Origins of human language in Southern Africa?

[jess tauber]

It would be interesting to know how the mapping of phoneme/gene diversity correlates with language internal distribution of semiotic resources (manual/body/facial gestures, analytically transparent sound symbolism, different morphosyntactic types, etc.). Would there be remnant signals here as well? The largest ideophone inventories tend to be found in Sub-Saharan African areas (though much rarer in 'Khoisan' itself), and in South, Southeast, and East Asia (with Uralic and Basque in Europe), and on into Oceania. Smaller but still significant sets are found in Mesoamerica and northern South America. 

Augmentative-diminutive shifting is much more popular around the Pacific Rim, but also is found in a number of East African Bantu languages, in South Asia, and Basque.

Root-level transparency can be found in a number of Khoisan languages- this is neither ideophonic nor augmentative-diminutive in nature, but seems to be differentially fused form-internal derivation. Some has to do with the level of moisture depicted, in an environment famous for easily obtainable water resources (and these have shifted around within Africa over many thousands of years), instrumental actions (for ex. one piece following clicks commonly associates with squeezing out contents from tubes, such as waste matter from intestines), general areas of an abstracted 'body' (dorsal, ventral and caudal, rostral), and so on. The latter is interesting in comparison to similar true morphology found in many Andaman languages. One never sees things like this in normal ideophones.

Jess Tauber
phonosemantics at earthlink.net