not to be abusive, was not an involved party, and made his block based on relevant diffs and the user's history. —Μετάknowledgediscuss/deeds 22:04, 10 April 2017 (UTC)
I would agree with Meta's assessment and also mention that I primarily block anons and rarely get involved with blocking registered users. In this particular case, I don't mind PapiDimmi's block given that every time the user started editing, I (several other admins) read all of their contribs because of PapiDimmi's unreliability and opinionation, which was a waste of everybody's time. —JohnC5 23:35, 10 April 2017 (UTC)
My impression is that they're no more than grade-school age and trying to bluff their way into being treated as an adult. Chuck Entz (talk) 03:17, 11 April 2017 (UTC)
Actually, I think we should link every important word in the definitions no matter how many times it is repeated. I always fix this when I see it. --WikiTiki89 15:25, 10 April 2017 (UTC)
@Wikitiki89: Interestesting. Of course, I can see an argument for both. Perhaps we should have a discussion about this elsewhere? —JohnC5 15:27, 10 April 2017 (UTC)

Edit request to Module:labelsEdit

Hi, could you replace line 153 with:

                                label = data.display or label

This is only a temporary change with little impact, but it should help me with tracking down problems elsewhere. —CodeCat 19:36, 12 April 2017 (UTC)

@CodeCat: done. As usual, tell me when to change it back. —JohnC5 20:01, 12 April 2017 (UTC)

Help with User:Gadrian01's editsEdit

@Angr, GuitarDudeness Hello. I'm not sure exactly who the Latin editors are. I think you're one, and maybe also Angr and GuitarDudeness, who I've pinged. Can you take a look at User:Gadrian01's edits? All of them are to Latin pages, and many if not all of them are wrong and should be reverted, but I'm not quite sure whether we should unilaterally roll them all back or go through them one by one. Thanks! Benwing2 (talk) 02:27, 25 April 2017 (UTC)

@Benwing2: An interesting case. The user clearly means well and seems to be focusing on Latin from the Vulgate Bible, but it is hard to tell whether some of the senses they add to entries are valid. They seem to have no regard for the POS of the lemmata, just dumping senses in willy-nilly. Chuck has been doing yeoman work keeping them in check. I don't know what to do about the user; they edit very infrequently to the point that communicating would seem of little use. —JohnC5 03:01, 25 April 2017 (UTC)

Why does 'sense' break if you specify a language?Edit

Every other damn thing breaks if you don't. Equinox 02:08, 27 April 2017 (UTC)

@Equinox: {{sense}} and {{qualifier}} were just Luacized at mod:qualifier. These now specifically only allow positional parameters. the |lang= parameter has never been used for anything in either template, and so now its absence is being enforced. I wonder if this will discourage people from using it in place of {{label}}... Who knows? I know this isn't much of an answer, but it's all I got. —JohnC5 02:14, 27 April 2017 (UTC)
I'm ambivalent about the removal of the |lang= parameter from {{qualifier}}, because a language code might be needed if the template were allowed to use subvariety labels from Module:labels/data/subvarieties. But at the moment the language code isn't used for anything in either of the two templates. — Eru·tuon 02:35, 27 April 2017 (UTC)
@Erutuon: That would represent a massive change in the purpose and usage of the {{qualifier}}. This template is not meant to categorize but merely to provide commentary. —JohnC5 02:38, 27 April 2017 (UTC)
@JohnC5: No, you misunderstand. The template should never categorize. All that the labels module would be used for is to standardize and link the dialect names that are commonly given using {{qualifier}}. — Eru·tuon 02:40, 27 April 2017 (UTC)
@Erutuon: When is {{qualifier}} used for dialect names? That is the job of {{label}}, {{alter}}, and {{a}} in different circumstances. —JohnC5 02:44, 27 April 2017 (UTC)
I'm not sure where all it's used that way, but many times at Wikisaurus:bathroom. — Eru·tuon 02:48, 27 April 2017 (UTC)
I would guess it's occasionally used that way in the various word list sections, which otherwise don't have a dedicated template for the purpose. — Eru·tuon 02:51, 27 April 2017 (UTC)
@Erutuon: Hmmm, intriguing. I feel we'd need to open this to wider discussion before making such a change to {{qualifier}}. At the moment, I am not particularly in favor of it. —JohnC5 02:56, 27 April 2017 (UTC)
Well, I did post about this in the Beer Parlour. See here. Maybe I should've pinged you. — Eru·tuon 03:02, 27 April 2017 (UTC)
At the moment, what I'm working on is making Module:qualifier add a tracking template when it's using a label that's currently in Module:labels/data/subvarieties. No change to resulting content yet. — Eru·tuon 03:05, 27 April 2017 (UTC)
@Erutuon: If we create {{dialect}} as @Jberkel suggested, then we could just move all of those usages of Module:labels/data/subvarieties over to that template. —JohnC5 03:12, 27 April 2017 (UTC)
I think {{lect}} would be better: shorter and can be used for things like chronolects. Chuck Entz (talk) 03:45, 27 April 2017 (UTC)
@Chuck Entz: Hmm, I agree. If the template is created, {{lect}} should be used. — Eru·tuon 04:01, 27 April 2017 (UTC)
I also agree. —JohnC5 05:02, 27 April 2017 (UTC)
Qualifier has always been the backup that you use when you have content that doesn't work in other templates. I oppose hijacking it for more specialized things. And given that people have done things in the past like leave notes for themselves in unused positional parameters, I find inclusion of Module:parameters to be rather annoying when there isn't an urgent need to stamp out misuse of specific parameters. Chuck Entz (talk) 03:56, 27 April 2017 (UTC)
@Chuck Entz: Respectfully, all the usages of second and third parameters I found for {{gloss}} and {{non-gloss}} have been things that the editors clearly meant to display, but since the values were in unused parameters, they were being omitted by accident. —JohnC5 04:46, 27 April 2017 (UTC)

Transliteration of उरुEdit

Here, the Sanskrit word उरु has been erroneously transliterated as "urú" while the correct transliteration should be simply "uru". The grave diacritic mark on the final u of the word is not only unnecessary but misleading as well. As someone whose native script is the Devanagri, I can say for a fact that there is no phonetic difference whatsoever in the pronunciations of the initial u and the final u that follows the r. Both are the short u, written as उ and when it is following र, it is written as रु, so the initial and the final vowels are the same sound. Using two different characters to represent identical sounds will only mislead people into thinking that these two vowels are pronounced differently while in fact they are not.

