NONHOMOPHONOUS HOMOGRAPHS.

Antoine Amarilli presents Non-homophonous homographs in French:

In this post, I present a list of French words which are spelled the same but pronounced differently. This is, in a sense, the converse of the much more frequent phenomenon of non-homographic homophones. [...] As it turns out, the phenomenon is surprisingly rare. Besides, it almost always occurs between words of different grammatical categories (except for “fils” and “plus“), which means that grammatical context should help to disambiguate.

There are three general categories—Indicative imperfect first person plural of a verb vs. plural of a noun (e.g., acceptions), Indicative present third person plural of a verb vs. adjective or noun (e.g., affluent), and Infinitive of a first group verb vs. nouns (often borrowed from English) (e.g., boxer)—plus “Miscellaneous cases” like as (a, as), “Clash between verb ‘avoir’ (‘to have’) indicative present second person singular, and noun ‘as’ (‘ace’, from Latin ‘as’).” I expect John Cowan already knows about it, but the rest of you may enjoy it (especially perhaps marie-lucie).

Comments

  1. Another case is that of liaisons which give information about the grammatical structure of a noun phrase: for instance, “des acheteurs de livres italiens” means “Italian book buyers”, ie. either “buyers of Italian books” or “Italian buyers of books”: when reading this, the presence or absence of liaison (respectively zitaljE~ or italjE~) will disambiguate to the first or second sense respectively, this being because liaison should not occur between a noun and an adjective unless they are immediate siblings in the grammatical tree.

    I had no idea. Does this actually correspond to how people talk?

  2. In fact I didn’t know most of it. I knew that in principle there were words in -ent that weren’t silent and so could collide with verbs in silent -ent, but I couldn’t have named a single one. About the -tions words, I knew nothing. The English borrowings in -er are obvious in hindsight, but again I didn’t realize there were any. I knew about maybe half the miscellaneous cases like est and fils, if that many.

  3. I meant I thought you’d know about the site, but I’m glad to have brought you something new!
    Does this actually correspond to how people talk?
    Yup. It comes natural if you’re French.

  4. marie-lucie says:

    des marchands de livres italiens
    Liaison between a noun and a following adjective is usually not done in casual speech. If I was presenting a formal lecture I might do the liaison for livres italiens ‘Italian books’, but in casual speech I would not. If there was ambiguity I might add, for instance, je veux dire qu’ils vendent des livres en italien or ‘I mean that they sell books written in Italian’ or …des livres faits en Italie ‘books made in Italy’ (like many art books, for instance), or … des libraires italiens ‘Italian booksellers’ (the latter without liaison either). The last one would be the most common meaning of des marchands de livres italiens.

  5. Some of these depend on the variety of French we’re talking about. Some people, for instance, say /by/ instead of /byt/; this is still usual in Quebec I think but not in France, malgré Littré, who said “cela ne vaut rien et est un effet de la tendance vicieuse que la prononciation a présentement à faire sonner les consonnes”.
    I’m guessing that the reinstatement of consonants in words like but and plus (which is indeed a “complicated mess”) is due precisely to the need to disambiguate homophones.

  6. m-l: Am I right to think that the liaison form of livres italiens is /livzi-/, and so the non-liaison form would be /livri-/?

  7. ” acceptions), ”
    The indicative imperfect first person plural of “accepter” is “acceptons”.
    I wonder if there is a verb “avier” – to fly an airplane.

