Where are French words stressed?

leolucas1980

Member
Brazilian Portuguese
I've lived in France and think that most French people don't even grasp what syllable stress is. It's an irrelevant concept in French, just as syllable tone in most European languages.
 
  • Wai Ho

    Senior Member
    Cantonese
    Bonjour! :)
    I have this lingering doubt (I hope it is not very stupid). All words of more than one syllable in all languages (or at least I suppose so?) have one syllable that is stressed more than the others. In English, a word's stressed syllable is shown beside the word in the phonetic pronunciation where a mark like an apostrophe appears before the stressed syllable. But in French I have this problem: I don't know which the stressed syllable is because the "apostrophe" doesn't appear in the phonetic pronunciation. I asked a friend who is learning French why this mark is not shown and he told me that it is because all words in French are stressed on the last syllable (?) Is he right? Then, I asked another friend of mine about this and she told me that that wasn't true and she gave as an example that "fenêtre" is not stressed on the last syllable. Do the diacritics indicate where the word is stressed like in Spanish?
    Thank you.
    Yes or no, he can be right. The apostrophe is not written in French IPA, because when we're reading a single word, for example, "télévision", we can change the stresses, we can put the stress on the beginning or at the end, also, we can pronounce all the syllables in mid tones. Otherwise, we need some stresses in a whole sentence, for exemple, "J'ai regardé la télévision", the "-dé" of "regardé" should be pronounced in a high tone, but the "-sion" of "télévision" can be pronounced with a half-high tone or a mid tone, not very high, because it's not a question. Remember, in the sentence, the important thing is the last syllable of words.
     

    berndf

    Moderator
    German (Germany)
    Yes or no, he can be right. The apostrophe is not written in French IPA, because when we're reading a single word, for example, "télévision", we can change the stresses, we can put the stress on the beginning or at the end, also, we can pronounce all the syllables in mid tones. Otherwise, we need some stresses in a whole sentence, for exemple, "J'ai regardé la télévision", the "-dé" of "regardé" should be pronounced in a high tone, but the "-sion" of "télévision" can be pronounced with a half-high tone or a mid tone, not very high, because it's not a question. Remember, in the sentence, the important thing is the last syllable of words.
    This discussion is about stress (that is dynamic stress) and not about tone. Those are different things and only loosely related.
     

    broglet

    Senior Member
    English - England
    I've lived in France and think that most French people don't even grasp what syllable stress is. It's an irrelevant concept in French, just as syllable tone in most European languages.
    I don't think it's irrelevant, it is just that syllable stress in French is not a property of each word per se but it can be just about anywhere (or nowhere) depending on the context. It is certainly the case that when confronted with a systematically accented foreign language such as English or Latin, most native French speakers seem to be completely at sea as far as accentuation is concerned. This is beautifully illustrated by Poulenc's bizarre accentuation in his wonderful Gloria. (His lau'damus 'te, benedici'mus te is particularly endearing :D )
     

    Penyafort

    Senior Member
    Catalan (Catalonia), Spanish (Spain)
    In Spanish post-tonic syllables were dropped too, ancla. There is no equivalent to encore/ancora in this language to make a comparison (in some old Spanish text I remember seeing encor but it could be dialectal),
    I'd say most if not all medieval samples you find of encor(a) or encar(a) are Aragonese or influence from it.
     

    Terio

    Senior Member
    Français (Québec)
    1. In French, the stress is always on the last syllable of isolated words : élé'phant, télépho'ner, générale'ment.
    2. If the word ends with a so called mute e, it does not form a syllable : é'tud(e), com'prendr(e)
    3. So, the sound [ə] is seldom stressed, but it can be, for example : dis-'le
    4. In a sentence, the only stressed words are the ones that come before a pause:
    Je mange une pomme chaque 'jour.
    Chaque 'jour, je mange une 'pomm(e).
    Chaque 'jour, si je 'peux, je mange une 'pomm(e).