I tried to correct this mistake by providing the correct transliteration but you have reverted my edits and switched back to the incorrect transliteration. It would be very much appreciated if you would let those who can actually read the Devanagari as their native script and those who have an authentic and first-hand knowledge of it take care of the transliteration of Devanagri texts. —This unsigned comment was added by (talk).

@ Thank you for you explanation. Unfortunately, you do not understand some of the basic principles of Sanskrit phonology. Vedic Sanskrit had a phonemic pitch-accent (see here) that is traditionally transcribed with a grave accent. This accent is crucial to the reconstruction of the Proto-Indo-European accent. As you you can see in the the entry for उरु in the Monier-Williams Dictionary, there was a pitch accent on the second syllable of the word. Our entries all show this accent, when it is known. I appreciate that you are editing Sanskrit entries, but please understand that I know quite a bit about Sanskrit. —JohnC5 02:26, 8 May 2017 (UTC)
@ Also, please create an account and start editing under it. It will give you much more credibility. You've made a fair number of small errors in your entries, and knowing what you are editing will help me teach you how to fix these errors. —JohnC5 02:29, 8 May 2017 (UTC)
@ I think you are confused between urū and urú. We used a modified IAST transliteration, so the first one would be उरू and the second one उरु with a second syllable having a higher pitch. See WT:About Sanskrit for some information. P.S. Devanagari is my native script. —Aryamanarora (मुझसे बात करो) 17:57, 9 May 2017 (UTC)

Descendant vs borrowingEdit

Should Persian چادر be listed as a "borrowing" of Sanskrit छत्त्र (chattra) or should it be a descendant? Im asking because borrowings usually happen with the borrowed words remaining largely intact whereas derivatives undergo significant phonetical and semantic changes. I know that Persian is not an Indo-Aryan language and thus cannot naturally derive terms from Sanskrit. However, for instance English sugar comes from Old French çucre and ultimately from Sanskrit शर्करा (śarkarā) but we do not classify sugar as a borrowing of çucre or of शर्करा (śarkarā). In Persian, there is not only a sound change but a significant change in meaning as well; while छत्त्र (chattra) means an umbrella, چادر means a blanket. So I'm contending that چادر be listed as a descendant of छत्त्र (chattra) while the terms originating from چادر such as English chador and Hindi चादर (cādar) remain borrowings. 2405:204:9487:1FF7:0:0:F5E:20A4 09:30, 10 May 2017 (UTC)

@2405:204:9487:1FF7:0:0:F5E:20A4: This is an interesting question. So, at some point between Old French and English, the word sugar was borrowed. I've updated it up and it appears that the word was borrowed from Middle French to Middle English. Now, if we had a Middle English entry for sugre, it would say that the word was borrowed from Middle French. From the perspective of the English entry, the English term is merely derived Middle French, not borrowed, because the word was not directly borrowed from Middle French to Modern English. For छत्त्र (chattra), at some point there was borrowing before it got to چادر; we just need to figure out at what stage. —JohnC5 15:11, 10 May 2017 (UTC)
Ideally, we would say Modern English sugar is inherited from Middle English X, borrowed from Middle French Y, inherited from Old French çucre, .... If you don't want to trace it all the way back, then you can just say it's derived. --WikiTiki89 15:56, 10 May 2017 (UTC)
Well, the etymology templates indicate the relationship between the word in the language entry (English sugar) and another word in the chain of derivation. So, while it's true that Middle French y would be inherited from Old French çucre, the templates do not get to indicate the most proximate relationship; they indicate the ultimate relationship, that between English sugar and Old French çucre, which is one of derivation, since inheritance and borrowing do not apply. I dislike this; it would be nice to have a complete chain of proximate relationships, from which a module could calculate the ultimate relationship between the word of the entry and each word in the chain of derivation, but that is not currently possible. — Eru·tuon 06:25, 11 May 2017 (UTC)

Proto-Iranian descendant of ĉasásEdit

It is stated in *ĉasás that *šahakáh and *šaháh are the Proto-Iranian descendants of ĉasás. Are you sure it is not *tsahakáh & *tsaháh instead? PIE *ḱ is palatalized in PII but it becomes a voiceless palato-alveolar sibilant ( (śa)) only in Indo-Aryan. Proto-Iranian turns PII *ĉ to *ts which then becomes /s/ in Iranian languages. Here's what Wikipedia says about this: —This unsigned comment was added by 2405:204:9487:1ff7::f5e:20a4 (talk).

