Wiktionary talk:About French/Archive 1

Definition from Wiktionary, the free dictionary
Jump to: navigation, search


There is currently no consensus on which IPA symbol to use for the French <r>. The following notes may be given:

  • [ʁ]
    – the voiced uvular fricative – represents how <r> is pronounced in ‘standard’ French. It is also used on French Wiktionary. However,
  • [ʀ]
    – the voiceless uvular fricative – is characteristic of a Parisian accent, and is therefore quite familiar/common on French TV broadcasts etc. It is also the symbol used by all the major French print dictionaries, including the Robert and the Trésor de la Langue Française.

Which one should Wiktionary use to represent the phoneme?


For the record, I favour /ʀ/. Widsith 08:39, 15 July 2007 (UTC)

I prefer the character for the Standard French phone [ʁ]. I think the reason that the Trésor de la Langue Française, the Robert, and others use [ʀ] is that it looks less strange (just a majuscule ‘r’) than [ʁ], so as they are both allophones (as are [r], [ɾ], and [χ])[1], they use the more familiar character (for the same reason that the COED uses [r] to represent [ɹ]/[ɻ]). I may, of course, be wrong, as these dictionaries may actually be prescribing the [ʀ] phone in place of [ʁ]; however, if this is the case, then it will probably be explained in one of their explanatory forewords, which I invite someone to cite and quote here if such is the case. † Raifʻhār Doremítzwr 13:40, 15 July 2007 (UTC)

I prefer [ʀ] for the reason that major print dictionaries use it — even if their reason is just that it looks more like 'R'. That said, I'd also be O.K. with trying to distinguish words that use [ʀ] from words that use [ʁ] in the Parisian accent (so, parapluie as [paʁaplɥi], but partie as [paʀti]), if y'all think that could be tenable. (I haven't thought about it too deeply yet, so feel free to shoot me down on that point.) —RuakhTALK 16:36, 15 July 2007 (UTC)

