Predicate Agreement Research Articles

<i>Pridёt Dvoje iz Larca:</i> an Experimental Study of Variable Agreement with Quantified Subjects

This paper presents an experimental study of predicative agreement in person and number with two types of quantified subjects containing a personal pronoun. Using two syntactic experiments with a rating task (the Likert scale), we compared the properties of the elective (dvoje iz nas ‘two of us’) and nominative (my dvoje ‘us two’) constructions regarding the preferred form of the verb in two configurations: with the preverbal and the postverbal position of the subject. Ordinal numbers (from dvoje ‘two of’ to semero ‘seven of’) and the quantifier word vse ‘all’ were compared as well. The results show that word order has a significant impact on the acceptability of different agreement strategies. For the elective construction, with the SV order, agreement by person and number of the quantifier is available, and the default agreement is rated as significantly less acceptable, while with the VS order, there are no significant differences between the ratings of the two agreement options. For a nominative construction with SV, only agreement by person and number of the pronoun is possible, while with VS, all strategies are rated as acceptable.

Variable agreement and the structure of coordination

Modern formal approaches to syntax propose different underlying structures for coordinate constructions. At the same time, none of them describe the entire set of properties of coordination, including variable predicate agreement in number (and gender). The goal of the paper is to define the internal organization of coordinate phrases. As the study of examples from the Russian treebank demonstrates, in addition to the previously known factors, the choice of agreement strategy is also influenced by the linear length of conjuncts and the distance between them and the verbal head. Additionally, the strategy of matching the nearest (and not necessarily the first) conjunct is relevant for Russian. We present and analyze the corpus data and conclude that there is no hierarchical organization in coordination. This lets us elaborate on some general ideas about syntactic derivation put forward by Noam Chomsky.

Symmetricity, postposition, and hydras: Experimental evidence against the ATB-analysis of partial agreement in Russian

We present the results of acceptability judgement study focused on partial predicate agreement in Russian. We test the predictions of a possible analysis of partial agreement assuming coordination on VP level with subsequent Across-the-Board movement of verbal heads to a structurally higher position. This analysis implies restrictions on partial agreement in the case of symmetrical predicates, SV word order, and coordinated subject being a head of a relative clause. Our experimental data support none of these predictions. We propose that this result argues in favor of the DP-coordination analysis of partial agreement in Russian.

The role of syncretism in the predicate agreement with coordinate constructions with the Russian correlative conjunction i…i

There are two strategies of agreement with coordinate constructions in Russian: standard agreement, in which the predicate takes plural form, and partial agreement in which the predicate agrees with one of the conjuncts. Both subjects having the same gender and number characteristics is one of the factors that make partial agreement more plausible. We explore whether in this case the acceptability of partial agreement is increased due to the syncretism effect. To study this issue, an experimental study was conducted. Our main hypothesis stated that the acceptability of partial agreement will be higher for the stimuli with the conjuncts of the same gender. The Russian correlative conjunction i…i was chosen for the stimuli as its semantic features are similar to the typical ones of constructions with partial agreement. The design included two independent variables (number of predicate and combination of conjuncts’ genders). Likert scale 1–7 was used to measure acceptability. The results show that matching features of conjuncts influence the acceptability of partial agreement in Russian. Although matching features significantly increase the acceptability of partial agreement, this agreement strategy is possible in non-syncretic sentences as well, what makes the observed effect different from the cases of resolution by syncretism. The possibility of partial agreement with coordinate constructions compound of conjuncts of different gender may also be interpreted as an argument in favour of post-syntactic model of agreement.