You can see User:Victar/Wiktionary:About Proto-Iranian#Consonants for a full account of PIE to Old Persian. Something strange is going on here (@Aryamanarora), because PII *ĉasá-ká-s should give PIr *ĉahakáh > OP *θahakaʰ, not PIr *šahakáh > OP *sahakaʰ. There are a lot of problems with this reconstruction, to be honest. —JohnC5 18:06, 10 May 2017 (UTC)
@JohnC5 I am confused as well. User:Victar/Wiktionary:About Proto-Iranian#Consonants does give OP s as a possible outcome of PIR though. There's still the issue of शश (śaśa) instead of the expected शस (śasa) that the Iranian cognate (MP sahōg) would correspond to. It would be great if we found an Avestan cognate though; I spent half an hour looking for one to no avail. —Aryamanarora (मुझसे बात करो) 18:13, 10 May 2017 (UTC)
@JohnC5 No, not at all... According to Victar's table, PII ĉ changes to both ts and š, not just ts, my bad. So ĉ to š to s in Old Persian is natural, I made a mistake. 2405:204:9487:1FF7:CDBE:1932:1535:4619 18:15, 10 May 2017 (UTC)
(edit conflict) @2405:204:9487:1FF7:CDBE:1932:1535:4619 No, you were originally right. I see the note Before *t, *dʰ, and *bʰ for PII ĉ to š now. Perhaps this reconstruction should be removed until we can verify it. —Aryamanarora (मुझसे बात करो) 18:18, 10 May 2017 (UTC)
@Aryamanarora I think why the noun is शश (śaśa) instead of the expected शस (śasa) is obvious: since both the syllables are sibilants, one influenced the other. 2405:204:9487:1FF7:CDBE:1932:1535:4619 18:26, 10 May 2017 (UTC)
@Aryamanarora This is not the first time a Sanskrit descendant is inconsistent; the PIE word for heart "ḱer" has given rise to हृदय while PIE ḱ does not usually correspond to Sanskrit ह. Such inconsitancies are quite common. 2405:204:9487:1FF7:CDBE:1932:1535:4619 18:26, 10 May 2017 (UTC)
we still can have PII *ĉasákás > PIr *tsahákáh > OP sahakaʰ, I don't see what the problem is? Just flip š with ts. 2405:204:9487:1FF7:CDBE:1932:1535:4619 18:32, 10 May 2017 (UTC)
In PIr, we represent [ts] with . However, PIr does not result in OP s, only PIr š before a consonant does. —JohnC5 18:47, 10 May 2017 (UTC)
Ok, so apparently OP θ became MP s, see *ĉatám. So PII *ĉasá-ká-s > PIr *ĉahakáh > OP *θahakaʰ > MP sahōg works. —JohnC5 18:54, 10 May 2017 (UTC)
As for the Sanskrit form, see this. —JohnC5 18:59, 10 May 2017 (UTC)
so I was right about one sibilant influencing the other then.. and the reconstruction is correct too! माधवपंडित (talk) 06:02, 11 May 2017 (UTC)
@JohnC5: Thanks for the cool link! —Aryamanarora (मुझसे बात करो) 18:20, 11 May 2017 (UTC)

Merging of sibilants in SanskritEdit

First off, I thank you for what you have done to solve our confusion with regards to the Indo-Iranian reconstruction *ĉasás. Your finding a source where this was addressed and also another example of such a phenomenon occuring has been helpful. There is one final discussion here: how do we know that PII *s and *ś merged in such cases in Proto-Indo-Aryan itself & not Sanskrit? I am theorizing that the PIA descendant of *ĉasás was *śasás which then became śaśá in Sanskrit. The source you have used as a reference mentions that the word was *śasás in "pre-Sanskrit". So a stage later than PII but before Sanskrit would be Proto-Indo-Aryan; these could be one of the very few differences between PIA & Sanskrit or else as we know Sanskrit and Proto-Indo-Aryan are so similar, they might as well be called the same. So the development could be: PII *ĉasás > Proto-Indo-Aryan *śasás > Sanskrit śaśá. All I'm saying is let the merge happen in Sanskrit. 2405:204:9487:1FF7:0:0:F5E:20A4 12:22, 11 May 2017 (UTC)

That's fine. Honestly, if we had some Prakrit examples, we could determine whether it is at the PIAr level or just Sanskrit. —JohnC5 17:06, 11 May 2017 (UTC)
@JohnC5: No, the Prakrit would be mostly useless. PIA *ś, *ṣ, *s all merge to *s in Sauraseni and Maharastri Prakrit and to *ś in Magadhi. —Aryamanarora (मुझसे बात करो) 18:23, 11 May 2017 (UTC)
yep, moreover, the Prakrits are descendants of the spoken form of Old Indo-Aryan/Vedic, I do not think they are Sanskrit's siblings. When Panini standardized Old Indo-Aryan, what he standardized remained Sanskrit while what was spoken by the masses evolved into simpler languages. What really would help is the dialect of Indo-Aryan from Mitanni as it is said to have been older than Vedic. 2405:204:9487:1FF7:0:0:F5E:20A4 01:56, 12 May 2017 (UTC)
This is not strictly true. Only Sauraseni Prakrit truly descended from Vedic Sanskrit. As seen here several distinctions are preserved in the Prakrits that disappeared in Vedic and later Sanskrit. —JohnC5 03:56, 12 May 2017 (UTC)

Swahili copulaEdit

The good news: thanks to your excellent help, Swahili entries are looking better than ever and I am deploying the new templates widely. The slightly unfortunate news is that the copula is so complex that a single table is simply not sufficient to cover it. Luckily, solving this is pretty simple: I just need three extra templates which excerpt rows from the main template. One should just show the present positive (with all persons and classes), one the present positive and negative (with all persons and classes), and one the present positive, negative, and relative (with all persons and classes). I would make them myself, but I don't know how to make them look like the main table, and I was hoping for aesthetic consistency. Thank you again for all your help with Swahili! —Μετάknowledgediscuss/deeds 08:19, 15 May 2017 (UTC)