Are [ʀ] and [ʁ] separate phonemes in the Parisian accent? † Raifʻhār Doremítzwr 16:47, 15 July 2007 (UTC)
Not at all; to quote from Diane Dansereau, Savoir dire: Cours de phonétique et de prononciation, D. C. Heath and Company (1990), ISBN 0-669-20996-1, page 185:
La consonne française [r] est sonore si elle est suivie d’un son sonore, comme dans les mots «riz» et «rue». Mais si elle est suivie d’un son sourd, elle est aussi sourde, comme dans l’expression «par terre». La variante sonore est de loin la plus commune. Comme nous l’avons déjà indiqué, la consonne [r] est aussi une fricative.
The French consonant [r] is voiced (lit. sonorous) if it’s followed by a voiced sound, as in the words riz ("rice") and rue ("street"). But if it’s followed by an unvoiced (lit. deaf) sound, it’s also unvoiced, as in the expression «par terre» ("on the ground"). The voiced variant is by far the more frequent. As we have already indicated, the consonant [r] is also a fricative [not just a liquid].
Should we not have separate symbols for allophones? (I can see an argument either way.)
RuakhTALK 17:18, 15 July 2007 (UTC)
We favour broad transcriptions in Wiktionary. Consonants in English are pronounced differently in initial, medial and final positions (see my postscript below) and sometimes depending on the following consonant: for example, "m" is pronounced differently in "comfort" (it is /ɱ/, Template:X-SAMPA, a w:labiodental nasal, that is, formed by putting the upper front teeth against the lower lip) from how it usually pronounced (by putting the lips together), as in "manage". We don't make these distinctions in Wiktionary, and, as far as I recall, the French monolingual dictionaries I have seen (Larousse, Hachette, Robert) do not make the distinction between the two sounds of the French "r" mentioned above. I don't think we should either. As for which is appropriate, presumably the contributors to the French wiktionary know what they are doing - they use the inverted R both initially ("rue") and before an unvoiced consonant ("par terre"). Perhaps we should take a poll of some modern French print dictionaries and go with the majority. — Paul G 08:39, 16 July 2007 (UTC)
PS: A further reply to Ruakh's point on using both R and inverted R: the test for whether distinct phones are needed is to look for minimal pairs (see also w:Minimal_pair). For example, the digraph "th" has two sounds in English: voiced (as in "the") and unvoiced (as in in "thing"). A minimal pair for these sounds is "thy"/"thigh". Different transcriptions are therefore needed for the two sounds (/ð/ and /θ/, respectively). English has aspirated and unaspirated consonants (compare "pin" and "spin") but no minimal pairs in this case; hence /p/ is used for both sounds. Likewise, I don't think there are any such minimal pairs in French as far as R or inverted R are concerned. So we should be using only one or the other. — Paul G 10:28, 16 July 2007 (UTC)
Re: "I don't think there are any such minimal pairs in French as far as R or inverted R are concerned.": Indeed not. As I said, they're context-selected allophones in dialects that have both. —RuakhTALK 16:06, 16 July 2007 (UTC)
By the way, lack of minimal pairs shouldn't be the only criterion. English has no minimal pairs between /h/ and /ŋ/, but I think we can all agree not to use a single symbol for both, because they're not just different phones, they're different phonemes. (I don't think [ʀ] and [ʁ] are comparable, mind; as they are the same phoneme. But it's important to be wary of simple-but-unreliable tests.) —RuakhTALK 16:17, 16 July 2007 (UTC)
Thanks for this page. I'm French, but I learn much from such discussions. I just want to add the following:
  • even Parisians associate R to the sound [ʁ] (in their mind, whatever the way they actually pronounce it)
  • using [ʁ] is never wrong, and not considered as strange, even in cases where [ʀ] is usual
  • for years, I have considered that the standard French R was represented by [ʀ]. But the only reason was that my dictionary was using this symbol. Authors are not specialists of phonetics, and might share the same impression, for the same reasons... And almost nobody is aware of the fact that two different sounds may be used.
  • of course, a choice must be made (this is implied by the use of // instead of [])
  • in my opinion, the best choice is [ʁ]. My reasons should be clear enough. Lmaltier 17:50, 18 July 2007 (UTC)
Based on the consensus that seems to be emerging I have tried to codify this on the project page. Could we get a bot to replace all the existing ʀs? Widsith 18:19, 18 July 2007 (UTC)
All participants to the above discussion should read the following document: http://venus.unive.it/canipa/pdf/HPr_04_French_fr.pdf (see and It makes things clear. However, note that, if you can understand the document without any problem, you must contribute to the Wiktionnaire... Lmaltier 17:50, 8 October 2007 (UTC)
I understand most of it (though I'm not familiar with a lot of the IPA symbols used in the detailed phonetic transcriptions, aside from what I could figure out from the text itself), but I'm not sure how worthwhile it is to our purposes here: we certainly can't use such detailed phonetic transcriptions, because most of our readers won't be able to read them, and the article doesn't really say much of note regarding broad phonemic transcriptions. Where it does have interesting things to say about broad phonemic transcriptions, I think we should be leery of defying common practice by following it instead; for example, given that nearly all current dictionaries write /ɛ̃/, do we really want to write /æ̃/ instead? (Even so, it was a very interesting read; thanks for the link!) —RuakhTALK 04:45, 9 October 2007 (UTC)
It was not a suggestion to follow it, of course, just an answer to questions raised in the discussion about R. Lmaltier 05:54, 9 October 2007 (UTC)
Oh, yes, O.K., sorry. Somehow I missed that you were pointing us to specific sections, and by the time I got to page ten I have to admit I was skimming a bit! I'm a bit confused, because that article says that [ʀ] denotes the voiced uvular "vibrant" (flap or trill), and in this discussion we've been assuming that it denoted the voiceless uvular fricative. I'm not familiar enough with how the IPA works to understand what's going on here; does the symbol actually have these two distinct meanings? Do we need to reboot this discussion in view of this? Sorry/thanks again. —RuakhTALK 06:40, 9 October 2007 (UTC)
I cannot say anything about "voiceless uvular fricative" and this kind of things (I'm not a specialist). The specialist uses [ʀ] (and explains why), but finds [ʁ] more "legitimate". So, there is no real need for a new discussion, I think, but others might think otherwise. Lmaltier 16:52, 9 October 2007 (UTC)

I must inform you that this whole discussion is based on the wrong presumtions. People should check their sources, e.g. Image:IPA chart 2005.png, before introducing false statements as in the “notes” at the top of this discussion. In the IPA, [ʁ] does indeed represent the voiced uvular fricative, but [ʀ] stands for the uvular trill, not the voiceless uvular fricative, which is properly transcribed [χ]. There is only one r-phoneme in French, which may be realized as a trill (alveolar [r] or uvular [ʀ]) or an uvular fricative [ʁ], depending on dialect. It is fundamentally voiced, but in many, if not most, dialects it is systematically devoiced before voiceless consonants (p, t k, s, f), becoming [r̥], [ʀ̥], or [ʁ̥] (i.e. [χ]), respectively. In a phonemic representation of French, I believe the use of a single basic symbol for this single phoneme to be for the best, but indicating devoicing via diacritic seems a good idea. For this symbol, we have a choice from the symbols for the three different realizations: [r], [ʀ], and [ʁ]. Now, [ʁ] (along with its devoiced allophone [ʁ̥]) is arguably the most common realization of this phoneme, for which reason the scholar quoted by Lmaltier above finds its symbol the most “legitimate” in a general system. Both [r] and [ʀ] are less common, but may be found in certain dialects and in specialized usage such as singing. Arguments for these include simplicity, originality, or conformity with existing dictionaries (as noted by others above). In any case, none is distinguished from another in any French dialect, so it doesn’t really make any difference which is used. I’d be okay with any of them, although I would prefer /r/ or /ʀ/. This really is a very similar question to that of the representation of English r as /r/ or /ɹ/, which has previously been discussed. – Krun 16:35, 14 June 2009 (UTC)

