Search

In every-day language use, two or more structurally unrelated constructions may occasionally give rise to strings that look very similar on the surface. As a result of this superficial resemblance, a subset of instances of one of these constructions may deviate in the probabilistic preference for either of several possible formal variants. This effect is called ‘constructional contamination’, and was introduced inPijpops & Van de Velde (2016). Constructional contamination bears testimony to the hypothesis that language users do not always execute a full parse of the utterances they interpret, but instead often rely on ‘shallow parsing’ and the storage of large, unanalyzed chunks of language in memory, as proposed inFerreira, Bailey, & Ferraro (2002),Ferreira & Patson (2007), andDąbrowska (2014).Pijpops & Van de Velde (2016)investigated a single case study in depth, namely the Dutch partitive genitive. This case study is reviewed, and three new case studies are added, namely the competition between long and bare infinitives, word order variation in verbal clusters, and preterite formation. We find evidence of constructional contamination in all case studies, albeit in varying degrees. This indicates that constructional contamination is not a particularity of the Dutch partitive genitive but appears to be more wide-spread, affecting both morphology and syntax. Furthermore, we distinguish between two forms of constructional contamination, viz. first degree and second degree contamination, with first degree contamination producing greater effects than second degree contamination.

In this article, we introduce the effect of “constructional contamination”. In constructional contamination, a subset of the instances of a target construction deviate in their realization, due to a superficial resemblance they share with instances of a contaminating construction. We claim that this contaminating effect bears testimony to the hypothesis that language users do not always execute a full parse while interpreting and producing sentences. Instead, they may rely on what has been called “shallow parsing”, i.e., chunking the utterances into large, unanalyzed exemplars that may extend across constituent borders. We propose several measures to quantify constructional contamination in corpus data. To evaluate these measures, the Dutch partitive genitive is taken under scrutiny as a target construction of constructional contamination. In this case study, it is shown that neighboring constructions play a crucial role in determining the presence or absence of the -s suffix among instances of the partitive genitive. The different measures themselves, however, are not construction-specific, and can readily be used to track constructional contamination in other case studies as well. ispartof: Folia Linguistica: Acta Societatis Linguisticae Europaeae vol:50 issue:2 pages:543-581 ispartof: location:POLAND, Poznan status: published