@Metaknowledge: Ok, I'll work on this at some point. Should these tables generate forms or should they just be called through named parameters? —JohnC5 14:38, 15 May 2017 (UTC)
Yeah, generating them like the table does would be good. It should be just like the main table, except only tha/thoset selected row(s) (along with the header row). —Μετάknowledgediscuss/deeds 18:40, 15 May 2017 (UTC)
@Metaknowledge: Is this approximately that for which you are looking? —JohnC5
Yes, but I think there's no need for the subtable to be collapsed. —Μετάknowledgediscuss/deeds 03:35, 18 May 2017 (UTC)
@Metaknowledge: I thought you might say that. Better now? —JohnC5 03:38, 18 May 2017 (UTC)
One more minor quibble: when expanded it is much wider than {{sw-conj}}. But I do like that it doesn't cut off "Positive" and friends like sw-conj does! —Μετάknowledgediscuss/deeds 03:42, 18 May 2017 (UTC)
@Metaknowledge: I'm not sure these are within my power to fix easily. First of all, I think your tables must display differently than mine. Maybe give me some screen shots? The distinction is that there is not internal table on the new one. I guess I could make them the same by having an uncollapsible table within the new tables, but then it would have the same layout issues that the main table does. Up to you. —JohnC5 03:48, 18 May 2017 (UTC)
Which phenomenon would you like a screenshot of? Anyway, it's not particularly important, just aesthetics. —Μετάknowledgediscuss/deeds 03:51, 18 May 2017 (UTC)
@Metaknowledge: I've made them all consistently crappy (Template:sw-conj/table-pres-p, Template:sw-conj/table-pres-pn, Template:sw-conj/table-pres-prn). You're welcome! ;PJohnC5 03:58, 18 May 2017 (UTC)
@Metaknowledge: Ok, I've placed the new templates at Template:sw-conj/sandbox. We can work more on the layout later if you'd like. —JohnC5 04:13, 18 May 2017 (UTC)
Thank you so much! As for the layout, I don't have much to say except that I think words shouldn't be cut off where possible. To the best of my knowledge, there is literally nowhere else online with these kinds of conjugation tables for Swahili, so I hope that Wiktionary will become a major resource as I add more vocabulary. Given that your templates are a principal selling point, I just want them to look professional. That said, it may be more work than it's worth trying to solve it. —Μετάknowledgediscuss/deeds 04:28, 18 May 2017 (UTC)
@Metaknowledge: the truth is that it's very easy to adjust the layout because any change that's made to Template:sw-conj/table, Template:sw-conj/table-pres-p, Template:sw-conj/table-pres-pn, and Template:sw-conj/table-pres-prn will be put everywhere. I'm just not good enough at css and html markup to make them as pretty as I'd like. Do you know any other editors who are mavens at layout whom we could entreat? —JohnC5 04:39, 18 May 2017 (UTC)
Oh man, I'm not really sure who's a maven... Besides the usual crew at the GP, I have a feeling @Jberkel would be good at this kind of problem. —Μετάknowledgediscuss/deeds 04:45, 18 May 2017 (UTC)
I'm happy to have a look at this, but travelling with unreliable internet connection so might take some time before I get to it. – Jberkel (talk) 19:28, 20 May 2017 (UTC)

@Jberkel: *nudge* —JohnC5 19:42, 31 May 2017 (UTC)


Hello! My edit on Iranian branch of "sek-" was right. why did you revert that? —This unsigned comment was added by Irman (talkcontribs).

@Irman: On closer inspection, I see you were correct. Apologies! —JohnC5 06:36, 5 June 2017 (UTC)
No problem! (Irman (talk) 06:38, 5 June 2017 (UTC))

PIE *(o)-úsEdit

Does a *(o)-ús suffix exist in PIE, or is there only an o-grade *-us suffix? --Victar (talk) 11:56, 21 June 2017 (UTC)

I'd be rather hesitant with anything that has final stress in the nominative singular. There is no known ablaut pattern that accounts for that. —CodeCat 12:31, 21 June 2017 (UTC)
@Victar: I believe Code means there's no ablaut pattern ending with *-us which has final stress since *-tós exists. The normal way of analyzing this would be something like proterokinetic *ó-us ~ ∅-éw-. What is this concerning in particular? —JohnC5 14:57, 21 June 2017 (UTC)
@JohnC5: I was curious when looking into Proto-Germanic *þahsuz. Also compare Proto-Indo-Iranian *bʰaHĵʰúš. --Victar (talk) 15:07, 21 June 2017 (UTC)
*tetḱ- looks really weird as a root. Roots don't normally have two plosives next to each other. —CodeCat 15:14, 21 June 2017 (UTC)
It's a thorn cluster. --Victar (talk) 15:17, 21 June 2017 (UTC)
That root is fairly well supported in the literature. It's normally considered to be a reduplication reanalyzed as a root. In the case of *þahsuz, why are you suggesting an oxytone? —JohnC5 15:26, 21 June 2017 (UTC)
@JohnC5: As per my query, I was curious if there was any evidence of *(o)-ús suffix, similar to *(o)-ós. Both *þahsuz and *bʰaHĵʰúš would fit. --Victar (talk) 16:03, 21 June 2017 (UTC)
@Victar: All cases of -ús in Sanskrit and Ancient Greek are considered leveling of the accent from *(é)-us ~ *(Ø)-éws. We make this claim because typologically *i and *u should not normally be accented. —JohnC5 16:52, 21 June 2017 (UTC)
Nevertheless, at least in the case of PII, they could just as well be from *(o)-ús, given the e/a-merger. Assuming u was unaccented in all cases seems just as much a theory as any. --Victar (talk) 16:58, 21 June 2017 (UTC)
It doesn't follow our understanding of either the nominal system of PIE phonology. *i and *u are both semi-vowels forced to be syllabic by phonetic properties. *e, *o, , and are the only vowels that truly take stress. It is much easier to say that a paradigm that has leveled towards the oblique in some cases than to say that the entire phonological system of PIE should be reconsidered. And again, I don't see any evidence why Proto-Germanic *þahsuz couldn't be *tótḱ-us. —JohnC5 17:09, 21 June 2017 (UTC)
I have no problem with a *tótḱus reconstruction. My inquiry wasn't about any specific examples, but I provided them upon request. I was just simply curious to the whys and whynots in reconstructing a *(o)-ús suffix. --Victar (talk) 17:18, 21 June 2017 (UTC)
Yeah, sorry if that came off a bit harshly. We don't like reconstructing *ŕ̥, *ĺ̥, *ḿ̥, *ń̥, , or ú except if there is exceptional proof and no other available explanation. It's like how we avoid *a unless completely necessary. This is because we use a fairly conservative phonological model and notation system. —JohnC5 17:30, 21 June 2017 (UTC)
No worries. Happy to have an informative conversation. --Victar (talk) 18:28, 21 June 2017 (UTC)
A thorn cluster is just any sequence of dental + velar stop, but you don't see that in roots because roots don't have adjacent stops. —CodeCat 15:29, 21 June 2017 (UTC)
Except for all of the exceptions like *tḱey- and *dʰgʷʰey-. —JohnC5 15:38, 21 June 2017 (UTC)