Vowel length[edit]

I wanted to add a comment to the effect that we shouldn't bother showing vowel length since it's not phonemic in French. But do people feel particularly that we should differentiate between long and short vowels? I don't really hear a difference in French anyway - I mean, I can hear it but it doesn't mean anything. Thoughts? Widsith 18:19, 18 July 2007 (UTC)

In theory, I think it's good to distinguish them, because there are many forms of French (mostly historical ones) where they're phonemic. In practice, I don't think we have the expertise here to do so, short of stealing pronunciations from other dictionaries. —RuakhTALK 19:07, 18 July 2007 (UTC)
Interesting – I hadn't thought about that. Widsith 07:35, 19 July 2007 (UTC)
Like Widsith says, vowel length is not phonemic in French, so usually it is not shown in dictionaries. Ruakh, I don't think it is possible to "steal" pronunciations from other dictionaries: if we happen to represent the same variety of French as other dictionaries and use the same pronunciation scheme, then inevitably we will render pronunciations identically to those dictionaries and it will be impossible to say whether or not we copied them. No copyright lawyer would have a valid case (IANAL). — Paul G 13:37, 19 July 2007 (UTC)
Call me old-fashioned, but I think it's theft whether or not a copyright lawyer could prove it in court. (And, I'm not sure I'd agree that it's not usually shown in dictionaries; the dictionary I use most, TLFi – le Trésor de la langue française informatisé, freely accessible online via http://www.cnrtl.fr/lexicographie/, does show it, and I've seen it shown in other dictionaries as well.) —RuakhTALK 14:45, 19 July 2007 (UTC)
I think you're mistaken to say that pronunciations can be stolen from other works of reference. The pronunciation of (for example) "oui" is /wi/. Now, did I steal that or compose it from memory? If I stole it, can anyone say which dictionary I stole it from? If it matches the pronunciation given in ten different dictionaries, does it mean that I've stolen it from all of them? Is there a case for saying that a new dictionary has stolen all of its pronunciations from one previously published?
Clearly not, otherwise there would be lawsuit every time a new dictionary is published. The pronunciation of "oui" is /wi/, which is a fact, and facts cannot be copyrighted. Whether I remembered this or copied it from a book is both immaterial and unprovable. However, if I copied the definition of "oui" from a dictionary, that would indeed be copyright theft, as dictionary definitions are the expression of the authors' ideas, which are copyrightable. — Paul G 16:53, 20 July 2007 (UTC)
The issue is more complex: Vowel length directly correlates with some vowel quality pairs (/a/~/A/, /o/~/O/, /E/~/3/), and which of the element is phonemic from a linguistic POV does not seem to be quite settled. Fact is also that most dictionaries are France-produced, so that contrast maintained elsewhere are frequently not indicated: some dictionaries already mark /a/ and /A/ as /a/! Circeus 19:20, 19 August 2007 (UTC)
What about when vowel length is phonemic in some dialects (notably the Québec dialect)? Some speakers make a distinction between, for example, 'mettre' (/mɛtʁ/) and 'maître' (/mɛːtʁ/). —Internoob (Talk|Cont.) 21:30, 17 March 2009 (UTC)
The only reason "maître" is usually shown as /mɛːtʁ/ is that there is no agreed symbol for the Quebec French ɛː, which is NOT just a long ɛ (I made an habit of using ɜ because it was suggested in a book I once read, which was written before that symbol was formalized for a different vowel). Vowel length is a relevant factor, but AFAIK only differenciates on its own where there is contraction: /dɑ̃/ -> "dans" /dɑ̃ː/ -> contraction of "dans la". Circeus 23:40, 17 March 2009 (UTC)
Should we indicate ɛː in our pronunciation sections, then? If so, how? —Internoob (Talk|Cont.) 21:14, 27 March 2009 (UTC)
Although it is not ideal, in the absence of a better recommendation ɛː is probably our bestoption, and it should of course only be indicated for Quebec pronunciations. Circeus 22:45, 27 March 2009 (UTC)

Conjugation templates[edit]

Conjugation notes[edit]

As mentioned at Wiktionary:Beer parlour, I've edited {{fr-conj-pre}} to add conjugation notes. The notes differ a bit depending whether the linking page is rompre or another page, so please check out both rompre and interrompre and let me know what you think. (This is not a widely transcluded template, by the way — those are the only two pages that include it — so please be bold in making changes to it.) —RuakhTALK 00:28, 19 August 2007 (UTC)

