DescriptionPublished as Coyote Papers: Working Papers in Linguistics from A-Z, Studies on Arabic, Basque, English, Japanese, Navajo and Papago
AbstractThis paper addresses the problem of what has traditionally been labelled Aux in Basque. The Guipuzkoan dialect of Basques has, in statements, a particle sequence which occurs to the right of the verb, in sentence final position. Two examples are given in (la) and (lb): (1) a. Miren itzultzen d a; Mary:abs pass:by:asp PART-Aux; 'Mary passes by.' b. Mirenek sagarrak jan z - it - u - (e) n; Mary:erg apples:the:abs eat PART -PARI -Aux –PART; 'Mary ate the apples.' The first particle in (la) marks the person of the subject of the sentence, and the first and second particles in (ib) mark respectively person and number of the object of the sentence; the last particle marks tense. According to the traditional analysis, the 2 particle complexes not only include person and tense markers, but above all the root of 2 verbs labelled Aux by reference to the notional category of help - ing verb. In (la) and (lb), the vowel labelled Aux is respectively identified with the root of the verbs izan and ukan. The root of izan is considered to be equivalent to 'be' when used as helping verb of intransitive verbs, like in (la); on the other hand, the root of ukan corresponds to 'have' when used with transitive verbs, like in (lb). Further, these 2 verbs are characterized as the necessary tools for indicating temporal distinction -under the label conjugation- as in jan ditu 'he has eaten them', and jan zituen 'he ate them'. Moreover, the traditional analysis only recognizes these 2 auxiliary verbs, which, by the way, also function as main verbs. For instance, (2) Mikelek katuak d - it –u; Mike:erg cat:the:abs PART-PART-V; 'Mike has cats.' However, this analysis fails to explain particle sequences in which the verbal root necessary for identifying the helping verb never appears. For example, the particle complex dizkiot 'I (have) them for him' does not contain the verbal root of the auxiliary verb ukan 'have'. In. fact, ukan never appears within particle sequences in which a double objective relationship shows up. The purpose of this paper will be to give an account of the vowel variation in (la) and (ib) conjointly with an explication of those cases which indicate a double objective relationship. In doing so, we will simply label Aux by means of a variable2 X. To illustrate, notice the change in the particle complexes of (la) and (ib): (1) a. d –a (PART-X) b. z - it -u-(e)n (PART--PART-X-PART) We shall argue that this segment does not correspond to an auxiliary verb form. It will be identified as an element whose formal properties depend upon the subcategorization of the verb. Furthermore, this paper meets another goal: it will provide evidence for identifying these particle sequences as an instantiation of the cross-linguistic category AUX as defined in (3): (3) Given a set of language internal analyses, in terms of constituents, those constituents which may contain only a specified (i.e., fixed or small) set of elements, crucially containing elements marking tense and /or modality will be identified as non-distinct. (Steele et al., (1981)) Furthermore, the AUX category has the following set of properties: 1. AUX is a constituent, 2. which occurs in first, second, or final position, 3. AUX contains a specified, i.e., fixed and small, set of elements, 4. which occur in a fixed order within the AUX constituent, 5. the membership of which set must include elements marking tense and /or nodality, but 6. it may include, as well, elements marking subject marking, subject agreement, question, evidential, emphasis, aspect, object marking, object agreement, and negation. The first 2 properties will not be discussed. From the outset, we assume that AUX is a constituent which may occur in initial, second from the beginning and final positions. In conclusion, we shall claim that the whole set of particles is to be called AUX, and not some part of it. Section 2 is concerned with the internal organization of the particle sequence. It must be stressed that this analysis is carried out on strictly synchronic grounds. Maybe, Section 2 will seem overemphasized. However, it is a logical consequence of the analysis being presented in this paper. The set of particle sequences identified with AUX are too easily treated as mere idiosyncratic forms no longer analyzable into smaller units. In this section, particle sequences in intransitive and transitive sentences will be first analyzed; then the analysis of the unlabelled segment X will be dealt with. In Section 3, we will point out the problems that the traditional auxiliary verb hypothesis poses. Given these problems, we will see how our approach solves them. Finally, in Section 4, a recapitulation of the argumentation will be presented which will require a revision of 2 empirical generalizations made in the Encyclopedia of AUX: the status of person marking as a non -definitional property must be reevaluated, and the particle sequence can contain indirect object markers for person and number. From this discussion, we can forsee the reason why the AUX identification in Basque is important. It will allow to establish a significant corelation between sentential constituents marked for case, and the markers which are part of AUX.