Agreement with coordinated subjects containing a personal pronoun: Experimental data from Russian

The article examines Russian predicative agreement with a coordinated subject containing personal pronouns. In cases when the conjuncts have different person features, the choice of person agreement is based on the person hierarchy; it is considered to be an almost universal mechanism. However, the corpus and experimental data presented in this article show that the person hierarchy is not strictly imposed in Russian. We consider the two types of coordinated subjects, mismatched in person: 1) я и Х ‘I and X’ / Х и я ‘X and I’, where X is a masculine proper name, and 2) я и ты ‘I and you’ / ты и я ‘you and me’. Six experiments were conducted using 1–7 Likert scale. The experiments presented sentences with different word order (SV(О) / (О)VS) and conjuncts (I and X / X and I / I and you / you and I). In addition, we included the predicate characteristics as separate factors in the experiments – temporal characteristics, unaccusativity/unergativity. In all experiments, 1st person plural agreement (for non-past tense) and plural agreement (for past tense) were rated most acceptable. Sentences with such agreement are also the most frequent in the collected corpus material. With (O)VS word order, first conjunct agreement also becomes possible. In non-past tense, 3rd person plural agreement is also acknowledged, even in experiments, in which none of the conjuncts bears the 3rd person feature. Examples with such agreement have also been found in the corpus. The acceptability level of possible agreement strategies depends on such factors as word order, conjuncts order, and agreement characteristics of the predicate

A network of allostructions: quantified subject constructions in Russian

AbstractThis article contributes to Construction Grammar, historical linguistics, and Russian linguistics through an in-depth corpus study of predicate agreement in constructions with quantified subjects. Statistical analysis of approximately 39,000 corpus examples indicates that these constructions constitute a network of constructions (“allostructions”) with various preferences for singular or plural agreement. Factors pull in different directions, and we observe a relatively stable situation in the face of variation. We present an analysis of a multidimensional network of allostructions in Russian, thus contributing to our understanding of allostructional relationships in Construction Grammar. With regard to historical linguistics, language stability is an understudied field. We illustrate an interplay of divergent factors that apparently resists language change. The syntax of numerals and other quantifiers represents a notoriously complex phenomenon of the Russian language. Our study sheds new light on the contributions of factors that favor singular or plural agreement in sentences with quantified subjects.

Verbal agreement in number in Iron Ossetic

This paper describes variability in Ossetic predicate agreement in number with different kinds of semantically plural subjects. Drawing upon the materials available in Ossetic corpora, as well as elicitation data, I come to a number of conclusions which cannot be found in previous literature. Transitivity proves to be the most significant factor determining number agreement in Ossetic. Only intransitive predicates allow non-agreement in number, and non-agreeing subjects are required to satisfy the conditions of indefiniteness and syntactic focality.

Оценка языковой способности нейронных моделей на материале предикативного согласования в русском языке

This study investigates the linguistic competence of modern language models. Artificial neural networks demonstrate high quality in many natural language processing tasks. However, their implicit grammar knowledge remains unstudied. The ability to judge a sentence as grammatical or ungrammatical is regarded as key property of human’s linguistic competence. We suppose that language models’ grammar knowledge also occurs in their ability to judge the grammaticality of a sentence. In order to test neural networks’ linguistic competence, we probe their acquisition of number predicate agreement in Russian. A dataset consisted of artificially generated grammatical and ungrammatical sentences was created to train the language models. Automatic sentence generation allows us to test the acquisition of particular language phenomenon, to detach from vocabulary and pragmatic differences. We use transfer learning of pre-trained neural networks. The results show that all the considered models demonstrate high accuracy and Matthew's correlation coefficient values which can be attributed to successful acquisition of predicate agreement rules. The classification quality is reduced for sentences with inanimate nouns which show nominative-accusative case syncretism. The complexity of the syntactic structure turns out to be significant for Russian models and a model for Slavic languages, but it does not affect the errors distribution of multilingual models.

Subject-verb Agreement in Constructions with Quantifiers in Russian

The agreement of subject and predicate in Russian is actually much less trivial than it might seem at first glance. This paper deals with the case when the subject is realized by a combination of a noun with a quantifier. I analyze a set of examples with the words двое, трое, пара, тройка, десяток, сотня, тысяча, миллион and миллиард where there is a variation in predicate number agreement. Using Random Forest, CIT and Logistic Regression algorithms I prove that collective (двое, трое) and non-collective (пара, тройка, десяток, сотня, тысяча, миллион, миллиард) quantifiers exhibit different patterns of agreement. The first group tends to trigger more plural agreement, while for the second one singular agreement is more typical. Moreover, the quantifier phrase position relative to the predicate can also influence the choice of number marker on the verb.