Naming convention[edit]

There's really not much pattern to the names of the various conjugation templates. This probably isn't a huge deal, and to some extent it's inevitable given the quirks of the various conjugations, but I think some approximation to consistency might be nice. Any thoughts? —RuakhTALK 00:28, 19 August 2007 (UTC)

Le gérondif[edit]

How do we want to refer to le gérondif (e.g. "en faisant")? Currently we're usually referring to it as the "gerund", but there are a lot of problems with that designation. Other designations I've seen in English include "gerundive", "adverbial gerund", "adverbial gerundive", "en participle", "adverbial participle", and simply italicizing the French term. Personally, I'm partial to the last of these, though I'd also support mostly sidestepping the issue by not mentioning them except when necessary, seeing as they're a fairly straightforward sum of en and the present participle. Then, in the rare cases where we do have to mention them, we can write something like "the gérondif (the verbal adverb, sometimes called the 'gerund' or 'gerundive')". —RuakhTALK 17:12, 21 August 2007 (UTC)

How about not including it in the table? It's not even a conjugated form to begin with, and I know of no native conjugation guides that use it. Circeus 20:27, 21 August 2007 (UTC)
As I said, I'd support that (though they are included in at least some native conjugation guides, such as Bescherelle's La Conjugaison pour tous). —RuakhTALK 21:37, 21 August 2007 (UTC)
They are not in their Art de Conjuguer (only the bare present participle), though. I think there's been a drive to use "en" in the conjugation books because that's by far the most common use for that verb form, but that still doesn't make the "en" part of the conjugation. Circeus 01:24, 22 August 2007 (UTC)
So can I assume that you support my suggestion to simply not mention le gérondif except when necessary, and when we do have to mention them, to write something like "the gérondif (the verbal adverb, sometimes called the 'gerund' or 'gerundive')"? —RuakhTALK 05:21, 22 August 2007 (UTC)
Yes. But an appendix is not even something that would work. The gerund is always en+présent participle. So just sending people to the gérondif entry (which I will write now) should be good enough. Circeus 15:17, 22 August 2007 (UTC)
I wrote an entry.Does it make sense? Circeus 15:48, 22 August 2007 (UTC)
k. :-) —RuakhTALK 17:37, 22 August 2007 (UTC)

We should definitely include it in some way. Perhaps you could outline why gerund is so problematic? Widsith 08:20, 22 August 2007 (UTC)

The "gerund" does not belong in a French conjugation table because it's a) not a conjugated form (it's a syntactic construction based on present participle preceded with "en") and b) it corresponds to syntactic uses ("gérondif" =/= gerund) distinct from those of the English gerund. Examples:
Les miettes tombant de la table était récupérée par le chat. ("Bits falling from the table were caught by the cat")
This is a French participe présent, it works roughly like the English one.
Je me suis fait mal en tombant de la table. (I hurt myself by falling off the table.)
This is a French gérondif construction. It does not correspond to a gerund, which i English happens to be identical to the present participle; it's a construction inherited from Latin.
Circeus 15:10, 22 August 2007 (UTC)
Re: "Perhaps you could outline why gerund is so problematic?": Certainly. Gerund is problematic here because the only languages to which the term is applied consistently are English and Latin, and in both, it refers to verbal nouns. There are some writers who apply the term gerund to other languages, such as French, but even among these there's no consistency: some use it to refer to le participe présent, others use it to refer to certain uses of le participe présent, and yet others use it to refer to le gérondif. —RuakhTALK 17:37, 22 August 2007 (UTC)

Yes I understand all that but still don't see quite why it's such a problem. But anyway, I have no problem with just removing that line from the table if that's what people want. Giving the present participle is surely enough. Widsith 15:29, 22 August 2007 (UTC)

That's exactly what I and Ruakh are saying above... Circeus 15:48, 22 August 2007 (UTC)
Gotcha. Let's do it. Widsith 08:06, 23 August 2007 (UTC)


Keene (talkcontribs) is modifying all the French conjugation templates to use {{fr-conj}}. I think this is a good idea, except that I think there should be an intermediate stage: I think we should have a {{fr-conj/avoir}} and a {{fr-conj/être}}, and the templates should use one of those based on {{{2}}}. ({{fr-conj/avoir}} and {{fr-conj/être}} would then use {{fr-conj}}, but supplying the compound forms as well.) This way, we can replace the "Use the present tense of avoir or être followed by the past participle" things with actual forms, which IMHO will look much better and be somewhat more useful. Thoughts? —RuakhTALK 01:57, 17 February 2008 (UTC)

