Prepositions: a Corpus-Based Study Dissertation

Home , 4C (Person of Interest), Deverbal noun, Razgovor, Sibirjak

The Semantics of Russian 'About' Prepositions: A Corpus-Based Study Dissertation

Presented in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in the Graduate School of The Ohio State University

Andrew James Kier, B.M., M.A., M.A. Slavic and East European Languages and Literatures The Ohio State University 2013

Dissertation Committee: Daniel Collins, Advisor Charles Gribble Brian Joseph

Abstract

This study examines, for the purpose of comparison, the semantic content of several important prepositions in modern Russian which translate ‘about, concerning’. It focuses primarily on the two primary prepositions o+LOC and pro, and includes the two secondary prepositions nasčet and po povodu, which also translate ‘about, concerning’. Historically, several scholars have acknowledged that o+LOC and pro express slightly different nuances, although the exact nature of these nuances has become less clear over time (Peškovskij 1928/2001,

Vinogradov 1947/2001). How has pro survived and remained productive, if the distinction between it and o+LOC has been fading for so long? Vinogradov predicted its demise in the 1940’s, yet it still enjoys a robust presence in both the spoken and written language of the present day. Furthermore, pro in the ‘about’ sense is historically neither rare nor limited to the recent past. Likewise, nasčet and po povodu (both fairly recent additions to the stable of Russian prepositions), which have both undergone subtle lexical shifts of their own in the past few

centuries (Čerkasova 1967), are also in need of a more articulated semantic description.

In this study, which consists of both a qualitative and a quantitative study, I have identified several selectional restrictions—the types of verbs and nouns

(and features within them) with which these prepositions occur—associated with the four ‘aboutness’ prepositions, in sample texts written between 1890 and

2009, from the Russian National Corpus. The quantitative study consists of

Fisher’s Exact tests performed on approximately 7,000 tokens (which consist of argument phrases with the ‘aboutness’ prepositions). Results indicate that in many instances, pro is associated with nouns which are animate, proper, unmodified, and non-deverbal.

A qualitative analysis of the types of verbs associated with ‘aboutness’

PP’s as complements reveals that pro is mostly limited to verbs whose

‘aboutness’ object referent codes the semantic role of Topic. In contrast, with verbs that depict an emotional state (bespokoit’sja ‘be worried’, etc.), the object referent codes the semantic role of Cause; with verbs that depict requesting or supplication (prosit’ ‘ask (for), request’, umoljat’ ‘beseech’, etc.) the object referent codes the role of Proposition. Object referents of pro, however, rarely

iii

occur in these semantic roles, being generally limited to the semantic role of

Topic; the verbs with which such a referent is associated depict one or more parts of the signal path of a message—speaking, writing, reading, inquiring, hearing (about), or cognitive activity. The performing of activities associated with a Topic object referent do not entail a change of state in any of the participants in the discourse, as is the case with emotional states and requests.

I have also examined ‘co-occurrence’ examples, in which o+LOC and pro

PP’s occur in the same syntagma or sentence, both subordinate to the same verb. Close readings of these co-occurrences show that in many of them, the semantic/pragmatic opposition between the o+LOC and pro object referents is such that the former are presented as being larger in scope, and more analyzed than the pro entities.

Dedication

To Bella: Always in our hearts, always in our memory.

Acknowledgments

I wish to thank my advisor, Professor Daniel E. Collins for his generous support, advice, encouragement, and enthusiasm for this project.

I would also like to thank my wife Alla for her constant encouragement, support, and patience.

Vita

July 27, 1965...... Born—Columbus, Ohio

1990...... Bachelor of Music, Virginia Commonwealth University

1995...... Master of Arts, Library Science, University of Arizona

1998—present...... Cataloger/Metadata Specialist OCLC, Inc., Dublin, Ohio

2006...... Master of Arts, Slavic and East European Languages and Literatures, The Ohio State University

Publications

Kier, Andrew J. (2006) “On The Origins of the Glagolitic Alphabet.” Ohio Slavic Papers: 6, Proceedings of the Second Graduate Colloquium in Slavic Linguistics. Columbus, OH: OSU Dept. of Slavic and E. European Languages and Literatures, pp. 177-189.

--- (2012) “Instruments of the Old Faith: Three Medieval South Slavic Healing Rites for Snakebite.” Poznanskie Studia Slawistyczne: 3. Poznan: Adam Mickiewicz University.

Fields of Study

Major Field: Slavic and East European Languages and Literatures

vii

Table of Contents

Abstract ...... ii

Dedication ...... v

Acknowledgments ...... vi

Vita ...... vii

List of Tables ...... xiii

List of Figures...... xv

Chapter 1: Introduction ...... 1

1.1 What this Study is about and What Prompted it (with a Preliminary Definition of ‘Aboutness’) ...... 1

1.2 Research Questions ...... 7

1.3 Goals of this Study ...... 8

1.4 Why this Study is Important, and its Transferability to Other Contexts ...... 9

1.5 Overview of Chapters in this Study ...... 10

Chapter 2: Scholarly Literature on the ‘Aboutness’ Prepositions ...... 12

2.1 Scholarly Literature on the Prepositions o+LOC, pro, nasčet, and po povodu ...... 12

2.1.2 O+LOC ...... 12

2.1.3 Pro (+ACC) ...... 19

2.1.4 Nasčet and Po povodu ...... 26

2.1.5 The Question of Style/Register...... 33

viii

Chapter 3: Methodology ...... 37

3.1 Overview ...... 37

3.2 Quantitative Corpus Analysis and the Russian National Corpus ...... 38

3.3 Criteria for Data Collection ...... 48

3.4 The Search Strings...... 50

3.5 The Search Results ...... 55

3.6 Criteria for the Datasets ...... 57

3.7 Parameters for the Statistical Analysis of the Datasets ...... 59

3.7.1 The Structure of Datasets in the Quantitative Study ...... 59

3.7.2 The Nature and Structure of the Quantitative Analysis ...... 61

3.7.3 Issues Pertaining to the Quantitative Study ...... 63

3.7.4 Specific Factors ...... 64

3.7.5 Explanation of Factor Groups and Factors ...... 65

3.7.6 Statistical Modelling in the Study ...... 70

3.7.7 General Limitations of Corpus Data ...... 71

Chapter 4: The Semantics of Prepositions ...... 73

4.0 General Overview of Prepositions ...... 73

4.1 Semantic Functions of Prepositions ...... 73

4.2 Polysemy and Metaphorical Extensions of Prepositions ...... 76

4.3 Russian Prepositions...... 82

4.3.1 Case and Semantics of Prepositions ...... 82

4.3.2 Grammatical vs. Semantic Case ...... 84

4.3.3 Semantic Roles ...... 85

4.4 The Area/Topic and Cause Extensions of ‘Aboutness’ Prepositions ...... 86

Chapter 5: Verbs/Situations Associated with ...... 89 the Four ‘Aboutness’ Prepositions...... 89

5.0 Overview ...... 89

5.1 Verbs of Speech/Oral Communication ...... 92

5.2 Verbs of Asking ...... 94

5.2.1 Prosit’/Poprosit’ with an ‘Aboutness’ PP ...... 96

5.2.2 Other Activities Related to Asking/Requesting ...... 97

5.3 Verbs of Cognition ...... 99

5.4 Verbs of Reading ...... 104

5.5 Verbs of Writing ...... 105

5.6 Verbs of Hearing ...... 106

5.7 Verbs of Emotional State (Psychological Verbs) ...... 108

5.8 A Few Other Verbs Used with ‘Aboutness’ Prepositions ...... 111

5.10 Summary ...... 116

Chapter 6: Quantitative Analysis of RNC Data: Results and Discussion ...... 120

6.1 Introduction ...... 120

6.2 Tests Between Verbs Plus Individual Prepositions ...... 121

6.3 Tests Between Aspectually Paired Verbs Plus Prepositions ...... 126

6.4 Verbs of Speech ...... 128

6.4.1 Verbs of Speech with Individual Prepositions ...... 128

6.4.2 Aspectually Paired Verbs of Speech ...... 129

6.5 Verbs of Writing ...... 131

6.5.1 Verbs of Writing with Individual Prepositions ...... 131

6.3.2 Aspectually Paired Verbs of Writing ...... 132

6.4 Verbs of Asking ...... 133

6.4.1 Verbs of Asking with Individual Prepositions ...... 133

6.4.2 Aspectually Paired Verbs of Asking ...... 134

6.5 Verbs of Cognition ...... 137

6.5.1 Verbs of Cognition with Individual Prepositions ...... 137

6.5.2 Aspectually Paired Verbs of Cognition ...... 141

6.5.3 The REMIND1 and REMIND2 Senses of Napominat’/napomnit’...... 142

6.6 Verbs of Hearing ...... 146

6.6.1 Verbs of Hearing, with Individual Prepositions ...... 146

6.6.2 Aspectually Paired Verbs of Hearing ...... 147

6.7 Verbs of Reading ...... 148

6.7.1 Verbs of Reading, with Individual Prepositions ...... 149

6.7.2 Aspectually Paired Verbs of Reading ...... 149

6.8 Summary ...... 150

6.8.1 First Series of Tests (Verbs with Individual Prepositions) ...... 150

6.8.2 Tests on Aspectual Verb Pairs (Second Series of Tests) ...... 153

6.9 Conclusions ...... 153

6.9.1 Pro ...... 153

6.9.2 Nasčet and po povodu ...... 156

Chapter 7: Co-occurrence of ‘Aboutness’ ...... 157

Prepositions in the Same Syntagma ...... 157

7.1 Co-occurrence of o+LOC and pro PP’s in the Same Syntagma ...... 157

7.2 Types of Relations in Co-occurrences...... 160

7.2.1 Co-occurrences in Coordinated Same-Clause PP’s ...... 160

7.3.1 Explicative Relationships between o+LOC and pro Object Referents ...... 172

7.4 Preposition Matching ...... 177

7.5 Summary ...... 183

Chapter 8: Summary and Conclusions ...... 185

8.1 Object of the Investigation ...... 185

8.1.1 Semantic Variation between o+LOC and pro ...... 185

8.2 Stylistic vs. Semantic Factors ...... 186

8.3 Main Results of the Quantitative Study ...... 187

8.4. Main Results of the Qualitative Study...... 188

References ...... 195

Appendix A: ...... 202

Results of Fisher’s Exact Tests on Datasets ...... 202

Appendix B: ...... 219

Tallies of Distribution of Noun Features ...... 219

xii

List of Tables

Table 3.1 Semantic Categories of Verbs in the Quantitative Study ...... 45

Table 3.2 Total Individual RNC Documents in each Time Period...... 49

Table 3.3 RNC Search String for govorit’ + o+LOC...... 50

Table 3.4 RNC Search String for o+LOC + govorit’...... 52

Table 3.5 RNC Search String for govorit’ + pro ...... 53

Table 3.6 RNC Search String for govorit’ + po povodu ...... 55

Table 3.7 Distribution of Total Tokens in each Dataset, by Verb Category ...... 59

Table 3.8 Factor Groups in the Quantitative Study ...... 65

Table 5.9 Object Referents of nasčet/po povodu in ‘Cause’ Role ...... 116

Table 5.10 Verbs that Favor Object Referents with o+LOC ...... 117

Table 6.11 Tests for dumat’ + o+LOC (left) vs. pro (right), by FG ...... 122

Table 6.12 Distribution of Tests for skazat’ + o+LOC and pro (Pd. A), by FG ...... 123

Table 6.13 Distribution of Tokens in First Series of Tests ...... 125

Table 6.14 Arrangement of Tests for čitat’/pročitat’ ‘read’ ...... 126

Table 6.15 Distribution of Significant Results in 2nd Series of Tests ...... 127

Table 6.16 Significant Results for FG Noun Animacy, rasskazyvat’/rasskazat’, o+LOC~pro ..... 128

Table 6.17 Significant Results for Verbs of Speech, by FG and Preposition (1st Series) ...... 130

Table 6.18 Significant Results for Verbs of Writing, by FG and Preposition (1st Series) ...... 132

Table 6.19 Significant Results for Verbs of Asking, by FG and Preposition (1st Series) ...... 135

Table 6.20 Significant Results for Verbs of Asking, by FG and Preposition (2nd Series) ...... 136

xiii

Table 6.21 Significant Factor Groups in Tests with podumat’ o+LOC~pro, Pds. A-D ...... 137

Table 6.22 Significant Factor Groups in Tests with znat’ o+LOC~pro, Pds. A-D...... 138

Table 6.23 Significant Factor Groups in Tests with znat’o+LOC~nasčet, Pds. B, C and D ...... 139

Table 6.24 Increase in Positive Correlation between pro and Animate Nouns, Pds. A-D ...... 140

Table 6.25 Distribution of REMIND1 (Speech) and REMIND2 (Cognition) Senses of napominat’ .... 142

Table 6.26 Distribution of REMIND1 (Speech) and REMIND2 (Cognition) Senses of napomnit’ ...... 143

Table 6.27 Fisher’s Exact Tests for napominat’~napomnit’ with o+LOC ...... 144

Table 6.28 Fisher’s Exact Test for napominat’~napomnit’ + pro ...... 144

Table 6.29 Significant Results for Verbs of Cognition, by FG and Preposition (1st Series) ...... 145

Table 6.30 Slyšat’ o+LOC~pro and slyxat’ o+LOC~pro, Period B ...... 146

Table 6.31 Significant Results for Verbs of Hearing, by FG and Preposition (1st Series) ...... 148

Table 6.32 Significant Results for Verbs of Reading, by FG and Preposition (1st Series) ...... 150

Table 6.33 Positive Correlation between pro and Animate Nouns ...... 151

Table 6.34 Positive Correlation between napisat’ + po povodu and Animate Nouns ...... 152

Table 6.35 Significant Results by Verb Category and FG, in Tests between o+LOC and pro..... 153

Table 6.36 Hierarchies of Noun Features Associated with Individuation (Andersen 2001) ...... 155

Table 7.37 Opposition of Co-occurring pro and o+LOC Object Referents ...... 184

Table 8.38 Verbs, Semantic Roles of Object Referents, and Speech Acts ...... 190

xiv

List of Figures

Figure 3.1 Organization of RNC Datasets in the Quantitative Study ...... 60

Figure 5.3 Simple Model of Message/Signal Transmission (Adapted from Cruse (2004)) ...... 90

Chapter 1: Introduction

1.1 What this Study is about and What Prompted it (with a Preliminary Definition of ‘Aboutness’)

This study examines, for the purpose of comparison, the semantic content of several important prepositions in modern Russian which translate ‘about, concerning’. It will focus primarily on the two primary prepositions o+LOC and pro1:

(1) On govoril o fizičeskix problemax. he-NOM talk-PST.M.SG about physical-LOC.PL problems-LOC.PL

‘He was talking about physical problems.’

(2) I potom ja stal dumat’ voobšče pro and later I-NOM become-PST.M.SG think-INF generally about

žizn’. life-ACC.SG.

‘And later I began to think generally about life.’

