I was always taught that a word contains as many syllables as it has vowels. By definition, a vowel is a sound that produces a syllable.

On the other hand, in English phonology, by definition, diphthong is two adjacent vowels which produce one syllable. I did not hear about diphthongs until I read some literature about English phonology.

I am aware about long vowels (which produce one syllable) and vowel hiatus (two consecutive vowels) which produce two syllables. So what is the proof that English diphthongs are actually two vowels which produce one syllable, as opposed to long vowels, hiatus, or a vowel followed by a consonant?

  • 8
    I am afraid you are mistaken. en.wikipedia.org/wiki/Syllabic_consonant. Pay no attention to the spelling of words, especially in English. Sounds fall into sets of varying sonority. From least to greatest: obstruents, liquids, nasals, vowels. Languages differ in where they set the threshold for what can be syllabic. In English, as my examples show, we allow everything but obstruents. The language Imdlawn Tashlhiyt Berber even allows obstruents to be syllabic.
    – Moss
    Aug 17, 2013 at 17:10
  • 3
    This is standard phonology. What teaching have you had (in what field)? "Even" is definitely two syllables (why is it spelled with two vowels?)
    – Moss
    Aug 19, 2013 at 18:42
  • 2
    I can only repeat (again!) what @Moss has said: this is standard phonology. Aug 21, 2013 at 2:08
  • 5
    If you're suggesting that it's only in linguistics in my country that this view is held, you're wrong. Any standard introductory text on phonology that discusses syllable structure will include discussion of syllabic consonants, the sonority hierarchy, etc Aug 22, 2013 at 4:27
  • 3
    What are the books you're reading that are telling you these things? Can you tell me of a textbook in phonology that says syllables can never have a consonant as nucleus? Try this online textbook on introductory English phonology--see p 105 re syllabic consonants. Aug 23, 2013 at 0:11

8 Answers 8


In thinking about this it's important to distinguish between phonetics and phonology.

Phonetically a diphthong is a sequence of two vowel targets, wherein the tongue starts at one vowel position and moves to another. For this reason it may sometimes be described as a combination of vowel and glide, but it's best to understand the articulatory facts of what it involves, movement from one tongue 'vowel' position to another, as opposed to the steady-state articulation found in simple 'monopthongal' vowels.

The evidence for this can come from simply hearing the vowel movement, as well as from various kinds of imaging such as spectrograms, ultrasounds, x-rays, etc.

Phonologically such an articulation can function as a single phoneme, as in the diphthongs of English. But a sequence of two phonemes consisting of vowel + vowel, or vowel + glide, can (eg in fast speech) be phonetically identical to a diphthong.

As for English diphthongs, a standard phonemic analysis of English will reveal a set of vowel phonemes. In every variety of English that I'm aware of some of those vowel phonemes will be realised as diphthongs, involving an articulatory movement from one vowel target to another, as revealed by the kinds of evidence already mentioned. Some of the evidence of the phonemic status of these diphthongs in English includes:

  • standard minimal sets: 'beet, bit, bet, bait, bite, boat, bout, bought, boot' etc
  • Diphthong as sole syllabic nucleus, eg 'how', 'boat', 'I'
  • Participation in same tense-lax contrasts as monopthongal vowels, eg 'divine-divinity', cf. 'serene-serenity'
  • What is the proof for your claim?
    – Anixx
    Aug 15, 2013 at 9:06
  • 1
    The proof is, as I say in my answer, through studies via spectrograms, ultrasounds, xrays, etc, which demonstrate that there are phonemes whose articulation involves movement between two vowel targets. The proof that these articulations function as single phonemes in some languages is specific to each language. Aug 16, 2013 at 3:27
  • what is proof that they function in English language as single phonemes?
    – Anixx
    Aug 21, 2013 at 8:36
  • 1
    @Anixx if you're talking about phonetics, a 'vowel plus glide' is just another description for a movement from one vowel target to another. If you're talking about phonology then xrays are irrelevant. Aug 23, 2013 at 0:16
  • 2
    It depends: are you talking about phonetics or phonology? Aug 23, 2013 at 5:53