But the compound forms have no more existence than the dozen or so compound you can form with any English verb, and we certainly do not list those... Circeus 02:55, 17 February 2008 (UTC)
I know; I'd actually prefer to omit them entirely. But if we're going to devote space to them, we might as well do so usefully. (I'm not suggesting that the forms be linkified, only that they be given.) And if we had English conjugation templates, I think they probably would include certain English compounds. —RuakhTALK 05:35, 17 February 2008 (UTC)


I've noticed that the symbol for glottal stop, [ʔ], is commonly used to represent a hiatus, or an h aspiré. [1] [2] [3] This makes the transcription potentially ambiguous, since a glottal stop is not an h aspiré. Should we allow this? Isn't IPA supposed to be unambiguous? —Internoob (Talk|Cont.) 00:10, 21 May 2009 (UTC)

That's because the hiatus is unrepresentable at the word level: it is a phonological process (triggered by the stringing of words to form a sentence, in this case). As such, it will always be impossible to represent without ad-hoc systems in a phonemic representation of only one word. Circeus 01:51, 21 May 2009 (UTC)
But we should use an ad-hoc system that does not involve an actual IPA symbol. ([ʔ] does not correspond to the normal pronunciation, and not everyone would agree that an underlying /ʔ/ is the correct phonemic representation). I would suggest instead creating a template for a prefixed "*" (or other symbol) that links to an explanatory note and adds these words to something like [[Category:French aspirated h words]]. I just had a look at the Wiktionnaire, and they seem to use [[fr:haïr#Prononciation|ʔ]], [[fr:homard#Prononciation|h]], or nothing at all, so let's not follow their example. CapnPrep 15:20, 14 June 2009 (UTC)
Agreed. I think that we should create templates for both h aspiré and h muet; that way, people won't assume that if we don't use the template for h aspiré that it means that it's muet. —Internoob (Talk|Cont.) 20:42, 21 June 2009 (UTC)
As an added note, to represent a vowel hiatus, proper way is with a period cf haïr -> /a.iʀ/. But I don't think that is usual for consonants. Circeus 00:19, 22 June 2009 (UTC)
How do we like this template? I'll make another {{muteh}} if we like it. —Internoob (Talk|Cont.) 18:38, 17 July 2009 (UTC)
I think it's kind of overkill. If anything, I'd rather just have a word or two (‘aspirated’, ‘non-aspirated’) which would perhaps link to a relevant explanation at Wiktionary:About French. But then, I actually like the use of /ʔ/. Ƿidsiþ 12:37, 18 July 2009 (UTC)
I made some changes. —Internoob (Talk|Cont.) 23:23, 22 July 2009 (UTC)
More changes. Now it looks more or less like fr:Modèle:h. —Internoob (Talk|Cont.) 18:57, 23 August 2009 (UTC)
We (the fr.wikt contributors) have just finished the final installation: fr:Modèle:h has been divided in fr:Modèle:h muet and fr:Modèle:h aspiré. Consequently I've created {{muteh}} on the same template, to keep an interwiki similarity. JackPotte 04:48, 24 August 2009 (UTC)
Moreover, I suggest to rename the current Category:French_words_with_mute_h and Category:French_words_with_aspirated_h, by respectively Category:French_terms_with_mute_h and Category:French_terms_with_aspirated_h, because we also have some expressions like "haut-parleur" (loudspeaker). JackPotte 05:01, 24 August 2009 (UTC)
But "haut-parleur" is a (compound) word, not a "term"! Circeus 20:08, 24 August 2009 (UTC)
Dear, according to term#Noun (2) it's the both. JackPotte 06:56, 25 August 2009 (UTC)

I see 3 people in favour of the change, (me, CapnPrep and Jackpotte,) 1 against, (Widsith,) and 1 unclear (Circeus). Should we make a policy? —Internoob (TalkCont.) 00:57, 27 August 2009 (UTC)

Sorry for the short time but I've just added the 2 templates in Wiktionary:About_French#Pronunciation (in the same time as the fr.wikt corresponding policy). Apart from that I'm waiting for your confirmation concerning the 2 categories names. JackPotte 04:38, 27 August 2009 (UTC)
Okay. Now we can revoke the green notices from the templates. I have no opinion concerning the category names. —Internoob (TalkCont.) 00:33, 28 August 2009 (UTC)
Let's change it tonight (UTC), and add a "nocat" option to the templates. JackPotte 07:52, 28 August 2009 (UTC)
"Nocat option"? I'm sorry, but I don't know what that is. :X Maybe someone else should do it. —Internoob (TalkCont.) 17:24, 28 August 2009 (UTC)
"|nocat=yes" inside {{asph}} or {{muteh}} would not classify in their categories. It would permit to use them in policies but it's a detail. JackPotte 20:58, 28 August 2009 (UTC)
Done. I just learned how to do that, and I hope it's right. —Internoob (TalkCont.) 02:44, 29 August 2009 (UTC)
Thank you it's now working as the fr.wikt one (Template:asph, Template:muteh), with the same categories names. Next step: deploy them into the hundreds of concerned pages. JackPotte 19:15, 29 August 2009 (UTC)
Apart from that, no news in Wiktionary:Beer_parlour#aspired_h_vs_silent_h. JackPotte 02:38, 30 August 2009 (UTC)

Reflexive verbs[edit]

Continued from Wiktionary:Beer parlour archive/2009/June#Pronominal verbs.

