Rethinking the genesis of the Romance periphrastic perfect

Abstract

The Romance languages all display periphrastic perfects that can be traced to Latin [habere“have” + noun + perfect participle]. A new survey of the Latin corpus reveals that this string had three distinct structures and values. I argue that the likeliest source of the perfects is a periphrasis denoting the achievement of a result or a persisting resultant state. This implies that the relationship between possessive and auxiliaryhabereis more complex than previously supposed. Finally, I examine the range of values that this periphrasis takes across the Romance languages. I maintain that the growth of the perfect at the expense of the preterite followed an orderly pattern, with requirements on the temporal denotation of the perfect successively relaxed.

The Old English Have-Perfect and Its Congeners

“…acc + perfect participle. acc> corresponded to three distinct syntactic structures with contrasting meanings: an adnominal type (example (32) below), an attained state type ((33)–(34)), and an affectee type ((35); de Acosta 2011): (32) longa nomina contortiplicata habemus long. acc.neut.p names.…”

“…At the same time, there is convincing evidence that the attained state type is not yet a perfect: its time reference strictly coincides with the tense (present or past) of habere , the subject of habere need not be the same as the agent of the verb appearing as a perfect participle and, most strikingly, the whole construction can be passivized. 8 It is also worth noting that instances of the attained state type in which habere clearly does not have possessive meaning are attested from the earliest literary Latin, some centuries before the emergence of the have -perfect (de Acosta 2011:160-162).…”

“…A problem with this hypothesis is that while the Latin be- perfect is attested from the earliest literary Latin, the have -perfect appears several centuries later, and indeed several centuries after the period of great Greek influence on Latin. In any case, Latin had all of the necessary ingredients to make a have - erfect, and it is perfectly possible to construct a Latin-internal account of this development (de Acosta 2011).…”

Lexical auxiliaries in Spanish: How and why?

“…During the incipient grammaticalization of the haber construction in Latin, the participle agreed with the object as well, a situation that continued into Old Spanish (cf. e. g.,Pinkster 1987 andAcosta 2011 on Latin and Macpherson 1967 on Old Spanish). Remnants of Latin object agreement persist in modern French and Italian, where there is gender and number agreement of the participle with clitic object referents whenever these precede the participle.…”