vowels make preceding consonants more audible?

[awer]

Hi,
I was wondering about the basic phonotactics for my conlang.
I was considering several options: 1. CV only, because I like the simplicity and wanted each such a pair to be one word/morpheme, taking in all the speech sounds I'm able to pronounce myself I was fascinated by the East Asian languages, I'm still wondering why they have so many single-syllable words like Vietnamese, but even Old Chinese did. That's pretty weird. Did they never have longer words? Does anyone have any resources on that?
2. any combination of sounds including single consonants as meaningful words/morphemes and any consonant clusters but only in the script, when pronouncing them an epenthetic schwa would make the speech fluent and comprehensible. I liked the freedom it'd give the language in terms of word formation, and conciseness, allowing single-letter words. But there was a problem distinguishing syllables when I made some short words, say, "na kme" and "nak me", considering how single letters can't usually be stressed in natural languages. I could change that in my conlang but it felt weird, unusual, I should get over this feeling though, because it's unreasonable. So we'd get /'na 'kəme/ in the former instance, and /'nakə 'me/ (allowing for /'nak 'me/) in the latter.

Anyway, I was wondering what makes some sounds (mainly vowels) more important than others (like stops)? The question popped back to me today when I read about the Miyako language (http://en.wikipedia.org/wiki/Miyako_language), a Japonic language that has words like "ss" and "ff". Is choosing phonotactics an arbitrary choice, or is there an optimal choice for easy speech production? Some languages tend to simplify their consonant clusters like Old Chinese → Modern Chinese, while others, like Polish, allow words like wzgląd and domostw even though its ancestor, Old Church Slavonic, is said to have only open syllables.

So, ultimately, do you think that consonants are more audible when followed by vowels? A word like 'ss' can't be even shouted so I wouldn't allow such a thing in my phonotactics. I'm just trying to figure out reasons for this.

[2+3 clusivity]

A word like 'ss' can't be even shouted so I wouldn't allow such a thing in my phonotactics.

Perhaps not by you as a non-native speaker. Don't be too quick. The link you gave shows that Miyako can be sung (https://www.youtube.com/watch?v=0Rjl0XOfhPM) another even shows an odd country music song.

Is choosing phonotactics an arbitrary choice, or is there an optimal choice for easy speech production?

You might look into http://en.wikipedia.org/wiki/Proto-Indo ... _structure and http://en.wikipedia.org/wiki/Sonority_hierarchy as one pole for syllable design. Alternatively, check out http://en.wikipedia.org/wiki/Georgian_l ... #Phonology. Even further afield, but akin to Miyako, look at http://en.wikipedia.org/wiki/Nux%C3%A1l ... #Syllables. It seems that languages can tolerate a lot of different syllable types.

Is it arbitrary? No. Certain designs seem to crop up more often, and the exceptions seem to cluster. Is there an optimal choice . . . uhhh . . . half-jokingly: CV. "Optimal" is a loaded word -- what type of language are you making, an EngLang? If so, go ahead with "efficiency," perhaps based on averages. If not, go with what you like because you are the arbiter and the goal maker.

Anyway, I was wondering what makes some sounds (mainly vowels) more important than others (like stops)?

I'm not sure Cs -- or as you note, stops -- are relatively less important. Many languages have a greater diversity of consonant phonemes, so, perhaps in terms of information structure, the C slots are more important because they allow greater contrasts. Consider a language like: http://en.wikipedia.org/wiki/Ubykh_language. Then again, you could argue back that a V change still providing just as much of a lexical shift as a change in any C and is therefore equal salient

Query: what is a vowel really and why should it be important? Perhaps -- as in Miyako, Nuxalk, and even English with it's syllabic <m, n, r, l>s -- it is better to think not about vowels but instead about the importance of what phonemes can occupy a syllable's nucleus.

[----]

phonemically vowel-less words in Ogami/Miyako just gain a final [ɯ] when shouted or sung; i.e. /s:/ > [s:ɯ]

[Curlyjimsam]

So, ultimately, do you think that consonants are more audible when followed by vowels?

I think so, yes. Final consonants are relatively likely to be lost (or merge with other consonant phonemes), and audibility is likely the most important factor.

[vokzhen]

Hi,
I was wondering about the basic phonotactics for my conlang.
I was considering several options: 1. CV only, because I like the simplicity and wanted each such a pair to be one word/morpheme, taking in all the speech sounds I'm able to pronounce myself I was fascinated by the East Asian languages, I'm still wondering why they have so many single-syllable words like Vietnamese, but even Old Chinese did. That's pretty weird. Did they never have longer words? Does anyone have any resources on that?

Why's it weird? Why have words longer than a syllable when you don't need them? Of course, you do sometimes need them, which is why a great many words in Chinese have been compounded and are made up of multiple syllables. Things like heavy reduction or complete loss of unstressed syllables, loss of final consonants, initial cluster reduction, and so on made words shorter. This didn't necessarily reduce information: qualities of final consonants can be transferred to the preceding vowel as tone, phonation, or both, cluster reductions may result in new sets of consonants (palatals, retroflexes) rather than mergers, and while I don't know how much it happened (or whether we can reconstruct that well) unstressed vowels can transfer their information to the stressed vowel via umlaut, vowel breaking, or other methods. If it does reach a point where there is too much information loss and communication is impeded by ambiguity, you can start getting compounding, which is what Mandarin is doing now (don't quote me on it, but I thought I heard someone on hear say that in connected speech, most lexical words are multisyllable compounds, and it's grammatical words and basic vocabulary that are only single syllables).

2. any combination of sounds including single consonants as meaningful words/morphemes and any consonant clusters but only in the script, when pronouncing them an epenthetic schwa would make the speech fluent and comprehensible. I liked the freedom it'd give the language in terms of word formation, and conciseness, allowing single-letter words. But there was a problem distinguishing syllables when I made some short words, say, "na kme" and "nak me", considering how single letters can't usually be stressed in natural languages. I could change that in my conlang but it felt weird, unusual, I should get over this feeling though, because it's unreasonable. So we'd get /'na 'kəme/ in the former instance, and /'nakə 'me/ (allowing for /'nak 'me/) in the latter.

Berber languages do something similar, where a number of words have no phonemic vowels but are broken up by an epenthetic vowel, its position and pronunciation depending on a set of rules. That said, keep in mind that homophones aren't a problem until they actually impede communication. For example, are "na kme" and "nak me" going to be ambiguous, or are they going to be obvious because na kme ran away yesterday, while nak me the dog? That is, they're different parts of speech so they won't get confused. If they are the same part of speech, are they likely to appear in the same contexts? And allophony might be able to help, as well. In English, we can come up with "the /naitreit/ is too high," but it will be an extremely contrived situation where it's not perfectly clear whether we're talking about chemicals (nitrate) or a hotel (night rate), and even so, while they have the same underlying phonemes, the allophony clears it up quickly.