Recently I've been thinking about the difference hiatus and diphthongs.

in my native language there are no phonemic diphthongs but phonetic diphthongs do occur e.g "კაი" ("okay") /kʼɑi/ [kʼɑɪ̯~kʼɑj].

Today I tried syllabifying the words "ინფორმაცია" (inpormatsia, "information") and "თერაპია" (terap'ia, "therapy") and the came out as something like [ɪn.pʰɔɾ.mɑ.ˈt͡siˑ.ɑ] [ˈtʰɛ.ɾɑ.pʼiɑ].

What clues should I use to distinguish between diphthongs and hiatus? or should I simply trust my own intuitions when syllabifying words containing two adjacent vowels?

1 Answer 1


The most important first step is being explicit as to what distinction you are positing. Terms like "hiatus" and especially "diphthong" are often over-applied to refer to different things. In the realm of a sequence of two or more adjacent vowel-like articulations V1 and V2, the pertinent questions are about prosodic organisation: are V1 and V1 in the same syllable, or in different syllables? If they are in the same syllables, are they the same w.r.t. the property that represents the vowel / glide distinction or are they different? Are they in the same subsyllabic constituent (usually, the nucleus) or is only one of them nuclear? W.r.t. weight-sensitive phonology, does the sequence behave like a long vowel (if there is vowel length); does it behave like a closed (heavy) syllable parallel to VC?

None of these questions is answered simply by looking for a presumed universal acoustic property signalling the answer to a question of prosodic analysis. Occasionally, in a language that has a contrast between bisyllabic [a.i] and monosyllabic [ai], there are audible cues that tell you which thing is being pronounced in a given word, but that will not be the case for Georgian. You will have to engage in a deep analysis of Georgian phonology and morphology to discover what analysis is the best (i.e. "are the segments both in one syllable, or are they in separate syllables?").

Reliance on the feelings of a biased individual is, uh, not a good way to answer the question on behavioral grounds, especially when the question being asked is "what is the best theoretical analysis?". Ideally, your experimental subjects would not know the research question that you are investigating, and your questions would not be structured to force (or suggest) a specific outcome. Asking "how many syllables are there in [X]?" is generally not a good strategy because syllable-identification and counting is a highly variable art that is specifically taught in school, and not necessarily the same way everywhere at all times (American English speakers don't agree as to whether "fire" and "towel" are 1 vs 2 syllables).

In lieu of clear phonological evidence for one analysis over the other, the best analysis is the one what doesn't specifically commit to a choice.

  • by hiatus I meant monosyllabic sequence of two vowels. Commented Aug 12, 2022 at 15:57
  • Alas, that is what is standardly considered to be a diphthong: hiatus is a sequence of syllables.
    – user6726
    Commented Aug 12, 2022 at 20:22
  • Oops, my bad I meant to say heterosyllabic instead of monosyllabic. Commented Aug 12, 2022 at 21:13

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.