My proposition is that reflexive verbs like se laver should only have their own page if not sum of parts. For example, I'd say that se laver is, because it's to wash oneself (note the red link). However something like se passer cannot be considered SoP (IMO) because it's usually translated by to happen. It can be SoP in other contexts: Ils se sont passés leurs notes (they exchanged notes, that is, passed them to each other). but the happen meaning is idiomatic. Mglovesfun 13:29, 19 June 2009 (UTC)

My general position is that simple pronominal (including all those that are found only as pronominal, such as s'arroger) can be dealt with at the verb with grammar labels, just as we do for e.g. impersonal verbs. When more words come in and we move into idiom territory, however, a separate entry with the pronoun is appropriate (évanouir, but s'évanouir dans la nature). This is consistent with the fact that there are virtually no verb that were pronominal from their inception. Virtually all started as non-pronominal verbs with only pronominal usage remaining today. Circeus 00:09, 22 June 2009 (UTC)

Accents and ligatures[edit]

Food for thought as it were. I'm gonna add siecle if we don't already have it, as most spellings with a grave accent and a lot of them with an acute are attestable in (early) Modern French. Is there any reason to exclude these, when they are spelt the same but an accent has been added later on. So decision and décision are the same word, the accent was added in the Modern French period (according to my sources). What do people think?

Ligatures it somewhat the opposite. As they're difficult to type, most of the time they miss them out. So cœur becomes coeur, and œil becomes oeil. This isn't just low quality sources either, I've seen them in hospitals and police stations and other "official" places. Note that the French Wiktionary no longer classes these as misspellings, but as 'typographical variants'. Do we want to do the same? Or we could go for redirects, but not for coeur because that's already got another language on it. Mglovesfun (talk) 12:16, 1 December 2009 (UTC)

I think we should have things arranged so that all hard-to-type correct forms can be found conveniently by someone typing without ligatures or accented characters. If it must take an extra click, OK. {{also}} has a role because it enables a user to avoid necessarily having to pagedown to the French language section in some cases. Can a bot be designed to add such entries/redirects or to make a list of French entries that could use them? The idea would be to add those where the accent- and ligature-less spellings did the least violence to the spellings used in the language (or in modern editions of the language). French would be a good test case for the general problem. This must have been addressed before, but I'd not participated.
I suppose that there might be no reason to have more than a redirect in many cases. One of the biggest disadvantages to redirects is that they either discourage would-be new contributors or lead them to make their contribution under the wrong spelling, but we discourage them in so many other ways that this seems trivial. DCDuring TALK 12:43, 1 December 2009 (UTC)
See m2; that sort of approach seems suitable here.  (u):Raifʻhār (t):Doremítzwr﴿ 00:50, 2 December 2009 (UTC)
Good idea, I don't consider these misspellings. Mglovesfun (talk) 21:15, 3 December 2009 (UTC)
What would be good wording that did not tag these as non-standard? m2={{form of|typographical variant|m2}}? It would be good to indicate that some these are less preferred in some contexts (because they run the risk of being misinterpreted) without requiring a full usage note. DCDuring TALK 22:17, 3 December 2009 (UTC)
Well, they’re non-standard in that they do not represent the standard; “defective spelling of” (which is used for Hebrew, for example) is an alternative.  (u):Raifʻhār (t):Doremítzwr﴿ 23:44, 3 December 2009 (UTC)
NOOOOOOOOO!!!!!!!! "Defective" doesn't mean "non-standard", it means "using fewer vowel letters"! (It's the opposite of "excessive", which means "using more vowel letters".) The Hebrew analogue of your ligature-heavy writing style would probably have a very high proportion of defective spellings, whereas plebes like me are more inclined to excessive ones. —RuakhTALK 23:55, 3 December 2009 (UTC)
¿Qué? OK, so I misunderstood what “defective spelling” meant. However, wouldn’t txtspk be a better analogue of defective spelling in English, whereas pronunciatory respelling systems that use a lot of digraphs (like o͞o and o͝o) would be a better analogue of excessive spelling?  (u):Raifʻhār (t):Doremítzwr﴿ 10:01, 4 December 2009 (UTC)
If we're talking strictly about the writing system, not about any cultural context, then yes: defective spelling is somewhat like text-speak, and I suppose excessive spelling is somewhat like the use of -ue in "analogue" and -ugh- in "doughnut" and so on. But thousands of years ago, when the writing system were new, spellings were completely defective; and the shift to less defective, more excessive spellings has been going on almost continually ever since. (It's not been a perfectly uniform transition — for example, the Dead Sea Scrolls used some excessive spellings that were still uncommon centuries later; and when Modern Hebrew was revived as a spoken language of daily life, some people turned to the Bible, rather than to the literary tradition, for spellings — but the overall trend is clear.) So, broadly speaking, a more conservative speller will use more defective spellings, while a more innovative one will use more excessive ones. —RuakhTALK 18:17, 4 December 2009 (UTC)
Understood; thanks. So, what do you suggest we should do with non-standard undiacriticked forms?  (u):Raifʻhār (t):Doremítzwr﴿ 14:59, 7 December 2009 (UTC)
The [[m2]] approach works for me. Though I might prefer something like "representation" rather than "form". —RuakhTALK 17:49, 7 December 2009 (UTC)
Representation / form / whatever; I’m fine with either.  (u):Raifʻhār (t):Doremítzwr﴿ 17:59, 7 December 2009 (UTC)
Did I mention that I created {{nonstandard spelling of}}. Mglovesfun (talk) 17:20, 7 December 2009 (UTC)
And IMO, that’s the best way of dealing with these variants. OTOH, you said that “most [of these undiacriticked] spellings…are attestable in (early) Modern French[; accents were] added in the Modern French period”, so mightn’t many of these be more legitimately marked as {{obsolete spelling of|…}}?  (u):Raifʻhār (t):Doremítzwr﴿ 17:59, 7 December 2009 (UTC)
More legitimate than what? Isn't that what I did with siecle? Mglovesfun (talk) 18:05, 7 December 2009 (UTC)
Yes, and than {{nonstandard spelling of}}.  (u):Raifʻhār (t):Doremítzwr﴿ 18:13, 7 December 2009 (UTC)
Right, sounds like a 'consensus' to me. Mglovesfun (talk) 20:24, 8 December 2009 (UTC)
I agree, these should be {{obsolete spelling of}}s. Ƿidsiþ 10:22, 9 December 2009 (UTC)