  8. “Des acheteurs de livres italiens” is actually a poor example, because in the casual speech of a great many (majority of?) native speakers the “liaison” consonant /z/ attached to “italiens” is more of a plural-marking prefix than the underlying consonant of the preceding word, and in ordinary non-standard spoken French this /z/ could in principle be due to “italiens” agreeing with the plural noun “Des acheteurs” or with the plural noun “livres”.
    Thus, in “Des acheteurs de farine italiens” a /z/ will often be prefixed to “italiens” in colloquial speech, in total defiance of the rules of the standard.
    Stephen Bruce: /by/ is indeed the normal realization of “but” in Québec, but I think that in this instance (and others) Littré is correct, the realization /byt/ is simply a spelling pronunciation and did not arise in order to prevent homophony: yes, /by/ is identical to /by/ “bu”, the past participle of “boire”, but since the former is a noun and the latter a past participle I cannot imagine a context where /by/ versus /byt/ would cause any kind of ambiguity.
    Back to “Des acheteurs de livres italiens”: as Marie-Lucie said, only in careful educated usage would the presence or absence of /z/ disambiguate the sentence.
    It would be better to modify the sentence and make the first noun singular, i.e. “Un acheteur de livres italien(s)”, where the /z/, if present, could indeed only refer to “livres”, *in standard and non-standard French alike*, whereas its absence, *in standard registers*, would point to the buyer being italian, as opposed to his books. The problem here is that standard French spelling here will eliminate the ambiguity: “Un acheteur de livres italiens” versus “Un acheteur de livres italien”: in the former the books are italian, and in the latter the buyer is. Hmmm.
    Here’s a better example (not original with me, I hasten to add!): “Un marchand de tapis anglais”, where “tapis” and “anglais” are not inflected for number in written French (ending as they do in S already). In standard and non-standard speech alike a /z/ in front of “anglais” would indicate that the salesman is of unknown nationality and that it is the carpets that are English. In standard French absence of /z/ in front of “anglais” would point to the salesman being English, whereas in non-standard French this absence would indicate ambiguity as to whether the carpets or the salesman are English.
    That is what makes French liaison a difficult topic, actually (and very difficult to master for adult L2 learners, alas): it is a phonological process in standard French, turning into a morphological process in non-standard French, and quite variably present, making the tools of sociolingistics indispensable to its study.
    And if some of you think this is too complicated, well, good news! I recently learned that there exists a French dialect which wholly eliminated all forms of liaison: this dialect is only spoken by the older generation and is quite under-studied, so if you want to do linguistic fieldwork in France without having to deal with that fine mess which liaison is, there’s hope!

  9. Much too common a phenomenon in Danish – probably because of the stød. I often get tripped up if a sentence goes another way than I expected it to. That happens in English, too, of course, but in that case it’s more likely to be an issue of wrong stress on the sentence level.
    I know I came across a multisyllabic word with two meanings that I’d never thought about before the other day, but I’ve already forgotten what it was.

  10. marie-lucie says:

    Jim: The indicative imperfect first person plural of “accepter” is “acceptons”.
    “Acceptons” is the first person plural of the indicative PRESENT tense of the verb ACCEPTER ‘to accept’. IMPERFECT forms for that person/number have -ions in all verbs, hence (nous) acceptions, where the t has the /t/ sound as in the infinitive and other tenses. If the verb stem already ends in i, there are two i’s, for instance: ÉTUDIER, stem is ÉTUDI- , so adding -ions gives (nous) étudiions ‘(we) were studying’, as opposed to (nous) étudions ‘(we) study/are studying’.
    These forms also occur in the SUBJUNCTIVE PRESENT of the same verbs, as in il faut que nous acceptions les résultats des élections ‘We have to accept the results of the elections’.
    To my knowledge, there is no verb AVIER in French.

  11. marie-lucie says:

    JC: Am I right to think that the liaison form of livres italiens is /livzi-/, and so the non-liaison form would be /livri-/?
    The casual, non-liaison form is indeed /livri-/, but although /livzi-/ is a possibility for a liaison form, the omission of /r/, a characteristic of casual speech, together with the /z/ of the liaison, a characteristic of formal speech, create a stylistic disconnect. This could be done on purpose of course, and perhaps it is normal for some people. In formal speech, as in reading aloud, I would say /livr∂zi-/.

  12. I think of reading aloud as an odd case for formal pronunciation; when I read aloud to either adults or children (which I do quite a lot), I aim for a colloquial pronunciation. Indeed, when the author writes “does not”, I may very well say “doesn’t”, unless it’s clear that the text needs that formal usage.

  13. marie-lucie says:

    JC, I agree with you, and I do adjust my speech to the circumstances of reading. I had in mind reading a prepared text in front of a professional audience (not of children). But I prefer to talk to them, not to read unless I have to.

  14. Yeah, reading a paper is one of those things I would never do. When I attend a conference, I submit a paper (sometimes just an abstract, depending), show slides, and give an extemporaneous talk.

  15. Etienne – could you ever say “Un marchand anglais de tapis” if you wanted to be crystal clear about the nationality of the marchand?