    Even a pretty long sentence may be stressed on only one syllable if it is prononced with no pause :

    Je mange une pomme bien mûre et bien rouge chaque fois que je peux en cueillir une dans le pom'mier.

    In words inherited from latin, the stress syllable is generally the same as in other romance languages. For example, if we compare with spanish:

    (un) 'term(e) / ( un) 'término

    (je) ter'min(e) / (yo) ter'mino

    (il) termi'na / (él) termi'nó

    But it is not the case with later borrowings :

    gra'cil(e) / 'grácil


     

    berndf

    Moderator
    German (Germany)
    n words inherited from latin, the stress syllable is generally the same as in other romance languages.
    This is mostly true in French as well. The particularity of French is that syllables after the stressed one got lost or were deduced to a terminal Schwa and then mostly lost. This is probably why stress eventually become non-phonemic because it became so regular.
    (un) 'term(e) / ( un) 'término
    Latin derived words like terminer, which seem to defy this rule, are usually re-borrowings and not inherited words.
     
    Last edited:

    Wai Ho

    Senior Member
    Cantonese
    I don't think it's irrelevant, it is just that syllable stress in French is not a property of each word per se but it can be just about anywhere (or nowhere) depending on the context. It is certainly the case that when confronted with a systematically accented foreign language such as English or Latin, most native French speakers seem to be completely at sea as far as accentuation is concerned. This is beautifully illustrated by Poulenc's bizarre accentuation in his wonderful Gloria. (His lau'damus 'te, benedici'mus te is particularly endearing :D )
    Yes, it's relevant, because when there's a stress, it's higher than elsewhere.
     

    Awwal12

    Senior Member
    Russian
    I have to note that "phonemic" isn't equal to "relevant". You can speak Russian using the basic allophones of the phonemes only, and you surely will be understood all right, but it would be positively painful to listen to. In Chuvash stress is decided by the phonemic composition of the word and is therefore non-phonemic (basically it falls on the last strong vowel, or, if there is none, on the first weak vowel), but an incorrect stress placement in normal speech would be instantly noted (even though it may happen in poems and songs for the reasons of the rhythm), and so on.
     

    berndf

    Moderator
    German (Germany)
    Yes, you are right. It is not necessarily the same.

    But as we have seen from the discussion here, native speakers don't perceive syllable stress within a word at all but they do perceive prosodic stress, i.e. the perceive the opposition between Clara and Clara but not between Clara and Clara. The latter is indeed irrelevant.
     

    broglet

    Senior Member
    English - England
    We are talking about technical terms in linguistics (phonology to be precise) here. This is not about "English lessons".
    With all due respect berndf it is not in the usual courteous spirit of this forum for a native German speaker to suggest, even if it were true, that he knows more about the meaning of English words than a native English speaker.
     

    berndf

    Moderator
    German (Germany)
    With all due respect berndf it is not in the usual courteous spirit of this forum for a native German speaker to suggest, even if it were true, that he knows more about the meaning of English words than a native English speaker.
    Again, we are not talking about the meaning of a common English word but of a technical term. That is one of the reasons, by scientists usually prefer artificial terms because they avoid such misunderstandings. If common terms or re-used as scientific technical terms they invariably have a narrower meaning than in common language.

    Can we end this discussion by defining relevant as shorthand meaning phonologically relevant for the purpose of this discussion.
    I had originally equated relevant with phonemically relevant but @Awwal12 had a good point that this might be a bit too narrow, That's why I nor propose the understanding phonologically relevant.
     
    Last edited:

    Linnets

    Senior Member
    Here in Italy there's a strong perception of French stressed on the last syllable; this is particular evident with Italian names pronounced by the French: Platinì, Bianchì, and so on.
     

    berndf

    Moderator
    German (Germany)
    In Italian, the stress is low or high?
    Stress is neither high nor low. Tone is high or low. Those are different things.

    The stressed syllable of a word in pausa has a low tone. In non-pausa, the stressed syllable has no phonologically relevant tone.

