Abstract
We apply experimental data to identify contexts in which Russian aspect is categorically determined (‘categorical contexts’) and contexts in which both aspects can appear (‘overlapping contexts’). Traditional analyses suggest that these contexts are distinct and that certain features are typical for each type of context. Such analyses tend to rely on constructed examples typically consisting of a single sentence. Our experiment, by contrast, is based on extensive (over 1100 words each) authentic texts (created by and for native Russians), and the identification of contexts is not a priori, but emerges from contexts normed by over 500 native speakers. We present two main findings: 1) the distinction between categorical contexts and overlapping contexts is scalar, not discrete; and 2) a multitude of factors distinguish the two types of contexts, modality in particular is prominent in overlapping contexts, a factor that has received little prior attention. Our data both confirms and significantly extends previous analyses by presenting empirically justified factors that distinguish categorical from overlapping contexts.
Аннотация
В статье на основе экспериментальных данных рассмотрено два типа контекстов: контексты, в которых русский глагол можно употребить только в одном виде, в совершенном или несовершенном (т.е. где вид задан однозначно, без вариантов, или ‘категорические контексты’) и контексты, в которых могут использоваться оба вида глагола (т.е. где области употребления совершенного и несовершенного видов пересекаются, или ‘пересекающиеся контексты’). Традиционно в аспектологии принято различать данные типы контекстов, считая, что для каждого типа свойственны свои характеристики. Подобные выводы до сих пор основывались в основном на искусственно сконструированных примерах, как правило, состоящих из одного предложения. В проведенном нами эксперименте, напротив, были использованы большие (более 1100 слов каждый) аутентичные тексты (созданные носителями русского языка и для носителей русского языка), где разделение контекстов не было задано изначально, а было выявлено на основе ответов более 500 информантов. В статье представлено два главных вывода: 1) категорические и пересекающиеся контексты не являются дискретными, отдельными друг от друга явлениями, а образуют единый континуум, шкалу; и 2) разграничение двух типов контекстов возможно благодаря целому ряду факторов. В пересекающихся контекстах особенно выделяется роль модальности, до сих пор недостаточно изученной как фактор выбора вида. Собранные нами данные подтверждают и расширяют результаты предшествующих исследований за счет описания эмпирически обоснованных факторов, позволяющих разграничить категорические и пересекающиеся контексты.
This is a preview of subscription content, access via your institution.
Notes
Cf. Dickey (2000, chap. 2, pp. 49–94). Note the perfective forms expressing capacity for potential actions in this example from V. Orlov’s novel Al’tist Danilov, cited by Nesset (2009, p. 70):
- (i)
Ni Danilovu, ni v sobstvennosti Muravlevym Kudasov ne nužen, odnako oni ego terpjat. […] Vse ravno on pridetpfv, izvinitsjapfv, i sjadetpfv za stol.
‘Neither Danilov, nor in particular the Muravlevs need Kudasov, although they tolerate him. Nevertheless, he comes, excuses himself and sits down at the table.’
- (i)
Cf. Bulygina and Šmelev (1992, p. 109) who offer a similar, but slightly more nuanced interpretation of this type of context, namely that the imperfective refers to a future event that is scheduled to occur, and this fact is about the present tense regardless of whether the event ever actually does occur.
‘Morphologically possible’ here refers to wordforms of the opposite aspect that would preserve the lexical meaning, tense, and modality of the original form. This means that certain types of verbs and verb forms were not tested in this experiment, among them perfectiva and imperfectiva tantum verbs, participles, gerunds, and -sja passives.
We are aware of the potential drawbacks of assigning numerical values to Likert-scale evaluations such as the one used in our study, particularly the fact that the distances between the evaluations on the scale might not be equal. However, there is growing evidence that the results of statistical tests where these values are treated as interval data yield very similar results to tests in which they are treated as ordinal data (see Endresen and Janda 2016 for examples and extensive theoretical discussion of this issue). Furthermore, the large size of our study increases the likelihood that individual differences among participants would be corrected for by the sheer mass of data. While Fig. 1 inevitably presents some distortion of the data, the overall picture of a continuum is accurate. Furthermore, we use the numerical scores only for the purpose of visualization here, not as input into any statistical model.
While at first glance it seems that there are three groups defined by the three quadrants that contain data, upon further reflection, we can justify conflating the top-left and bottom-right quadrants. These two quadrants share the feature of having one aspect that is rated highly by most participants vs. the opposite aspect that is dispreferred by most participants. The only difference between the two is that in the bottom-right quadrant our participants agreed with the author of the original text, rating his / her choice also as the only one that is highly preferred, whereas in the top-left quadrant the majority of participants disagreed with the author of the original text. If we regard the author’s choice as just one rating among many, then the top-left and the bottom-right belong to the same group, namely the group of test items for which the choice of aspect is more or less categorical. However, since only 2% of the data appears in the top-left quadrant and most of that data is not strongly categorical, we leave those contexts aside in the remainder of this study.