  16. marie-lucie says:

    Vanya: could you ever say “Un marchand anglais de tapis”
    Personally, I would not say that. I might say un marchand de tapis, un Anglais or un marchand anglais, un vendeur de tapis. English allows a lot of modifiers before a noun, as in an English carpet dealer, an English Persian carpet dealer, etc. Translating such phrases into French causes poor translators to try to form similar noun phrases, and the result is anglicized French such as the one you suggest. In any case, in French the last word in a phrase is usually the most important, so that un marchand anglais de tapis would emphasize tapis, not anglais.

  17. Vanya: like Marie-Lucie I would avoid “Un marchand anglais de tapis”: it indeed sounds a little clumsy. In spoken French (assuming lack of liaison is not enough to indicate that the salesman is English) I think I would simply add enough of a pause between “anglais” and “tapis” to make it clear that the sentence is to be parsed “(Un marchand de tapis) anglais” and not “Un marchand de (tapis anglais)”.
    Marie-Lucie, John Cowan: I personally can’t stand people who read out loud from a prepared text at conferences: it strikes me as a disrespectful waste of the audience members’ time. It seems to me that the whole point of presenting something at a conference (rather than submitting it directly to a journal) is to get some feedback and explore some possibilities (if not during question period, then perhaps over lunch or during a coffee break). Thus, when you present something accompanied by a slide-show or hand-out it is vital that the audience “get” something meaty from your talk that is not present on the slides or the hand-out. The same thing goes in class: if you give your students in writing everything you are going to be saying you fully deserve the lack of attention which will inevitably follow. Here too, it is a matter of respect for the audience, be it made up of students or of fellow researchers: if you just read something out loud the message you are sending is: My mind is made up, here is what I think, accept it or ignore it, either way your thoughts, observations, ideas and/or feedback are irrelevant.

  18. English allows a lot of modifiers before a noun, as in an English carpet dealer, an English Persian carpet dealer, etc.
    Perhaps this is why one of the stereotypes of francophones in the conlang and auxlang communities is that in their languages they want to have a separate word for everything. There is, of course, the obvious retort that anglophones already have sixty separate words for everything, to go with all their varieties of religion, and are consequently incapable of memorizing a substantial vocabulary for any other language.

  19. I do embroider my talks using my slides as notes (and sometimes find myself explaining something at length that is presented on the next slide). But I don’t publish the slides in advance, only afterwards.
    Also, in most of my talks I make sure that the slides are headlines, that is, complete sentences, not mere NPs. The last talk I gave was exceptional, because I was using limericks as examples, so most of the slide was occupied by the current example.

  20. mollymooly says:

    The word “heteronym” means “non-homophonous homograph” but might just as easily mean “non-homographic homophone” (or indeed “non-homograhic non-homophone”). “Heterophone” (and “heterograph”) would be better, but seem to exist mainly on Wikipedia.

  21. marie-lucie says:

    I did not mean to give the impression that I read my conference presentations. There are however some circumstances where reading aloud is appropriate, such as when including verbatim quotes. Usually those episodes of reading are very short.

  22. M-L
    “”Acceptons” is the first person plural of the indicative PRESENT”
    Dammit, I knew that! Thanks.

  23. David Marjanović says:

    From the “language and culture” thread:

    Duden Herkünfstwörterbuch

    Herkunftswörterbuch.
    From the Ellis Island thread that turned to Ossetic names in Hungary:

    Iazyges

    Those don’t seem to have left any traces beyond Antiquity.

    in the casual speech of a great many (majority of?) native speakers the “liaison” consonant /z/ attached to “italiens” is more of a plural-marking prefix than the underlying consonant of the preceding word

    Hence Internet joke spellings like bonjour mes zamis.
    This thread:
    German has a few cases. Some involve phonemic stress (umgehen), others loans (Rock).

    I think of reading aloud as an odd case for formal pronunciation; when I read aloud to either adults or children (which I do quite a lot), I aim for a colloquial pronunciation. Indeed, when the author writes “does not”, I may very well say “doesn’t”, unless it’s clear that the text needs that formal usage.

    That would never have occurred to me. I guess this is diglossia speaking: being Standard German, written German is far removed from any colloquial pronunciation where I come from.

    I personally can’t stand people who read out loud from a prepared text at conferences

    Neither can I. Fortunately, PowerPoint has made this a very rare phenomenon!

