Abstract :
[en] In every-day language use, two or more structurally unrelated constructions may occasionally give rise to strings that look similar or even identical on the surface. As language users sometimes employ shallow parsing, the processing of these strings may short-circuit (Ferreira, Bailey and Ferraro 2002; Ferreira and Patson 2007; Dąbrowska 2014). As a result of this interference, a subset of instances of one construction may deviate in their formal realization. This effect is called constructional contamination (Pijpops and Van de Velde 2016). Pijpops and Van de Velde only investigated a single case study in depth, namely the Dutch partitive genitive. We will now test whether constructional contamination is an occasional rarity, or constitutes a more wide-spread phenomenon in language. To do so, we take under scrutiny three new case studies. We only present two here, because of space constraints. The first deals with weak preterites in Dutch, which may be subject to constructional contamination from an enclitic 2sg pronoun in the present. Such enclitic pronouns, present in a number of dialects, are realized as -de or -te, and are homophonous to the weak suffix, as in (1) (Vosters 2012). Vosters showed that in these dialects, the weak preterite, e.g. graafde in (2), was more prevalent than in other dialects, thereby rebuffing the alternating strong form, e.g. groef in (2). Furthermore, under the definition of Pijpops and Van de Velde (2016: 543–544), we expect a lexical differentiation in the effect. In particular, we will test the hypothesis that in those dialects displaying clitic realization, verbs that more often exhibit an enclitic second person will prefer the weak preterite more strongly than other verbs (cf. Vosters 2012: 244, Fn. 16). (1) Contaminating construction: enclitic present, 2nd person singular pronoun Vandaag graaf-de een put. (Vosters 2012: 242) today dig-2SG.PRS a hole ‘You will dig a hole today.’ (2) Affected instances of the target construction: preterite Gisteren groef/graafde ik een put. yesterday dug I a hole ‘I dug a hole yesterday.’ The second case study concerns the competition between the bare infinitive and the long infinitive with te (‘to’). Normally, Dutch posture verb auxiliaries zitten (‘sit’), liggen (‘lie’) and staan (‘stand’) cannot grammatically take the bare infinitive, if they are finite. A peculiar exception to this rule are the instances where the auxiliary is indicative present plural in a subordinate clause (Haeseryn et al. 1997: 970; Klooster 2001: 61). Here, the bare infinitive is possible. The reason for this is that these particular cases contain a chunk that is superficially identical to the IPP-construction as in (4), where the bare infinitive is grammatical and frequent. Based on constructional contamination, we further hypothesize that, the more instances of the finite posture verb auxiliaries superficially resemble IPP-constructions, the more often they will exhibit bare infinitives (Van de Velde 2017: 60–68). (4) Contaminating construction: infinitivus-pro-participio Ze hebben de hele dag zitten kletsen. they have the whole day sit chatter ‘They have spent the whole day chattering.’