Entry Request: Proto-Indo-European *ḱongʰ-Edit

Howdy John!!! This interesting PIE root word has no entry yet. It could be either *ḱongʰ- or *ḱon- and means a shell. I request you to create it if possible.

Descendants: it has Latin congius and Anct. Gk. κόγχος, κογχύλη etc as descendants and also is the ultimate source of English terms conch & cockle. Thanks a lot! 2405:204:93A6:C276:0:0:D6D:18B1 08:58, 28 June 2017 (UTC)

Beekes does not think that is the origin of κόγχη (kónkhē) and its derivatives are from PIE but instead from Pre-Greek. —JohnC5 23:00, 28 June 2017 (UTC)

Re: {{la-IPA}}Edit

Would it be a good idea to show some stages of Latin, just like {{grc-IPA}}? --kc_kennylau (talk) 09:34, 31 July 2017 (UTC)

@Kc kennylau: I've wanted that for a while, but it's hard to know which stages to show. What did you have in mind? —JohnC5 14:57, 31 July 2017 (UTC)
A very rough idea: Old Latin (pre 350BC), Classical Latin (1 AD), Late Latin (200 AD), Vulgar Latin (400 AD). The feasibility remains to be discussed. --kc_kennylau (talk) 15:44, 31 July 2017 (UTC)
@Kc kennylau: I would support this, but would like to know what literature would be used to define these. Also, we treat Old Latin as a separate language, especially as there are some opaque sound changes between Old and Classical Latin (intervocalic s (/z/) to r). —JohnC5 15:58, 31 July 2017 (UTC)
Then just rename the first one as Early Classical Latin :P (as is done in caussa) --kc_kennylau (talk) 19:28, 31 July 2017 (UTC)
I'm not particularly convinced on that one. —JohnC5 19:42, 31 July 2017 (UTC)
Old Latin might be speculative. I wonder, based on the odd vowel changes between it and Classical Latin, whether its diphthongs were really pronounced as a surface reading of the spelling would suggest. For instance, I wonder if oi was really pronounced /oi/, given that it changed to ū. (I would love to find out if anyone has explained how this sound change, and ohters, worked.) But I wonder the same about Classical Latin oe and Attic Greek οι (oi), and everything I've read seems to take those literally. But this may not be relevant to the proposal. — Eru·tuon 20:00, 31 July 2017 (UTC)

Hausa sortkeyEdit

Since I saw that you edited it, I thought I'd ask you for help. I want ' to be ignored in sorting except when it precedes y, in which case 'y should be treated as a single letter that comes after y. (And ƴ should be treated as being identical to 'y.) I don't know how to get it to work right. —Μετάknowledgediscuss/deeds 20:20, 13 August 2017 (UTC)

@Metaknowledge: I believe that the behavior you seek is already in place. Just remember to give the pages a good hard purge if they look out of place. —JohnC5 03:13, 14 August 2017 (UTC)


Hey John, could you have a look at *pstḗn for me? It's a bit of a complicated entry and I can use another set of eyes. Thanks. --Victar (talk) 06:13, 18 August 2017 (UTC)

Sanskrit gha-Edit

I'm trying to work out the etymology of Sanskrit: घसति (ghasati, to eat, devour, consume) and related terms, but am left scratching my head. gh- requires *g⁽ʷ⁾ʰ- or *g⁽ʷ⁾H- before a back vowel, but -a- open syllable requires *-e- or *-n̥-. Any ideas? See User:Victar/Reconstruction:Proto-Indo-European/gʷh₃nes-. --Victar (talk) 16:51, 23 August 2017 (UTC)