1 Pro always governs ACC; o may also govern ACC where it has the meaning ‘up to, against’.

In addition to o+LOC and pro, I am including two secondary prepositions in my discussion and analysis, nasčet and po povodu (both non-primary and governing GEN), which also translate ‘about, concerning’:

(3) Oni vam uže skazali nasčet demagogii. They you-DAT.PL already tell-PST.PL about demagogy-GEN.SG

‘They already told you about demagogy.’

(4) Ja rešila vam napisat’ po povodu I decide-PST.F.SG you-DAT.PL write-INF about

davnix sobytij. long-ago-GEN.PL events-GEN.PL

‘I had already decided to write to you about long-ago events.’

Specifically, I will explore the semantic differences between these four prepositions (though with a focus on the relationship between o+LOC and pro, which are by far the most robustly attested). Although most dictionary definitions of pro, nasčet, and po povodu define them in terms of o+LOC, in actual usage each of these prepositions occupies a slightly different space within the semantic field of ‘aboutness’ (as I will refer to it). My aim is to identify the semantic, as opposed to stylistic, factors which condition their variation. The fact that some words naturally occur in the same environment can be attributed to “a need for semantic coherence: the meaning of virtually any word limits the range of its

possible non-anomalous syntagmatic partners” (Cruse 2011: 82). If meaning entails a choice of one word over another, “it is equally true that meaning entails a limitation in the choice of accompanying items. This selective pressure exercised by a word on its partners is variously referred to as selectional restrictions or selectional preferences” (ibid.). In the present study I will identify the specific selectional restrictions of the four ‘aboutness’ prepositions – the types of verbs and nouns (and features within them) with which these prepositions occur – in sample texts written between 1890 and 2009, from the

Russian National Corpus.

In a verbal argument phrase, an ‘aboutness’ preposition describes a relationship between the object referent of the preposition and the situation described by the verb, in which the referent functions as the topic of the information being conveyed during the situation: talk, write, read about the election. A topic can be defined as the mental abstraction of any set of related facts, events, experiences, etc. concerning an entity or idea. Verbs associated with ‘aboutness’ prepositions typically denote communicative processes

(speaking, writing, reading, asking, etc.) or mental and psychological processes and states (thinking, knowing, forgetting, worrying, etc.). ‘Aboutness’ in prepositions can represent a continuum of meanings, each of which ascribes a different level of intensity or ‘depth’ to the relationship between the situation and

the object referent. The English ‘aboutness’ prepositions about, on, and of are members of one such continuum:

(5) Think about these things. (6) Think on these things. (7) Think of these things.

In (5), about conveys a sense of basic consideration or awareness of these things, while in (6), on conveys a more ponderous attitude; “serious, profound mental activity requires on” (Dirven 1993: 88). In (7), of is antithetical to on; to think of something is simply to be aware of it or recall it in passing. Thus, about represents (in English) a neutral (or unmarked) ‘aboutness’ term, on as an intensified form, and of as attenuated. It is for this reason that I feel ‘aboutness’, which evokes an unmarked, generic sense, is the most suitable term for the present study.2

While most scholarly and pedagogical works on Russian grammar describe o and pro individually, very few acknowledge a relationship between the two beyond that of a generic synonymy. “Although we feel some sort of inner difference between these synonyms, we have difficulty defining exactly what it is”

(Peškovskij 1928: 319).

2 Another possibility is ‘topicality’, which invokes the ‘area/topic’ semantic extension of a preposition. However, as I will explain, several prepositions in this study are not limited to an ‘area/topic’ sense.

Some have placed a style dichotomy between o+LOC and pro (Greč

1827, Vostokov 1874, Ušakov 1935, etc.); pro is sometimes—but not always— labelled as “colloquial” or “prostorečie” (‘simple speech’). The issue of strictly semantic content is rarely invoked, having been squelched in the wake of stylistic pronouncements. Historically, several scholars (which I discuss below) have acknowledged that o and pro express slightly different nuances, although the exact nature of these nuances has become less clear over time (Aksakov 1880,

Peškovskij 1928, Vinogradov 1947/2001). Likewise, nasčet and po povodu (both fairly recent additions to the stable of Russian prepositions), which have both undergone subtle lexical shifts of their own in the past few centuries (Čerkasova

1967), are also in need of a more articulated semantic description.

I became interested in this question after noticing many instances of o and pro occurring within the same discourse, sometimes even within the same sentence, which ruled out the possibility of a purely stylistic explanation.

Additionally, I found abundant evidence of pro PP’s as arguments of verbal participles and gerunds (deverbal adverbs, which are typical of neutral or formal rather than colloquial style. This suggests that traditional explanations which assign a colloquial status to pro simply do not reflect the reality of the situation. I was curious to discover under which specific semantic conditions this variation between o and pro operates. Several years ago, I was introduced to the Russian

National Corpus, presently the largest online Russian language corpus.

Realizing its value as a tool by which the conditioning factor(s) for the variation between o and pro might be revealed, I undertook a pilot project as a preliminary step, which I will discuss in subsequent chapters of this study.

As I will demonstrate in this dissertation, the nuances between o and pro

(as well as the other two prepositions in the study) are semantic in nature, not syntactic; they can occur in the same type of syntactic environment. As structural words, prepositions connect full lexical items (nouns, verbs, and adjectives), imparting the kind of relational information between the constituents of a clause which is characteristic of grammatical rather than lexical morphemes.

The four ‘aboutness’ prepositions in this study share this syntactic function.

The semantic opposition between o and pro has persisted for quite some time, yet according to some (Peškovskij 1928/2001, Vinogradov 1947/2001), this distinction has been in a state of gradual weakening since at least the mid- nineteenth century. How has pro survived and remained productive, if the distinction between it and o+LOC has been fading for so long? Vinogradov predicted its demise in the 1940’s, yet it still enjoys a robust presence in both the spoken and written language of the present day. Furthermore, pro in the ‘about’ sense is historically neither rare nor limited to the recent past; for example, the majority of attestations of pro in the medieval Novgorod gramoty (which date from the late twelfth through the late fourteenth centuries), are in the ‘about’ sense.

The alterity of pro in its relation to o+LOC appears to be grounded in the assumption that the two are stylistically polar opposites, though in reality there has been a great deal of incursion of pro into non-colloquial territory. In fact, the boundaries between “literary” and “colloquial” styles are quite porous. I will present the case that, based on textual evidence, pro is in fact colloquial in name only. Since the nineteenth century, grammars and dictionaries have been inconsistent in how they label pro for style; some label it as prostorečie, and others, razgovornyj ‘colloquial’, while many others do not label it at all.

Therefore, I feel it is time for a reevaluation of this issue. Happily, the availability of an extensive (and diachronic) online corpus has made this possible.

1.2 Research Questions

In this study I will address the following questions:

1. Can a synchronic variation study of syntactic environments in which o+LOC and pro appear (individually) reveal the conditions under which these prepositions alternate?

2. If so, how can we describe and explicate the strictly semantic (as opposed to stylistic) differences between these prepositions?

3. If a semantic difference between o+LOC, pro, nasčet, and po povodu can be demonstrated, how is this difference reflected diachronically? Has their relationship changed over the past century, or has it remained static?

1.3 Goals of this Study

This study, which consists of a qualitative and a quantitative portion, aims to identify and explain the nature of the semantic differences between four prepositions in modern Russian which translate ‘about, concerning’: o+LOC, pro

(+ACC), nasčet (+GEN), and po povodu (+GEN). I will examine the various consistencies and inconsistencies in how this particular topic has been addressed in scholarly literature and works on Russian grammar since the nineteenth century.

The quantitative study consists of a diachronic analysis of randomly selected verbal argument phrases containing these four prepositions, from the

Russian National Corpus3 (RNC), from works created between 1890 and 2009. I believe the RNC is an excellent corpus for a study of this scope; the fact that the

RNC is very extensive, and is available online, makes it ideal for a study requiring a large amount of data and the means by which to extract it rapidly (via the website’s built-in search engine). The qualitative and quantitative analyses

3 Russian National Corpus. Website http://www.ruscorpora.ru. Viewed 11/1/2012.

thus complement each other. It is my intention to fill in some of the existing gaps in the scholarship regarding the lexical similarities and differences which exist between these four prepositions. The emphasis throughout will be on the relationship between o+LOC and pro, which are by far more frequent than nasčet and po povodu.

1.4 Why this Study is Important, and its Transferability to Other Contexts

This study examines the nature of the differences between o+LOC and pro from a standpoint which has long been marginalized and underrepresented in the scholarly literature – that of their semantic, as opposed to stylistic, content.

To my knowledge, the quantitative corpus study included in this work is the first study of its kind to be undertaken with the goal of comparing the semantic environments of these two prepositions. In this way, this discussion offers a new perspective by challenging the traditional prescriptive and stylistically-motivated assumptions, which textual evidence of the past 120 years clearly demonstrates to be obsolete.

Of the many descriptions of o+LOC and pro, very few (Peškovskij

1928/2001, Vinogradov 1947/2001) specifically address the semantic opposition between these two prepositions. Vinogradov 1947/2001 predicted the demise of pro, probably anticipating that its semantic conflation with o+LOC (as he

understood it) would soon render pro redundant. However, it is obvious that it has survived and is thriving at the present time, more than half a century after

Vinogradov’s pronouncement. If we accept that speakers generally do not tolerate exact synonymy, then it is difficult to accept that the two prepositions have conflated in meaning, since clearly, pro maintains a productive existence at the present time.

Prepositional synonymy and nuance is a feature which is not, of course, limited to Russian. Accordingly, this study will be of interest to those studying or researching this phenomenon in other languages, particularly in contexts which involve the perception of stylistic differences between two (or more) terms.

1.5 Overview of Chapters in this Study

Chapter 1 of this study is an outline and explanation of the issues which motivated my research questions. In Chapter 2, I survey scholarly literature on this issue. Chapter 3 is an explanation of the methodology I use for the quantitative study of RNC texts from 1890-2009. Chapter 4 is a discussion of the general semantics of prepositions, focusing on aspects of it which are most salient to this study—metaphorical extensions, the trajector~landmark relationship, the depiction of spatial relations, case, and semantic roles. In

Chapter 5, I examine the types of verbs associated with the four ‘aboutness’ prepositions. Chapter 6 is an explanation of the results of the quantitative analysis of RNC data, and what they indicate. Chapter 7 is a discussion of examples in the RNC in which o+LOC and pro co-occur within the same syntagma. In Chapter 8, I present an overall summary, and discuss the conclusions I have reached.

Chapter 2: Scholarly Literature on the ‘Aboutness’ Prepositions

2.1 Scholarly Literature on the Prepositions o+LOC, pro, nasčet, and po povodu

The four ‘aboutness’ prepositions (o+LOC, pro, nasčet and po povodu) addressed in this study have possessed various senses over the course of time.

The nature of the changes they have undergone sheds light on the roles and functions of these prepositions in the language in its present stage of development.

2.1.2 O+LOC

2.1.2.1 The ‘About’ Sense

The primary preposition o is attested in all the Slavic languages and derives from Proto-Indo-European *ambhi, cf. Greek amphi ‘around, about’, Latin ambi- ‘around, about’ (Watkins 2000: 3). The preposition o can govern LOC or

ACC. With ACC, it indicates movement and contact against a surface:

(1) Ne spotknis’ tut o porog NEG stumble-IMP.2SG here on threshold-ACC.SG

‘Don’t stumble on the threshold here.’

With LOC case, o denotes the subject or focus of communication, of cognition, or of an emotional state: govorit’/dumat’/bespokoit’sja o ‘talk about/think about/be worried about’. This preposition also governs LOC in certain other senses— temporal, quantitative/qualitative, and a few others. However, these senses had all ceased to be productive by the early twentieth century and are preserved only in frozen expressions.

The ‘aboutness’ sense of o+LOC is shared by all the Slavic languages; its archaic nature is evident from the fact that “[o]nly the oldest, primary prepositions take the locative” (Timberlake 2004: 181). Thus, the meaning is attested in

Russian Church Slavonic:

(2) Č’to glagoleši o tebě samem4 what-ACC say-PRS.2SG about you-LOC.SG self-M.LOC.SG ‘What do you say about yourself?’

Old Russian examples can be found from the very beginning of East Slavic literacy, e.g., Novgorod gramoty:

(3) ...ko mne kažit’ o vs’mo

4 John 1:22. In Slovar’ russkogo jazyka XI-XVII vv. Vyp. 12. 1987 (7). (All translations mine.)

to I-DAT report-INF about all-N.LOC.SG ‘... to report to me about everything’ (no. 61, c.1240)

(4) ...popecąlisą o mne look after-IMP.2SG about I-LOC ‘... look after me’ (no. 765, c.1240)

(5) ...ne zabud’ l’va o pozъvě5 NEG forget-IMP.2SG Lev-ANM.ACC about summons-LOC.SG ‘... do not forget Lev in connection with the (judicial) summons’ (no. 122, c.1410)

O+LOC has had its current meaning of ‘about’ for many centuries. In argument phrases, it is associated with verbs which depict the encoding and communicating of oral and written information (govorit’ ‘talk, speak’; pisat’ ‘write’, sprašivat’ ‘query’, etc.) and the decoding of information (slyšat’ ‘hear’; čitat’ ‘read’, etc.). Additionally (and in the same sense), it is associated with verbs of cognition (dumat’ ‘think’; znat’ ‘know’; zabyt’ ‘forget’, vspomnit’ ‘remember’, etc.) and of emotional state (volnovat’sja ‘worry’; bespokoit’sja ‘be bothered’). With these verbs, the NP of o+LOC refers to the topic, information about which is being conveyed (through speech, writing, or requesting) or mentally manipulated

(through thinking, knowing, remembering, etc.) by means of the activity or state.

5 Drevnerusskie berestjanye gramoty. Website http://gramoty.ru. Viewed 2/4/2012.

2.1.2.2 Spatial Sense

Slovar’ russkogo jazyka: XI-XVII vv. (1987 and 1995) lists a total of six separate senses for o+LOC, including its original spatial meaning of

‘around/near’, which denoted circumferential location or proximity:

(6) Iōann imąše rizu svoju ot John-NOM have-IMPFT.3SG coat-ACC.SG own-F.ACC.SG from

vlas vel’blǫžd’ i pojąs ousmen hair-GEN.PL camel’s-M.GEN.PL and belt-ACC.SG leather-ACC.SG

6 ō čreslěx svoix. about waist-LOC.PL own-LOC.PL ‘John had a coat made of camel hair, and a leather belt around his waist.’ (Buslaev 1858: 307)

(7) Novgorod’ci že staša tv’rdo Novgorodian-NOM.PL and become-IMPFT.3PL steadfast-N.SG

o knjazi Romaně. behind Grand Duke-LOC.SG Roman-LOC.SG ‘And the Novgorodians became steadfast beside Grand Duke Roman.’ (ibid.: 319)

6 Matthew 3:4.

2.1.2.3 Temporal Sense

The temporal sense of o+LOC denotes an approximate time period within which an event (specified by the OP) occurs. This construction is typically used with cyclical events such as holidays:

(8) O petrovi dni lonščii7 around Petrov-M.LOC.SG Day-LOC.SG last-M.LOC.SG ‘Around Petrov Day last’ (Novgorod gramota no. 463; c. 1300)

The event can also be non-cyclical (i.e., a one-time event), such as a wedding.