Template:fr-conj again[edit]

I have a couple of suggestions to make about Template:fr-conj to make it more friendly to our readers.

The statement under Compound gerundive that says "Use the gerund of avoir or être followed by the past participle" is really quite superfluous. It would be far clearer to print the actual past gerundive form, as in

{{#switch:{{{aux|}}}|être=en [[étant]] [[{{{pp}}}]]|avoir=en [[ayant]] [[{{{pp}}}]]|en [[ayant]] [[{{{pp}}}]] '''''or''''' en [[étant]] {{{pp}}}}}

which gives, for donner,

en ayant donné

or, for sortir,

en étant sorti

or, for a verb with no auxiliary specified (partir is just an example here),

en ayant parti or en étant parti

Also, why do we put the message saying "Use the ... tense of <aux> followed by the past participle" under compound verb tenses, rather than printing the actual forms? I notice that Circeus (talkcontribs) mentioned, in the discussion above, that we don't list English compound tenses in full. However, this comparison cannot be made directly. All good French dictionaries (at least bilingual ones), and the French Wiktionnaire, print these forms in full. English dictionaries don't even print verb tables (or, if they do, they contain only irregular past forms). And, of course, as they say on Wikipedia, Wiki is not paper.

Any thoughts? This, that and the other (talk) 09:49, 9 December 2009 (UTC)

I really don't see much need to include all the compound forms, but if we're going to, then I really think the current approach is the worst possible way. Printing the actual forms would be a huge improvement. —RuakhTALK 03:33, 14 December 2009 (UTC)

Dots in IPA[edit]

For quite a while now, we've been using dots in our French IPA when copied from fr:, such as /lyk.se/ (luxer). Not that its of much importance, but is this a standard practice? on fr:WT:W they've been discussing getting a bot to run by and delete all of these as it's "nonstandard" - any opinions? I think it's of a very low priority. Mglovesfun (talk) 11:59, 19 December 2009 (UTC)

Those dots represent syllablic boundaries. Whilst rarely necessary, they are sometimes useful (as in the case of the Latin īdōlolatrēs), and are thus best retained.  (u):Raifʻhār (t):Doremítzwr﴿ 21:24, 19 December 2009 (UTC)
Deal. Mglovesfun (talk) 10:47, 20 April 2010 (UTC)

Invariability debate[edit]

