Talk:Isochrony

Need for clarification
I do not understand the part on Origins of the Distinction. If a language has short, CV, syllables, it can cram more per time interval and go under the threshold of 330 ms. This seems to make it harder to distinguish a rhythm based on syllables and should push these languages toward a stress-based rhythm. On the opposite, a language with complex, and longer syllables, should have the possibility to go for either a syllable or a stress-based rhythm. But the paragraph seems to say the opposite.

Also, there are no references in this part.

204.52.215.75 (talk) 18:46, 5 November 2008 (UTC)Roberto Zamparelli


 * I'm just guessing, but my expectation is that a language with just CV syllables will have a syllable-based rhythm because all syllables are approximately of equal length. However, in a language where syllables vary from CV to CCCVVCCCC, that will not work. kwami (talk) 19:30, 5 November 2008 (UTC)

One of you people wanna, I dunno, explain what the hell "CV" means in the main article rather than just dropping your insider jargon on us, the lowly masses? Ashwinr (talk) 18:37, 27 October 2011 (UTC)

Removed following babble from main page:

--24.22.151.182 00:36, 12 November 2005 (UTC)william arnold montegue24.22.151.182 00:36, 12 November 2005 (UTC)--

Disbomber 02:10, 12 November 2005 (UTC)

Unclear because of terms which are used without explaining them for the non-technically-familiar. What is "/N/"? It also took a second to realize that C and V were consonant and vowel. Can someone please rewrite this section to elucidate these more clearly? SteubenGlass (talk) 21:07, 1 August 2009 (UTC)

Removal of 'Changes in timing' section
I have removed the following, as it suggests that Spanish is normally stress-timed. It is, however, of interest, & I shall try to find another place for it: ==Changes in timing== These patterns can change over time or be borrowed from other languages. For example, Mexican Spanish, due to its phonetic Nahuatl substratum, shows a marked tendency towards stress timing, which makes it sound as if influenced by American English prosody. There are reports of Mexican people pronouncing "los Estados Unidos" as two "syllables", which actually means the speaker marks two beats or stress peaks (over /ta/ and /ni/), in the same way that e. g. an Argentine Spanish speaker would mark the two syllabic peaks in a word like "pompón". The pervasive vowel reduction and shortening found in English is in part a consequence of stress timing; Mexican Spanish under this influence shows signs of vowel shortening as well. - Removed by Rothorpe 22:28, 30 September 2007 (UTC)

Merge of subsidiary articles
I'm strongly in favor of the merge of Syllable-timed language, Mora-timed language, and Stress-timed language into this article. There are a lot of bigger problems (such as poor referencing and scholarly controversy), but I think those would be easier to address if everything is in one article instead of four. Furthermore, the whole concept of a stress timed language (and likewise for the other two) is controversial (per cites in Timing (linguistics)), and it would be easier to address this debate if we don't, by our division into articles, assume one answer or the other. Kingdon (talk) 17:00, 24 May 2008 (UTC)


 * I have completed this merge. The old talk pages are still at Talk:Stress-timed language and Talk:Syllable-timed language (I think leaving them there, with these links from this talk page, is proper procedure, although I don't do this often enough to be sure). There was no talk page at Talk:Mora-timed language. There remain some passages which lurch back and forth a bit awkwardly.  That is, one sentence/paragraph assumes the distinction between stress-timed and syllable-timed, and the next sentence/paragraph throws cold water on that distinction, and then maybe the next switches back, but I wasn't attempting a complete rewrite, just getting these articles merged without doing anything too horrible to the text. Kingdon (talk) 04:55, 1 June 2008 (UTC)

proposed move to Isochrony
I propose to move this page to Isochrony, which is more specific and actual normal usage to refer to the idea that languages split up time into regular intervals, which can be mora, syllable or word. The current use Timing(linguistics) is ambiguous and unspecific, it could refer to timing of turns for instance. Jasy jatere (talk) 09:31, 14 November 2008 (UTC)

Brazilian Portuguese a syllable-timed language?
AFAIK, Brazilian Portuguese is also a stess-timed language just as European Portuguese, albeit with much less intensity on the stressed syllable and less reduction of the non-stressed syllables. However, vowel reduction is extremely common in Brazilian Portuguese (reducing /o/ to /u/, /e/ to /i/ and in some positions /a/ to /3/), and it definitely sounds very different from Spanish in what refers to the length of the syllables when they're pronounced. The text cites no reference to that quite polemical statement, considering that Portuguese is generally considered a stress-timed language, no matter if it's the European or the Brazilian variant.201.9.164.207 (talk) 20:47, 22 May 2010 (UTC)

http://www.scielo.br/scielo.php?pid=S0102-44502000000200006&script=sci_arttext "Syllable-Timing in Brazilian Portuguese": Uma Crítica a Roy Major