Cohen’s \(d = |0.6|\) is interpreted as a medium-large effect size, and it means that although 76% of the two groups overlap, 73% of the items in the mini-quadrant lie above the mean for the top-right quadrant, and there is a 66% chance that an item taken at random from the mini-quadrant will have a value higher than that of an item taken at random from the top-right quadrant. For interpretation of Cohen’s d, see https://rpsychologist.com/d3/cohend/.
The text each example comes from is indicated by an abbreviation cited in Table 1.
The acceptability of the imperfective stojal ‘stood’ is likely enhanced when participants give positive ratings for the imperfective alternatives for other verbs in the sentence as well. In other words, stojal ‘stood’ looks better when ‘excellent’ or ‘acceptable’ ratings are given to these underlined imperfectives: …kotoromu prostostanovilos’ploxo, onvyzyvalskoruju, [originalpostojal / non-originalstojal], ždal. Participants rated both perfective and imperfective options for all verbs where both were morphologically possible. All of these choices are not represented in citations of the examples in this article in order to improve readability. The presence of chains of verbs is a fact of authentic Russian that was included in our experiment. However, the specific influence of these chains on the rating of aspectual forms is a topic for further research and as such goes beyond the scope of the present article.
Šmelev, A., & Zaliznjak, A. (2006). Aspect, modality, and closely-related categories in Russian. Paper presented at the Slavic Linguistics Society Conference at Indiana University.
References
Bondarko, A. V. (1971). Vid i vremja russkogo glagola (značenie i upotreblenie). Moskva.
Bondarko, A. V. (2002). Teorija značenija v sisteme funkcional’noj grammatiki. Na materiale russkogo jazyka. Moskva.
Bondarko, A. V., & Bulanin, L. L. (1967). Russkij glagol. Leningrad.
Bulygina, T. V., & Šmelev, A. D. (1992). Identifikacija sobytij: ontologija, aspektologija, leksikografija. In N. D. Arutjunova (Ed.), Logičeskij analiz jazyka. Modeli dejstvija (pp. 108–114). Moskva.
Dickey, S. M. (2000). Parameters of Slavic aspect. A cognitive approach. Stanford.
Divjak, D. (2009). Mapping between domains. The aspect–modality interaction in Russian. Russian Linguistics, 33(3), 249–269.
Endresen, A., & Janda, L. A. (2016). Five statistical models for Likert-type experimental data on acceptability judgments. Journal of Research Design and Statistics in Linguistics and Communication Science, 3(2), 217–250.
Forsyth, J. (1970). A grammar of aspect. Usage and meaning in the Russian verb. Cambridge.
Jakobson, R. (1971[1932]). Zur Struktur des russischen Verbums. In R. Jakobson (Ed.), Selected Writings. Vol. II: Word and language (pp. 3–15). The Hague, Paris.
Janda, L. A. (2004). A metaphor in search of a source domain: the categories of Slavic aspect. Cognitive Linguistics, 15(4), 471–527.
Janda, L. A., & Lyashevskaya, O. (2011). Grammatical profiles and the interaction of the lexicon with aspect, tense and mood in Russian. Cognitive Linguistics, 22(4), 719–763.
Janda, L. A., & Reynolds, R. J. (2019). Construal vs. redundancy: Russian aspect in context. Cognitive Linguistics, 30(3).
Maslov, Ju. S. (2004[1948]). Kategorija soveršennogo / nesoveršennogo vida v slavjanskix jazykax. In Ju. S. Maslov, Izbrannye trudy: Aspektologija. Obščee jazykoznanije (pp. 71–141). Moskva (First published in Izvestija AN SSSR. Otdelenie literatury i jazyka, 7(4), 303–316. 1948.)
Nesset, T. (2009). Metonymy of aspect / aspects of metonymy. Scando-Slavica, 55, 65–77.
Padučeva, E. V. (1996). Semantičeskie issledovanija. Semantika vremeni i vida v russkom jazyke. Semantika narrativa. Moskva.
Reynolds, R. J. (2016). Russian natural language processing for computer-assisted language learning (Doctoral Dissertation, UiT The Arctic University of Norway). Tromsø.
Šatunovskij, I. B. (2002). Nesoveršennyj vs. soveršennyj vid v imperative (k probleme načala). In N. D. Arutjunova (Ed.), Logičeskij analiz jazyka. Semantika načala i konca (pp. 267–309). Moskva.
Šatunovskij, I. B. (2009). Problemy russkogo vida. Moskva.
Timberlake, A. (2004). A reference grammar of Russian. Cambridge.
Trubetzkoy, N. S. (1931). Die phonologischen Systeme. Travaux du Cercle Linguistique de Prague, 4, 96–116.
Trubetzkoy, N. S. (1939). Grundzüge der Phonologie (Travaux du Cercle Linguistique de Prague, 7). Prague.
Zolotova, G. A., Onipenko, N. K., & Sidorova, M. Ju. (2004). Kommunikativnaja grammatika russkogo jazyka. Moskva.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Janda, L.A., Endresen, A. & Reynolds, R.J. Aspectual opposition and rivalry in Russian are not discrete: new evidence from experimental data. Russ Linguist 43, 249–271 (2019). https://doi.org/10.1007/s11185-019-09217-7
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11185-019-09217-7