Defining a diphthong as a sequence of two vowels may not be correct. A diphthong is a sequence comprising a vowel and a glide, at least, according to the Summer Institute of Linguistics: http://www-01.sil.org/linguistics/GlossaryOflinguisticTerms/WhatIsADiphthong.htm

A glide, or semivowel, is defined at https://en.wikipedia.org/wiki/Glide_%28linguistics%29 as “a sound, such as English /w/ or /j/, that is phonetically similar to a vowel sound but functions as the syllable boundary rather than as the nucleus of a syllable.” In other words, a sound that is like a vowel, but non-syllabic.

When the glide comes first in the diphthong, we have words such as British “new” [nju], in which [j] as in "yes" [jɛs] is the glide.

When the glide comes last in the diphthong, we have words like English “aisle” [aɪ̯l], in which [ɪ̯] is the glide.

Note the little curve symbol under the <ɪ>, yielding <ɪ̯>. That little curve, called an inverted breve, means that the vowel sound is non-syllabic, i.e. that it’s a glide.

Incidentally, vowels aren’t the only sounds that can be syllable nuclei. Consider syllabic consonants, such as the [̯̯̯̯l̩] in “bottle” [bɑdl̩]. Also see the examples mentioned with this definition of syllabic consonants: http://www-01.sil.org/linguistics/GlossaryOfLinguisticTerms/WhatIsASyllabicConsonant.htm

  • Sorry, but who said that y and w in English are semivowels if everywhere else they are consonants?
    – Anixx
    Aug 15, 2013 at 9:04
  • 1
    Right, especially in Welsh and Swedish w and y respectively are consonants :-)))
    – Manjusri
    Aug 15, 2013 at 9:41
  • Neither semivowels nor consonants can be syllable nuclei, but I'd be interested to know more about the terminological issues surrounding the term "glide" or "semivowel." I still need to figure out how to ask the question, though. Aug 16, 2013 at 2:42
  • 'Semivowels' are usually defined as non-syllabic segments which have a vocalic articulation, so it's part of their definition that they are not syllabic nuclei. And doesn't your comment '...nor consonants can be syllabic nuclei' contradict your answer where you say 'Consider syllabic consonants...'? Aug 16, 2013 at 3:32
  • 1
    @Anixx No I did not say that. I think you're confusing phonetics and phonology. These two levels have to be kept separate, which is what I did in my answer. Aug 23, 2013 at 0:25

In fact, loosely speaking, there is just one continuous vibration or wave in one syllable with one, two or more vowels. Those syllables with more than one vowel are just due to the writing system, and it just designates the location of tongue when the tongue moves, not saying that two vowels are put together to form part of one syllable or one syllable.

You may record the sound of several syllables with more than one vowel by computer and take a look at a diagram of its spectrum or wave with Praat, they are all continuous.

But if you do the same experiment on several syllables when you speak continuously, you may almost always find there is a break or almost weakest energy between diagrams of wave or spectrum of the syllables or two vowels that sound as two syllables.

  • 1
    I found a handy PPT that introduces the concept of reading spectrograms, talks about the characteristics of diphthongs in spectograms, and gives a few examples.
    – acattle
    Aug 16, 2013 at 2:46
  • @acattle,Excellent,but to tell you the truth,I have not read any such introductory material,although I think there must be some research done on such a question.Thank you for your URL Aug 16, 2013 at 3:39

You asked if English diphthongs could be analyzed instead as long vowels, but that doesn't make sense. Long vowels don't change in quality (phonologically) over their duration. But you are asking about sounds such as "hey", "no", "cow", "toy", which clearly do change. Actually, I realize you probably have a confusion of definitions between linguistic long vowels and "long vowels" as taught in elementary school.

Could it be vowel hiatus? Well that would mean that each vowel is treated as belonging in a separate syllable. But in the example words I gave above, we English speakers clearly treat those as monosyllabic.