(Tempo-silábico em Português do Brasil: a critic to Roy Major)

''ABSTRACT: This paper reintroduces the discussion about stress-timing in Brazilian Portuguese (BP). It begins by surveying some phonetic and phonological issues raised by the syllable- vs stress-timed dichotomy which culminated with the emergence of the p-center notion. Strict considerations of timing of V-V units and stress groups are taken into account to analyze the long term coupling of two basic oscillators (vowel and stress flow). This coupling allows a two-parameter characterization of language rhythms (coupling strength and speech rate) revealing that BP utterances present a high-degree of syllable-timing. A comparison with other languages, including European Portuguese, is also presented. The results analyzed indicate that Major's arguments for considering Brazilian Portuguese (sic) as stress-timing are misleading. '' -

''According to rhythm, languages are classified in syllable-timed and stress-timed. Japanese is probably the most perfect example of a syllable-timed language, but French and Brazilian Portuguese are also syllable-timed while Russian and English are markedly stress-timed languages.''

http://www.sk.com.br/sk-reduc.html —Preceding unsigned comment added by Linda Martens (talk • contribs) 20:47, 12 October 2010 (UTC)


 * According to your source, Plínio Barbosa does not say it is syllable-timed, he says it has mixed characteristics, criticizing Roy Major works who said it is stress-timed. Read what is written in your very source.

''Não por serem desprovidas de sentido, mas por serem aproximações de uma realidade bem mais complexa. Através da metodologia apresentada aqui concluiremos que o PB é de tipo misto: silábico e acentual.''


 * Barbosa not only said it has mixed-characteristics, but also had his work revised by another linguist, which the sourced you removed from the article O Troqueu Silábico no Sistema Fonológico (Um Adendo ao Artigo de Plínio Barbosa, there she presents many stress-timed characteristics such as vowel reduction; stressed syllables longer than unstressed ones; Haplology; degemination, syncope and elision.


 * And you also removed a source which says that the Isochrony changes according to the speech rate and to the different dialects inside Brazil.--Luizdl (talk) 04:44, 13 October 2010 (UTC)

Explanation/Clarification
Under the Stress timing section, it states that "At fast speech rates, Brazilian Portuguese is close to British English, while in slow speech rates is close to Peninsular Spanish." But what does this mean? Is Brazilian Portuguese stress-timed at fast speech rates and syllable-timed at slow speech rates? If so, wouldn't that be a clearer way to state this. If not, I don't know what this could mean, and it definitely needs rewording/clarifying. Wierdw123 (talk) 04:08, 17 December 2010 (UTC)


 * I changed this to reflect your comment. and further precision would be great. Metal.lunchbox (talk) 20:27, 6 July 2011 (UTC)

Latin and Ancient Greek - mora-timed?
Can't Latin and Ancient Greek be safely said to (have) be(en) mora-timed languages? They had long and short vowels, which retained their length regardless of their being stressed or not. Poetry in them was written in quantitative verse forms, which, when red aloud, violates all sense of emphasis the respective words would have in prose. Attemps to apply these metres literally to modern Romance or Germanic languages would fail miserably (therefore we use ancient meters emphatically today), which clearly establishes the difference between both stress-timed Germanic languages (in which all long vowels are stressed), and syllable-timed Romance languages (in which long vowels simply don't exist). Steinbach (talk) 16:03, 15 November 2011 (UTC)

Correlation between distinctive length and timing
The article says very little about any correlation between a language being not syllable-timed, and having contrastive length of either vowels or consonants. It would seem to me that a syllable-timed language could hardly remain a steady rhythm if it contrasts for example 'ta' with 'ta:'. The need for contrastive length would necessarily 'mess up' the timing. For this reason it seems doubtful that Finnish is syllable-timed; it has both long vowels and consonants, and to my ear the Finnish rhythm seems very irregular and 'jumpy' which probably means it's mora-timed rather than syllable-timed. And another point worth looking at is whether stress and/or timing seem to correlate with lengthening of vowels in open syllables. As far as I know most Romance languages do this, and all modern Germanic languages (historically stress-timed) have undergone such a lengthening in the past 1000 years. CodeCat (talk) 14:16, 1 February 2012 (UTC)

Origin of the differentiation section under a critical eye
I'm highly skeptical of this section. It's uncited, both in the psychology and the linguistics, and its reasoning is dubious.