The only thing that I can come up with that might make sense is PIE *g⁽ʷ⁾h₃e-/*g⁽ʷ⁾h₃n̥- > pre-PII *gʰa- > PII *gʰa-. --Victar (talk) 18:50, 23 August 2017 (UTC)
@Victar: I would scrap this particular effort. Rix's proposal is all analogical, and as Cheung and Mayrhofer point out, the lack of non IIr. cognates makes this one impossible to sort out without a fair bit of guesswork. I would just have this as an IIr. set of roots and then mention in their etymologies the problems with reconstruction. —JohnC5 19:14, 23 August 2017 (UTC)
Regardless, a better understanding the development of gha-CV in Sanskrit, which exists elsewhere, and well as the development of *n̥ and *h₃ in PII benefits us all. Side note: Kobayashi reconstructs the development as PIE *n̥ > pre-PII *ᵃn > PII *ãⁿ. This is at odds with *ǰʰatás, but easily rectified with pre-PII *ᵉn. --Victar (talk) 19:35, 23 August 2017 (UTC)
@Victar: Oh I surely appreciate the research you've done. I'm just saying that I don't think you're going to come out the other side with a PIE entry, which is unfortunate. Are there other examples we have of palatalization before *n̥? How about the other syllabics (*l̥, *r̥, *m̥)? —JohnC5 19:41, 23 August 2017 (UTC)
PII *ǰʰatás is the best parallel I've found so far for *g⁽ʷ⁾ʰn̥-. I'll keep looking for more, specifically *g⁽ʷ⁾h₃-.
That entry is part of series of entries I'm working on, bits and pieces of which should be worthy of entries:
User:Victar/Reconstruction:Proto-Indo-European/gʷeh₃-, User:Victar/Reconstruction:Proto-Indo-European/gʷerh₃-, User:Victar/Reconstruction:Proto-Indo-European/gʷh₃nes-
--Victar (talk) 19:55, 23 August 2017 (UTC)
Some comments:
  • You're reconstruction of *gʷh₃ḗw-s is incorrect. Lengthened did not experience laryngeal coloration. Some people do propose *gʷéh₃-u-s ~ *gʷh₃-éw-s instead of *gʷṓws, but this is currently a minority opinion.
  • I'd say everything under *gʷh₃nes- is far too speculative. It requires too much reanalysis without any direct evidence of the internal structure nor any extra-IIr. evidence of such an irregularly shaped formation.
  • *gʷerh₃- can be a root, and I'll admit that it is tantalizingly similar to *gʷeh₃- (if there is sufficient evidence for this root). The claim, however, that this root arose from metathesis of an unattested *r/*n-heteroclite reanalyzed as a root stretches the imagination.
I don't mean to sound harsh, and I appreciate your research thus far. I'm just worried you are trying to connect several unrelated forms. —JohnC5 20:19, 23 August 2017 (UTC)
@CodeCat, Victar, Vahagn Petrosyan; pings all around! —JohnC5 20:22, 23 August 2017 (UTC)
Again though, those entries are irrelevant to my query, other than being a possible example (and I would never publish them in their current state). Please treat them as such and ignore them otherwise. I'm strickly asking about the development of gha-CV in Sanskrit. --Victar (talk) 20:33, 23 August 2017 (UTC)
Looking in Mayrhofer, the only ones I can find come from *g(ʷ)ʰn̥- and the like. —JohnC5 20:43, 23 August 2017 (UTC)
Right, thus *ǰʰatás, which gives us हत (hata), but not **ghata, which corroborates that becomes e normally in pre-PII. --Victar (talk) 20:54, 23 August 2017 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── Here are some examples I found of *g⁽ʷ⁾ʰVCV/*g⁽ʷ⁾ʰn̥CV:

  1. PIE *g⁽ʷ⁾ʰe-CV > pre-PII *gʰʲe-CV > PII *ǰʰa-CV > Sanskrit haCV: हसति (hasati) < *gʷʰés-e-ti
  2. PIE *g⁽ʷ⁾ʰo-CV > pre-PII *gʰā-CV > PII *gʰā-CV > Sanskrit ghāCV: [Term?] < [Term?]
  3. PIE *g⁽ʷ⁾ʰow-CV > pre-PII *gʰāw-CV > PII *gʰāw-CV > Sanskrit ghoCV: घोषयति (ghoṣayati) < *g⁽ʷ⁾ʰows-éye-ti
  4. PIE *g⁽ʷ⁾ʰn̥-CV > pre-PII *gʰʲe-CV > PII *ǰʰa-CV > Sanskrit haCV: हत (hata) < *gʷʰn̥-tó-s
  5. PIE *g⁽ʷ⁾ʰu-CV > pre-PII *gʰu-CV > PII *gʰu-CV > Sanskrit ghuCV: घुष (ghuṣa) < *g⁽ʷ⁾ʰus-éh₂
  6. PIE *g⁽ʷ⁾h₂ey-CV > pre-PII *gʰʲey-CV > PII *ǰʰay-CV > Sanskrit hayCV: हयति (hayati) < *gʷh₂éy-e-ti
  7. PIE *g⁽ʷ⁾h₂i-CV > pre-PII *gʰʲi-CV > PII *ǰʰi-CV > Sanskrit hiCV: हित (hita) < *gʷh₂i-tó-s

I'm open to other theories, but -h₃n̥- is the only one that I've been able to come up with:

  1. PIE *g⁽ʷ⁾h₃n̥-CV > pre-PII *gʰa-CV > PII *gʰa-CV > Sanskrit ghaCV: घसति (ghasati) < *gʷh₃n̥s-e-ti

--Victar (talk) 05:58, 24 August 2017 (UTC)