Optional agreement as successful/failed Agree

AbstractWe investigate optional predicate agreement in Santiago Tz’utujil (Mayan). Several generalizations emerge: (i) inanimate arguments base-generated as complements control agreement optionally; (ii) some animate arguments base-generated as complements control agreement optionally; (iii) all arguments base-generated as specifiers control full agreement obligatorily. We propose that two conditions must be met for the operation Agreeto succeed, resulting in the exponence of all the features of the agreement controller. First, a goal must be visible (bear the right feature). Second, a goal must be accessible (be in the right structural position). If one or both conditions are not met, Agreefails, but the derivation converges and 3sgagreement is exponed. While Agreeis deterministic, surface optionality arises when the operation fails. We use optional agreement to diagnose the syntactic structure of understudied constructions in Mayan (nominalizations, Agent Focus). We discuss microvariation, highlighting methodological considerations that arise when assuming an I-language approach.

Developmental stages challenging cross-linguistic transfer: L2 acquisition of Norwegian adjectival agreement in attributive and predicative contexts

This study presents cross-sectional data on adjectival agreement in second-language (L2) learners of Norwegian with four different first languages (L1s). The target language has full noun phrase agreement between article, adjective and noun, and the source languages represent different agreement conditions, similar to or different from the target language. Sixteen learners participated in the study, and their oral production of adjective agreement was analysed individually. Two hypotheses were proposed. First, learners will develop adjectival agreement in a piecemeal way and follow the developmental stages predicted by Processability Theory (Pienemann, 1998), with attributive and predicative agreement implicationally ordered. Second, learners with adjective agreement in the L1 will transfer that into the L2, whereas learners without agreement in the L1 will not use agreement. Under the first hypothesis, we expect the learners to be distributed along a developmental scale, with some learners applying agreement in attributive positions only and others applying agreement in both the attributive and predicative positions. Under the second hypothesis, we anticipate a difference between the groups: Learners with agreement in their L1 will mark agreement in all contexts where it occurs in the L1, whereas learners who do not have agreement in their L1 will fail to mark agreement overall. The comparison demonstrates larger differences within the L1 groups than between the L1 groups. This suggests a gradual acquisition of agreement, with the agreement features and positions emerging one by one rather than being transferred from the L1.

Czech analytic agreement and the characteristics of littlepro

AbstractThis study deals with subject predicate agreement in Czech and the structure of a null subject. First, it introduces the concept of Null Subject Languages as it has been used in a generative framework. Then it shows the complex feature content of the Czech subject-predicate agreement morphology, arguing that the data repeatedly signal a division of the agreement features into two clear cut feature sets and two distinct agreement domains: (a) the lower, lexical domain insidevP containing the nominal n[±φ]feature set (Gender and N-Number), and (b) the higher, functional domain on the T head, which contains the functional D[±φ]feature set (Person and D-Number). The study relates the analysis to the present day microparametric concept of a Null Subject Parameter as in, e. g., the studies in Biberauer (2010), accepting the hypothesis of the checking of the subject-predicate agreement with two levels of predicate (separate headsvand T). It proposes that with pronominal subjects, the φ feature values can be re-set with respect to the speech act information located in the Aboutness Topic. Given the several levels of agreement, each of which can be licensed by a zero morpheme, the study proposes thatprois a complex entity consisting of two separate parts, each endowed with a distinct set of φ features.

Resolving polite conflicts in predicate agreement

When an agreement controller contains conflicting semantic and formal features, the predicates that agree with it can differ with respect to the type of features they reflect. With the second person pronoun used for politely addressing a single referent, finite verbs always agree in (formal) plural number, while participles in some languages may or must agree in (natural) singular number and gender. Building on the claims from the previous literature that the honorific pronoun formally encodes both the grammatical features (plural number and person) and the natural gender and singular number, the analysis proposed in this paper derives the variation in predicate agreement by combining the formal tools of feature geometries, separate probing and variation in the ordering of Agree operations. The main claim advanced by the paper is that predicates differ with respect to the φ-features they probe for, and the order in which this probing applies.