First of all, we should fact-check the numbers it claims divide the perception of separate beats, since there are many folk facts and numbers in "pop psychology" that the author could be using. Also, it would be nice to know if this all-too-convenient claim that this threshold can be used to take the pronunciation times of consonant clusters and predict whether it will be syllable-timed or stress-timed. (Who even knows if those figures for pronunciation times of consonant clusters are real? Double articulation means the length of pronunciation doesn't always correspond to the length of the cluster...) It also presupposes that the default is syllable-timed and the reason for stress-timed is that the syllables are too irregular (and slow?). But do we have any empirical data of this typological trend?

But also, it just doesn't make much sense. Its argument is that languages with highly variable consonant cluster lengths will be stress-timed, while ones with syllables whose structure (and thus whose duration) is more regular will be syllable-timed. So what does perceiving distinct beats vs. a continuous sound have to do with it? Is it trying to claim that in syllable-timed languages, the beats come so rapidly that they are heard as one continuous beat? Where's the evidence of that? Moreover, do the beats in syllable-timed languages come faster than stress-timed languages? The article doesn't say anything about that; in fact, under "stress-timed languages", it suggests that in one language's slow speech it behaves as a syllable-timed language! Moreover, since the article mentions that research attempting to show that the syllable durations in syllable-timed languages has not produced compelling evidence. So where does this supposed regularity come from?

All in all, this smells strongly of folk explanation: it's uncited, the numbers are too convenient, it haphazardly incorporates data from other fields of science, it's uninformed in terms of knowledge of the field... Looking at the history, it was added by an unregistered user with few other edits (and none that lend credence in linguistics), and it was never really investigated.

I'm writing this for discussion, but I'm also fairly convinced about this. If there's no good reply when I check this again in a day or two, I will remove the section myself. Flipping Mackerel (talk) 23:37, 19 October 2012 (UTC)


 * And away it goes. Flipping Mackerel (talk) 17:05, 21 October 2012 (UTC)

Modern Greek
Standard Modern Greek is a SYLLABLE-timed language. There are many references to this in academic papers. It is not in dispute. If you want to remove it (Lfddr) find a reference! — Preceding unsigned comment added by Interista (talk • contribs) 18:32, 22 July 2013 (UTC)
 * You're right, there are papers citing Dauer and papers citing people that cited Dauer. Truly, many references. 'Typically quoted' might've worked, but you've now rephrased it to 'is'. No go. See Arvaniti 2007a p. 55 onwards. — Lfdder (talk) 19:09, 22 July 2013 (UTC)

Southern dialects of Italian claim
The source article for this statement needs to be checked by an Italianist or a competent typologist: Some southern dialects of Italian, a syllable-timed language, are effectively stress-timed.[16] What Italianists refer to as dialetti italiani 'Italian dialects' are in principle Italian only in a geopolitical sense, minor(ity) languages cognate to Italian and other Romance languages and not in any way construable as dialects of Italian (which are referred to as italiano regionale 'regional Italian'). — Preceding unsigned comment added by 75.134.23.211 (talk) 17:01, 28 August 2013 (UTC)

That is true; and in fact the statement refers almost certainly to a southern regional variety of Italian, the one that is spoken in central Puglia, from Bari up to the town of Matera. People who grow up there [such as myself] find it stressful, lifelong, to reproduce a normal italian flow of syllables. They tend to strongly compress the duration of all sequences of unstressed syllables, creating schwas and consonant clusters. The unevenness of their speech sounds almost foreign, and often funny, to other italians. Maulon69 (talk)
 * I'm doubting if Italian is a syllable-timed language at all. How can syllables in Italian words be equally long if there are double consonants?--2.246.34.124 (talk) 01:35, 5 March 2014 (UTC)

In italian, the vowels before double consonants are extremely short, as are all vowels in closed syllables, while the vowels in open syllables are long. So, for example, in the words “calde”, “cade”, and “cadde”, the syllables “cal-”, “ca-“, and “cad-“ are approximately of the same length, with the second being almost “caa-“. Such adaptation of vowel lengths is totally instinctive and is the reason why, for example, italians find it slightly difficult to pronounce the neo-latin word “càustico” [“caustic”], in which the first syllable is artificially too long (while “caùstico” would pose no problem). [The aforementioned three words mean: “warm“ (feminine singular), “he/she/it falls”, and “he/she/it falled”, respectively.] Maulon69 (talk)
 * As far as I know, vowels are allophonically lengthened in open syllables, so that there is no difference in length in the end. That is, there's either a long consonant with a short vowel, or a short consonant with a long vowel. This is a very common phenomenon across languages, and also occurs in for example Swedish. CodeCat (talk) 04:12, 5 March 2014 (UTC)