@Victar: Very interesting; thanks for this. If your evidence is true, then your solution also doesn't look good, since palatalization happens through laryngeals and from syllabic nasals. Unless you want to make an argument about rule ordering (for which you'd need a lot more evidence), then your evidence seems to show that the outcome of *g⁽ʷ⁾h₃n̥-CV should be haCV. The best solution is that that is a IIr. root that was innovated or borrowed or by some yet un-proposed process. One thing that could work could be *g⁽ʷ⁾ʰosHeti, since Brugmann's law wouldn't apply, but the formation would be odd. —JohnC5 06:10, 24 August 2017 (UTC)
Also, I suppose a tudati formation *g⁽ʷ⁾ʰsH-é-ti might have the cluster broken with an epenthetic vowel, but again a weird thought. —JohnC5 06:19, 24 August 2017 (UTC)
Oh, I just noticed the *h₃ potentially coloring the vowel coming from *n̥. Again you'd need to provide a lot of diachronic rule ordering. Also, what is the evidence for *g⁽ʷ⁾h₂ey-CV? —JohnC5 06:35, 24 August 2017 (UTC)
The laryngeal is aspirating the *g⁽ʷ⁾- and the syllabic nasal is independently vocalizing. Palatalization could only happened after the vocalizing of syllabic nasals, so it's pretty clear the order of that.
Thank goodness you finally understand my argument! How have we not been on the same page this whole time?! Yes, I'm saying that haCV is the conventional development for *g⁽ʷ⁾Hn̥-CV, but *g⁽ʷ⁾h₃n̥-CV is different because it's altering the vocalization of the following syllabic nasal:
PIE *g⁽ʷ⁾h₃n̥-s-e-ti > pre-PII *gʰoseti > pre-PII *gʰaseti > PII *gʰasati > Sanskrit घसति (ghasati)
--Victar (talk) 06:48, 24 August 2017 (UTC)
Unfortunately, comparative evidence points to laryngeal coloration being one of the oldest changes during/after the PIE. You're claiming that laryngeal coloration maintained functionality as a synchronic process from late PIE into early pre-IIr. and then that the laryngeals also maintained their distinction till after the syllabic nasals became oral vowels. That is very elaborate. Also, why does it not work with *h₂? That's what has been causing me so much confusion. Neither pre-IIr. *a or *o caused palatalization, so how did *g⁽ʷ⁾h₂ey-CV get palatalized? Again, whence did the reconstruction of *g⁽ʷ⁾h₂ey-CV come? —JohnC5 06:57, 24 August 2017 (UTC)
1. Proposing the laryngeals coloring persisted into PII, let alone pre-PII, is not a new theory and has been proposed many times in various papers.[1]
2. The vocalization of syllabic nasals could have happened very early, perhaps even starting in late-PIE. As mentioned, Kobayashi reconstructs the PII development as PIE *n̥ > pre-PII *ᵃn > PII *ãⁿ.
3. What you're missing is that palatalization happened in pre-PII, not PII, and in pre-PII *e was still retained. Or am I misunderstanding your confusion?
4. The development of *gʷh₂ey appears to be : *gʷeh₂- (to go) > *gʷh₂éyti > *gʷh₂ey- (to go) > हयति (hayati, to go)
--Victar (talk) 16:40, 24 August 2017 (UTC)
So the paper you've sited (Ollett) points to the fact that *gʷh₂éyti or any root with *h₂- or *h₃-coloring the vowel should not result in palatalization. Indeed something as simple as *g⁽ʷ⁾h₂es- or *g⁽ʷ⁾h₃es- should be sufficient. With that in mind, the LIV and Mayrhofer associate हयति (hayati) and हित (hita) with an IIr.-only root *ǵʰey-. I guess I should have asked about this data before. Yeah, the second and third laryngeals are not permeable to palatalization. —JohnC5 18:34, 24 August 2017 (UTC)
If we can get to *g⁽ʷ⁾h₂es- or *g⁽ʷ⁾h₃es- being responsible for *ghas-CV, my query is complete. And theoretically, the same coloring process should work for *g⁽ʷ⁾h₃n̥s-. --Victar (talk) 19:24, 24 August 2017 (UTC)
True, though *g⁽ʷ⁾h₃n̥s- is extremely unlikely. —JohnC5 20:11, 24 August 2017 (UTC)
Not as a root, but the outcome of a root like *gʷh₃ens-.--Victar (talk) 21:13, 24 August 2017 (UTC)
You've never explained where the *s in *gʷh₃ens- supposedly comes from. To be perfectly blunt, if you put *gʷh₃ens- into the mainspace I will rfv it immediately as wild, un-sourced speculation. —JohnC5 21:24, 24 August 2017 (UTC)
John, I feel as though you have barely listened to my this whole correspondence. If you had, you wouldn't be making such threats. --Victar (talk) 21:48, 24 August 2017 (UTC)
I've read your whole argument several times. I've had trouble understanding it because multiple times you've made needlessly complex assertions or just plain incorrect research. Several times you've gone around mainstream scholarship for the sake of your pet project of jury-rigging together several similar looking but probably unrelated roots. Not everything in PIE fits together like a beautiful puzzle. I'll admit that it took me a bit to cotton on to what your theory was, but that doesn't mean that everything is the extended-metathesized-reanalyzed-unattested-denominative-cum-root of some other semantically similar root. In this project we put up verifiable, citable scholarship, and if there is not cogent, well-respected explanation for something that can be sourced, then we don't make entries for it. In some ways this project is the cutting edge and in others we are woefully conservative. Reconstruction is one of the conservative sides. I'm going to ping @CodeCat, Angr, Chuck Entz to ask whether they think I'm being unreasonable, but even if I've not understood your argument, there still isn't a source in the world which can support your proposed forms in this case or anything similar to them. Just because something can be made to work does not mean we accept it, especially something that flies so wildly in the face of Indo-European's phonology and morphology. —JohnC5 22:04, 24 August 2017 (UTC)
John, if you did listen to me, you would have heard me say twice that these userspace entries, in their entirety, are not intended to be converted to mainspace entries. I welcome and beseech your critique and that of others of my mainspace entries, but these are my personal project on my personal sandbox and to criticize me for creating them and exploring the ideas therein is unwarranted. So please, retrain from the unjustified and condescending lectures. I respect the heck out of your vast knowledge, but I'll certainly think twice about bouncing ideas off of you in the future if all I'm met with is irritation and obstinance. --Victar (talk) 00:58, 25 August 2017 (UTC)
I'll admit that I overreacted, and for that I apologize. You've been doing excellent work on this project for a while now, and it is wonderful to have someone, who is so focused on reconstruction, working on IIr. etymologies. You're About Iranian page is a masterful resource, which I consult often. On the other hand, this is a rarefied set of topics, with which we are engaging and which require extreme stricture if we are meant to be take seriously. It is my fault for taking your personal project wrongly, but I will continue to advise restraint in your work. It is far better to propose something controversial in discussion and have it be rejected or (more often) wither from lack of interest than to make large changes only to have a pitched battle erupt. I'm also sorry for being condescending in my responses to you; you did not deserve that. On the other hand, I hope you appreciate why I'm being hard on you: you do such good work that it pains me when you go out on odd excursions. Some of your proposals are the stuff of scholarly papers and paradigmatic confrontations, not dictionary entries. When I see you making an entry in your userspace, my main worry is what might creep into the mainspace undiscussed. I hope you will not omit me from you discussions for I find your research very interesting. —JohnC5 03:53, 25 August 2017 (UTC)
Thanks, John. I know you're just looking out for the integrity of the project, and it would certainly be at my loss to lose your input. --Victar (talk) 07:14, 26 August 2017 (UTC)