Предикативное согласование со словами «ряд», «половина», «часть», «множество» в современном русском языке

Предикативное согласование со словами «ряд», «половина», «часть», «множество» в современном русском языке

Предикативное согласование с квантитативными сочетаниями, включающими слова половина или треть и показатель приблизительности

This paper focuses on predicate agreement with quantified phrases that include bolee, svyse ‘more’, okolo ‘about’, menee ‘less’ and the nouns polovina ‘a half’, tret’ ‘a third’, and numerals. This study based on data from the National Russian Corpus describes the modern standard of predicate agreement with quantified phrases containing okolo, bolee + poloviny / treti. The data show that a predicate is most likely to occur in the singular with ‘bolee, okolo + numeral’ and in the plural with ‘okolo, bolee + polovina / tret’ ’. The author proposes an answer to the question of why the standards of agreement for quantified phrases with similar structures are different. The factors that influence the form of the predicate are also examined.

Age-Related Differences in ERP Components during Korean Honorific Sentence Processing Depending on Listener’s Social Status and Predicate Agreement

배경 및 목적: 한국어가 가지는 언어형식인 경어법 처리과정을 청년층과 노년층을 대상으로 ERP 실험을 통해 살펴보고자 한다. 방법: 청년층 23명(만20-40세), 노년층 21명(만60-80세), 총 44명이 본 연구에 참여하였다. 실험 문장은 호칭+목적어+서술어 3어절 문장으로, 서술어 종결어미를 변화시켜 구성하였다. 호칭에서 나타나는 높고, 낮은...

The Agreement between Conjoined Subjects and Predicate: Croatian Church Slavonic Corpus Analysis

The abundance of grammatical categories in Slavonic and their overlap are particularly evident in the agreement between conjoined subjects and predicate. When they are accompanied by agreement conditions, such as word order and animacy in Slavic languages, different agreement patterns, dependent also on concrete context and speaker, are to be expected. In this paper the study of the agreement between conjoined subjects and predicate is based on an analysis of the medieval Glagolitic Croatian Church Slavonic corpus. Number, gender, and person are grammatical categories, i. e., features of conjoined noun phrases and predicate agreement. The analysis includes noun phrases conjoined by coordinating and some non-coordinating conjunctions as well as noun phrases conjoined by a gradational ‛not only [. . .] but also’ structure. Comitative and reciprocal noun phrases are included as well. The research in the given corpus shows that the conjoined noun phrases with predicate agreement can be syntactic (predicate showing agreement with one conjunct) or semantic (predicate showing agreement with all conjuncts). Syntactic agreement appears as the so-called contact agreement (predicate showing agreement with the closest conjunct) and as distant agreement (predicate showing agreement with the most distant conjunct). Semantic agreement is applied mostly in accordance with G. G. Corbett’s resolution rules for Slavic languages. However, the analysis shows that some resolution rules for number should be revised due to dual number. Although absent from the majority of contemporary Slavic languages, it is precisely in historical Slavic idioms that dual number reveals its identity, highlighted in agreement study as well.

Predicate Agreement with Quantifier Phrases Containing the Words Polovina or Trett and an Approximative Marker

This paper focuses on predicate agreement with quantified phrases that include bolee, okolo (“about, more than”) and the nouns polovina (a half), tret’ (a third), and numerals. This study based on data from the National Russian Corpus describe the modern standard of predicate agreement with quantified phrases containing okolo, bolee poloviny/treti. The data show that a predicate is most likely to occur in the singular with “bolee, okolo numeral” and in the plural with “okolo, bolee polovina/tret’”. The author proposes an answer to the question of why the standards of agreement for quantified phrases with similar structures are different. The factors that influence the form of the predicate are also examined.

Predicate Agreement with the Quantifier Phrase with Approximate Quantity

This study deals with predicate agreement with quantifier phrases that contain mentioning of an indefinite or an approximate quantity. The author analyzes the frequency of occurrence for singular and plural predicates based on data from the Russian National Corpus during the period of 2000-2010 and investigates factors that influence the predicate choice. The author concludes that the singular agreement is typical in the concerned expressions and is determined mainly by the meaning of indefiniteness. There are several factors that favor plural predicates. Also, some restrictions of predicate variation are revealed.

