----- Original Message -----From: fournet.arnaudSent: Saturday, December 01, 2007 12:25 PMSubject: Re: Re: Re: Re: [tied] Anser (was: swallow vs. nighingale)
A.F :Khoisan Cognates :============> BeiJing Dao4============Proto-Khoisan : *de [˜] Meaning : many Sandawe : *dē=> BeiJing duo1============ ==Proto-Khoisan : *diʔa [˜] Meaning : egg Sandawe : *diʔa=> BeiJing dan4Is it not Fascinating !?that it works.=========***It does not work.In order to seriously propose cognates, regular rules of correspondence must be formulated and demonstrated with numerous examples.In the examples cited above, while *dao and dào appear initially worthy of a further look (provided one is willing to overlook that BeiJing /d/ is actually [t]), the other two examples agree (?) only in their initials with there is no correspondence, at least apparently, in what follows the initial.With relatively short words, the purely random correspondence of initials in two languages, is not proof of anything.Patrick Ryan***Now as far as Arabic is concerned,this language displays a very high level ofsegmental "instability" :Verbs meaning to cut :batta, batara, barata, batala, balata, sabata, bataka.r and l are both infixes and suffixes.And there are hundreds of examples like that.Most affixes can appear anywhere :rashsh : sprinkle watert?a-rashHamâ : to be angryHa-t?-amHamm : blackHama-t?a : black bloodIt is always hard to know which two consonants might be the "real" root.============ =***The process described above is not one I have ever seen described in a text on PAA or Semitic. No expert in either mentions "segmental instability" as far as I know.If this is the view of an expert ("real" root), evena minority view, I would like a reference.
I have no idea what "t?" is supposed to mean in an Arabic word if not a sequence of /t-?/. Perhaps the writer above means Humatun, blackness, where the final -t is the feminine inflection.Patrick Ryan***