The project investigates sensitivity to the selectional properties of the matrix verb in L1, L2 and L3 English. It has been established that a structural prediction triggered by the matrix verb favors a clause complement (1a, the eventive complement) over the DP complement followed by a restrictive relative clause (1b, the entity complement). If the participants are offered full relative clauses (1b), the parser needs to adjust its analysis at the complementizer that.
(1a) Bill saw [CP (that) [DP the friend of the neighbor] [VP was drinking coffee]]
(1b) Bill saw [DP the fired of the neighbor [RC that was drinking coffee]]
At the same time, the eventive complement can also have a form of the Small Clause in English (2a). In this case, there should be no structural conflict at any place in the sentence, if the participants are presented with the reduced relative clause (2b).
(2a) Bill saw [SC [DP the friend of the neighbor] [VP drinking coffee]]
(2b) Bill saw [DP the friend of the neighbor [RC drinking coffee]]
Sentences (1b) and (2b) are ambiguous for the relative clause modification i.e., the doer of the drinking can be either the friend or the neighbor. In (1a) and (2a), only the grammatical subject of the subordinate clause the friend (of the neighbor) can be the doer of the drinking.
We assume that sentences like in (2b) will be interpreted as The friend was drinking coffee because (2a) and (2b) have identical word orders, and (2a) favors this interpretation. In (1b), the interpretation The neighbor was drinking coffee, should be preferred because it typical for English. The effect of the initial prediction for (1a) will be reduced, if not annulled, at the complementizer that.
Native languages may also have an effect on Ln processing. Therefore, the L1s of our participants vary by the amount of the property under investigation. English has the properties explained above, and it is the only language of our participants in Chicago (USA). For our Norwegian-English L2ers in Trondheim (Norway), the properties described above are similar across both languages. Portuguese, the L1 for our Portuguese-English L2ers in Lisbon (Portugal), has the property illustrated in (2a) and (2b). Armenian, the L1 for our Armenian-English L2ers in Yerevan (Armenia), has the property illustrated by (2a) and (2b) too, but the word order in the target sentences is reverse. Our Spanish-Catalan speakers of L3 English in Barcelona (Spain) have the sentence like (1a) ambiguous between the eventive complement and the ambiguous relative clause in their L1s.
Marina Sokolova, PhD, Norwegian University of Science and Technology, Norway. Project head.
Research interests: human language processing, syntax-prosody interface, multilingualism.
Jess Ward, PhD student, University of Illinois at Chicago, USA. Project coordinator.
Research interests: phonological and syntactic L2 / L3 acquisition, current focus – heritage speaker L3 acquisition, pronoun disambiguation, and eye-tracking.
Andreia Janeiro, BA student, University of Lisbon, Portugal. Assistant for data collection.
Research interests: bilingualism, syntax, language processing
Prof. Shushanik Paronyan, Higher Doctoral Degree in Philology, Yerevan State University, Armenia. Project host at the Department of English for Cross-Cultural Communication.
Research interests: English philology, multilingualism
Gayane Hovhannisyan, BA student, Yerevan State University, Armenia. Assistant for data collection.
Research interests: art, culture, bilingualism.
Research interests: multilingualism, cognitive decline, aphasia
Ingrid Oyervides, PhD student, University of Barcelona, Spain. Assistant for data collection.
Sergey Sokolov, MA (equivalent). Technical consultant.
Research interests: human psychology, language processing, computer modelling, research management.
