No CrossRef data available.
Published online by Cambridge University Press: 15 March 2011
In his pioneering article “The Patterning of Root Morphemes in Semitic” (1950), J. H. Greenberg tabulated the frequency of every possible pair of Arabic consonants in each possible position (I–II, I–III, II–III) over a total of 3775 triliteral verbal roots, alternative statistics, for positions I–II only, were presented by J. Kurytowicz (1973), who, unlike Greenberg, excluded weak roots (i.e. those containing semivowels or geminated consonants) and admitted denominative verbs in the simple stem. Tabulations for Hebrew were prepared by K. Koskinen (1964), who included nominal roots, and excluded weak roots. The analyses by these three scholars have won deserved renown, but still more patterns can be detected in the rich data. Despite the differences of approach, the data are used here without modification, so that the results can be readily verified.