8

We are gearing up for the new semester at the Thai university where I teach English. One course I’ll be helping out with is on English pronunciation. In the unit on sentence stress, the course textbook introduces the idea of stress-timed vs. syllable-timed languages: in the former type, sentence rhythm is determined by units each based around one stressed syllable, while in the latter, each syllable receives an equal beat in ‘machine-gun’ fashion. In stress-timed languages, vowels in unstressed syllables are often reduced (as with the schwa in ‘the’, English being stress-timed.) Presumably vowel reduction happens less if at all in syllable-timed languages. The contrast in sentence rhythm is illustrated by two pictures: a row of people of different heights walking in groups of two or three for stress-timed, and a row of soldiers of equal height marching in formation for syllable-timed.

The book presents English as a stress-timed language, and students learn to differentiate content words (which receive stress) from function words (which are generally unstressed.) It mentions in passing that Thai, by contrast, has syllable-timed rhythm. Is this correct?

The Wikipedia entry on ‘isochrony’ lists Thai with the stress-timed languages. It cites a study by Dauer (1983). However, the findings of this study (I don’t have access to the full text) seem to be that there is no empirical phonetic basis for grouping languages according to syllable/stress/mora-based rhythm in this way.

Two 2008 entries on the blog ‘Language Log’ by Mark Liberman also question the reality of stress vs syllable rhythm. Liberman measures the speech rhythms of a Spanish/English bilingual and finds that while his Spanish speech is more rhythmically regular than his English speech, this is due mainly to Spanish having fewer consonant clusters and thus more uniformity in duration of articulating each syllable. In Liberman’s view there are other motivations for a more uniform rhythm, it’s not a general feature for any given language.

Consulting the bibliography of Thai linguistics in the Routledge Tai-Kadai volume doesn’t turn up anything that directly addresses the question of whether Thai is stress- or syllable-timed.

I’m an English native speaker who speaks both Thai and Mandarin pretty well. Mandarin is supposed syllable-timed. If we look at the rhythm at the level of individual sentences, it’s not hard to find an example that contradicts the general characterization. The sentence ‘His car broke down’ can translate as follows into each language:

Mandarin: 他的車子壞了. Tāde chēzi huài le.

The rhythm here seems to be trochaic (stressed-unstressed).

Thai: รถยนต์ของเขาเสีย Rót yōn khŏng kháo sĭa.

Each syllable receives more or less equal stress. (I’m using ĭ and ŏ to represent rising tones here.)

So in this small example, Mandarin seems stress-timed, Thai syllable-timed, not the other way around.

I have two questions here.

  1. Is there a scholarly consensus about the status of Thai – is it stress-timed, syllable-timed, or something else? What should we tell our students, who are native speakers of Thai, in this regard?

  2. Could I try to measure this myself using Praat or some other easily accessible tool? How would I go about doing this? (I am not a linguist, and although I’ve downloaded Praat, I have never used it for anything serious.)

  • For a language to be stress-timed, it is to have at least some syllables unstressed, but in Thai writing system it is impossible to write a syllable without marking its tone, which can be a good argument for Thai to be syllable-timed. – Yellow Sky Dec 27 '13 at 11:10
  • @YellowSky: I don't think you can infer so much about a (spoken) language from its writing system. The Thai writing system isn't even phonetic. – hippietrail Dec 27 '13 at 15:20
  • @hippietrail - Why isn't it phonetic? Every Thai sound and tone has its own rendering in Thai script. I agree, it's complicated and often uses historical spelling, but the tone is consistently shown in every syllable. – Yellow Sky Dec 27 '13 at 20:39
  • 2
    Thanks for the responses guys! I agree that we probably shouldn’t make inferences about timing based on the writing system alone. – neubau Dec 28 '13 at 1:03
  • Re Thai words that might have unstressed syllables, there are plenty of candidates – my sample sentence might be misleading in that way. For instance polysyllabic borrowings from Pali/Sanskrit and ‘sesquisyllabic’ words with a CəCV(C) shape. The latter definitely have a unstressed-stressed pattern, but for the former I’m not so sure. – neubau Dec 28 '13 at 1:03
8

I cannot answer this as a linguist. However, being a native Thai speaker, and being someone who educated himself about linguistics much enough to have some idea on stress-timed vs. syllable-timed (acoustic phonetics is one of my particular interests), I believe that I can provide some good information on Thai's prosody (specifically stress and rhythm).

To summarize, in my opinion, Thai language is neither syllable-timed nor stressed-timed. A good metaphor for understanding this is a "rounded rectangle". A rounded rectangle is neither a rectangle nor a circle. Having said that, I do believe that Thai language is more syllable-timed than stressed-timed.


