Moving beyond the researcher's individual introspection: Perspectives from the key-tolok alternation and the nominative-accusative alternation of Korean resultatives

Relying on introspection (i.e. the researcher's own intuition and judgment) has been the normal practice particularly in generative linguistics. However, especially when faced with crucial data for the development of a linguistic theory or for the right characterization of a linguistic phenomenon, it is important to rely on more scientific data. Against this background, this paper conducts two acceptability judgment tests and shows that none of the claims made in the literature as to the alternation between -key and -tolok and the alternation between the nominative and the accusative case with respect to Korean resultatives is so accurate. By methodologically relying on a number of subjects and taking into consideration both the factor of whether there is a distinct nominative-marked NP from what the main verb is predicated of and the factors of the category of the main predicate and the category of the -key/-tolok predicate, the experiment on the key-tolok alternation reveals that, generally, -key and -tolok can alternate with each other only when the -key/-tolok predicate is a verb. Meanwhile, by taking into consideration not only the factor of the category of the key-predicate but also the factor of whether the NP thematically predicated of by the key-predicate is also thematically an argument of the main verb, the experiment on the nominative-accusative alternation shows that the alternation is truly allowed only when the key-predicate is formed with an adjective plus the -ci suffix and that the use of the accusative case is the norm in co-argument key-sentences and the use of the nominative case is the norm in non-co-argument key-sentences. Through these two case studies, this paper stresses the importance of not relying purely on the researcher's own introspection in linguistic research.