Another "syllable-timed language"
Someone has just added to the list of so-called syllable-timed languages on the grounds that "Mandarin is more syllable-timed than Cantonese". Such statements are without value unless there is some evidence to cite. It seems unfair to delete this right now, because so many languages are mentioned in this article without citations. It really needs a thorough overhaul to get rid of unsupported material. RoachPeter (talk) 06:37, 21 September 2013 (UTC)

Opening line
The statement on the first line, "Isochrony is one of three aspects of prosody, the others being intonation and stress", doesn't make sense. Intonation and stress are prosodic variables, whereas isochrony is a rhythmical characteristic of some languages (as the lines below make clear). It would make better sense to say "Rhythm is one of three aspects of prosody, the others being intonation and stress. One type of rhythm is isochronous rhythm, etc.". I don't know where the idea of there being only three aspects of prosody comes from. So much needs putting right in this article! RoachPeter (talk) 19:46, 5 March 2014 (UTC)

Clarify the difference between Duration vs Temporal Duration
It appears that the usage of the word 'temporal duration' is peculiar to this article. What is meant by temporal duration? Isn't duration temporal in general? Quick Google search only resulted in this article and websites which scrape the Wikipedia.

I think it would be a great addition to this article if someone could add sources/explanation or perhaps a link to the word 'temporal duration' in the article.

Thanks --Criticpanther (talk) 17:09, 28 April 2016 (UTC)
 * I agree, and have changed 'temporal duration' to 'interval' in the lead. Incidentally, 'temporal duration' is also used in the article on Time perception, and that should perhaps also be changed. I am also unhappy about the use of 'equal' in this exposition - we are really talking about durations which are relatively constant, not equal to something. RoachPeter (talk) 08:30, 29 April 2016 (UTC)

Vowels in Japanese
How common/official is the notion that “Japanese does not have long vowels or diphthongs but double vowels”? As far as I know, the combination of a (C)V and a V syllable with the same vowel is considered a “long vowel” although, strictly speaking, it is just two syllables which, as they are atomic from the perspective of Japanese grammar, cannot offer “long” vowels. However, from this perspective, one cannot speak of “double” vowels either, as the first syllable might be a CV one and, being atomic, there is nothing “doubled”, e.g. in とう (just two different syllables, but read tō). On the other hand, if you apply a phonetic perspective, the vowels are long, because they are pronounced this way. --77.187.177.187 (talk) 11:00, 2 October 2016 (UTC)

Welsh is syllable-timed?
I don't think so. My knowledge of Welsh is quite limited, but the source doesn't support the claim that it's "commonly quoted as [an example] of syllable-timed languages". The authors' conclusion is that "Welsh does not show syllable-timing and may be foot-timed in the sense of the quantitative feet defined here. [...] The timing structure described in this contribution differs both from syllable-timing and foot-timing in the conventional isochrony-based sense." (emphasis mine) There's nothing in there about others calling it syllable-timed, either. If Welsh is going to stay in the list, you need a better source. I haven't checked the others, only this one because I was under the impression it was stress-timed. ♫ ekips39 (talk) ❀ 06:34, 15 November 2018 (UTC)

I've fixed this to say that Welsh is not in fact syllable-timed, as per the source. Feel free to improve it further. ♫ ekips39 (talk) ❀ 01:15, 25 November 2018 (UTC)

Isochrony = prosodic typology ?
Hi, I was reading the French page Typologie rythmique des langues and what is said there is exactly the same subject as this page. Then, I tried to link both pages together. But I failed because they are already linked to different items: → English with wikidata:Q968561 : isochrony → French with wikidata:Q33218678 : prosodic typology

So, 2 possibilities : → isochrony = prosodic typology : we have to merge Q968561 with Q33218678 → isochrony =/= prosodic typology : we have to rename de French page and link it to this English page. What do you think?

PS: feel free to correct my grammar or my orthograph

Lethario (talk) 16:58, 2 October 2019 (UTC)
 * They are not the same. The study of isochrony can be regarded as a subset of prosodic typology, along with whether a language has contrastive stress/pitch/tones (though I don't know what this is called), but they are neither equivalent nor coordinate. In fact the French article is titled typologie rythmique, not typologie prosodique, and discusses isochrony only. So I think Q33218678 should be merged with Q968561, on the condition that the English label "prosodic typology" be removed. Nardog (talk) 17:56, 2 October 2019 (UTC)
 * Done. Nardog (talk) 18:00, 2 October 2019 (UTC)
 * I tried to link the French article but this has been rejected by Wikidata admin Lethario (see also wikidata:User talk:Andreasmperu). As far as I can see typologie rythmique is the same as isochrony (I guess one could say the former is the study of the latter, but that's like saying physics is the study of physics), and they should be merged (it is simply user-hostile to not link the French article and en:Isochrony, which bear striking resemblance), but you might have to convince the admin. Nardog (talk) 00:49, 5 October 2019 (UTC)

