Participles, Adjectives, and the Role of Argument Structure

The focus of this paper is a construction which is surprisingly rarely scrutinised: the impersonal passive of the intransitive. Although it sometimes receives a brief mention in discussions of the passive, and although the impersonal passive label is often wrongly given to morpholexical impersonal constructions, there are as yet no thorough analyses of the impersonal passive available for any language. In this paper, I offer an analysis of this construction in Polish, where it is made up of a tensed auxiliary or copula, and a participle commonly referred to as the passive participle. 1 The impersonal passive of the intransitive – an introduction A example of an impersonal passive of the intransitive in Polish is given in (1), with the assumption that no neuter singular referent can be found in the context of this sentence which could be interpreted as the antecedent of its ‘dropped’ subject. The agent in impersonal passives, downgraded to an oblique, is optional and frequently left unexpressed. The -n-/-tparticiple has the SG.N ending -e used in situations when agreement breaks down, such as here, where there is no subject for the participle to agree with: (1) By!o codziennie sprz"tane (przez firm#). was.3SG.N daily clean.PART.SG.N (by company) ‘There has been cleaning every day (by a company).’ Sentence (1) is an example of a predicative use of the -n-/-tparticiple in an impersonal construction, i.e. a construction without a subject. Its personal counterpart is the common personal passive, where the participle has to agree with its subject in gender and number; depending on the grammarical gender of the subject, the participle will have one of the following endings, in the singular: masculine -y, feminine -a, or neuter -e; and in the plural: masculine human -i, or other than masculine human -e. In Table 1 below, (1) is repeated in (3) and shown next to its personal counterpart in (2). Although the subject of the personal variant can bear any number or gender, in the table below it is illustrated with a noun of neuter gender and singular number (this minimises the number of variables for an easier comparison of the examples). In the typological literature, a participial personal passive is sometimes referred to as an ‘objective resultative’. This is in contrast with another type of construction, the so-called ‘possessive resultative’, in which the participle is a member of a secondary predicate which is part of the clausal object of a personal active verb ‘have’. The bottom row of Table 1 shows a relatively familiar personal ‘possessive resultative’ in (4), and a virtually unstudied impersonal ‘possessive resultative’ in (5). It is worth noting that Polish possessive resultatives have a more neutral possessive interpretation than their English translations. In the absence of an oblique agent, Polish possessive resultatives do not exclude the interpretation that the possessor may have been the agent.