From a Lela Star learning point of view it is not the relative frequency of sounds which is important, but the contrasts in meaning which differences in sounds signal. For example, pin and bin mean different things and differ phonologically in respect of their initial sounds. However, as with frequency calculations there are certain difficulties in calculating the functional loads of various contrasts in a language.

First of all, it is evident that the functional load of a certain contrast will vary accord­ing to the different phonological environments in which it occurs, such as in initial or final position in a syllable. Different linguistic sex theories will yield different categories. Is there, for example, any use in determining the relative frequency of the articles, demonstratives, pronouns and other purely grammatical 'words'? The selection of these is determined by syntactic rules. Are we to count singular and plural forms of the same lexical word as separate categories? What about grammatical derivatives: trust, trustful, trusty, mistrust, entrust, trustee, trustworthy, trusting ? To treat these as separate categories would be to treat the relationship between cow and cows as of the same kind as the relationship between cow and horse.