Dal’ (all editions) notes that it indicates a duration of time:

(9) My o Pasxě vsegda v derevně We during Easter-LOC.SG always in village-LOC.SG

byvaem be/visit-PRES.2SG ‘We are always in the village during Easter.’ (Dal’ 1905: 1466)

Ušakov 1935-40 considers this sense to be obsolete and prostorečie.8

7 Drevnerusskie berestjanye gramoty. Website http://gramoty.ru. Viewed 2/4/2012. 8 Ušakov, D., ed. 1934-1940. Tolkovyj slovar’ sovremennogo russkogo jazyka. Website http://ushdict.narod.ru. Viewed 2/4/2012.

2.1.2.4 Quantitative/Qualitative Sense

Another use of o+LOC was to specify the number of parts or attachments to an item. This sense gradually became obsolete by the early twentieth century, surviving in a handful of frozen phrases such as the following:

(10) stol o trëx nogax table-NOM/ACC.SG about three-LOC.PL legs-LOC.PL ‘a three-legged table’

In addition to having an enumerative sense, Dal’ (all editions) provides evidence of o+LOC portraying its object as a purely qualitative attribute:

(11) xram o zolotoj glave church-NOM/ACC.SG about golden-F.LOC.SG cupola-LOC.SG ‘a church with a golden cupola’

(12) peči o gluxix dverkax9 ovens-NOM/ACC.PL about deaf-LOC.PL door-LOC.PL ‘ovens with hermetically-sealed doors’

Thus the quantitative and qualitative functions can be seen as two sides of the same coin; in both senses, the OP referent denotes a salient property or attribute of the host noun. However, the phrases which have survived to the present all portray a quantitative sense:

9 Dal’, V., ed. (1863-1866) Tolkovyj slovar’ živogo velikorusskogo jazyka Vladimira Dalja. (1st ed.) Website http://slovari.yandex.ru/~книги/Толковый%20словарь%20Даля. Viewed 12/15/2011.

(13) palka o dvux koncax stick-NOM.SG about two-LOC.PL tips-LOC.PL ‘a two-edged sword’

2.1.2.5 Other Senses of o+LOC

In addition to the senses of o+LOC discussed above, some scholarly works also mention various other meanings in which this preposition can be used. Most are from biblical or religious (i.e., Church Slavonic) texts:

(14) o xlěbě i vodě on bread-LOC.SG and water-LOC.SG ‘on bread and water’ (Lunt 1969: 31)

(15) o Xristě brat in Christ-LOC brother-NOM.SG ‘a brother in Christ’ (Aksakov 1880: 146)

(16) ...položiti dušu svoju o ljudex put-INF soul-ACC.SG own-F.ACC.SG for people-LOC.PL ‘to lay down His soul for (the sake of) the people’ (Slovar’ russkogo jazyka: XI-XVII vv.: 7)

2.1.3 Pro (+ACC)

2.1.3.1 The ‘About’ Sense

Pro is one of only three prepositions in modern Russian that govern the

ACC exclusively.10 As with o, it has had several discrete senses over the course of its long history. Unbegaun (1957: 274) notes that “[t]he prepositions which govern more than one case are, naturally, those which tend most of all to lose their distinctness of meaning.” It is therefore somewhat ironic that o (which governs both LOC and ACC) currently has several distinct meanings, while pro has shed all of its senses except ‘about’. He asserts that “pro is gradually losing ground to o [+LOC]” (ibid.). Yet in spite of predictions to the effect that pro is

“fading away, having been displaced by o+LOC” (Vinogradov 1947: col.: 558), its demise has not by any means come to pass at the time of writing (2012).

There are around twenty identifiable attestations of pro in the Novgorod birchbark gramoty, sixteen of which represent the ‘about’ sense, three the ‘for/for the sake of’ sense, and one the causal sense (‘because of’). This suggests that

‘about’ has been the primary sense of pro for a long time:

(17) a pro sela sama vesi11 and about villages-ACC.PL self-F.NOM.SG know-PRS.2SG ‘and you yourself know about the villages’ (no. 944, c.1180)

10 The two others are skvoz’ ‘through’ and čerez ‘through, over’. 11 Drevnerusskie berestjanye gramoty. Website http://gramoty.ru. Viewed 2/4/2012.

(18) prikažzivaješi pro ribi give instruction-PRS.2SG about fish-ACC.PL ‘you are giving instruction(s) about the fish’ (no. 99, c.1340) (ibid.)

In most dictionaries and scholarly literature on Russian grammar since the second half of the nineteenth century, ‘about’ is listed as either the primary or the only sense of pro, and is described as being synonymous with o+LOC (Aksakov

1880, Peškovskij 1928, Filin 1961, Ožegov (all editions), Evgen’eva 1984,

Kuznecov 2000, Švedova 2007). An exception is Dal’s Tolkovyj slovar’ (all editions), which gives ‘for (the sake of)’ as the primary sense. Offord (1996: 289) notes that pro is “more or less synonymous with o+prep...”, but does not elaborate on this. Bondarenko (1961: 72-73) does not list them in a table of synonymous prepositions; nor does Russkaja grammatika (1982: 38). A few works indicate that some degree of semantic difference does in fact exist between o+LOC and pro, but they do not elaborate on it (Aksakov 1880,

Peškovskij 1928/2001). Aksakov (1880: 137) asserts that with pro,

[the] nuance of distance and in general, of purpose, is weakening, and pro generally denotes only a moral aspiration to somewhere [ustremlenīe nravstvennoe kuda nibud’], but sometimes even this meaning is weakly felt: On dumaet vse pro svoego druga [‘He always thinks about his friend’]; Pro tebja govorili [‘They were talking about you’].

Vinogradov (1947/2001: 572) mistakenly cites Aksakov’s assertion as pertaining to o+LOC and pro, when in fact, Aksakov was comparing the ‘for the sake of’ and the ‘about’ sense of pro, which I will discuss in 2.1.3.2.

Vinogradov (1947: col.) commented on what he believed to be a nuance distinguishing the two prepositions: pro “expresses a nuance of relation directed not at the essence [suščnost’] of the object (this nuance is a property of the preposition o), but at its surface [poverxnost’], a relation, as it were, which captures the object only externally, revolving around it.” He illustrated this with the following examples:

(19) rasskazat’ o novyx rabotax po fizike tell-INF about new-LOC.PL work-LOC.PL on physics-DAT.SG ‘tell about new works (publications) on physics’

(20) rasskazat’ pro novye raboty po fizike tell-INF about new-ACC.PL work-ACC.PL on physics-DAT.SG ‘tell about new works (publications) on physics’ (ibid.)

By showing the prepositions in two contexts which are identical except for the prepositions themselves, Vinogradov assumes that readers are able to recognize a semantic difference between the two examples, based only upon the substitution of one preposition for the other. However, all this really shows is that

the other non-preposition constituents in (19) and (20) are “normal collocates”

(Cruse 2011: 182) of both o and pro. Assuming that all the other constituent items in (19) and (20) are all construed in the same way, there is nothing other than the prepositions themselves which determines the difference between (19) and (20). Clearly, a proper understanding of the opposition as described by

Vinogradov requires that we determine the selectional restrictions of o and pro which will reveal that opposition. My approach in this study is to examine certain features within the types of nouns and verbs with which these prepositions typically occur, with the goal of identifying the co-occurrence preferences of the latter.

2.1.3.2 Other Senses of pro

If pro ever had a non-metaphorical, spatial meaning in Russian at some point in the past, there is no attested evidence of it. Of the handful of discrete senses for which written evidence does exist, however, all but the ‘about’ sense had become obsolete by the late nineteenth century. The majority of the obsolete senses of this preposition portray the object as the final cause of the activity:

(21) Est’ zapas, no ne pro vas be-PRS.3SG reserve-NOM.SG but NEG for you-ANIM.ACC.PL

‘There is a reserve, but [it is] not for you.’

(22) izgotovit’ kušan’e pro gostej prepare-INF dish/food-ACC.SG for guests-ANIM.ACC.PL ‘prepare a dish for the guests’ (Barsov 17: 225).

Aksakov (1880:137) describes this sense of pro as indicating “a purpose [or] a destination in the future, the relationship of the object as a purpose for the other object. Here, the object becomes the purpose, somewhat distant; but the relationship is direct.” His description of the ‘about’ sense of pro proceeds from this; he perceives the ‘nuance of distance’ between the Agent and the

Beneficiary of the action to be present in its ‘for the sake of’ sense to have weakened in the ‘about’ sense of pro, along with the nuance of a ‘moral aspiration’ of the agent, in carrying out the action.

Pro at one time could also denote ‘in exchange/payment for’:

(23) emu Martynu pro to moe he-DAT.SG Martin-DAT for that-N.ACC.SG my-N.ACC.SG

prodanoe dvorovoe město sell-PTCP.PST.N.3SG yard-N.ACC.SG site-ACC.SG ‘to Martyn, [in exchange] for the plot of land which I sold’ (Slovar’ russkogo jazyka XI-XVII vv.: 94. c. 1659).

(24) mja ieste žalovalě pro svoie

I-ENCL.1SG be-PERF.2PL do a favor-F.PERF.DU for ones-F.ACC.PL

kouně12 money-ACC.PL ‘you have done me a favor for your money’ (no. 949, c. 1360).

At present (2012), the ‘for (the sake of)’ sense of pro survives in only a handful of frozen phrases: pro zapas ‘for reserve’; pro sebja ‘to oneself’ (e.g., čitat’ pro sebja ‘read to oneself’).13 Phrases with this sense “have the character of fixed expressions, for example: kopit’ pro zapas [‘save up for reserve’]; in free use, such constructions are not appropriate to the modern literary language and are archaic.” (Grammatika russkogo jazyka 1960: 180).

In another early usage of pro, the object represented the reason or cause for the activity:

(25) I bjaxu vsi meži soboju v and be-IMPF.3PL all-M.NOM.PL amongst self-REFL.INS in

raznosti i v nenavisti velicě i difference-LOC.SG and in hatred-LOC great-F.LOC.SG and

vraždovaxu drug na druga pro feud-IMPF.3SG other-NOM.SG on other-ANM.ACC.SG about

uděly otčiny knjaženia svoego

12 Drevnerusskie berestjanye gramoty. Website http://gramoty.ru. Viewed 2/4/2012. 13 Although in most cases pro sebja denotes ‘to oneself’, it can also mean ‘about oneself’.

appanages-ACC.PL patrimony-GEN.SG principality-GEN.SG own-

svoego. own-M.GEN.SG ‘And all [brothers] were in difference [of opinion] amongst themselves and in great hatred and bore each other enmity over the appanages belonging to the patrimony of their principality.’ (Slovar’ 1995, vyp. 20: 93)

However, by the nineteenth century, only the ‘about’ and the ‘for (the sake of)’ senses were still productive.

Another sense of pro is found in Western dialects of Russian, described by Dal’:

(26) idti pro ljudi go-INF to people-ACC.PL ‘visit people, be guests’ (Dal’ 1907: col.)

He defines this as “v ljudi, v gosti” [‘be guests’] (ibid.). Normally, ljudi ‘people’

(ACC pl.) would take the animate form ljudej since it is an animate noun with an

ACC-governing preposition. However, the use of the inanimate ACC form resembles the so-called ‘Second Accusative’ (Worth 1998), or ‘inclusive’ form:

(27) idti v gosti go-INF in(to) guests-ACC.PL

‘go (to someone) as guests, be/become guests’

Semantically, the Second Accusative denotes an animate OP as becoming a member of an inclusive group of people such as pilot, soldier, secretary, guest, etc. Although theoretically, any animate noun may be used, “[w]hat is constant is that these expressions always involve becoming something” (Corbett 2008: 19).

As I mentioned above, this presents a syntactic conflict between case and animacy: in idti v gosti, the OP gosti ‘guests’ has the (historically) inanimate form of the ACC plural. “However, the preposition v ‘in(to)’ takes the ACC or LOC.

The ACC plural of animate nouns like letčik ‘pilot’ is as the GEN, that is, letčikov”

(ibid.). The normal ACC pl. of gost’ ‘guest’ is gostej, since it is an animate noun.

However, since the preposition v+ACC cannot be used with an animate noun object, its object takes the inanimate ACC plural form gosti. In Dal’s example, pro follows the same pattern as v+ACC ‘in(to)’, although v+ACC is the norm in a

Second Accusative construction.

2.1.4 Nasčet and Po povodu

Nasčet and po povodu each derive from a prepositional phrase – na sčet

‘on account’ and po povodu ‘by cause/reason’ – which over time became fully lexicalized and univerbated. Nasčet (+GEN) is one of a large number of prefixal prepositions in Russian, which are “etymologically a prefix or preposition and a case form of a noun.” (Timberlake 2004: 177). Po povodu (+GEN), on the other

hand, is a convert preposition, belonging to “an open group of words or phrases that are used as prepositions, in that they occur with a noun and impose a semantic operation on the noun” (ibid.). Both nasčet and po povodu are relative newcomers in the language, having arisen in the mid-eighteenth century.

2.1.4.1 Nasčet

As a prepositional phrase, na sčet ‘on account (of)’ was first used in constructions of the type “otnesti kakoj-libo rasxod na čej-nibud’ sčet [‘include any expenditure at someone [else’s] expense’]” (Čerkasova 1967: 158). Its use as a preposition “is observed in the first half of the eighteenth century” (ibid.).

The development of an abstract meaning is connected with the use of na sčet in constructions of the type “On vsjakuju meloč’ otnosit na svoj sčet [‘He takes every detail into account, takes little things personally’]”, in which “the combination of otnosit’ [‘take, relate’], constituting the basic medium of the idea of reference, appears with ... na sčet in a qualitatively different meaning (‘to attribute relative to someone/something, concerning someone/something’)”

(ibid.). By taking certain matters (such as the comments or actions of others) personally, one assumes that those things pertain directly to—i.e., are about— oneself. Perhaps this is what allowed a shift in the meaning of na sčet to occur, from signifying a relationship not between the head noun and sčet ‘account’, but

between the head noun and an additional entity, with na sčet indicating the relativity between the two:

(28) razmyšlenija na sčet upravlenija thought-NOM.PL concerning governance-GEN.SG

imenij14 estates-GEN.PL ‘Thoughts concerning the governance of estates’ (Čerkasova 1967: 159)

In any event, by the nineteenth century nasčet had become a preposition in its own right, denoting “in regard to/with respect to” (Čerkasova 1967: 159). In this way, the syntactic function of a PP with nasčet appears to be attributive in nature:

(29) nasčet medikamentov byl trebovatelen15 regarding medicine-GEN.PL is-M.PST.SG strict-M.NOM/ACC.SG ‘He was strict regarding medicines.’ (ibid.)

