With Tunisia being so much in the news of late, one constantly hears (on the BBC Word Service, for instance) the British variants of the adjective and substantive derived from the toponym Tunis, namely Tunisia and Tunisian. In British English, s (stem-final z being rare) before a derivational suffix beginning in a front vowel––like –ia or –ian––remains unpalatalized while optionally undergoing “voicing” (actually, laxing) intervocalically, hence the pronunciations [tunɪziə] and [tunɪziə], where American English changes the stem-final consonant to a palatal [ʒ], hence [tuníʒə] and [tuníʒən] (note also the tense stressed vowel in the American version, where British has a lax vowel). The same differences (mutatis mutandis) hold for words like Parisian and Asian.

The systematic upshot is a semiotic one. British English does not mark nominal derivation here beyond adding a suffix, whereas American English does, in the form of the marked obstruent {ʒ} < [z] and the marked tense vowel [í]. This is a good illustration of a widespread phenomenon in language, particularly frequent in dialectology, whereby identical contexts allow of diverging morphophonemic treatments and produce variation across languages and language families as well as dialects. In this particular case, one could adjudge British English to be (expectedly) more conservative than American, since the former chooses to preserve the phonetic identity of the final obstruent and the stressed vowel of the deriving base in the derivative, whereas the latter changes them to their marked counterparts, thereby choosing to underscore the derivative’s semiotic status––its hierarchical value––at the expense of phonetic uniformity between base and derivative.