Diachronic Slavonic syntax: Gradual changes in focus (review)

Reviewed by: Diachronic Slavonic syntax: Gradual changes in focus Mila Vulchanova Björn Hansen and Jasmina Grković-Major, eds. Diachronic Slavonic syntax: Gradual changes in focus. Munich: Sagner, 2010. 208 pp. [Wiener Slawistischer Almanach, 74.] This volume is symptomatic of a recent trend in grammar studies marked by a revived interest in diachronic research from a theoretical perspective. This trend has been visible in the initiative to arrange a forum for linguists working within language diachrony using a formal analytical framework (mostly generative), which came to be established as the Diachronic Generative Syntax Conference (with its 14th meeting in 2012). This kind of work has been supported by the creation and wide availability of numerous electronic resources such as historical corpora, large scale databases of manuscripts for the study of language history. While most recent research has focused on the history of Germanic and Romance languages, with few exceptions (e.g., the recent publication Development of language through the lens of formal linguistics, 2010) research on Slavic diachrony is still relatively limited, sporadic, and scattered. In this respect the current volume is a welcome resource for linguists interested in the development of Slavic languages. It is the result of the conference "Diachronic Syntax of the Slavonic Languages: Gradual changes in focus" held in Regensburg in late 2008 and contains many of the papers presented at that conference, duly reflecting the diversity of topics and ideas aired there. The main organizing idea of the volume is the assumption that the processes that lead to major and observable changes in language are marked by smaller, gradual, and often subtle, transitions, which sometimes even point to conflicting trends. While the contributions all address various aspects of Slavic language history, they reflect a number of theoretical approaches, from formal (generative) to Construction Grammar, language typology, and grammaticalization theory. The book includes 17 papers covering Russian, Czech, and Polish, as well as the almost extinct Ruthenian and Lower Sorbian. Thematically, four of the papers report results and analyze data obtained from language corpora (Bartels, Eckhoff and Haug, Krasovitsky et al., and Rabus), [End Page 299] two address negation (Dočekal and Veselinova). Other papers look at the development of verbal categories such as gerunds (Bjørnflaten) and the perfect (Jung), properties and constituents of nominal expressions (Karlík for a specific category of adjectives and Fried for adnominal participial adjectives), and clause-level categories, such as modal adverbs (Hansen), the syntax of perfect auxiliaries (Migdalski), predicative possession (McAnallen), and reflexive constructions (Lazar). Some papers address more general aspects of diachronic change, such as the grammaticalization of nominal paradigms (Rappaport), the driving forces in the history of Slavic syntax (Večerka), and transitivity and syntactic structure (Grković-Major). The papers that draw on corpus data all share a common theme: the ways in which data extracted from historical corpora may shed light on changes that have been under way or have become visible over shorter (from a diachronic perspective) segments of time (e.g., 50 to 100 years). In this respect, the paper by Krasovitsky, Baerman, Brown, Corbett, and Williams entitled "Morphosyntactic change in Russian: A corpus-based approach" is of particular interest for the topic of the volume. In this paper the authors trace gradual changes in predicate agreement in Russian over the past two centuries. The syntactic contexts included in the survey are base orders, where the subject phrase precedes the verb phrase, expletive (impersonal) structures, where oblique phrases occur in clause-initial position, and sentences where the predicate precedes the (logical) subject. In addition, the semantic nature of the referents is taken into consideration, and animate and inanimate phrases are reported separately in the statistical analyses. Based on data from a large corpus of literary works, the following intriguing generalizations emerge: a radical increase of plural agreement with conjoined NP subjects; a similar, even farther reaching, change for quantified expressions containing the lower numerals (dva 'two', tri 'three', and četyre 'four'), but a very different pattern in the case of nominal expressions quantified by higher numerals (such as pjat' 'five' and above and quantifiers such as neskol'ko 'several' and malo 'few'. These patterns can be explained by the nature of the quantity...