Although constructions with nasčet are “widely represented not only in the literary language [of the nineteenth century], but also in colloquial and dialectal”

(Čerkasova 1967: 159), this preposition appears to have slipped into less formal

14 Pushkin, A. (1829) Roman v pis’max. (RNC) 15 Leskov, N. (1880) Kadetskij monastyr’. (RNC)

use by the end of that century, and into the twentieth (Ogienko 191416, Ušakov

193517, Kuznecov (2000: 603)), although as with pro, there is no universal agreement on the stylistic status of nasčet.

2.1.4.2 Po povodu

Originally a prepositional phrase, po povodu became a separate preposition sometime in the mid-eighteenth century, having a causal sense

(‘because of’).18 This process took place “between the 1740’s and ‘50’s”

(Čerkasova 1967: 73). By that time, “a causal meaning had already become fundamental” (ibid.: 72) in po povodu; the object referent of po povodu represented the motivation behind, or grounds for, the action. This was entirely logical, as the preposition po+DAT by itself can denote causality (po bolezni ‘due to illness’). Over time, however, the causal sense of po povodu in certain contexts shifted toward what Čerkasova (1967) describes as a ‘deliberative’ (i.e.,

Area/Topic) sense: “The semantics of governing words had a decisive influence on the lexical meaning of po povodu; in po povodu, appearing with verbs of speech and thought, the causal meaning receded into the background. Such use

16 While he does not actually label nasčet as colloquial, Ogienko 1914 does assert that po povodu is “relatively better than nasčet” (140). 17 Ušakov, D., ed. (1934-1940) Tolkovyj slovar’ sovremennogo russkogo jazyka. Website http://ushdict.narod.ru. Viewed 2/4/2012. 18 The noun povod translates ‘cause, reason’.

of po povodu (in a causal-deliberative [pričinno-deliberativnoe] meaning) had already been predominant in the eighteenth century” (ibid.: 74).

In other words, the object referent of po povodu can be seen as semantically both Topic and Cause, because it prompts an act of communication or cognition which focuses on it. This is illustrated by the following example:

(30) Senat imel rassuždenie po povodu senate-NOM.SG have-m.PST.SG discourse-ACC.SG regarding

razoslannyx ukazov. 19 send around-PST.PTCP.GEN.PL decree-GEN.PL ‘The senate had a discussion concerning/about the decrees sent around.’ (ibid.)

This ‘causal-deliberative’ sense appears to occupy a semantic field which has features found both in the area/topic and cause/reason senses: the referent of po povodu is the motive for the activity, not just as the topic around which it proceeds. In (28), the senate’s discussion was prompted by the earlier action, causally connecting the two events.

In spite of this shift toward an area/topic sense, po povodu can still be used with a causal meaning in the modern language, though this is considered nonstandard:

19 Polnoe sobranie zakonov Rossijskoj imperii s 1649 goda (1801).

(31) [speaker 1]: U menja gorlo bolit. [speaker 2]: Po povodu čego?

[speaker 1]: ‘My throat hurts.’ [speaker 2]: ‘From what?’

Example (31) is part of a conversation that I overheard in March, 2012. In response to Speaker 1’s assertion that he has a sore throat, Speaker 2 asks what the cause is (e.g., being among sick people, the cold weather, etc.).

Eventually po povodu came to be used with verbs of speech and cognition in a strictly ‘area/topic’ (deliberativnyj) sense, generally with no implication of causality. However, despite this subtle shift in meaning, enough of a semantic difference between po povodu and o+LOC (which is also considered deliberativnyj) existed, such that the two could be employed to contrast one another. An example of this is found in N.A. Dobroljubov’s critique of Gončarov’s satiric novel Oblomov (1859) concerning a lethargic young nobleman (Il’ja

Oblomov) who feels no motivation to align himself with the norms of society (or even to get out of bed). Dobroljubov states that

readers will probably not be particularly troubled if we do not grieve over concerns of whether such-and-such a phrase corresponds to the character of the hero and his position, or that it was necessary to rearrange a few words in it, etc. Therefore it seems not at all objectionable to be engaged more by the the content and meaning of Gončarov’s novel, though of course, genuine critics will again reproach us

that our article is written not about [o+LOC] Oblomov but about [po povodu] Oblomov.20

In an endnote, editor Jurij Oksman notes that ‘true critics’ refers to Apollon

Grigor’ev and his imitators, “who accused critics of the revolutionary democratic camp of insufficient attention to the peculiarities of the external and internal structure of a work”. This implies that such critics would accuse Dobroljubov of obsessing over the wider implications of Oblomov’s character, to the detriment of details about Oblomov the man, within the sphere of the novel. Thus, it is evident that po povodu signified (to Dobroljubov) a wider scope of ‘aboutness’ than did o+LOC.

2.1.4.3 Other ‘Aboutness’ Prepositions

In addition to the ‘aboutness’ prepositions mentioned above, there are several others which can also be said to convey the idea of ‘aboutness’ to a certain, though limited, degree: po+DAT (lekcija po nauke ‘a lecture on science’), otnositel’no ‘relative(ly), relative to’, and kasatel’no ‘concerning, regarding’’21.

The main reason I have not included them in the present study is that overall

20 Dobroljubov, N. (1859) “Čto takoe oblomovščina?” Website http://az.lib.ru/d/dobroljubow_n_a/text_0022.shtml. Viewed 3/7/2012. Cited in Dobroljubov, N. 1970. Izbrannye literaturno-kritičeskie stat’i. Ju.G. Oksman, ed. Moskva: Nauka. 21 They are derived from the verbs otnosit’[sja] ‘refer’, and kasat’[sja] ‘touch, concern’).

they are very rare in the RNC. As I will explain in more detail in Chapter 3, effective quantitative analysis requires datasets of no fewer than ten tokens each, and none of the above prepositions is attested enough in the RNC to allow for the creation of useable datasets from them. In addition, each of these terms is very specialized in scope: po+DAT denotes a discipline or area of study.

Finally, all three words are used primarily in adnominal constructions.

2.1.5 The Question of Style/Register

In discussing lexical synonymy, sooner or later the issue of style or register will arise. Both are systems of opposition; they are meaningful only when comparing the degree of similarity between two or more lexemes. While lexical synonymy involves denotation, style/register is a system of connotation whereby a given style/register (literary, colloquial, etc.) favors the use of a given lexeme.

In language, “there is a constant tendency to avoid full equivalence” (Sazonova

1963: 76) of meaning. However, as far as exact synonyms, “there are very few in Russian; usually they are single-root words: produmat’~obdumat’ [‘think’], etc.”

(ibid.: 77). In a group of lexical synonyms, “the basic word within such a group gives the most basic understanding, and being neutral in usage, is the dominant

[dominanta] of that group (Sovremennyj literaturnyj russkij jazyk 1988: 18). All

members of a group of synonyms “coincide in meaning, and each member...has its own additional specific features which define its particular place in the language, particular sphere of use in speech, different from the other members of the same group” (Sazonova 1963: 71). Style is one such feature.

Style (stilistika) in modern Russian is arranged into several spheres of usage. In the present study the most salient of these are neutral, colloquial, and prostorečie (‘simple speech’). Neutral style “is the means of expression and vocabulary which can be used in any sort of oral or written speech. A large part of the literary vocabulary is stylistically neutral” (Sazonova 1963: 4) which is to say, stylistically unmarked. Colloquial style represents everyday, conversational speech. Prostorečie (‘simple speech’) denotes nonstandard usage.

Sazonova 1963 claims that “it is very important to emphasize which one is the stylistically neutral” (77). By contrast, Palevskaja 1964 questions the validity of the assumption that every group of synonyms must have a dominant, since criteria for deciding which lexeme is the most ‘basic’ representative are difficult if not impossible to establish. “[I]t is not always the case that one word is the

‘default’ as far as meaning. The relationship between words in a synonymous group is not only stylistic – it may the case that all words in the group are in fact stylistically neutral, yet have the same various nuances of meaning” (56). In the group of four ‘aboutness’ prepositions in the present study, o+LOC and po

povodu have been seen as stylistically neutral; pro is sometimes labelled as colloquial, as is nasčet (Ogienko 1914, Ušakov 1935, Kuznecov 2000).

In the scholarly literature of the past two centuries pro is sometimes, though by no means always, labelled as colloquial, and occasionally as prostorečie. In literary texts, it is not uncommon to find it as an argument in participle and gerund clauses, which are distinctively noncolloquial forms. If pro is indeed colloquial, it has behaved in a very uncolloquial manner, and for a long time.

Much of the secondary literature does not comment on the stylistic value of pro (Lomonosov 1755, Barsov 1788, Kurganov 1793, Slovar’ 1789-94, Greč

1845, Opyt obščesravnitel’noj grammatiki russkago jazyka 1853, Dal’ (all editions), Aksakov 1880, Stojan 1916, Peškovskij 1928, Avanesov 1945,

Grammatika russkogo jazyka 1960, Bondarenko 1961, Kuznecov 2000).

However, a few authors consider it to be either colloquial (Ušakov 1935), prostorečie (Greč 1827, Vostokov 1874), or both (Vinogradov 1947/2001).

The main argument I wish to make is that the issue of style should not drive a discussion which focuses specifically on the semantic content of o+LOC and pro. Furthermore, the fact that o+LOC is considered stylistically neutral does not mean that it is a polar opposite of pro. Rather, it means that o+LOC is essentially free of stylistic connotations; in whichever environment it is used, it

remains stylistically neutral (unmarked). Thus it is problematic to focus on style as an explanation for what is in essence a semantic issue. This is especially salient for the environments in which both o+LOC and pro occur within the same syntagma:

(31) Nabrexala emu pro delovuju poezdku lie-F.PST.SG he-DAT.SG about business-F.ACC.SG trip-ACC.SG

V Italiju i o prodannoj za in(to) Italy-ACC and about sell-PST.PTCP.F.LOC.SG because of

dolgi mašinu.22 debt-ACC.PL vehicle-ACC.SG ‘She lied to him about the business trip to Italy and about the vehicle sold for debts’.

In Chapter 7 (Co-occurrences of o+LOC and pro within the same syntagma), I will examine other RNC examples similar to (31).

22 Simonova, D. (2002) Poloveckie pljaski. (RNC)

Chapter 3: Methodology

3.1 Overview

In this work, in order to answer the research questions identified in

Chapter 1, I will offer both a qualitative and a quantitative analysis of the four

Russian ‘aboutness’ prepositions (o+LOC, pro, nasčet, and po povodu). As noted in Chapter 2, there is a paucity of scholarly works which compare o+LOC and pro lexically; in particular, there do not appear to be any quantitative studies of the differences among the prepositions denoting ‘about’. What has been written in the course of the past century and a half has been mostly of an intuitive or even prescriptive nature, unsubstantiated by any systematic analysis of actual usage. For this reason, I believe a qualitative and quantitative study of these prepositions, using a corpus of sufficient size and scope, will enhance the current understanding of this topic.

In Chapters 1 and 2, I described the methodology I am using for the qualitative analysis of these prepositions; in the present chapter, I will detail the methodology for their quantitative analysis.

One common method for determining the semantic differences between two or more words judged to be close in meaning is to compare their usage in their typical syntactic environments—i.e., their co-occurrence patterns. As

Tagliamonte (2006: 193) observes, “A distributional analysis, a.k.a. factor by factor analysis, is all about finding out how a context (independent factor) constrains the use of the (dependent) variant.” Accordingly, as will be detailed in the next section, the quantitative portion of this work comprises various distributional analyses of randomly selected contexts from the Russian National

Corpus (RNC) in which the prepositions occur.

3.2 Quantitative Corpus Analysis and the Russian National Corpus

An inductive, or quantitative, analysis of the occurrences of a given word within a population (such as a written corpus) can reveal patterns of co- occurrence with other constituents: “probably the most relevant type of data to emerge from corpus studies concerns patterns of co-occurrence among words— what co-occurs with what, and with what frequency” (Cruse 2011: 216). In the present study, I am examining the prepositions o+LOC, pro, nasčet, and po povodu in the context of the syntactic environments in which they typically occur

(with verbs depicting the communication or cognition of the topic denoted by the

OP referent).

The ideal tool for an empirical study addressing the semantic nature of these prepositions is a large corpus like the Russian National Corpus (hereafter

RNC), which consists of texts by many authors, representing a variety of genres, and provides a multitude of regular environments in which the prepositions occur.

A corpus consisting of a single text reflects only one author’s individual usage of the word or words in question—and this is assuming that all four prepositions occur with enough frequency within that single text to ensure a viable analysis in the first place. For quantitative tests to be statistically viable, there has to be a minimum of 10 tokens; as noted by Milroy and Gordon (2003: 164), “If the number of tokens is lower than 10, there is a strong likelihood of random fluctuation, while a figure higher than 10 moves toward 90 percent conformity with the predicted norm, rising to 100 percent with 35 tokens.”

The RNC (http://ruscorpora.ru) is at the present time (April, 2012) the largest online corpus of the Russian language, consisting of over 180 million words. The stated goal of the RNC is to provide a “representative and well- balanced collections of texts” containing “all the types of written and oral texts present in the language (various genres of fiction, journalistic, academic, and business, as well as dialectal and sociolectal, texts).” The proportion of text types in the RNC corpus is determined by their share in real-life usage at the time the corpus was compiled.

The texts in the RNC represent standard Russian works written between the mid-eighteenth century and 2009. Its reflection of real-life usage means that datasets drawn from it will contain no obvious imbalances or shortages of target words, which might then result in misleading statistical results or in a lack of sufficient data for a well-rounded analysis. For example, in a corpus made up entirely of works from scientific journals, verbs of emotional state are not likely to occur very frequently. Therefore, I believe that the RNC’s textual variety, along with its overall size and diachronic arrangement, make it an ideal corpus for the present study.

The quantitative study is based on the parameters of (1) choice and (2) degree of synonymy. First, Russian speakers have a choice among the several prepositions denoting ‘about, concerning’ in syntactic constructions of the type govorit’/dumat’/sprosit’/pisat’/čitat’ o ‘speak/think/ask/write/read about’. As noted in Chapter 1, the prepositions o+LOC, pro, nasčet, and po povodu are the most frequently occurring ‘aboutness’ prepositions in the modern Russian lexicon; they are the only prepositions attested with sufficient robustness for a quantitative study.

Second, there are certain constraints acting upon this choice, which I believe are motivated by the degree of synonymy among the prepositions. They are not complementaries, since the conceptual area of ‘aboutness’ cannot

realistically be divided into binary opposites. Therefore, the prepositions must be either absolute synonyms or near-synonyms.