As someone has already pointed out in the comments above, Thais pronounce every word in syllables, because that is what we were taught since the primary school (equivalent to Grade 1 in the US). This is also strengthened by our poems. There are many sub-types of classical Thai poems. Each and every one of these sub-types will have a set of specific rules for the rhymes and number of syllables in each verse. This can explain why no one would label someone who speak Thai in a very syllable-oriented manner as incorrect. He will be perfectly understandable; he just might sound a little too stiff and might sound unnatural. But that's all. This is quite similar to when a native English speaker speaks in monotone. No one would accuse him for speaking sub-standard English; he just sounds uninteresting.

However, Thai natural speech (informal speech) can sound quite close to a stress-timed language, to the point that it might be mistaken as a true stress-timed language. It is also true that if you observe native Thai speakers carefully in their natural speech, they will group words into two or three or four or five or even more syllables, and each group will have one or more quasi-stressed syllables. (I use the term quasi-stressed because those stressed syllables are generally not very dominant.) And, having all that said is still a very simplification at best. There are no exact rules for stressing in Thai. However, without any explicit rules, Thai people can nonetheless speak approximately the same way somehow.

One way to understand Thai's stressing system is to compare it to English's intonation system. It can vary very widely from one person to another, or one region to another. And, except for those obvious rules (such as, in English's intonation: end questions with rising tone, or end declarative sentences with low tone--which are not particularly true in real English speech), you can confidently say that there is no definite rule at all.


Another interesting point about Thai prosody is that it keeps changing. A highly respectful person once said that he had noticed that Thai people today speak differently from Thai people yesterday. I researched into old video clips, and I've reached the same conclusion. One of my observation is that the more you go back in time, the more obvious the syllable-timed aspect of Thai language is. (For example, you can try watching this old commercial Thai ad: ถ่านไฟฉายตรากบ.)

It is my belief that Thai become lesser syllable-timed (in other words, more stress-timed) through our exposure to Western cultures, especially through education, technologies, music, and movies. And through my own observation, most younger Thais could sound quite stress-timed like in their natural speech without being aware of it themselves.


To provide a partial answer to your first question ("What should we tell our native Thai speaking students"), I believe that you should treat Thai as a syllable-timed language, because that will align with their prior education better. If your students have minimal exposure to English prosody, the concept of stress-timing will be invaluable for them and will help them to be on the right track.

As for your second question (about using Praat to verify this phenomenon yourself), I personally found that another software, WaveSurfer, which is a small program and has a straightforward user-interface, is a perfect fit for my personal needs. You can find WaveSurfer at TMH KTH or at SourceForge.

| improve this answer | |
  • Great to have a thoughtful native speaker’s answer, thank you Khun Damkerng. ‘Quasi-stressed syllables’ is a nice way to put it, and I (as a non-native) would have the same intuition. And I agree that it’s a simplification, as I mention it seems that quite a few linguists have debunked the tendency to place languages in one or another timing category. – neubau Dec 30 '13 at 4:32
  • Re your comments on the timing of modern informal speech becoming more stress-based – my line of thought was that the village people (ชาวบ้าน) in the old days were largely illiterate and didn’t use much formal Indic-derived vocabulary, and thus their speech would naturally be syllable-timed (consist mainly of native monosyllables). Now everyone knows and can use that formal language, so there is less of a diglossic situation and timing is shifting to something more stress-based. – neubau Dec 30 '13 at 4:33
  • Re the ad – very interesting! But since it (the man speaking at the end, not the song) is a sales pitch, I’m not sure if we can take it as representing the everyday speech of the time. For me it brought to mind the chant of a street hawker. But I’ll listen again more carefully. – neubau Dec 30 '13 at 4:34
  • 1
    About how to present this to students, I agree that letting sleeping dogs lie is probably best. – neubau Dec 30 '13 at 4:34
  • 1
    ... One of the impact is a noticeable number of Thais in this generation will shift their [ช] toward [sh], [ท] toward [t], [ส] toward [s] (actually it's more complicated than this). Before this shifting, we've already have a problem with our [ร] and [ล]. Nowadays, you can find many who speak ครูครับผมเครียด gingerly without no [ร] sound at all! Elders usually sigh privately, knowing that they can't do anything much. Moreover than that, [ร] starts to pop up in places it's not supposed to, especially those offensive words such as, มรึง, กรู, ตรู, etc. I found the shifting uncomfortably strange. – Damkerng T. Dec 30 '13 at 5:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.