    Swedish phonemically distinguishes high and low tone.

    In most European languages, tone distinguishes sentence types and has nothing to do with stress. Rising tone is a comma, strongly rising tone a question mark, falling tone a full stop and strongly falling tone an exclamation mark.

    EDIT: The kind of stress we are taking of here is dynamic syllable stress. This is the distinction between more or less forcefully pronounced syllables within a word that can affect the meaning of a word (not in French, though), like in English, where project and project are different words. The former is a noun and the latter is a verb.
     
    Last edited:

    Awwal12

    Senior Member
    Russian
    In most European languages, tone distinguishes sentence types and has nothing to do with stress.
    I can say that in Russian tone surely isn't defined by stress, and yet stressed syllables tend to be focal points of intonational patterns, that is, tone has something to do with stress after all. I suppose it's more or less the same in most European languages.
     
    Last edited:

    berndf

    Moderator
    German (Germany)
    ... in Russian ... yet stressed syllables tend to be focal points of intonational patterns... I suppose it's more or less the same in most European languages.
    Yes, tonal information usually coded in stressed syllables. In my language it is the same. The tonal difference between
    Das ist eine Dàmpflok!
    Ist das eine Dámpflok?
    is on Dampf-, the stressed syllable of Dampflok and not on the last syllable of the sentence, i.e on -lok.
     
    Last edited:

    DearPrudence

    Dépêche Mod (AL mod)
    IdF
    French (lower Normandy)
    They are sometimes stressed in England as well, and maybe even more so :p
    What about a rendezvous in a chateau?
     

    berndf

    Moderator
    German (Germany)
    According to Warnant, yes, long vowels do exist in French: teillage is pronounced [tɛjaːʒ(ǝ)]. As far as tempête is concerned, there's no long vowel: [tɑ̃pɛt(ǝ)].
    At some time in the development of modern French there was phonemically relevant lengthening of the vowel before a lost s and the ^ was more than pure orthographic convention. This manifested in minimal pairs like maître and mettre. I have asked around quite a bit but not found any French person who still maintained this distinction. But in other French speaking regions this distinction might still be alive.
     

    Terio

    Senior Member
    Français (Québec)
    The distinction is still alive in Canadian French : mettre ≠ maître ; prète ≠ prête ; faite ≠ fête ; trempette ≠ tempête. The long vowel even tends to diphtong to [aj], though it is considered a vicious prononciation to be corrected in formal speech.
     

    Swatters

    Senior Member
    French - Belgium, some Wallo-Picard
    The /ɛ/-/ɛː/ distinction is also alive and well in Belgium, with the distinction usually purely marked by length, except before a nasal where the long vowel gets nasalised (penne /pɛn/ [pɛn] vs. peine /pɛːn/ [pɛ̃ːn]). There's a stigmatised tendency to raise non-nasalised /ɛː/ to [eː] for some speakers (same for the other long mid-low vowels, so that tempête can surface as [tɒ̃ˑpeːt̪], gueule as [ɟøːl] and mode as [moːd̪]).

    To come back to stress, length can interact with it in Belgian French, so that the normal phrase-final declarative intoneme (that normally consists of a fast pitch rise in the last few ms of the pretonic syllable then a slow drop over the tonic syllable) can spread over the last two syllables when the penult is long or underlyingly long (= mid-high and nasal vowels), especially when the final is short (say, "aussi", "elle se présentera" or "un détenu"): the pitch rises over the whole long penult then drops over the whole last syllable. This can make the penult seem more prominent than the final syllable.
     

    Wai Ho

    Senior Member
    Cantonese
    The distinction is still alive in Canadian French : mettre ≠ maître ; prète ≠ prête ; faite ≠ fête ; trempette ≠ tempête. The long vowel even tends to diphtong to [aj], though it is considered a vicious prononciation to be corrected in formal speech.
    Fête = fight
     
    Top