Reinterpretation

REINTERPRETATION John Haiman University ofManitoba A morphological peculiarity ofHua, a language ofPapua New Guinea, is that object and possessive pronouns with a certain class of verbal and nominal roots are infixed rather than prefixed. This is shown to result from a combination of two reinterpretations , Watkins' Law and the analytic leap, both identifiable as instances of abductive reasoning. The Hua example is particularly instructive in showing how a change from prefixation to infixation, essentially a morpheme metathesis, could have occurred gradually, and in suggesting a systematic source of counter-examples to the putative universal that the order of morphemes in a word is fixed. The mechanism of reinterpretation can, then, be described ; but functional explanations that have been proposed are only descriptions of tendencies—exceptions to which, though frequent, are unpredictable. The necessary motives of reinterpretation remain unknown. One morphological peculiarity uniquelydistinguishes Hua fromthecloselyrelated languages surrounding it in the Eastern Highlands of Papua New Guinea: the fact that personal-pronoun morphemes with a small class of verbal and nominal roots, rather than being invariably prefixed (and thus following the general rule), are infixed. The origin ofthis anomaly presents a problem in historical reconstruction.1 This article describes the process of reinterpretation whereby the anomaly arose. A 3sg. form, susceptible to more than one analysis by virtue of a null 3sg. marker, provides the basis for visible restructuring of its entire paradigm, in accordance with the proposal of Watkins (1962:178 et passim), known to its adherents as 'Watkins' Law' (cf. Arlotto 1972:156). This mechanism is thus a specific instance of the more general model of reinterpretation outlined in Andersen 1973. The Hua example is also instructive in furnishing a counter-example to the proposed universal (Perlmutter 1971) that the order of morphemes in a word is 1 The Hua language, spoken by about 3,100 people in the immediate vicinity of Lufa station, is identified by Renck 1975 as the westernmost form of Yagaria, a group of closely related dialects belonging to the Kamano subfamily of the East Central family of the Eastern New Guinea Highlands stock (Wurm 1971). Wurm's basis for genetic affiliation is lexicostatistical (542, 552); but he does provide a checklist oftypological features shared by most ofthe members of the stock: the existence of (la) phonemic tone, (lb) consonantal prenasalization, (Ic) dual number, (Id) medial verb forms which occur in two forms depending on whether the subject of the following verb is identical with the subject of the medial, (Ie) an interrogative suffix on verbs; and the absence of (2a) gender distinctions, (2b) prefixes in general (except for the negative marker), and (2c) a distinction between the 2nd and 3rd persons except in the singular. Hua shows all these features, deviating from the norm only in having portmanteau suffixes. Most adult Hua speakers are fluent in one or more of the neighboring languages, particularly Gimi, spoken to the south and southwest (a member of the Gimi-Fore subfamily of the East Central family), and Siane, spoken to the north and the northwest across the Tua river (a member of the Siane-Yabiyufa subfamily of the same East Central family). Previous treatments of Hua (Haiman 1972, 1975, 1976a,b) are based on fieldwork undertaken overa period offouryears, 1971-75, supported by a series ofgrants from the Australian Research Grants Council. A preliminary version of §§1-2 of the present paper was read at the LSA Winter Meeting in 1975. 1 am grateful to my audience at that meeting, and to Harold Koch and Arnold Zwicky for a number of improvements. 312 REINTERPRETATION313 fixed. A similar counter-example in Turkish suggests that the mechanism by which such anomalies arise may be a systematic source of such counter-examples. The simplicity and plausibility of the Watkins-Andersen model of reinterpretation is not in question. Nevertheless, it should be noted that an explanation for any change ideally should demonstrate not only its mechanics—how it was possible—but also its motivation : why it was necessary. When the latter is missing, we are in possession of laws which account for the direction of a change 'une fois qu'il pleut' (cf. Kurylowicz 1949), but none accounting for the change itself...