Naslov
Native and non-native processing of Italian subject pronouns: Evidence from eye-tracking

Autori
Kraš, Tihana ; Sturt, Patrick ; Sorace, Antonella

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Skup
Architectures and Mechanisms for Language Processing (AMLaP) 20

Mjesto i datum
Edinburgh, Ujedinjeno Kraljevstvo, 03.09.2014. - 06.09.2014

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
L1 processing ; L2 processing ; subject pronouns ; Italian ; eye-tracking

Sažetak
Being a null subject language, Italian allows both null (omitted) and overt (expressed) subject pronouns in finite clauses. According to the Position of Antecedent Strategy (PAS) (Carminati, 2002), in intra-sentential contexts the null pronoun is biased towards an antecedent in the subject position, while the overt pronoun is biased towards an antecedent in a non-subject position. The PAS is based on the notion of discourse prominence, defined in syntactic terms: null pronouns are thought to prefer the most prominent antecedents in discourse, antecedents in the subject position being considered as the most prominent ones. Evidence for the PAS has been provided in self-paced reading studies with native speakers featuring pronouns in forward anaphora contexts (Carminati, 2002 ; Filiaci, 2011). // In an eye-tracking study we investigate whether native speakers (n=28) and proficient non-native speakers whose L1 is English (n=28) process Italian subject pronouns in accordance with the PAS in backward anaphora contexts. Participants read 36 complex bi-clausal sentences with null or overt pronouns in the subordinate clause. The pronoun matched either the subject or the object of the main clause, depending on its gender (Quando lei/lui/Ø è entrata/-o in ufficio dopo pranzo, Adriana ha salutato Roberto con un grande sorriso, ‘When she/he/null entered[fem/masc] the office after lunch, Adriana greeted Roberto with a big smile’). Note that the participal in the initial clause agrees in gender with the subject pronoun (entrata/entrato --- entered[fem/masc]), so that gender matching between the pronoun and the main clause subject can be manipulated regardless of whether the pronoun is overt or null. The backward anaphora context is theoretically interesting because the PAS predicts that the main clause subject (Adriana) will be ignored as a potential antecedent for the overt pronoun due to the hypothesized object preference. // For the native speakers, first pass reading times on the main clause subject plus auxiliary (Adriana ha) show a reliable gender mismatch cost for the null pronoun conditions (mismatch: 320 vs. match: 289 msec), but not for the overt pronoun conditions (mismatch: 310 vs. match: 318 msec), suggesting that the native speakers attempt to assign the null pronoun immediately to the main clause subject, but ignore this position as a potential antecedent for the overt pronoun. Conversely, at the object region (Roberto) a first-pass mismatch cost is found for the overt pronoun (i.e. slow times when the pronoun mismatches with the object ; mismatch: 266 vs. match: 243 msec), while the null pronoun conditions do not differ (mismatch: 237 vs. match: 240 msec), suggesting that the native speakers have waited to assign the overt pronoun to this position. // In contrast, the non-native speakers show a very different pattern. First-pass reading times show a mismatch cost for both null and overt pronouns at the subject-plus auxiliary region (null: 385 vs. 362 msec ; overt: 379 vs. 355 msec), resulting in a main effect of matching, with no reliable differences at the object region. Therefore, the non-native speakers try to assign both pronoun types immediately to the main clause subject. // To conclude, the native speakers follow the PAS with both null and overt pronouns, while the non- native speakers violate the PAS with overt pronouns, possibly relying on their L1 processing strategies and possibly demonstrating some general properties of L2 processing. Additionally, the native speaker results for the overt pronoun show that a discourse constraint (i.e. the PAS) can lead the processor to ignore the first structurally licit potential antecedent position (i.e. the main clause subject) during active dependency formation. // REFERENCES Carminati, M. N. (2002). The processing of Italian subject pronouns. Unpublished PhD thesis, University of Massachusetts at Amherst. Filiaci, F. (2011). Anaphoric preferences of null and overt subjects in Italian and Spanish: A cross- linguistic comparison. Unpublished PhD thesis, The University of Edinburgh.

Izvorni jezik
Engleski

Znanstvena područja
Filologija

POVEZANOST RADA