  24. I think of reading aloud as an odd case for formal pronunciation; when I read aloud to either adults or children (which I do quite a lot), I aim for a colloquial pronunciation. Indeed, when the author writes “does not”, I may very well say “doesn’t”, unless it’s clear that the text needs that formal usage.
    Yes, I do the same thing.

  25. I’m the author of the linked post. Thanks for featuring me here!
    Etienne: Thanks for your reworked example using “anglais” and your remarks about this liaison not being common in colloquial French, I edited my post to reflect this. (I kept “meubles” rather than “tapis” because the fact that “tapis” doesn’t change between singular and plural is not crucial here, and I feel that “meubles” is a stronger incitation to liaison than “tapis” (I wouldn’t know how to justify this.))

  26. marie-lucie says:

    a3nm: I feel that “meubles” is a stronger incitation to liaison than “tapis” (I wouldn’t know how to justify this.)
    I agree with you, but I am not sure why either! “Meubles” shares a phonological form with with “livres” (with -CR(∂)/-CL(∂) final cluster), but tapis does not. In the plural tapis italiens/anglais, the liaison sounds more pedantic (or hypercorrective) than it does with meubles/livres and words with the same pattern. Perhaps the liaison sounds better after unstressed syllables (where the final vowel is schwa) than after stressed syllables (with “full” vowels)?

  27. There are a few of these in English. I can only think of two at the moment:
    1. “I resent your letter” with a Z-sound in the second word means something like “I am offended by the tone of your letter”, but with an S sound, it means “I forwarded your letter to a third party”. (Maybe I should have used e-mail rather than letter: much more likely to be forwarded these days.)
    2. A joke I ran across somewhere on the web: How do you tell the difference between a physicist and a plumber? Ask them to say ‘unionized’. The one who pronounces it ‘un-eye-on-ized’ is the physicist, the one who says ‘yoon-yun-ized’ is the plumber.
    Not quite the same thing, but I know five examples where a proper name and a common noun are spelled the same in English except for the capitalization and have different pronunciations. Here’s the quiz I made for middle-schoolers a few years ago – it works quite well as a collective effort with a large enough class:
    1. A book of the Bible and something we all need.
    2. A world capital and a type of legume.
    3. An ancient battle and something you would find in a delicatessen.
    4. A town in France and an English adjective.
    5. A nationality and a useful thing to have around the house.
    How long will it take for the Languagehat community to solve all of these? Less than two hours is my guess.

  28. Oops. Just noticed this thread is five days old. Perhaps more than two hours, if everyone has moved on to fresher ones.

  29. Michael: as long as the page is open, people will still check it and potentially comment. The mere lapse of time is nothing. Your #1 and #5 I got instantly (I was rather expecting #5). As for #2, I looked over a list of national capitals (including capitals of self-governing territories) without finding anything, but I don’t know what a “world capital” is: New York City is not the political capital of anything, but is sometimes called the world financial capital (jointly with London).
    I will add as #6 a German city and a group of invertebrates (or a famous event in that city and something you wouldn’t want to subsist on).

  30. Trond Engen says:

    I too took no. 1 and no. 5 immediately.. I know I know no. 4 but that doesn’t mean I remember it. Same with no. 3, but this I looked up in Wikipedia’s list of battles. I think I know no. 2 but there’s an issue with the spelling.

  31. Well, are you going to give the answers, so I can tell you whether you’re right? I know what #6 is: Worms, as in “Diet of”, which always makes American students giggle in History class. Thanks for the addition.
    As for #2, “world capital” is vague, and I meant “national capital”. Come to think of it, it’s still capitalized when used of a legume, and is an adjective specifying a particular kind of that class of legume, which must be named, so this example is kind of cheating. But it is pronounced differently when used of the legume. Does that help?
    I am glad to see that I’m not the only one who reads old posts. (Besides spammers, I mean.) It really does make a good ‘warm-up’ quiz for middle-schoolers or high-schoolers, especially (of course) in a Geography class.
    Here’s a similar exercise for an English class: Fill in the blanks in the following sentence with the correct word from this list: 1. grisly, 2. gristly, 3. grizzly, 4. grizzled.
    “The ________ veteran of the Park Police blanched when he saw the ________ scene left by the ________ bears: there was nothing left of the unfortunate tourists except ________ bits.”
    One of these days I’m going to make a whole set of such exercises and post them on the web with automatic grading. (I already have the software.)

