This is a relaxed version of this question Are there any languages that only allow CV syllables? asking for strictly CV-languages. Here I want to know if there are languages with the phonology CC*VV*, allowing for consonant clusters initially and diphthongs or even triphthongs finally. Examples of such languages are welcome, too.
-
1Probably still highly doubtful. The point raised in the accepted answer that the existence of languages requiring an onset consonant is doubtful still applies here. You’d have to first find a set of languages that unambiguously require onset consonants, which in itself is not an easy endeavour.– Janus Bahs JacquetCommented Feb 18, 2023 at 14:31
-
Your formula is a bit unclear. Languages with only CV syllables are included in "allow CC*VV*", selecting zero. No language allows syllables with CCCCCCCCCCCCCCCCCCCCCC onsets. I assume you mean "with one obligatory consonant (vowel) and any number of further optional consonants (vowels)". Is that correct?– user6726Commented Feb 18, 2023 at 16:05
-
Simplified formula by omiting superfluous brackets.– Sir CornflakesCommented Feb 18, 2023 at 20:24
-
1In Lushootseed, like German, any word beginning with a vowel gets a glottal stop before it. The only difference is that it's transcribed as a consonant in Lushootseed, but not transcribed at all in German.– jlawlerCommented Feb 19, 2023 at 17:36
-
1Some varieties of the Hmong language are very close to the type you're asking about.– Yellow SkyCommented Feb 19, 2023 at 18:26
1 Answer
The Bantu language Kerewe is almost an example, because there are no codas at all, and all syllables have an onset except in utterance-initial position where there are also V-initial syllables.
The biggest challenge is finding iron-clad obligatory-onset languages. All candidates seem to have codas as well. In Tigrinya, all syllables have an onset; likewise Arabic syllables always have an onset. There are words (morphemes) in Arabic which can be argued to underlyingly have an initial vowel, but on the surface they have a glottal stop. Lushootseed, Klamath and Tonkawa are similar in being surface obligatory-onset languages, and in these languages there is no evidence for putative glottal-stop insertion (since no possibly word initial morpheme begins with a vowel). Klamath does have vowel-initial roots, but they are always precede by some C-final prefix. In all of these languages where the specter of glottal-stop insertion exists, there is also an autonomous glottal stop phoneme.
I know of two languages which are good candidates for obligatory-onset which have [ʔV]-initial words, but which do not otherwise have phonemic /ʔ/. Both Khoekhoe and Angas have this property. This initial glottal stop is phonetically robust and does not delete phrasally after a consonant (nor is there any vowel fusion), for example Angas [ʔās] 'dog', [ʒwál ʔās] 'small dog'. Following typical phoneme-elimination reasoning, one would simply say that there is no glottal stop in Angas of Khoekhoe, and therefore there are vowel-initial words although they are pronounced with a glottal stop. Thus these languages could either be obligatory-onset languages, or not, depending on whether one dismisses glottal stop in glottal stop initial words. Since all of these languages have codas, they are not examples of the desired type.
The Sucite language (Senufo) has obligatory onsets, and while glottal stop is a phoneme of the language, it does not seem to appear in word initial position so the obligatory onset is not purchased by inserting glottal stop before all initial vowels. However, it does allow /n/ as a coda consonant, so it too does not count. Another possible disqualifier is that not all syllables have vowels (there are syllabic nasals), so it depends on whether "V" means "vowel" or "nuclear segment". An example is [m̀pī] 'rabbit'. If [m̄] is a V, then this is an onsetless syllable. If [m̄] is a C, then this is a V-less syllable. In fact, syllabic nasals constitute the entire content of syllables, and are never preceded by a regular-C onset.