Huh? Now it disagrees with its own etymology, and why is the ˀ in between the vowels? It should be at the end of diphthongs, not the middle. —Rua (mew) 20:12, 25 August 2017 (UTC)

@CodeCat: Sorry, for the confusion. I'm in the middle of splitting out sections of *plew- into *pleh₃(w)-. I'm moved that as part of the change. —JohnC5 20:21, 25 August 2017 (UTC)
Ok, but even so, it should be *plauˀtei in our notation. —Rua (mew) 20:38, 25 August 2017 (UTC)
@CodeCat: Did we decide that *eHw should become PBS *auˀ? Derksen reconstructs *oˀu, but I may have forgotten a conversation we've had before. —JohnC5 01:08, 26 August 2017 (UTC)
I'm a bit confused by that as well. Maybe something should be added to the BSL About entry. --Victar (talk) 07:16, 26 August 2017 (UTC)
There's no difference among the Balto-Slavic languages. The only thing that matters is the presence of absence of acute in a given syllable. This is why we denote it as ˀ, it's meant to indicate a suprasegmental feature rather than a phoneme. See also w:Proto-Balto-Slavic. Derksen seems to think it was a phoneme, but that position seems limited to the Leiden school and is also tied to the glottalic theory. Our own reconstructions are more conservative, based more on the actual Balto-Slavic evidence and less on conjecture. For the same reason we don't reconstruct a distinction between short a and o, the distinction isn't preserved anywhere in Balto-Slavic. —Rua (mew) 10:15, 26 August 2017 (UTC)
As for this term in particular, is an acute attested in Balto-Slavic? If not, then there shouldn't be one in Proto-Balto-Slavic either. —Rua (mew) 10:17, 26 August 2017 (UTC)
@CodeCat: Derksen seems to imply that there are acute forms. I'm fine with your proposed alterations otherwise. —JohnC5 02:14, 27 August 2017 (UTC)

Latin audioEdit

Why did De Vaan write that the verb belongs to the third conjugation? --kc_kennylau (talk) 17:34, 26 August 2017 (UTC)

@Kc kennylau: I dunno. It seems plausibly like a typo since it mirrors other fourth conjugation nouns in derivation and form. —JohnC5 02:23, 27 August 2017 (UTC)


Hi, I see you've moved the Faliscan to its own script, so thanks for that. But there's a pretty random link to the Faliscan in the English section now. Did you perhaps intend to put it in the Latin section under "descendants"? Lingo Bingo Dingo (talk) 09:38, 29 August 2017 (UTC)

@Lingo Bingo Dingo: That was a mistake. I created that link so that I could get the correct title for the Faliscan page, but then I forgot to remove the link afterwards. Thanks for the heads up! —JohnC5 22:28, 29 August 2017 (UTC)


Hi! Where are we with this? --Barytonesis (talk) 15:24, 5 September 2017 (UTC)

@Barytonesis: It's effectively done. Indeed, what we have now could be put into production and then updated later. I was just hoping @Erutuon would finish up the notes and categories. At t he moment, I need to be studying Sanskrit and therefore have been working one Sanskrit related modules. Depending what Erutuon thinks, I can put what we have in production. —JohnC5 18:39, 5 September 2017 (UTC)
@JohnC5: I'm in favor. The modules are functional now, and we can finish the notes and add the rest of the categories later. — Eru·tuon 18:55, 5 September 2017 (UTC)
@Barytonesis: We are live. @Erutuon, could you create all the categories we do currently have? Also, please monitor CAT:E for a bit to make sure nothing is broken. —JohnC5 19:40, 5 September 2017 (UTC)
Nice! I hope we haven't missed too many cases. We'll see. --Barytonesis (talk) 19:48, 5 September 2017 (UTC)
@Barytonesis, Erutuon: Apparently we need to deal with how non-singular adjective tables work (see δῐσχῐ́λῐοι (diskhílioi)). I'm gonna be out of town for several days, so I may have spotty ability to help with this. It should not be too bad, however. —JohnC5 20:36, 5 September 2017 (UTC)
The dative plurals of nouns in -ys, -yos (like ὀφρῡ́ς (ophrū́s), ὀφρῠ́ος) lack a pitch marker. --Barytonesis (talk) 21:35, 5 September 2017 (UTC)
Fixed. — Eru·tuon 21:44, 5 September 2017 (UTC)
And δῐσχῑ́λῐοι (diskhī́lioi) is fixed. — Eru·tuon 05:15, 10 September 2017 (UTC)
Return to the user page of "JohnC5".