So the real tricky question is whether these sequences are true VV diphthongs or actually VC and CV glides. The tricky thing is that all the glides in my dialect of English (Western Canadian) end on either a [j] or [w] sound, which, when I say it that way would mean these aren't really diphthongs. But [j] can also be interpreted as [i] and [w] as [u].

To me it is instinctive that they are phonological diphthongs, but that is probably not a good enough answer. But ask yourself, do you consider the "y" in "fly" to be a legitimate vowel sound in English or merely an [a] sound followed by a [j]? It is curious that 4 out of the 5 names of our vowels in english are actually pronounced as diphthongs. As Manjusri hinted at, an understanding of moras would be helpful. Briefly, a diphthong in any language must have two moras (same with long vowels), however a glide belongs in either the onset (no mora) of a syllable or in its coda. Then things get complicated. Depending on the language, and depending on sound, the coda may or may not carry a mora. Unfortunately I think English is ambiguous about this. If it could be shown that the purported [j] or [w] at the end of a syllable had no mora then that would be proof that it is really a glide sitting in the coda rather than the second half of a diphthong, sitting in the nucleus. I don't think you can find such proof though.

  • Actually, in Dutch long and short vowels do change in quality over their duration.
    – Manjusri
    Aug 16, 2013 at 21:14
  • Maybe they do incidentally (phonetically). If it is a noticeable change to a non-linguist native speaker then I would say it is a diphthong, not a long vowel, linguistically speaking, regardless of how Dutch teachers describe it. (See the link in my answer.) I don't know the phenomena you are referring to in Dutch though.
    – Moss
    Aug 16, 2013 at 21:17
  • Actually I would say "long vowel" isn't that well-defined, at least not outside linguistics. For instance many of the long vowels in languages such as Finnish and Hungarian are felt by native speakers to be "pure" but turn out to phonologically have more than one target, which makes them diphthongs. Even in English people feel that the "a" in "hate" is a long vowel for instance. English speakers have trouble making pure vowels when learning new languages as adults and are likely not to even know the words "phoneme" or "diphthong". Aug 17, 2013 at 8:52
  • 1
    I'm sure there is all sorts of confusion about what a long vowel is outside of linguistics. I don't think you are correct however about what you are saying about Finnish. I am pretty familiar with its phonology. "Each short monophthong has a long counterpart with no real difference in acoustic quality." en.wikipedia.org/wiki/Finnish_phonology But it also claims that long vowels are percieved as two separate vowels which would make them a case of vowel hiatus.
    – Moss
    Aug 17, 2013 at 17:18
  • 1
    It looks like I misremembered about long vowels in Hungarian and Finnish. All I can find now is that in Hungarian the quality changes as well as the length but long vowels do not diphthongize. Aug 18, 2013 at 4:27

The other answers have raised many important aspects why diphthongs are a useful concept, and I'd like to add some points that I personally feel are crucial in this debate.


I'm not sure insisting on a proof of the existence of diphthongs will help you understand why many use the concept diphthong to analyse language. There is no proof that they exist because they don't. They are an abstraction (just like the phoneme), a way of analysing or describing language and the same phenomenon can be described in many different ways.

There isn't any proof of gravity either - if an apple falls down, and you insist it's because of gravity I could still say it's because it is the nature of apples to fall down to Earth or that God likes apples to fall down etc. and we'd be no wiser. However, you could say that your theory of gravity explains a whole lot of other things at the same time, such as pears falling down and celestial bodies rotating around each other. In this sense gravity is a very elegant concept because it explains many phenomena at the same time.

Diphthongs are not combinations of monophthongs

In the same way I can point out how diphthongs allow a more elegant description of (for example) English phonology. There are 12 monophthongs in English (RP). If we allow combinations of two monophthongs we should expect to find all or almost all such combinations - 12!/(12-2)! = 132. But there are only 8 diphthongs in English (RP).

Of course we could come up with a number of rules stating that such an such monophthongs cannot occur together. We would need so many such rules that our description of English phonology would become bloated. Diphthongs are a more elegant abstraction for describing the sound system of English.