  32. Sure. #1 is Job/job, #2 is Lima/lima, #3 still eludes me despite looking at a list of battles before 301 CE (the nearest I could get were the Battles of Salamis and Tunis), #4 seems faintly familiar but searching for it is hopeless, and #5 is Polish/polish.

  33. Trond Engen says:

    I’m pretty sure no. 3 is Salamis. It would be nice to beat John to no. 4.

  34. So far, so good. Yes #3 is Salamis/salamis. The latter isn’t found much in the plural, but I figure it’s like grass, fish, and cheese: if the emphasis is on the different kinds that exist, you can say “grasses of the Great Plains”, “fishes of the North Atlantic”, or “that deli has an amazing selection of cheeses and salamis”.
    Broad hint for #4: a fellow grad student years ago came back from France with a T-shirt naming the town she had visited. She found that she couldn’t wear it around Chicago because construction workers would point and laugh and repeat the English word. They had no idea it was a town in France, and supposed to be pronounced differently: they just thought she was boasting about the quality of (how to put this?) the contents of the shirt.

  35. #4 is nice.
    I must think differently than John and Trond, I got 2, 3, and 4 with no problem but was puzzled by 1 and 5 until John posted his answers.
    How is Salamis supposed to be pronounced? I’ve always pronounced it like the deli item, without really thinking that this was correct.
    I think in high school history class we just called the Diet of Worms ‘the diet of worms’. I guess it’s supposed to be ‘vorms’. What about Diet? ‘deet’?

  36. I’ve always heard the battle as SAHL-uh-Miss – pretty much the same vowels and rhythm as ‘colonist’ – the meat as suh-LAH-meez.

  37. I say SAL-uh-miss for the battle (first syllable rhyming with “pal”).

  38. marie-lucie says:

    I got all of them (including #6) except Nice:nice! I was thinking of towns, but Nice is a sizable city (over 300,000 people, and “greater Nice” has almost 1 million). I was also thinking of names ending in y, and I thought that the answer might be Sacy:sassy, but Sacy is tiny.

  39. Will: it’s “vohrmss”, more or less. There was a U.S. military base there until 1999, and long ago I heard a story about a soldier who was moving back to the States with his family. When they got there, however, the movers refused to unpack any of their boxes, as they were plainly marked WORMS and HANDLE WITH CARE!
    When I told my mother this joke, it took her some time to get it: she simply couldn’t see Worms as worms despite being bilingual. I suspect this might be the case for m-l as well.
    Salamis seems strange to me: I would have to say varieties of salami or salami sausages as the case may be.
    A short list of German-English interlingual homographs (as they are called): Gift, Billet, Gross, Stern, Mantel, Mist, Stock, Rat, Brief, Box, Regal and uncapitalized arm, dick.

  40. David Marjanović says:

    A short list of German-English interlingual homographs (as they are called): [...] Mist

    Ah, yeah. Some 10 or 15 years ago a whiskey called Irish Mist was sold over here. Big mistake.

    Rat

    The lengthening of vowels in monosyllabic words started in the southwest of the German-speaking area and never reached the north. It’s evidently still a bit foreign in Berlin: in the bus (not the subway, but the bus) the station Rathaus Steglitz (town hall of one of the districts of Berlin) is announced as if spelled Ratthaus, heavily implying that the politicians are all rats…
    (Conversely, the lengthening of open syllables started in the north and never reached the Swiss dialects.)

    Box

    That’s a loanword from English into German; and assuming a British value of the HOT vowel, it’s pronounced as similar to the original as possible.

  41. David Marjanović says:

    …Sorry, that’s not the issue. The issue is that it refers only to certain types of boxes.

  42. marie-lucie says:

    JC, are you saying that perhaps I could not see Nice as “nice”? I was not looking at a map, and I was thinking of names ending in y (there are many in France, but not so much in the South) which might look like English adjectives, so the right solution did not even occur to me.

  43. (My original comment seems to have vanished here.)

    David: I meant Box as in ‘boxing’, presumably from French boxe, which is ultimately from English.

    m-l: Yes, that’s what I meant.

Speak Your Mind

*