As Milroy and Gordon (2003: 154) note, “The heart of the linguistic analysis of a variable is uncovering the factors that favor and disfavor the associated variants.” Having taken into account the significantly higher frequency in the RNC of o+LOC compared to the other three prepositions, and the fact that Russian dictionaries usually define pro, nasčet, and po povodu in terms of (i.e., as synonyms of) o+LOC, I have designated pro, nasčet, and po povodu as the associated variants of o+LOC, which I consider the unmarked

‘aboutness’ preposition. According to Lyons (1977, vol. 2: 306), “[T]he formally marked member [in a group of related terms] tends to be more restricted in its distribution (i.e. in the range of contexts in which it occurs) than the formally unmarked member.” The semantically marked lexeme is “more specific in sense than the corresponding semantically unmarked lexeme” (ibid.: 307). As I will discuss in chapter 4, pro, našcet, and po povodu cannot be substituted for o+LOC in every environment; each has a narrower distribution than o+LOC. It is for this reason that o+LOC can be seen as semantically unmarked, and the other three prepositions as marked variants of o+LOC.

As noted above, the present study examines the four prepositions as they occur in arguments of verbs. This proceeds from my research question whether the semantic content in any of the verb types typically associated with the

prepositions acts to constrain any of these prepositions. The following examples are representative of the types of constructions used in this analysis:

(1a) Ja xotel vam ... rasskazat’ ob odnom svoem proekte.23 ‘I wanted to tell you about one project of mine.’

(1b) A kto sejčas pro samo stroitel’stvo pomnit?24 ‘But who now remembers [anything] about the construction itself?

(1c) Sokolov pisal po povodu novogo naznačenija K.P. Jakovlevu.25 ‘Sokolov wrote to K.P. Jakovlev about the new appointment.’

(1d) Pomnju, ja čital nasčet vzaimnoj tjagi antipodov. ‘I remember, I was reading about the mutual attraction of antipodes.’26

The bolded portion of each sentence represents one token of a relevant construction—that is, one verb + one PP (including any noun modifiers within the

PP).

In a pilot project (Kier 2009) involving 2,400 tokens, I established the verb types that co-occur with o+LOC and pro in the RNC; I categorized them according to their basic semantic properties into the following classes, which I am also using in the present study: verbs of speech/oral communication (hereafter,

23 Tynjanov, Ju. (1928) Smert’ Vazir-Muxtara. (RNC) 24 Ančarov, M. (1979) Samšitovyj les. (RNC) 25 Port, G. (1911) Pis’ma. (RNC) 26 Dovlatov, S. (1986) Čemodan. (RNC)

verbs of speech); verbs of asking; verbs of writing; verbs of cognition; verbs of hearing; verbs of reading; and verbs of emotional state. For the present study, I have omitted verbs of feeling/emotional state from the statistical analysis, because the rarity with which they occur with pro, nasčet and po povodu in the

RNC would prevent statistically viable results. However, I include them in the qualitative discussion (see Chapter 8).

All four of the ‘aboutness’ prepositions can be used either as arguments of verbs (2a–d), or adnominally (3a–d):27

(2a) govorit’ o vyborax ‘talk about the elections’

(2b) govorit’ pro vybory ‘talk about the elections’

(2c) govorit’ nasčet vyborov ‘talk about the elections’

(2d) govorit’ po povodu vyborov ‘talk about the elections’

(3a) razgovor o vyborax ‘a conversation about the elections’

(3b) razgovor pro vybory ‘a conversation about the elections’

(3c) razgovor nasčet vyborov ‘a conversation about the elections’

(3d) razgovor po povodu vyborov ‘a conversation about the elections’

In the present study, I have limited the statistical analysis to verbal arguments first of all for methodological reasons: in my preliminary efforts to identify tokens

27 2a-3d are my own examples.

through the RNC search engine, I discovered that the search algorithm

Noun+Preposition+Noun brought up a huge number of false hits (on the order of tens of thousands); the amount of effort involved in isolating the valid tokens was disproportionate to the anticipated results. In contrast, the search engine had a much more reliable return of positive hits with the algorithm

Verb+Preposition+Verb, though even here I had to check each individual token for its validity; ultimately, I was able to use approximately 80% of the returns.

Another justification for this limitation is the fact that the nouns that can take ‘aboutness’ arguments are, generally speaking, secondary to the verbs investigated in this study. A large proportion of the nouns that take prepositional

‘aboutness’ arguments are, in fact, derived from these verbs. The remainder are textual conveyors, so they imply acts of communication or cognition.

For the most part, the adnominal PP’s (e.g., kniga o nauke ‘a book about science’) involve the Area/Topic sense of the preposition; they can also be used in several other senses, all of which are also found with verbal argument PP’s.

Therefore, the full semantic range of prepositional ‘aboutness’ can be discussed on the basis of verbal arguments alone, so the omission of adnominal PP’s is not a subject for concern.

The four prepositions in the study are compatible with a large number of individual verb types within the broad categories above. Consequently, it was necessary to limit the number of verb types used in this study in order to keep

this study to a reasonable size and ensure that each individual verb type has a dataset of adequate size for statistical analysis. To this end, I determined the

RNC frequencies of a number of verbs that take an ‘aboutness’ PP from each of the above-mentioned semantic categories. They were chosen on the basis of having sufficient frequency with which to create datasets having at least ten tokens. The quantitative study covers the verbs listed by semantic category in

Table 3.1 (total contexts as of March 20, 2012). (Several additional verbs are discussed in the qualitative portion of this work; see Chapter 8.)

Verbs of Speech Total Contexts (1890-2009)

govorit’ ‘talk-IPFV’ 331,793

pogovorit’ ‘talk-PFV’ 13,082

skazat’ ‘say, tell-PFV’ 427,101

rasskazat’ ‘tell, relate-PFV’ 37,438

rasskazyvat’ ‘tell, relate-IPFV’ 39,527

dokladyvat’ ‘report-IPFV’ 3,507

doložit’ ‘report-PFV’ 6,368

Verbs of Writing Total Contexts (1890-2009

pisat’ ‘write-IPFV’ 83,833

napisat’ ‘write-PFV’ 56,168 Continued

Table 3.1 Semantic Categories of Verbs in the Quantitative Study

Table 3.1 continued

Verbs of Inquiring/Asking Total Contexts (1890-2009 rassprašivat’ ‘inquire-IPFV’ 3,842 rassprosit’ ‘inquire-PFV’ 956 sprašivat’ ‘ask (about)-IPFV’ 37,362 sprosit’ ‘ask (about)-PFV’ 99,783 prosit’ ‘ask (for), request-IPFV’ 42,976 poprosit’ ‘ask (for)’, request-PFV’ 19,706 spravljat’sja ‘ask (about)-IPFV’ 2,701 spravit’sja ‘ask (about)-PFV’ 6,838

Verbs of Cognition Total Contexts (1890-2009 dumat’ ‘think-IPFV’ 133,026 podumat’ ‘think-PFV’ 52,317 napominat’ ‘remind-IPFV’ 17,124 napomnit’ ‘remind-PFV’ 9,459 pomnit’ ‘remember-IPFV’ 57,195 vspominat’ ‘remember, recall-IPFV’ 21,426 vspomnit’ ‘remember, recall-PFV’ 34,605 znat’ ‘know-IPFV’ 288,794 uznavat’ ‘find out-IPFV’ 15,921 uznat’ ‘find out-PFV’ 54,698 zabyvat’ ‘forget-IPFV’ 11,047 zabyt’ ‘forget-IPFV’ 39,721 pozabyt’ ‘forget-PFV’ 2,768 Continued

Verbs of Hearing Total Contexts (1890-2009 slyšat’ ‘hear-IPFV’ 47,842 uslyšat’ ‘hear-PFV’ 24,594

slyxat’ ‘hear-IPFV’ 4,955

Verbs of Reading Total Contexts (1890-2009 čitat’ ‘read-IPFV’ 56,999 pročitat’ ‘read-PFV’ 25,732

These verb categories are the central organizing factor for the datasets as a whole. A token consists of an entire verbal argument—a single occurrence of a verb with an ‘aboutness’ PP. The PP consists of one of the four ‘aboutness’ prepositions plus the noun it governs, as illustrated in (4a–d):

(4a) My rasskazali o rabote.28 ‘We talked about work’

(4b) Ne zabud’ pro Mišu! ‘Don’t forget about Misha!’

(4c) Ja žaleju, čto ne mog ran’še napisat’ po povodu recenzii na moju knigu. ‘I’m sorry that I couldn’t write earlier about the review of my book.’

28 (4a)-(4d) are my own examples.

(4d) Nasčet Ivana rassprašival. ‘[He] asked about Ivan.’

The bolded words in each of the above sentences comprise the specific elements in the analysis—a verb plus a preposition plus a governed noun phrase.

3.3 Criteria for Data Collection

In all, I gathered approximately 8,400 individual tokens. For present purposes, a text used for a token must be in the form of an argument phrase

(following the examples given above). It must consist of a verb from one of the six semantic categories given in Table 3.1; an ‘about’ preposition (o+LOC, pro, nasčet, or po povodu); and a noun object of the preposition (plus any modifiers).

All tokens in this study are taken from works created between 1890 and

2009 in the RNC corpus of written texts; there are no restrictions on the genre or type of work from which the tokens are drawn. I divided the RNC into four subcorpora, each comprising a thirty-year time period: 1890–1919 (Period A),

1920–1949 (Period B), 1950–1979 (Period C), and 1980–2009 (Period D). The year 1890 was taken as the terminus a quo because preliminary searches within the thirty-year time period of 1860–1889 had returned too few hits to be viable for statistical study. If a work was written between a range of years, both years must

fall within the same thirty-year time period; they may not overlap into an adjacent time period. For example, a work with a date range of 1966–1975 is acceptable, but one with 1970–1984 is not, because it straddles two time periods.

The total number of individual documents, along with the total word count in each time period, is given below:

Time period Documents Word count A (1890-1919) 22,251 24,838,997 B (1920-1949) 4,611 36,612,818 C (1950-1979) 7,102 30,951,542 D (1980-2009) 38,993 87,742,154 Total for all periods 72,957 180,145,511

Table 3.2 Total Individual Documents in RNC in each 30-year Time Period

The menu choice ‘select subcorpus’ on the RNC web site allows a user to customize a set of texts by specifying various parameters such as title or author of work, year(s) of creation, and genre. The only parameters I specified were the beginning and end years of each thirty-year period, in ‘Date of creation’ in the

‘Main parameters’ section. I saved each subcorpus so that it would be readily available for all searches.

3.4 The Search Strings

Using the RNC’s lexico-grammatical search engine, I carried out searches within each thirty-year time period for each verb plus each preposition. Table 3.3 illustrates the search string entered for govorit’ ‘talk’ + o:

Word Gramm. features Semantic features

говорить v sem sem2 Distance: from 1 to 8 Word Gramm. features Semantic features

о pr sem sem2 Distance: from 1 to 8 Word Gramm. features Semantic features

S,loc sem sem2

Table 3.3 RNC Search String for Verb govorit’ with Preposition o+LOC (in that Order in Sentence)

In the first column, the lemma говорить ‘say, speak, talk-IPFV’ specifies a search for all contexts of this verb in any of its forms. In the ‘Grammatical features’ column, ‘v’ instructs the search engine to search for contexts in which the given lemma is a verb.29 ‘Pr’ specifies ‘preposition’ for the second word in the string.

29 By specifying ‘v’, the search will also return participle (adjectival) forms of the verb.

The column ‘Semantic features’, with ‘sem’ checked, directs that only the primary dictionary meaning be searched, since the secondary definition of o (with ACC) is

‘[up] against’, which is not a desired hit. The bottom left column of the category

‘Word’ remains empty; this allows any word fitting the syntactic specification in the middle column to be returned. ‘S,loc’ specifies that the word can be any LOC case noun. Because o has an allophonic variant ob before an initial vowel, the above search was also done for that form. In addition, since the ratio of total contexts of o to ob in any given time period is approximately 5:1, I have kept this proportion in the datasets whenever possible. Thus, in a dataset of 30 tokens, at

30 least six are with the form ob.

In its entirety, the search string given in Table 3.3 will return all texts within the given time period in which the three elements specified occur between one and eight words’ distance from each other in the order verb-preposition-noun.

Preliminary test searches of various word distances showed that the upper limit of eight was sufficient to catch enough valid hits while still remaining a manageable size. Test searches of distances of greater than eight words did not yield more hits than those of up to eight words; hence a distance of eight words or less proved to be sufficient.

30 I did not search for the remaining allophonic variant obo, since it only occurs before the first-person singular personal pronoun (obo mne ‘about me’) and forms of the pronominal adjective ‘all’ (obo vsem, obo vsej, obo vsex ‘about all-M/N, F, PL).

The word order of an argument phrase in Russian is not limited to post- verbal. Depending upon certain factors related to topic emphasis, it may also be pre-verbal. Therefore, I also carried out searches according to the following parameters (Table 3.4):

Word Gramm. features Semantic features

о pr sem sem2 Distance: from 1 to 8 Word Gramm. features Semantic features

S,loc sem sem2 Distance: from 1 to 8 Word Gramm. features Semantic features

говорить v sem sem2

Table 3.4 RNC Search String for Preposition o+LOC with Verb govorit’ (in that Order in Sentence)

When I began conducting searches in the RNC for tokens for this study, I found that the ratio of contexts with V–PP word order to those with PP–V was approximately 2 to 1; accordingly, I have kept this proportion in the datasets whenever possible. Thus, in a dataset of 30 tokens, about 20 have V–PP word order, and 10 have PP–V word order.

For datasets with pro, the following search string was used (Table 3.5):

Word Gramm. features Semantic features

говорить v sem sem2 Distance: from 1 to 8

Word Gramm. features Semantic features

про pr sem sem2 Distance: from 1 to 8

Word Gramm. features Semantic features

S,(acc|gen) sem sem2

Table 3.5 RNC Search String for Verb govorit’ with Preposition pro (in that Order in Sentence)

‘S,(acc|gen)’ specifies the search to return all ACC nouns (including animate

ACC, whose morphology coincides with GEN), since pro governs ACC. As with o+LOC, a second search must be carried out for the word order pro-N-V. Note that under ‘Semantic features’, only ‘sem’ is checked. This is because the secondary definition of pro, ‘for the sake of’, though obsolete, occurs in several

fixed expressions such as pro zapas ‘for emergency’ and pro sebja ‘to oneself’; we are concerned here only with its ‘about’ sense.

Nasčet uses the same parameters as pro except that the grammatical features for the noun object are ‘S,(gen|acc)’ since nasčet governs GEN.31 Po povodu, on the other hand, requires a slightly different search string than the other three prepositions. Although po povodu functions as a single lexeme, the

RNC search engine does not recognize it as such and will fail to return any results if it is entered as a preposition (pr). Therefore, the search string must specify povod as a DAT sg. noun:

Word Gramm. features Semantic features

говорить v sem sem2 Distance: from 1 to 8

Word Gramm. features Semantic features

повод S,dat,sg sem sem2 Distance: from 1 to 8

Word Gramm. features Semantic features

S,(gen|acc) sem sem2

31 Both nasčet and po povodu require ‘S,(gen|acc)’ for the OP. This returns GEN as well as animate ACC results.

Table 3.6 RNC Search String for Verb govorit’ Plus Preposition po povodu (in that Order in Sentence)

This search string will then return all contexts of povod as a DAT sg. noun. Even then, not every hit will contain po povodu in the function of a preposition; as a noun, povod can be used in contexts other than as one element of a complex preposition. However, those hits which show po povodu in the correct context can easily be ascertained when examining the complete range of search results.