Sverre pointed out that syllables must contain a single vowel and if there is movement from one vowel target to another then the syllable contains a diphthong. You argued that

One can define a vowel through syllable, or syllable through vowel

and I agree that there is a certain circularity here. But syllables also have a definition independent of that. A syllable consists of a sonority (basically, loudness) peak and has sonority troughs (little loudness/acoustic energy) as boundaries. This solves the circularity of the syllable -> diphthong argument. And if the vowel contained in a syllable has two articulatory targets it is a diphthong.

  • Re second bold title: you compared use of diphthongs to a combination of vowels. Why not compare it to a combination of vowel+consonant? Re third title: why you consider diphtong a vowel rather than vowel+another sound?
    – Anixx
    Aug 25, 2013 at 19:39
  • You can analyse /eɪ, aɪ, ɔɪ/ as combinations of monophthong + /j/ and /aʊ, əʊ/ as monophthong + /w/. But in many dialects of English there are /ɪə, eə, ʊə/ and there is no consonant that can replace the second target here. For dialects that lack these three diphthongs (such as some descriptions of General American), the other diphthongs can indeed be analysed as vowel + consonant.
    – robert
    Aug 25, 2013 at 19:52
  • @robert If you wanted to, you could use a "nonsyllabic schwa" /ə̯/ as the second element in centering diphthongs. (Now, I don't know why you would ever want to, except to prove a point, but you could.)
    – Draconis
    Mar 30, 2019 at 19:55

By definition, a vowel is a sound that produces a syllable.

And what is the definition of a syllable? Taking dictionary.com as an example, syllable is defined as "an uninterrupted segment of speech consisting of a vowel [...] (with or without preceding or following consonant sounds)".

I'm sure you can see where this is going.

  • One can define a vowel through syllable, or syllable through vowel, but as your quote agrees, a syllable must contain a vowel.
    – Anixx
    Aug 22, 2013 at 14:53
  • What I wanted to point out is that these definitions are problematic because the two terms define each other. By defining A as B and B as A, we are no closer to understanding what A and B are.
    – Sverre
    Aug 22, 2013 at 16:01
  • This doesn't account for syllabic consonants, unless you definite such consonants to be vowels in otherwise vowelless syllables. But then you're really making it difficult for yourself. Aug 24, 2013 at 16:30
  • @hippietrail: Is that comment directed at me?
    – Sverre
    Aug 25, 2013 at 12:38
  • @Sverre: I don't know - just for whoever was following the ideas here. Aug 25, 2013 at 15:38

A diphthong is a moving vowel, in other words the articulatory apparatus undergoes a significant change of status during the articulation of. The organs of articulation change position; so that you can say a diphthong consists of three constituents, the origin, the destination and the trajectory. Diphthongs are often regarded as being two vowels, di+phthong. They are indicated symbolically by a symbol that indicates the origin, followed directly by a symbol that indicates the target or destination. The diphthong, however, is really the trajectory between these two. There are three basic types of falling diphthong in English: those that target [j], such as "eye", those that target [w], such as "ow", and those that fall in towards the centre of the vowel space (the so-called centring diphthongs), such as "air". A centring diphthong can be analysed as a diphthong of one of the other two types, followed contiguously by an R-sound, though for non-rhotic speakers the actual R merges with the centring diphthong and becomes inaudible. [i:] & [u:] could also be analysed as diphthongs, viz. [ij] & [uw]. If that were the case, we could list some twelve diphthongs in English. There are four which target [j]: [ij], [ej], [aj] & [oj], as in bee, bay, buy boy. There are four which target [w]: [uw], [ew], [aw] & [ow], as in moo, sew, cow & bowl. And the are possibly four centring diphthong, if we can count "ur" as a centring diphthong that begins at the centre and targets itself. There are "ear", "oor" as in "tour", "air" & "ur". How to symbolize these without the I.P.A. characters I don't know.

