The explanation of homonymy in the lexicon of Arabic

The phenomenon of homonymy appeared at the conceptual beginnings of Ar
abic grammar. In the chapter devoted to the relationship between form
and lexical meaning in the kitâb by Sîbawayhi we read that it is possi
ble to find in the spoken language of the Arabs "two concordant forms
although the meanings differ, when you say: wajadtu? alayhi (I felt re
sentment against him), which comes from al-mawjidatu (resentment) and
wajgadtu (I found) if you look at wijdana l-? allati (to find the lost
animal). Examples of this kind are numerous". Throughout the centurie
s nobody has been able to provide a satisfactory explanation of this p
henomenon.
Standard linguistic studies restrict themselves to the level of the ro
ot, whereas this phenomenon can only be accounted for at the submorphe
mic level. As posited in the Theory of Etymons and Matrices (TME), the
minimal phonetic units of the lexicon are not phonemes but phonetic f
eatures and the minimal units of sound and meaning are not morphemes b
ut matrices composed of vectors of phonetic features correlated to a n
otional invariant.
In this book we account for some fifty roots (in the traditional meani
ng of the term). Such a large number cannot be dismissed as just a han
dful of convenient examples chosen for the purpose. Within the framewo
rk of TME, homonymy is thus an ordinary phenomenon which is predicted
and explained by the theory and every case of homonymy will have an ex
planation once the entire lexicon has been described.