Yes, that old chestnut. What is a French invariable noun? There are at least two possible meanings. cinq cannot have a plural, or at the very least it's proscribed, while nez can have a plural, but it's nez. Do we really need Template:fr-noun-inv? Probably yes we do, if we can agree how to use it, as invariable and uncountable are not the same thing. But it needs some work, too. Mglovesfun (talk) 14:47, 21 December 2009 (UTC)


Per the French Wiktionary, I've marked the date for Modern French entries at 1600 (see fr:Modèle:=frm= which was not created by me, I should add). Similarly WT:AEN says English words have to be attested post 1500. Mglovesfun (talk) 22:55, 10 January 2010 (UTC)

Wiktionary:About French/Todo[edit]

New "project" page for French, similar ones exist already for German and Spanish. Mglovesfun (talk) 13:25, 15 January 2010 (UTC)

Wiktionary:Tea room#Etienne[edit]

Input needed. Mglovesfun (talk) 10:59, 27 January 2010 (UTC)


There's a very interesting theory that -bt- in words like obtenir is pronounced /pt/. I don't think it is, otherwise the first syllables of obtenir and optique would be identical, and I don't think they are. In the audio file in obtenir, he seems to pronounce /bt/, but we have /pt/! Mglovesfun (talk) 13:03, 19 April 2010 (UTC)

Names of languages[edit]

Names of languages (français, anglais et al.) behave like proper nouns but are in all lowercase. Can we consider them as proper nouns, just like Spanish does? See español for an example. Mglovesfun (talk) 13:09, 19 April 2010 (UTC)

I don't know if they behave like proper nouns; you can't say *"j'aime anglais" or *"anglais se parle en angleterre", for example. —RuakhTALK 13:56, 19 April 2010 (UTC)
I guess the argument (whether I endorse it or not) is that russe and finnois aren't uncountable, they're just always singular. Mglovesfun (talk) 10:46, 20 April 2010 (UTC)
To bear in mind, you also can't say "j'aime France" --Rising Sun talk? contributions 10:48, 20 April 2010 (UTC)
You can't say "I love United States" in English either (formerly, Congo, Sudan and Ukraine always took the article). In French most proper noun of countries are arthrous instead of anarthrous (the primary exception IIRC being masculine singular islands: Madagascar, Saint-Pierre-et-Miquelon, Aruba, Cuba, Haïti), with the reverse being the case in English (Where the arthrous countries are plurals or collectives: United Kingdom, Comores, Soviet Union). Circeus 14:40, 21 April 2010 (UTC)

I don't agree to label any French language name as a proper noun because:

  1. w:Wikipedia:No_original_research: all French dictionary are already publishing them as some common nouns (1, 2 and 1 Quebec example).
  2. All proper nouns have a capital letter in French (unless maybe the Unix commands).

Moreover, I've never heard about "les russes de Moscou et Saint-Pétersbourg sont différents" but "le russe" or "les accents russes". Hence we would need some serious attestations to introduce this countable concept. JackPotte 11:03, 21 April 2010 (UTC)

Sounds quite like a consensus. Yes, dictionaries always listed them as common nouns (or just nouns, if you want to be picky about it). Mglovesfun (talk) 14:42, 21 April 2010 (UTC)
Yes, they are considered as common nouns in French. Months, too, and days, and programming languages, and card games, and demonyms (despite the capitalization). Generally speaking, the sense of proper noun is the same in English and in French, but its application is slightly different (due to different traditions). Lmaltier 20:59, 28 June 2010 (UTC)
Late response, however, I am a little confused by what you mean by they behave like proper nouns. One cannot say « J'aime français. », one is forced to say « J'aime le français. », equally: « Je hais l'anglais. » in place of « Je hais anglais. ». However, one can say « Je parle français. », here, the 'le' is facultative. --Île flottante 05:13, 1 September 2010 (UTC)
In conclusion, one has to know French quite fluently to know how to deal with these subtleties. --Actarus (Prince d'Euphor) 22:07, 1 September 2010 (UTC)

Reflexive head words[edit]

Looking at marier (this version), it stops half way and puts a second head word se marier (the reflexive form of marier). I don't like this, it's the sort of thing they're slowly phasing out on Wiktionary. My personal preference is to put {{reflexive|se marier}}. I don't think you can assume the reader knows that the reflexive form is with se (or s') so it's good to put it. Does anyone have a preferred format? Mglovesfun (talk) 11:00, 28 June 2010 (UTC)

How can we do "se marier" when "se promener" isn't with this template? I agree to use it but we might respect a certain order ;). JackPotte 11:28, 28 June 2010 (UTC)
What template? Mglovesfun (talk) 11:30, 28 June 2010 (UTC)
{{reflexive}} indeed. JackPotte 16:57, 28 June 2010 (UTC)
I suspect the reason is unwarranted deference to the original creator of the page, who inappropriate used a convention common in fr:wikt. Circeus 01:33, 1 July 2010 (UTC)