  • 2
    While your answer is totally correct, I think OP is asking "why do we call [aj] a diphthong in English, when we call it a vowel plus a consonant in Japanese?"
    – Draconis
    Mar 30, 2019 at 19:54

The other answers here are excellent and give good evidence. But here's another way of looking at it, from a phonological viewpoint.

Phonologically, the distinction is really between "syllabic" (acts as a nucleus) and "nonsyllabic" (doesn't act as a nucleus).

The division into "consonant" and "vowel" is pretty arbitrary, and you can't always draw a nice line between the two. In Chinese, fricatives can be syllabic; in English, nasals and laterals can be. Many languages have pairs of phonemes that are exactly the same except one is syllabic and the other isn't (such as /i/ and /j/). In PIE, *i and *j are reconstructed as allophones: is that phoneme a consonant or a vowel? In the context of this answer, when I say "vowel" I mean "something on the IPA vowel diagram", and when I say "consonant" I mean "something on the IPA consonant diagram"—but either of those can be either syllabic or non-syllabic.

Sometimes, it's pretty clear that you have a vowel + glide sequence, with two separate phonemes. For example, in Japanese, /j/ acts just like any other non-syllabic consonant: you can have /aka/, and /aja/, and even /a.ika/, but not */ajka/. So in Japanese, it makes sense to analyze [aj] as an underlying /a/ (syllabic) plus a /j/ (non-syllabic).

Sometimes, though, it's pretty clear that you only have a single phoneme. In English, the sequence [aj] acts just like any other syllabic phoneme. It can come before all sorts of codas, as in geist and likes, where other syllabic+non-syllabic sequences can't. And the [j] never comes "detached", even to satisfy the Maximal Onset Principle: in my dialect, for example, your is [jɹ̩], but liar is [laj.ɹ̩]. So it makes sense to say that /aj/ is a single phoneme in English, because that analysis is more elegant and explains all the data nicely.

And that's what a diphthong is: a series of multiple different phones, that acts as a single phoneme. The same can be said for affricates: phonetically, [tʃ] is a stop followed by a fricative. But if you call it an affricate, you're saying that those two sounds act as a single phonemic consonant. (In Swahili, for example, /tʃ/ patterns with the stops: it acts like a /k/ or a /t/, not like a sequence of consonant-plus-consonant.)

P.S. Like with anything in phonology, there's no indisputable "proof" of this. Phonemes aren't something we can see or measure directly. Instead, I'm saying that this theory, with diphthongs in it, explains the English evidence in a nice way. Maybe someday someone will come up with a different theory with no diphthongs involved! But for now, this theory is the most solid one I know of, and that's sort of the general consensus among linguists as a whole.

  • (If you downvoted, please explain why?)
    – Draconis
    Mar 30, 2019 at 20:14
  • /i/ and /j/ cannot be described as being the same except syllabic or not. That they are somehow associated in many languages and written with the same letter is simply a coincidence. "In Chinese, fricatives can be syllabic; in English, nasals and laterals can be" - consonants cannot be syllabic in ANY language. The very definition of vowel is that it produces a syllable.
    – Anixx
    Mar 30, 2019 at 20:19
  • 2
    @Anixx (Also, look at spectrograms of [i] versus [j]. Their similarity isn't a coincidence.)
    – Draconis
    Mar 30, 2019 at 20:51
  • 2
    @Anixx: Have you looked into syllabic nasals (/m/ and /ŋ/) in Cantonese? I'd be pretty surprised if you can justify there being a vowel there, phonetically or phonologically... And in any case, there's more than one definition for most linguistic terminology. Until you make the definition that bans consonants from being syllabic the standard one, you can't really say he's wrong. Mar 30, 2019 at 23:28
  • 1
    @Anixx As I've said in my answer, the vowel vs consonant distinction isn't really a useful one. In the context of the answer, a "vowel" is something on the IPA vowel chart, while a "consonant" is something on the IPA consonant chart. Either can be either syllabic or non-syllabic, or a single phoneme can even be both depending on the context (as in English, Swahili, PIE).
    – Draconis
    Mar 31, 2019 at 2:10

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.