Not considering things like vocabulary or syntax (if that is possible), what phonological differences make the most telling distinction between two dialects, is it the vowels or the consonants?

Informally, among English dialects, it seems that between any two things called dialects the major differences are the vowels (and the consonants seem to stay the same).

But I've heard (unsupported of course as is my personal judgement) that in other dialect families, it is a change in consonant (and not the vowel that marks the dialect difference) (the classic example being the 'shibboleth', an s/sh difference.

(of course this may be a false dichotomy, and there are other phonetic differences that are not about vowel/consonants, like say prosody or accent).

This question is somewhat motivated by my other question about lexical sets; that is an analysis technique for managing multiple varieties purely by vowels. Maybe that method is useful in general (in other sets of language varieties), or maybe it is just helpful for English, or maybe it is just one tool, and other tools are more or less successful (and I'd like to know to what degree).

  • vocabulary, syntax and morphology can also differ greatly from one dialect to another
    – Bozho
    Sep 16, 2011 at 14:05
  • 1
    and unless we have statistical data, we can't tell for sure whether it's vowels or consonants. I also feel vowels are more common, but I can't be affirmative
    – Bozho
    Sep 16, 2011 at 14:06
  • 2
    I would expect this to be very different from language to language. In my experience with Spanish and German when I guess a speaker's origin it's due to their consonants but with English both. Southern English glottal stops, American rhoticity, etc. Sep 16, 2011 at 15:39
  • @hippietrail: that's very much a good kind of response I'm looking for, that maybe my informal reflection is specific to English dialects (that maybe vowel changes are more common there) than in other dialect groups).
    – Mitch
    Sep 16, 2011 at 15:42
  • Is this question supposed to include accent differences, or only dialects?—because some of the things mentioned seemed to apply to accents rather than dialects (like h-dropping in English). I could be wrong.
    – Cerberus
    Sep 16, 2011 at 15:43

2 Answers 2


There is no "primary" difference. Some examples from across the spectrum:

  • Stops: one of the main differences people consciously identify between British English registers is the nature of stop releases. Fully released stops, even in word final position is characteristic of more deliberate and higher register speech, whereas stop glottalization or lack of release is common in lower registers.
  • Fricatives: one of the most salient markets of the Taiwanese dialect of Chinese is the merger of alveolar sibilants with their retroflex counterparts.
  • Approximants: broadly speaking, a major and easily identifiable difference between the Scottish dialect of English and RP is that in Scottish English, there is much less /r/-dropping.
  • Vowels: vowel examples are probably the most common, but I wouldn't go as far as to say that they are the "primary" difference due to all the above examples. Probably the easiest isolated example of vowels being a cue of dialect is Canadian diphthong raising.

Moreover, each of the above can all contribute equally to non-native dialects:

  • Stops: the inability of CV or CVN languages to adapt to CVC languages is a huge cue for non-nativeness; another prevalent example is when speakers of languages with obligatory word-final devoicing like German and Russian make homophonous word pairs like bat and bad (not exactly a stop contrast, since it's mostly vowel length, but it's complicated).
  • Fricatives: since so many fricatives sound alike, fricatives are incredibly frequently pronounced differently for non-native speakers. Examples are substitution of /s/ for the cross-linguistically rare dental fricative found in English and the use of the velar /x/ in place of /h/ by speakers from languages without [h]. /h/ is also a common source of non-native accents. Since [h] is basically just loud breathing, it's non-distinctive in many languages. English speakers have a hard time adjusting to languages with contrastive word-final /h/, for instance, and French speakers have a hard time adjusting to languages with contrastive word-initial /h/.
  • Approximants: conflation of /r/ and /l/ by Japanese and Korean speakers of English is all too well studied.
  • Vowels: speakers will very often import vowel qualities from their native languages. For instance, English speakers learning any language with the pure high back rounded [u] will substitute the English centralized gliding equivalent.

There are also prosodic differences that influence our judgment of dialects, but I'm not experienced enough with them to advise on it intelligently.

In summary, differences between dialects are marked in so many ways including both vowels and consonants. There is no "primary" difference. That question is as unanswerable as the question "What is the primary sound in languages? Vowels or consonants?".

  • Thanks, Steven, for an excellently detailed answer. Such good answers make it easier to complain though! My quibbles: 1) I think that non-native language difficulties are not so relevant (I think native varieties tel more). 2) No doubt some single distinctions stand out more than others (say rhotic vs non-rhotic) because they are easier to describe and notice. I am wondering (and I feel like it is true but don't know that) if you take a list of all differences between dialect X and Y (AusE and NZE, or CanE and AmE), wouldn't the set of vowel differences be the overwhelming majority?
    – Mitch
    Sep 16, 2011 at 14:59
  • To continue...That is, if r-dropping happens, aren't a lot more vowels probably different, too? (and r-dropping could be simply a change in vowel-coloring by r's).
    – Mitch
    Sep 16, 2011 at 15:00
  • I'm personally resistant to distilling complicated observations into generalizations in general (ha!). There are numerous complications to trying to find a "primary" difference. You mention that one could just list all the differences, but how do you sum them? Relative frequency or salience? What about varying degrees of the differences? I think that this question is not unlike the tired "Is the Sapir-Whorf hypothesis true". A weak version is so obviously true (vowels play a big part in dialect), and a strong version is so obviously false (vowels play the only part in dialect).
    – Steven
    Sep 16, 2011 at 16:16
  • By the time you've finished inevitably qualifying and defining your generalization, you'll already have pointed out so many specific examples and exceptions that you'd have had a more productive and enjoyable process had those examples been the subject of your inquiry in the first place.
    – Steven
    Sep 16, 2011 at 16:17
  • Yes, I realize my question is an oversimplification, superficial, and even though ultimately quantifiable, too complex (and the individual questions much more easily addressable). But it's a good thing to have the more refined questions come out (as I think you've helped do).
    – Mitch
    Sep 16, 2011 at 16:27

I must agree with Steven Xu, "primary difference" is the wrong question to ask. In addition to Steven's list, dialects also differ in:

  • Word order
  • Stress and rhythm (Jamaican English IIRC is syllable-timed)
  • Intonation (does sentences go up or down at the end, up or down to mark a question, is the intonation pattern before the focused word equal or different to the pattern after the focused word and what the intonation is on the focused word itself). This is one of the most obvious ones in my L1, Norwegian.
  • Number of cases
  • Vocabulary
  • etc.

There are as many dialectal differences as there are differences between languages and generally a more interesting and unanswered question is: where is the cutoff between dialect and language. The old joke that "A language is a dialect with an army and a navy" doesn't hold since German German and Austrian German aren't considered two different languages, and there's a movement underway to reunite Brazilian Portuguese and European Portuguese (makes for a bigger market, that.)

  • I think the OP means "primary phonological or phonetic difference" specifically. Sep 16, 2011 at 19:16
  • Maybe the title of the question should be changed. However, it is a wide-spread misconception that dialects are only about sounds and vocab...
    – kaleissin
    Sep 16, 2011 at 19:21
  • Ah I see. I'll ask for clarification... Sep 16, 2011 at 19:23
  • @kaleissin: The content of the question is all about phonology; I'll fix the title. As to the difference between dialect and language, I think the example of the varieties of English (with the explanation in other comments) shows the level of difference I am asking about.
    – Mitch
    Sep 16, 2011 at 21:13
  • With the changed title only the intonation-part and stress-rhythm-part is relevant in my answer (as these are part of phonologyl). Is the answer worth changing or would it be better to add another answer? Hm, decisions, decisions.
    – kaleissin
    Sep 16, 2011 at 21:34

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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