Диссертация (1136198), страница 15
Текст из файла (страница 15)
In total,these findings let us reconsider the question which properties of features are crucial foragreement attraction in production and in comprehension.Keywords: agreement, gender, attraction, production, comprehension, Russian139November 2016 | Volume 7 | Article 1651Slioussar and MalkoGender Agreement Attraction in Russian1. INTRODUCTIONto be the marked value of number feature2 , and the asymmetryis attributed to the fact that attractors with a marked feature aremore disruptive. Hence it is known under the name of “pluralmarkedness effect.”However, the concept of markedness is not widely agreedupon.
Different authors adopt different theoretical approachesand different tests to determine marked and unmarked featurevalues [including frequency, presence of a non-zero affix, defaultuse of a form (e.g., in impersonal sentences), various semantictests etc.; see Haspelmath, 2006]. It is impossible to evaluatethem looking only at singular and plural.
To figure out whichof these properties may be relevant for the asymmetry betweenfeature values (and whether it makes sense to attribute it tomarkedness in a particular theoretical framework), it is crucialto look at other features systems. As we will show below,Russian gender is interesting in this respect because the resultsof different markedness tests do not converge, letting us teaseseveral approaches apart.1.1. The Phenomenon of AgreementAttractionGrammatical agreement is one of the most basic linguisticoperations.
It is well-known, however, that it is not alwaysaccurate. In the last 20 years many studies have looked at socalled agreement attraction errors, exemplified in (1). In (1a) theverb agrees not with the head of the subject NP key1 , but withanother, embedded NP cabinets (we will further call such NPs“attractors”). In (1b) the verb in a relative clause agrees with thesubject of the matrix clause.(1)a. The key to the cabinets were rusty (Bock and Miller,1991).b.
The musicians who the reviewer praise so highly willprobably win a Grammy (Wagers et al., 2009).Agreement attraction errors are observed in spontaneousspeech and in well-edited texts. They have also been studiedexperimentally, mostly in English, but also in French, Spanish,Italian, Dutch, German, and some other languages (Bock andMiller, 1991; Vigliocco et al., 1995, 1996; Pearlmutter et al., 1999;Anton-Mendez et al., 2002; Hartsuiker et al., 2003, to name justa few). The first accounts suggested that the verb simply agreeswith the linearly closest noun (Jespersen, 1924; Quirk et al., 1972;Francis, 1986, a.o.).
However, later studies demonstrated thatagreement attraction is a structural phenomenon. For example,Vigliocco and Nicol (1998) showed that people make attractionerrors producing questions, e.g., “Are the helicopter for theflights safe?.” Various factors that influence attraction have alsobeen identified. However, the overwhelming majority of studiesfocused on number agreement in the languages where numberhas only two values: singular and plural. It is not clear to whatextent these results can be generalized to other cases.In this paper, we analyze subject-predicate gender agreement.Gender attraction has been investigated only in a few studies,and mostly in Romance languages, which have two genders. Wereport one production and three comprehension experimentson Russian, a language with three genders.
To the best of ourknowledge, this is the first comprehension study looking atagreement attraction in a non-binary category. Below we presentseveral findings from the research on number agreement, whichwill be most important for our study, and different accounts ofattraction. Next, we review the few existing studies on genderattraction, providing rationale for the present work.1.1.2. Parallel Results in Production andComprehensionExperimental studies demonstrated that attraction exists notonly in production, but also in comprehension. In production itmanifests itself as agreement errors. In comprehension attractionerrors have been observed to trigger more grammaticalityjudgment mistakes and to provoke less pronounced effects inreading time and EEG studies than other agreement errors.
Inother words, people perceive ungrammatical sentences as if theywere grammatical or had a minor violation. This is often called a“grammaticality illusion.”The results from production and comprehension are largelyparallel (in particular, significant attraction effects are observedonly with plural attractors).
This is often used to conclude thatthe mechanism of attraction is the same in both modalities. Wewill come back to this problem discussing our findings becausewe did not observe parallelism that we expected based on theprevious studies.1.1.3. Debate on Ungrammaticality IllusionsWe just mentioned that in comprehension, attraction causesgrammaticality illusions, making ungrammatical sentences moreacceptable. Can it also lead to ungrammaticality illusions, andmake grammatical sentences less acceptable? For example, ifpeople tend to miss agreement errors in sentences like (2a), dothey sometimes see non-existent errors in sentences like (2b)?As we show below, different approaches to attraction makeopposing predictions about ungrammaticality illusions, so this isan important question.1.1.1.
Plural Markedness EffectIn all studied languages, attraction effects were found to beasymmetric. They can be observed when the head is singular,and the attractor is plural [as in (1) above], but are muchweaker or virtually non-existent in the opposite configuration.In the majority of agreement attraction studies, this asymmetryis explained in terms of feature markedness.
Plural is assumed(2)Several studies (e.g., Nicol et al., 1997; Pearlmutter et al., 1999)suggested that ungrammaticality illusions do arise. However,Wagers et al. (2009) demonstrated that at least on-line findingsmay be artifactual (they might be due to the fact that processing1 Here and further, the following standard symbols are used: N, noun; NP, nounphrase; P, preposition; PP, prepositional phrase; V, verb; M, masculine gender; F,feminine; N, neuter.Frontiers in Psychology | www.frontiersin.orga.
The key to the cabinets were rusty.b. The key to the cabinets was rusty.2 Notably, in semantics there is an ongoing debate whether singular or plural is thedefault (e.g., Sauerland et al., 2005; Farkas and de Swart, 2010).240November 2016 | Volume 7 | Article 1651Slioussar and MalkoGender Agreement Attraction in Russianplural nouns carries an additional cost compared to singularones, not to any aspects of subject-verb agreement processing).This hypothesis can be tested by analyzing some cases wherethis problem does not apply, and we do so in the present studylooking at gender agreement 3 .The more plural the subject NP, the higher the possibility ofchoosing a plural verb.
In such accounts there is no way to avoidungrammaticality illusions: if the agreement controller can bemis-construed or ambiguous, there is no way to restrict such misconstruals to only ungrammatical sentences. They happen evenbefore we encounter the verb, i.e., even before it is clear whetherthe sentence is or is not grammatical.Now let us turn to the retrieval account (Solomon andPearlmutter, 2004; Lewis and Vasishth, 2005; Badecker andKuminiak, 2007; Badecker and Lewis, 2007; Wagers et al., 2009;Dillon et al., 2013).
Research on memory suggests that theamount of material a person can hold in a ready-to-process stateis extremely limited (McElree, 2006; Cowan, 2001). Thus, it canbe hypothesized that when we reach an agreeing predicate, thesubject needs to be reactivated. This reactivation can be done viaso-called cue-based retrieval (Lewis and Vasishth, 2005; McElree,2006): we query the memory with a set of cues (e.g.,“number:plural,” “case: nominative” etc.) and select an element thatmatches the maximum number of cues.This process is not error-free, and the retrieval account arguesthat attraction arises at this stage.
![](https://s.studizba.com/z.php?f=/templates/si/i/noavatar.png&w=100&h=100&t=1"){kind=avatar}