3.5 The Search Results

Once a complete search string is entered for a verb, preposition, and noun within a given time period, the RNC returns hits of all contexts that satisfy the search criteria. This is illustrated in (5), from Period C (1950–1979):

(5) говорить v distance: from 1 to 8 from о pr distance: from 1 to 8 from S,loc Found 1 339 documents, 5 451 contexts. Pages: 1234567891011 next page 21. Юрий Трифонов. Обмен (1969) [disambiguated] All examples (2) Хотя говорить о таких мелочах, поверь, очень неприятно. [Юрий Трифонов. Обмен (1969]32 [disambiguated] Он говорил о смерти и о том, что не боится ее. [Юрий Трифонов.

32 “Though to speak about such trifles, believe [me], is very unpleasant.” Trifonov, Ju. (1969) Exchange. (RNC)

Обмен (1969]33 [disambiguated]

Excerpt (5) represents two hits, both occurring in the 1969 novel Obmen by Jurij

Trifonov. I have underlined the target words—the verb, preposition, and noun.34

Although I include modifiers of the noun object (if present) as part of the token, these are not highlighted in the search results, since they were not specified in the search string.

Each page (screen) of the search results lists ten documents, each of which contains one sentence containing the target words. The hits are listed chronologically, from the most recently written text to the oldest. Some search strings, like govorit’ + o in (5), return many thousands of hits, while others return very few or none at all within a given time period. For example, znat’ ‘know-IPFV’

+ nasčet returned a total of 42 separate contexts in Period B (1920-1949).

However, of these 42 hits, only twelve contexts turned out to be suitable for use as tokens, so this particular dataset consists of only twelve tokens.

In situations where the total number contexts returned was quite high, as with govorit’ + o, collecting thirty useable tokens was not difficult. However, some searches did not return enough contexts to yield a full thirty tokens. If the

33 “He spoke about death and about the fact that he did not fear it.” Trifonov, Ju. (1969) Exchange. (RNC) 34 These appear in an orange font on the actual web site.

verb and/or preposition has a low overall frequency within the specified time period, the search will return only a small number of hits.

Because certain time periods (especially Period C) have low overall word counts compared with others, datasets in those time periods have correspondingly fewer tokens. Nevertheless, no dataset used in this study contains fewer than ten tokens.

3.6 Criteria for the Datasets

I created the search algorithms used in this study with the goal of returning enough hits containing the three target words (verb + preposition + noun) to ensure an adequate number of tokens for each dataset. Nonetheless, it was to be expected that not all hits would contain useable data. Examples of false hits are sentences where the verb and PP are not concatenated syntactically, or in which a form homonymous to a target word is used: O kak tjaželo žit’! ‘O, how difficult to live!’; Ja vsegda byl pro-Putin ‘I was always pro-Putin’, etc. There were also hits containing the phrase po ėtomu povodu ‘on this occasion’.

Because this represents an entire PP, not a preposition, po ėtomu povodu performs a different function than the lexicalized phrase po povodu. Thus, it was necessary to scan through each body of search results visually to find useable hits.

As noted above, all the tokens in the quantitative study involve the same syntactic construction—a verbal argument with a prepositional phrase. The OP must be a noun (or substantivized adjective such as bol’noj ‘sick person, patient’). In instances where a sentence has more than one OP, I counted only the first noun for the quantitative analysis. For example, from a hit such as Ona govorila o nauke, o ximii, i o sporte ‘She was talking about science, about chemistry, and about sports’, the actual token consists solely of Ona govorila o nauke. However, in the qualitative portion of this study, I will discuss constructions in which two conjoined PP’s occur with two different prepositions

(see Chapter 7).

Within each dataset, I avoided using more than one token by the same author whenever possible. However, because certain subcorpora are fairly small

(such as Period A), it was sometimes necessary to allow more than one token by the same author to ensure that the total for that dataset would be adequate. In those situations, I did not use more than two tokens by the same author, and, whenever possible, I took them from two different works separated by at least ten years.

Category Tokens

Verbs of Speech 2,219 Verbs of Cognition 3,158 Verbs of Hearing 664 Verbs of Asking 1,377 Verbs of Writing 571 Verbs of Reading 401 Total: 8,390

Table 3.7 Distribution of Total Tokens in each Dataset, by Verb Category

3.7 Parameters for the Statistical Analysis of the Datasets

3.7.1 The Structure of Datasets in the Quantitative Study

Figure 3.1 illustrates the structure of the datasets in this study from the top

35 down, using rasskazat’ ‘tell-PFV’ + pro as an example :

35 All other datasets follow this format as well.

Figure 3.1 Organization of RNC Datasets in the Quantitative Study

Each set of tokens at the bottom of the chart represents a single dataset for a specific time period (A, B, C, or D). Each dataset is a discrete unit. For example, the dataset for a given verb with o+LOC from Period C is compared only with the same verb (or its aspectual partner) with pro, nasčet, or po povodu from Period

C. After carrying out tests on the individual datasets for each time period, I compare the four datasets comprising each verb plus preposition for changes in values which could indicate a semantic shift in the preposition over time.

The data in this study do not represent a single sample population but rather many small ones, each consisting of no fewer than ten and no greater than thirty tokens. Milroy and Gordon (2003: 165) comment that “the ideal [number of tokens per variable] appears to be around 30, but if this cannot be attained a

figure as much as possible in excess of 10 is a sensible goal.” I have arranged the verbs in the study into semantic groups mainly for the purpose of carrying out statistical tests using these categories as variables. Thus, ‘(semantic) verb category’ is one of the predictor variables.

3.7.2 The Nature and Structure of the Quantitative Analysis

The essence of quantitative testing (distributional analysis) is the assessment of how the choice process is influenced by the various independent

(explanatory) factors whose specific combinations define the context in which a choice between variant forms is presented. These tests measure the likelihood that the relationship between the variables is not due merely to chance.

As part of the analysis, independent factors are arranged into various factor groups. As Tagliamonte (2006: 105) observes,

[a] factor group is some aspect of the context ... which affects whether or not a variant occurs. Each factor group can also be thought of as a hypothesis about what influences the choice process. In this way, a factor group is also a constraint on the dependent variable.

The first step in an analysis of variables (ANOVA) involves testing the dependent variables (tokens with o+LOC as the preposition) against each individual factor

group within the set of independent variables (tokens with pro, nasčet, or po povodu).

For the present study, frequency counts of the four prepositions carried out as part of the pilot project (Kier 2009) indicate that in the RNC o+LOC occurs far more frequently than pro, nasčet, and po povodu combined. Between 1890 and 2009, it occurs roughly eighteen times as often as pro, 74 times as often as nasčet, and 120 times as often as po povodu. Thus, regardless of the composition of the sample datasets, these proportions do not change. However, the frequencies of the prepositions, in themselves, reveal nothing about the semantics of the prepositions, nor about the nature of the other constituents with which the prepositions occur in that population. A frequency count, in itself, does not measure anything beyond how often a given form occurs within a certain population.

Until now, no quantitative tests have been undertaken specifically on the four Russian ‘aboutness’ prepositions that are the focus of this work, to my knowledge, other than basic frequency counts within a textual population (not the

RNC). To address the possibility of covariance or correlation between the prepositions and certain features within the constituents with which they typically occur (verbs and nouns), a deeper analysis is required, one which takes these constituents into account. Only in light of the results of such an analysis, supplemented by a qualitative analysis, can one formulate reasonable inferences

about co-occurrence, trends, and causality, where applicable. As William Haas has rightly insisted, “No judgements should be elicited concerning isolated words: language in use is always contextualized, so intuitions should be sharpest in respect of words in context” (cited in Cruse 2011: 218).

3.7.3 Issues Pertaining to the Quantitative Study

Through a study of extant literature on a topic (such as prepositional synonymy), one can reveal, or at least extrapolate, which factors are relevant for the variation existing in a particular environment or, at least, which factors have mattered to other researchers in past studies. Armed with this background information, one can then decide whether or not these factors are applicable to the present investigation.

As noted above, the ‘aboutness’ prepositions in Russian have thus far not been analyzed quantitatively, beyond basic frequency counts. Indeed, the only comment of substance on the matter is by Vinogradov (1947/2001), concerning what he understood as an ‘essential vs. superficial qualities’ division between o+LOC and pro, respectively. However, Vinogradov made no reference to the types of verbs or nouns with which o and pro typically occur; his

‘essential~superficial’ division exists outside of any context beyond the two prepositions themselves.

Since the previous scholars have provided no contexts or evidence for their explanations, we must look for some. We know that these prepositions occur with verbs and nouns, but we do not know which features within those verbs and nouns are reliable indicators of the choice of one preposition over the other, or indeed, if any such features are reliable indicators. Because of this, it is necessary to examine the other constituents quantitatively as well as qualitatively) for any influence they might have on the choice of preposition. If any influence can be shown, it might prove to be a reflection of an

‘essential~superficial’ division, or it might not. Here it is appropriate to quote the very apt dictum of J.R. Firth that “You shall know the meaning of a word by the company it keeps” (cited in Cruse 2011: 215).

3.7.4 Specific Factors

In the quantitative study in the present work, a token consists of a verb phrase that includes a prepositional argument phrase. For each individual test carried out on a given token, the dependent variable consists of o+LOC (e.g., rasskazyvat’ pro ‘tell-IPFV about’). The independent (predictor) variables are the verb and the prepositional object noun. I have selected several factors (which reflect certain features within the independent variables) for testing, arranged into

factor groups of two factors each. Each of these factors was found to show statistical significance in the dataset used in my pilot project (Kier 2009):

Factor group Factor 1 Factor 2 OP noun formation non-deverbal deverbal OP noun semantics Common vs. proper common proper Countability count mass Animacy inanimate animate Modification of noun unmodified modified Verb aspect imperfective perfective

Table 3.8 Factor Groups in the Quantitative Study

3.7.5 Explanation of Factor Groups and Factors

As outlined in Table 3.8, I have assigned one or more factor groups to each of the independent variables. In the following subsections, I will discuss each factor group in detail.

3.7.5.1 Deverbal~Non-deverbal Nouns

The first factor group shown in Table 3.8 is OP noun formation, with the two factors non-deverbal and deverbal. Productive deverbal nouns (nomina actionis) in Russian are “derived by suffixation of verbal roots; they are … formed

with an augment {-ij-} added to what looks like the passive participle” (Timberlake

2004, 216). Examples are suščestvovanie ‘existence’ (from suščestvovat’

‘exist’), vpečatlenie ‘impression’ (from vpečatlit’ ‘impress’), and udalenie ‘removal’

(from udalit’ ‘removal’). There are several types of deverbal nouns in Russian, only some of which are productive in the modern language. Of these,

Timberlake regards the {-ij-} type as “the most transparent” (ibid.). In my pilot project (Kier 2009), deverbal nouns of this type were shown to be statistically significant within the dataset for that study; therefore I am including them as a factor group in the present study as well.

3.7.5.2 Common~Proper Nouns

The second factor group for nouns represents the semantic division of common~proper nouns. For purposes of this study, I have included collective nouns (such as gruppa ‘group’) with common nouns, since both refer to a class of entities. A common noun usually refers to a certain class of entities (tables, women, jobs) or to an instance of a certain class (a table, those women, my job).

Proper nouns in Russian designate “individual items which are included in a class of homogeneous items, but by themselves do not bear any special

reference to this class”36 In other words, a proper noun/name refers to a unique entity; it tells us which thing is in question, without giving any other information about it, and “renders the referent unique within the domain” (Cruse 2004: 396).

Examples of proper nouns are personal and geographical names, titles of published works, companies, etc.

3.7.5.3 Count~Mass Nouns

The third factor group for nouns represents the semantic division of count~mass nouns. A count noun (or countable noun) is a common noun that can have a singular or plural form and thus can be modified by a numeral: mal’čik

‘boy’, sad ‘garden’. A mass noun, on the other hand, represents an indivisible unit (usually expressed in the singular), which cannot be counted: pesok ‘sand’, xolod ‘the cold’, steklo ‘glass’. For the purposes of this study,I treat collective nouns as a subtype of count nouns, given that they readily pluralize and can be modified by numerals: narod ‘people’.

36 Russkaja grammatika. Akademija nauk SSSR, Institut russkogo jazyka. Website http://www.rusgram.narod.ru/1121-1146.html. Viewed 9/1/2012.

3.7.5.4 Animacy of the Noun Object

The fourth factor to be considered is animate~inanimate. Examples of animate nouns are čelovek ‘person’, Petr ‘Peter’, kot ‘cat’, doč’ ‘daughter’, while stul ‘chair, stool’, janvar’ ‘January’, vremja ‘time’, derevo ‘tree’ are inanimate. I count whatever animacy is given in the token form. If the inanimate form is given for something which should correctly be animate, I count it as inanimate, since that is the form used by the speaker. This occurs in only two tokens.

3.7.5.5 Weight of Modication (Premodifiers) of the Noun Object

The fifth factor group represents the presence or absence of any premodifiers of the noun object—modifiers occurring before the noun. A modifier semantically limits the referent of its syntactic head. For example, an adjective plus a noun “prototypically restricts the domain designated by the noun alone to a subpart, and designates a subset of the entities denoted by the noun alone”

(Cruse 2011: 311).

Although perhaps the most common type of modifer, adjectives are not alone in this function—determiners, adverbs, particles, and other constituents can also modify nouns. For example, in Russian, a entire clause can function as a premodifier:

(6) [M]nogie čitatelej ... uže znajut o many-NOM.PL readers-M.ANIM.ACC.PL already know-PRES.3PL about

novoj bazirujuščejsja na ėkspertnom new-F.LOC.SG base-PRES.PTCP.REFL.F.LOC.SG on expert-M.LOC.SG

principe, sisteme.37 principle-LOC.SG system-LOC.SG ‘Many readers already know about the new system based on an expert principle.’

Therefore, for present purposes I define a modifier as any syntactic element in the phrase whose absence would produce a noticeable change in the semantic content of the PP. However, I have limited this variable to two factors, unmodified~modified, for the reason that the initial tests are of binary variation; this means that the variables must reflect an opposition of exactly two factors. I am considering the following constituents as modifiers: 1) cardinal numerals and other quantifiers, even if morphologically they decline as nouns; 2) adjectives; 3) demonstratives; and 4) possessives. While most of the adjectives in my dataset are preposed, there are several tokens in which an adjective immediately follows the noun it modifies. In these cases, I count the noun as modified (e.g., pro doč’ ee ‘about her daughter’). I am not considering a title attached to a person’s name (general Grabbe ‘general Grabbe’) as a modifier, but rather as an apposite noun.