Can syllable-timed languages have long vowels?
Long vowels are commonplace in mora-timed languages (i.e. Japanese, Finnish) and stress-timed languages (i.e. most Germanic and Slavic languages) but it generally is not in syllable-timed languages I have been hearing that some such languages like Korean have long vowels. Wouldn't that make it mora-timed?Razzamatazz Buckshank (talk) 22:50, 16 March 2020 (UTC)

English narration audio file is inaccurate
The audio file which features an English speaker narrating the “Stress-timing” section of this article is no longer accurate to some of the text in the end of the section. This section seems to have been last edited in April 2020. Perhaps a new recording can be made to reflect the current text? JGard4159 (talk) 13:33, 2 September 2021 (UTC)

Ambuguous expression?
The article says 'More recent research claims that the duration of consonantal and vocalic intervals is responsible for syllable-timed perception' What does this mean? — Preceding unsigned comment added by 130.43.66.230 (talk) 22:41, 4 September 2021 (UTC)
 * Frankly, I have no idea what is meant. It reminds me of the time someone on the radio stated confidently that weather was caused by the climate. LynwoodF (talk) 22:32, 6 September 2021 (UTC)
 * I couldn't find an explanation on what it means exactly from skimming the abstract of the dissertation it cites. From how I read this line it's pretty much just saying that the phonetic duration of sounds determined whether a language is perceived as syllable-timed or stress-timed. This doesn't seem like a useful statement, it almost seems like a tautology. If we have three people here, none of which can figure out what it means, I vote to just remove it.--Megaman en m (talk) 22:40, 6 September 2021 (UTC)
 * The dissertation is clearly a valuable contribution to the study of a very contentious subject, but I cannot see that the statement in the article sums up what it says in any useful way. So I am inclined to agree with you. LynwoodF (talk) 09:59, 7 September 2021 (UTC)

Māori is mora-timed
I believe that Māori is (predominantly) mora-timed, with an admix of syllable timing, but would like an opinion from an expert before adding it to the list on the page. In particular, does this generalise up the phylogenetic tree: Tahitic, Central Eastern Polynesian, Eastern Polynesian, Polynesian. If so it would be good to say so. Nick Mulgan (talk) 05:18, 9 December 2022 (UTC)

Catalan is stress timed?
That's what this article says, but that's not what I've read. In papers where a strictly binary classification is used (syllable vs. stress timing), most articles place it on the first camp. If a spectrum is used, it's still classified as more syllable than stress-timed.


 * This article calls it syllable-timed: https://revistes.urv.cat/index.php/rile/article/view/13/15
 * This article starts with the hypothesis that it's an intermediate language, but ends up classifying as syllable-timed: http://prosodia.upf.edu/home/arxiu/publicacions/prieto/prieto-et-al_phonotacgic-phrasal-properties-speech-rythm-ca-sp-en.pdf

My impression as a native speaker is that, if English rhythm is a 1 and Spanish is a 10 in the spectrum, Catalan would be a 7 or 8. Only very specific subdialects might be classified as actually stress-timed, such as rural inland Majorcan.

I also suspect stress timing was much more widespread 400 to 1200 years ago (when all the syncope, apocope and reduction happened), but I have no articles to back it.

Another hypothesis I have is that much of Eastern Catalan's so-called vowel reduction (which appeared in the Late Middle Ages, when the language was supposedly stress-timed) has been reanalyzed as apophony (since the modern Eastern colloquial register allows for some instances of unreduced /e/ and /o/ as in class e  and seny o ). This would endanger an important part of the stress timing hypothesis. The only active reduction phenomenone nowadays would be unstressed /a/>[ə]. 139.47.33.66 (talk) 19:21, 5 June 2023 (UTC)


 * I thought this matter was put to bed years ago. I recall finding evidence that, although there is unstressed vowel reduction in Eastern Catalan, there is no clear evidence of stress timing. However, I can find no evidence of this article being amended. I am old now and maybe I just failed to do anything about it and then forgot about it. I suggest you make whatever changes that you, as a native speaker, think appropriate. LynwoodF (talk) 11:43, 7 June 2023 (UTC)


 * Incidentally, it strikes me that you are working in the field. I studied under Paul Russell-Gebbett at Nottingham 60+ years ago. I understand he is still remembered in Catalan linguistic circles. LynwoodF (talk) 11:52, 7 June 2023 (UTC)