Back to list

Detail of contribution

Auteur: Eric PEDERSON

Titre:
Irrealis or 1-place disjunction


Abstract/Résumé: The use of irrealis markers in disjunctive coordination has been long observed. Recently, Mauri (2008) argues for a fairly clean distinction between dedicated disjunction markers and general irrealis markers which might also be used to express alternative or disjunctive semantics. However, while an irrealis marker might be generally used monoclausally, seeing coordinated clauses or NPs with this same marker does not mean that each clause is simply marked as irrealis. It is entirely possible that the use of irrealis marking in this particular way has conventionalized into a “true” disjunctive construction with precise disjunctive semantics. An essential characteristic of both disjunction is that it is a two (or more) place operators: Or Ravi is going to the market clearly indicates that there is an additional understood proposition most likely explicitly in the speech context. Irrealis markers are generally understood as one place operators both structurally and semantically. However, many languages conventionally concatenate two clauses both marked as irrealis. For example, consider the “open” disjunction in Dyirbal (Dixon 1972:363) which uses the particle yamba ‘perhaps, might be’: yaqa guya buran / gilabayqi mipa / yugur yamba / yaygal yamba / “I saw a fish, what was it down there? -it might have been a barramundi, or it might have been a red bream” Or the Tamil dubitative clitic =oo: ravi vaaraan=oo sita vaaraaL=oo “Ravi is-coming=maybe [or] Sita is-coming=maybe.” Given an analysis of, e.g., English or as essentially about reduced realis of the two disjuncts, the primary distinction between a European-style disjunction construction and examples such as these is that constructions with or cannot serve as one place operators. Operators such as Dyirbal yamba and Tamil =oo can readily attach to single or multiple hosts, essentially making them N-place operators. As a further example, consider disjunctive constructions which derive from a negative marker. Negatives are inherently one place operators. To express a strict choice between two clauses (something like exclusive disjunction), some languages have grammaticized a negative with a nominalizer, quotative, or some other such construction. For example, the nominalized archaic negative copula allatu serves as an exclusive “either-or” disjunction marker in Tamil: ravi vaaraan allatu sita vaaraaL “Ravi is coming, not so, Sita is coming” Again, we see the birth of a two place disjunctive construction following naturally from the conversion of a one place irrealis operator (negation) into something which can take multiple arguments.