37 Kostandov, L. (1971) “Nado iskat’”. In Literaturnaja gazeta, 5/5/1971. (RNC)

3.7.5.6 Verb Aspect

While the previously discussed factors relate to the noun object of the preposition, the factor perfective~imperfective relates to the inherent aspect of the verbal lexeme. Although every verb in the study is either imperfective or perfective aspect, not every verb has an aspectual pair that is also attested in the dataset (boltat’, molčat’, slyxat’, etc.). In addition to the (first) series of tests, which measures the distribution of noun factors, I have also carried out single- level tests with aspectual verb pairs, but only in configurations where they have at least one factor in common:

rasskazyvat’ (imp.)+o ~ rasskazat’ (perf.)+o yes rasskazyvat’ (imp.)+o ~ rasskazyvat’ (imp.)+pro yes rasskazyvat’ (imp.)+o ~ rasskazat’ (perf.)+pro no

3.7.6 Statistical Modelling in the Study

For each test in the present study, the null hypothesis (H0) states that the choice of preposition has no correlation with any of the factors; that is, the factor groups listed in Table 3.7 do not influence the choice of preposition. In accordance with accepted sociolinguistic practice, the results are presented “in terms of percentage use of each form” (Milroy and Gordon 2003: 145)—for example, the percentage of animate~inanimate noun objects occurring with rasskazat’ pro in Period C.

The tests in this study involve binary variation—the measurement of one variable. For these tests, I have used Fisher’s Exact Test (two-tailed).38 This test determines the exact P-value of a given variable. By contrast, the Chi- square test, the most common test of statistical significance, requires that no individual cell have a value of less than 1, and that at least 80% of the cells have a value of at least 5. In many cases, the values within my datasets fall outside these parameters; therefore, Fisher’s Exact test is better suited for these particular data.

3.7.7 General Limitations of Corpus Data

One point I wish to address regarding the general utility of written corpora is that there are limits to the conclusions that can be asserted about the data. As

Cruse (2011: 216) notes, “as far as lexical semantics is concerned, probably the most relevant type of data to emerge from corpus studies concerns patterns of co-occurrence among words—what co-occurs with what, and with what frequency.” Relying on frequency data as the sole indicator of the meaning of a word or words is one drawback in a corpus-based approach. Since computer programs can only recognize word forms (ibid.), individual examples of usage must be examined by a speaker who is aware of the range of possible construals

38 VassarStats: Website for Statistical Computation. Website http://vassarstats.net). Viewed 2/8/2012.

and other pragmatic inferences—which is to say, that speaker recognizes that the word is not just an isolated form, but is part of a larger environment or context. Without the benefit of a deductive analysis of the material as well,

“frequency data cannot be directly related to meanings” (ibid.).

I intend to avoid the issues highlighted by Cruse by analyzing quantitative data within the larger context of the environments in which they occur. As I have mentioned, each token in this study consists of an ‘aboutness’ preposition in a complete argument phrase. The datasets I have created are sample populations of the entire RNC. The large overall size of the RNC certainly does not in itself prevent a whole-corpus study from being carried out; in fact, its own dedicated search engine could easily accomplish this). However, the use of sample data— specifically, of the size and in the categories in which I have arranged my datasets—allows attention to be directed toward qualitative issues such as word construal, i.e., the forms behind the frequencies.

When a correlation is observed between two variables, it sets up the possibility, but not necessarily the probability, that variation in one variable is causally linked to variation on the other. Before concluding that a causal relationship exists, however, qualitative analysis of other possible factors explaining the relationship is absolutely necessary.

Chapter 4: The Semantics of Prepositions

4.0 General Overview of Prepositions

4.1 Semantic Functions of Prepositions

Before addressing my specific research questions regarding the lexical semantics of the prepositions o+LOC, pro, nasčet, and po povodu, it is necessary to discuss some general issues in the semantics of prepositions, and in particular, what they imply about the roles of other constituents in the verb phrase, because, as I will show in Chapters 5, 6, and 7, each of the four prepositions in question may frame the relationship between those constituents in somewhat different ways. Essentially, prepositions are lexical items that indicate how individual clausal constituents (typically, verbs and nouns) relate to each other conceptually and semantically. They are employed “for the expression of spatial, temporal, causal, telic, possessive ... and other relationships between objects, or of such relationships of objects to activities, states, and qualities” (Vinogradov 2001: 555), e.g., locative (in the yard), temporal (on Wednesday), manner (with patience). At the clausal level, they

“express the relationships of nouns to significant words belonging to other parts

of speech (including a noun to other nouns): vyšel iz lesa [‘He came out of the forest’], daleko ot goroda [‘far from the city’]” (Sovremennyj russkij jazyk [SRJ]

1988: 249). For example, the phrase the book on the table shows the physical position of one entity (book) relative to another (table).

Unlike nouns and verbs, prepositions “largely constitute a closed set”

(Cruse 2011: 321). Their relational meaning “consists in a small number of very abstract, i.e., simplified geometric and topological properties” (Zelinsky-Wibbelt

1993: 354). However, in spite of the observation that “people seem never to have taken prepositions seriously” (Jackendoff 1973: 345), a wide range of complex and increasingly abstract senses can, in many cases, be derived from a single basic spatial meaning. Thus the conceptualization of abstract concepts such as time, cognition, instrumentality, and causation (for example) becomes easier when their lexical depictions draw upon the basic spatial arrangements of the physical world. “[I]n exploring the organization of concepts that, unlike those of physical space, lack perceptual counterparts, we do not have to start de novo

... the mind does not manufacture abstract concepts out of thin air ... it adapts machinery that is already available” (Jackendoff 1983: 188-89).

Prepositions may be of either primary (underived) or derived origin39.

Most primary prepositions have a spatial meaning, or one or more meanings

39 Examples of derived prepositions are Engl. between, instead of, toward and Russ. po pričine ‘because of’, krome ‘except’, vopreki ‘in spite of’.

“arguably derived from a spatial one” (Cruse 2011: 321). O+LOC and pro are primary prepositions, while nasčet and po povodu are derived from the prepositional phrases na šcet and po povodu. As such, neither of the latter two has a spatial sense. When present, a primary preposition’s spatial sense represents its fundamental, non-metaphorical meaning. This basic sense is also reflected (to various degrees) in the metaphorical extensions deriving from it.

“The transfer of meaning from one domain to another is most productive with spatial expressions because of their expressive force: our most basic experience is spatial” (Zelinsky-Wibbelt 1993: 380).

Prepositions play an important role in the depiction of states of affairs (or scenes), making possible a flexible conceptualization of the physical space which surrounds us. Our direct experiences with the world around us “provide the basic structures which, through analogy, metaphor, and metonymy, enable us to conceptualise ever more abstract domains” (Taylor 1993: 151). As archetypal indicators of spatial relations, prepositions are “extensively employed in the construal of relations in more abstract domains” (ibid.). Typically, the fundamental meaning of a (primary) preposition describes a location in, or a path through, physical space. The various metaphorical senses of a preposition also reflect, to some degree, a ‘place’ or ‘path’ sense as they proceed outward from the source meaning. The fundamental meaning is the one sense that “is not derived by extension from any other sense in the network” (Cruse 2011: 323).

However, it is not always the case that all semantic extensions of a preposition develop directly from its core meaning; a given extension can “serve as the base for further extensions which cannot be derived directly from the source meaning”

(ibid.: 322).

4.2 Polysemy and Metaphorical Extensions of Prepositions

For the semantic connection to remain relevant between a preposition’s fundamental meaning and its extensions, an adequate description of metaphor

“must allow that the primary or original meanings of words remain active in their metaphorical setting” (Davidson 1978: 31-47). One can establish the polysemy of a given preposition by determining which information is basic, “and which information is evoked, specialized or constrained by a specific use in a specific context” (Zelinsky-Wibbelt 1993: 355). This can be approached by arranging the various senses onto a map which shows the senses that the preposition develops as it moves outward from its fundamental spatial meaning, with each extension more abstract than the previous one. The productiveness of the metaphorization process “clearly shows its linguistic motivation and communicative function: metaphors are created in order to understand one thing

in terms of another. The less-known is understood and communicated in terms of the well-known” (Zelinsky-Wibbelt 1993: 380).

As each language has its own stock of prepositions (and each preposition its own set of individual senses), the nature and composition of these extensions—i.e., their degree of granularity—varies. In any given language, each preposition has its own set of tendencies for usage in certain senses, and its own set of constraints against use in certain others. For example, Russian pered+INST ‘before/in front (of)’ carries a temporal sense (pered zarëj ‘before dawn’), while pod+INST ‘under’ does not. As Dirven (1993: 84) notes, the

English preposition at has found use in at least six separate extensions (time, state, area/topic, means/manner, circumstance, and cause/reason). By contrast, its approximate Russian equivalent in the spatial sense, u+GEN, has two extensions: possessive (Vinogradov 1947/2001: 563) and ablative (which

Bondarenko 1961: 70-71 refers to as ob”jektnyj ‘object-oriented’).

As noted above, metaphorical extensions of prepositions do not emanate from the source domain “in any haphazard way[,] but follow a path of gradually increasing abstractions” (Dirven 1993: 76). The temporal sense is usually cited as the first extension from the spatial domain (see, e.g., Bondarenko 1961,

Vinogradov 1947/2001, Dirven 1993, Cruse 2011). As Jackendoff (1983: 189) observes, “Prepositions of time are on the whole identical to spatial expressions.”

Space and time are conceptualized in much the same manner; for example, we speak of points in time, and of cyclical events as coming around.

In his analysis of twelve English prepositions (at, on, in by, with, through, about, under, over, from, off, and out of), Dirven (1993) discusses the following extensions, or domains; these can be applied to Russian as well.

 Space (source domain): in the house, on the floor  Time: in two weeks, on Tuesday  State: in trouble, under arrest  Area/Topic: specialize in, think about  Manner: with aplomb  Means: through hard work  Instrument: walk with a cane  Circumstance: by accident, at these words (I left)  Cause: be excited about, suffer from

This model places the meanings of prepositions in a continuum, whereby the depicted scenes show increasing abstraction in what they depict:

 Space (Source domain): Depiction of an actual 2-D or 3-D spatial relation, with no abstraction

 Time: Conceptually related to Space, but condensed into a 2-D depiction (i.e., depicts points on a 2-D timeline)

 State: Depicts a generic event/condition which lacks participants and agency

 Area/Topic: Depicts an agent carrying out a communicative or cognitively activity (speaking, thinking, etc.) that focuses on an external entity qua

topic, or depicts an entity which is a result of a communicative/cognitive process (book, lecture, question, etc.) of which the topic-entity is an attribute (a book about sports)

 Manner/Means/Instrument: Depict various configurations of “orientation, contact, enclosure or connection between two entities, situations or concepts” (Dirven 1993: 89)

 Circumstance: Depicts a state/event/condition that has an actual or potential effect on the agent’s performing of the activity

 Reason: Depicts a volitional activity performed in response to an external entity or situation

 Cause: Depicts a nonvolitional state (usually emotion-based) which is a reaction to external entity or situation

Although Dirven’s continuum follows a logical structure, it is obvious that not every given preposition (in English, Russian, or any other language) is semantically suitable for use in every extension. Even synonymous (or nearly synonymous) prepositions may follow different paths. For example, one preposition may develop three separate senses, while its synonym may be limited to only two. Additionally, certain senses of a preposition may become obsolete and archaic over time. The four prepositions that are the subject of this study are no exception.

To my knowledge, there is no classification of Russian prepositional meanings that is as semantically precise as Dirven’s classification. For example,

Bondarenko (1961: 70-73) offers an outline of the semantic categories of

Russian prepositions, which he divides into two supercategories: ob”jektnye

otnošenija ‘object-oriented relations’ (pro zverej ‘about beasts’, dumat’ o rabote

‘think about work’), and otnošenija obraza dejstvija ‘relations of the form of activity/procedure’ (pro sebja ‘to oneself’), i.e., argument and adverbial uses, respectively. Bondarenko’s more specific classifications are essentially semantic—point in space, point in time, reason/cause, purpose, form of activity, referent, and determinant. The last four categories each cover a bewilderingly wide variety of semantic relations. Bondarenko provides no justification for these broad categories, and he presents them in an order that is apparently random; for example, he places pričinnye ‘cause/reason’ directly after ‘time’ but does not show how a causal sense could evolve from (or coexist as a polyseme with) a temporal one. For present purposes, it is most important to note that

Bondarenko classifies ‘aboutness’ as a form of ‘purpose’, just like going somewhere ‘for’ (i.e., ‘to obtain’) something. This is erroneous. The relation between the constituents in an ‘aboutness’ construction (e.g. Ona rasskazala mne o babuške ‘She told me about her grandmother’) does not in any way entail or imply a purpose or goal. The activity rasskazat’ ‘tell, recount’ depicts simply the communication of a topic. There is no semantic element of purpose or goal in the act of communicating a topic.

The most influential classification of Russian prepositional meanings seems to be Vinogradov’s list of twenty possible extensions from the spatial

sense (1947/2001).40 Like Bondarenko’s taxonomy, Vinogradov’s classification does not have any evident logical ordering to make it possible to see how one meaning can become another. Evidently, Vinogradov would place into the category of deliberative (deliberativnye) prepositional meanings, “for the indication of content, reason for judgement, item of thought, speech, feeling (with verba sentiendi et declarandi)” (Vinogradov 1947/2001: 564), which is roughly analogous to Dirven’s ‘area/topic’. The term deliberativnyj is defined in the

Slovar’ lingvističeskix terminov (1960)41 as a “form of expression ... indicating that the speaker is pondering over the decision to be taken”. While this description obviously encompasses cognitive processes such as thinking and pondering

(i.e., the mental deliberation over a thought or idea), Vinogradov’s appropriation of this term is unwarranted, as it obviously cannot be applied to any of the other types of activities associated with ‘aboutness’, such as talking, reading, writing, emotional states, or requesting, and so on. Furthermore, by using deliberativnyj in connection with the cognitive activities of thinking/pondering, Vinogradov is evidently relying on a pragmatic assumption—that behind every thought lies a purpose or goal—to explain a semantic context which, in reality, does not entail either of these features.

40 Vinogradov (1947/2001: 561) notes that A.A. Šaxmatov had defined and systematized these extensions in his Sintaksis russkogo jazyka (2. ed., pp. 552-3). 41 Website http://dic.academic.ru/dic.nsf/maruso/1389/deliberativo. Viewed 1/2/2012.

In the remainder of this study, I will refer mainly to Dirven’s extension map, as its arrangement reflects a logical progression from concrete to abstract senses. Moreover, it has greater explanatory value for the relationship between meanings like ‘cause’ and ‘topic’, which are crucial for my analysis.

