Ray Brown wrote:
>Let's see. If we don't use the pulmonic airstream, then we are left with 
>two other possibilities: glottalic (or pharyngeal) airstream; lingual (or 
>velaric) airstream.
>
>(a) Let's consider the glottalic airstream first:
>If the airstream is egressive (i.e. air flows out of the body) then we 
>produce what are commonly called ejective stops. As the airstream never has 
>chance to pass the larynx these stops must be voiceless. The IPA symbols 
>are exactly the same as fr ordinary (i.e. pulmonic) voiceless plosive 
>followed by a raised comma. If we're not going to have pulmonic consonants, 
>then we as well simply use |p t k|.
>
>If the airstream is ingressive (i.e. is drawn into the body), then we have 
>implosive consonants. When the glottis is released and the airstream drawn 
>inward there tends to be some vibration of the vocal chords, so these 
>sounds tend to be voiced when they occur in natlangs (not uncommon, e.g. in 
>many African languages). the IPA symbols are the same as corresponding 
>(pulmonic) plosives with a hook added above. We could, therefore, simply 
>use |b d g|.
>
>(b) the lingual (velaric) airstream:
>If the airstream is egressive, it's known as spitting, which is, maybe, not 
>a sound we'd want to include     ;)
>
>Sounds made with an ingressive lingual airstream, however, do occur in 
>natlangs and are normally called 'clicks'. If we're content to restrict 
>ourselves to the three found, e.g. in Zulu & Xhosa, then we can use the 
>same spelling as that found in those languages, i.e.
>|c| = dental click; |q| = post-alveolar click; |x| = lateral click.
>These languages use |ch, qh, xh| for aspirated clicks, |nc, nq, nx| for 
>nasal clicks and |gc, gq, gx| for voiced clicks - but......
>
>The production of clicks, however, does involve two types of airstream. The 
>primary articulation of the click is made by an ingressive velaric 
>airstream, but there is a secondary pulmonic egressive airstream. For 
>example, the initial sound of _Xhosa_ is a lateral click with secondary 
>articulation of aspirated velar plosive. So it depends, I guess, on how 
>strict one wishes to be in interpreting "exclusively uses nonpulmonic 
>phone(me)s."
>
>>
>>Another thing was that I felt that each phoneme behaved as a syllable,
>
>That I do not understand. All the above sounds are contoids (_phonetic_ 
>consonants); they cannot possibly behave as syllables.

I can imagine something like a Salishan (?) or other NW Coast Amerindian 
language (using some of your symbols)--
[p'ts'tl'xk'] meaning maybe 'the canoe is going downstream' or whatever 
;-)))) though I don't know how long one could carry one like that, as, 
basically, it involves holding one's breath :-))