4.3 Russian Prepositions

4.3.1 Case and Semantics of Prepositions

4.3.1.1 General Notions Regarding Case

The lexical content of cases is an important factor in the understanding of the functions of prepositions. A preposition “not only supports and strengthens the meaning of cases, it also supplements, specializes, and complicates them in this or that government” (Vinogradov 1947/2001: 555). Meaning in prepositions assumes the form of “an inkling, which becomes clear only with case; in this way, the governing case ... is not a randomly inherent property but a consequence of meaning” (Aksakov (date unknown) in Vinogradov 1947/2001: 556). Generally, case is a system in which dependent nouns are marked “for the type of relationship they bear to their heads” (Blake 2001 in Kittilä 2011: 2). Among other relationships at the phrasal level, case “marks the relationship of ... a noun to a preposition” (Kittilä 2011: 2). Kittilä points out the close relation between

case and preposition, in that both “express similar functions, e.g. coding semantic roles. However, the two concepts are not identical and there are formal differences between them” (ibid.). As affixes, case markers are attached “tightly to their host” (ibid.), while prepositions are generally independent items. As such, the latter do not usually “trigger any changes in the nouns they modify”

(ibid.). Although prepositions and cases can both be used for the expression of various semantic roles (such as Instrument or Location), “there are obvious differences in the nature of the roles typically expressed by cases and ... by adpositions” (ibid.: 4). Generally, prepositions are semantically more specific,

“whereas cases are more abstract in nature (especially if a language has both)”

(ibid.). It is typical for cases to code the central participants in an utterance (like agent and patient), while prepositions code “more peripheral roles” (ibid.) which are connected with locations or indirect participants such as Beneficiary (She baked a cake for her daughter). And although the metaphorical senses of prepositions are more abstract than their source meaning, the extensions are generally also narrower in scope than case meanings. For example, INST case codes (broadly) the role of Instrument, whether prepositionless (“I zolotom i molotom” [‘With (i.e., by means of) gold and hammer’] (Jakobson 1936: 78)) or with a preposition (s zolotom i molotom ‘with gold and hammer’). The first example shows gold and hammer as the means for performing some action or

reaching a goal, while the second example shows them as (comitative) instruments, which do not originate within the agent42.

4.3.2 Grammatical vs. Semantic Case

Grammatical (or direct) cases are represented typically by NOM and ACC; other cases such as LOC and GEN are semantic (or oblique). Semantic cases are generally “semantically richer and thus less dependent on the verb” (Kittilä

2011: 4). In contrast, grammatical cases “typically code core grammatical relations such as subject and object” (ibid.). Semantically they are “rather vacuous and...derive their meaning largely from the verb” (ibid.). The semantic independence of prepositions, in contrast to that of cases, means that a preposition “usually retains its function irrespective of the verb it appears with”

(ibid.). This continuum is illustrated by Kittilä:

(2) Grammatical cases ---> Semantic cases ---> Adpositions verb-dependent verb-independent

42 The progression from abstract to concrete can be seen clearly in the Manner/Means/Instrument continuum of metaphorical extensions.

To a certain extent, the differences in the semantic specificity of these elements

“can ... be explained by the fact that cases (especially semantic ones) often originate from adpositions” (ibid.: 5).

4.3.3 Semantic Roles

The term semantic roles denotes the different functions that the participants in a scene play with respect to the situation in that scene. For example, the performer of a volitional activity is usually referred to as the Agent of that activity. Semantic roles are not bound to syntactic roles like subject, object, and indirect object, especially in light of the general semantic flexibility of prepositions; for example, a syntactical direct object is not necessarily also a

Patient: in Čerez nego mnogie postradali ‘Because of him, many suffered’, the prepositional object (nego ‘him’) functions as a Cause, a role that otherwise can be conveyed by the NOM, GEN, or INST case.

Although a prototypical Agent is animate (because s/he acts deliberately),

“an agent-like cause may not be animate: The wind rattled the windows” (Cruse

2011: 289), in which case it can be referred to as a Force, since it has no consciousness and is therefore incapable of agency, which must be deliberate. It can be stipulated that the role of “Experiencer can only occur in connection with a process or action, where animacy is crucially involved: Mary terrified John and

John heard the noise” (ibid.). This highlights one of several differences between area/topic (He read about the economy) and cause (He [was] worried about the economy). In the first example, the agent is depicted simply as reading (news, etc.) about the economy. The activity of reading does not entail that the reader must also experience what s/he is reading about, whereas being worried does entail this, since worrying implies the experiencing of emotion. Thus in the latter case the reader is an Experiencer, not an Agent.

4.4 The Area/Topic and Cause Extensions of ‘Aboutness’ Prepositions

A verb associated with an Area/Topic PP depicts a situation involving communication or cognition (speaking, writing, thinking, etc.) organized around an argument (the object referent of the PP) representing a topic. The situation denotes as a medium or process through which the agent either transmits the topic as a message, or manipulates it through a cognitive activity or state. In an adnominal construction like a book about cats, book represents a textual conveyer whose contents are organized around the topic of cats.

The Cause extension represents the outermost pole in the semantic continuum of prepositional senses. In the role of Cause,43 the ‘aboutness’ object

43 A Cause can represent either an animate or inanimate entity.

referent has agent-like properties (produces an effect on a patient or experiencer) but is not itself an agent, since it is not carrying out an activity. For example, the object referent can be depicted as the cause of an emotional/psychological state (worry, grief, excitement, etc.) in an Experiencer:

On volnovalsja o buduščem ‘He was worried about the future’. The

Experiencer is the grammatical subject of the reflexive verb.

The nonvolitionality of Cause stands in contrast to the volitionality required by a verb with a PP in the Area/Topic metaphorical extension. “The notion of mental movement is found with verbs denoting speaking and thinking

(think/doubt about): these activities are seen as movements over a mental area.

The same notion underlies emotional state predicates which denote emotional states caused by wandering over certain causes (excited/crazy about)” (Dirven

1993: 82). Dirven also notes that in English, the “causal about ... [is] very strongly related to the ‘area’ meanings of [this] preposition” (ibid.: 95). A Cause is essentially a Topic which sets up a change of state in the participant(s) in the discourse predication.

Certain aspects of emotional states such as worry, grieve, etc., are comparable to cognitive activities (think, remember, know, etc.), however, an emotional state is typically understood to have a cause which lies outside the experiencer’s own mind. To be worried about something is to not only be aware

of a particular external entity, but to have one’s emotions affected by that awareness.

Chapter 5: Verbs/Situations Associated with the Four ‘Aboutness’ Prepositions

5.0 Overview

In arranging into specific semantic categories the set of verbs with which o+LOC, pro, nasčet, and po povodu most often occur, it is useful to consider the specific meanings of the verbs with which ‘aboutness’ PP’s occur, since some meanings depend upon the context. For example, the core meaning of slyšat’

‘hear’ concerns the aural perception of sound: Ja slyšal šum ‘I heard a noise’, i.e., ‘I aurally perceived sound waves as they travelled through the air’. However,

Ja slyšal o vašix uspexax ‘I’ve heard about your successes’ metaphorically extends the basic meaning into ‘learn, come to know (about)’ and carries no necessary implication of aural transmission.

Most verbs found with the four prepositions under investigation describe one or more parts of the process by which information is encoded, transmitted, and decoded by the participants within the discourse or statement. In its entirety, this process represents a signal path:

Figure 5.2 Simple Model of Message/Signal Transmission (Adapted from Cruse (2004))

As shown in Figure 5.3, in an act of communication a sender encodes a message, which is transmitted (by the sender or an intermediary). The message is then decoded (by the recipient or an intermediary), and the recipient receives it. Verbs of speech and of writing focus on the encoding and dissemination of information, while verbs of hearing and of reading focus on the decoding and mental processing of information.

In the context of an argument phrase with an ‘aboutness’ PP, the informational content of this message refers to an entity of some sort. An agent

(the sender) disseminates information about this entity (the knowledge/awareness of which resides in the agent/sender’s cognition) to the recipient through the media of speech or writing. This information becomes part of the receiver’s knowledge/awareness when the recipient decodes (interprets)

the information—by hearing it spoken or by reading it. Speaking, writing, hearing, asking, and reading are means by which knowledge/awareness, in the form of informational content, is transmitted between various sentient participants. Accordingly, some verbs connected with the semantic field of communication describe the encoding and transmitting of the information, while others focus on its receiving and decoding. In this study, I have arranged the given verbs into categories which reflect their physical manifestation, for the most part. The exception to this is the placement of verbs of asking into their own discrete category, rather than with verbs of speech; I have done this because, although asking is typically performed through the medium of speech, it (asking) has an additional semantic component not present in the ordinary process of speech: the expectancy of information from the patient (the person being asked about something). Asking, unlike speaking, focuses on the potential response of the patient. In this way, it is more complex than speaking.

As I will show, some of the activities described by the verbs in this study fit into the above model of signal transmission, while others do not.

5.1 Verbs of Speech/Oral Communication

I have categorized the verbs govorit’/pogovorit’ ‘talk’, skazat’ ‘say’, rasskazyvat’/rasskazat’ ‘tell, relate’, boltat’ ‘chat’, dokladyvat’/doložit’ ‘report’, molčat’ ‘be silent’, and tolkovat’ ‘explain, expound’ as verbs of speech and oral communication (hereafter ‘verbs of speech’). They describe various processes of communicating information, typically through the medium of speech, although not always in sensu stricto; they can also refer to metaphorical communication:

Dejstvija govorjat gromče slov ‘Actions speak louder than words’. In the context of a signal path, they encompass both the encoding and the transmitting of the information.44 The other part of the signal path—the receiving and decoding of information—is described by verbs of reading (§5.4) verbs of hearing (§5.6).

Of the Russian verbs of speech, the most frequent words, govorit’/skazat’,

“denote speech activity either as the process of speaking or as conveyance of information, a communication” (Merzon and Pyatetskaya 1983: 9). They present speaking as a generic activity, and focus only on the activity itself, whereas the other verbs in this category describe the conveyance of (spoken) information.

“[R]asskazyvat’-rasskazat’ means to convey something verbally, to narrate, relate something” (ibid.: 44), which encompasses both the encoding and transmitting of information:

44 Logically, if information is encoded, it must also be transmitted; therefore, activities such as speaking assume both parts of this process.

(1a) Ja potom rasskazala Dmitriju ... ob obstojatel’stvax našego znakomstva.45 ‘I later told Dmitrii Petrovich ... about the circumstances of our acquaintance.’

(1b) [I] dolgo ešče, oxotno, rasskazyvaet mne pro “strašnyj xarakter” popovna ...46 ‘[A]nd for a long time, eagerly, he tells me about the “dreadful character” of the priest’s daughter ...’

(1c) A pomnite ... kak odin oficer rasskazyval nasčet perepravy?47 ‘Do you remember ... how one officer was telling about the [river] crossing?’

As these activities center on the communication of a specific topic, they require some sort of object in the form of an ‘aboutness’ PP. Similarly, the aspectual pair dokladyvat’/doložit’ ‘report (on)’ also describe the communication of a specific topic, but more in the capacity of prepared information:

(2a) Soldat každyj den’ ... dokladyval o želanii putnikov videt’ velikogo Gudvina ...48 ‘Every day the soldier ... reported on the travellers’ desire to see the Great Goodwin.’

(2b) Ja ved’ v Morskom ministerstve rassčityval doložit’ tolkom nasčet bronenoscev.49 ‘In the Naval Ministry, I thought I was going to report properly on aircraft carriers.’

45 Berg, R. (1983) Suxovej. In Znanie sila. 2003. (RNC) 46 Gippius, Z. (1923) Zadumčivyj strannik (o Rozanove). (RNC) 47 Stadnjuk, I. (1956) Maksim Perepelica. (RNC) 48 Volkov, A. (1939) Volšebnik Izumrudnogo goroda. (RNC) 49 Ratušinskaja, I. (1998) Odessity. (RNC)

Molčat’ translates ‘be/remain silent’. This can also imply a refusal to speak about something:

(3) A ja im pro maz’ molču. Začem starost’ju da bolest’ju ixnee vesel’e portit’!50 ‘I’m not going to tell them about the ointment. Why spoil their happiness with old age and sickness?’

In this situation, molčat’ is not a state but an activity—an incomplete communicative action between a sender and a recipient. This represents a signal path (albeit an incomplete one), because the activity is transitive—directed toward a recipient (the dative pronoun im ‘[to] them’).51

5.2 Verbs of Asking

This category consists of verbs which denote inquiring about information.

Sprašivat’/sprosit’ denote “ask[ing] somebody a question with the purpose of getting an answer or explanation” (Merzon and Pyatetskaya 1983: 28).

Rassprosit’/ rassprašivat’, in contrast, represent a more detailed inquiry:

50 Kuvaev, O. (1975) Kto-to dolžen kurlykat’. (RNC) 51 Unlike in Russian, ‘be silent’ can only be expressed intransitively in English.

(4a) A ja dejstvitel’no xotela sprosit’ tebja ob odnoj vešči.52 ‘And I really wanted to ask you about one thing.’

(4b) Potom otec sprašival menja pro moi otmetki.53 ‘Then (my) father asked me about my grades.’

(4c) [J]a xotela rassprosit’ Kostju popodrobnee ob ix žizni.54 ‘I wanted to ask Kostja more about their life.’

(4d) O.I. dolgo rassprašivala otca pro byvšie u nego so mnoj razgovory.55 ‘For a long time, O.I. questioned her father about his conversations with me.’

Spravljat’sja/spravit’sja denote inquiring about or checking on the condition of an entity or situation:

(5a) Zvonila že Aleksandra ... spravljat’sja ob otce i soobščat’ o zdorov’e careviča.56 ‘Aleksandra ... would call ... to ask after (her) father and to report on the health of the tsar’s son.’

(5b) Prixodili čužie spravit’sja nasčet ee zdorov’ja, prinosili prosfory, govorili, čto za nee segodnja počti vo vsex cerkvax služili molebny.57 ‘Strangers came to inquire about her health, they brought prosphoras; they were saying that today in almost all churches, prayers were offered for her.’

52 Kin, V. (1928) Po tu storonu. (RNC) 53 Allilujeva, S. (1963) Dvadcat’ pisem drugu. (RNC) 54 Medvedev, V. (1957) Barankin, bud’ čelovekom! (RNC) 55 Rediger, A. (1918) Istorija moej žizni. (RNC) 56 Rasputina, M. (1960) Rasputin: počemu? (RNC) 57 Chekhov, A. (1895) Tri goda. (RNC)

Evidence from RNC texts shows pro to be strongly disfavored with spravljat’sja/spravit’sja.58 I will discuss this in more detail in §5.7.

5.2.1 Prosit’/Poprosit’ with an ‘Aboutness’ PP

In contrast to sprašivat’/sprosit’ and rassprašivat’/rassprosit’, prosit’/poprosit’ denote the addressing of someone “with the purpose of obtaining something the speaker wants” (Merzon/Pyatetskaya 1983: 32):

(6a) Esli rebënok prosit konfet, nado emu snačala dat’.59 ‘If the child is asking for candy, you need to give it to him at first.’

(6b) Poprošu vrača o snisxoždenii.60 ‘I’ll ask the doctor for leniency.’

These verbs can take a direct object argument (ACC or GEN), or an argument in the form of an ‘aboutness’ PP. The latter is typically a situation or activity as opposed to an entity, and it takes the form of a deverbal or quasi-deverbal noun: prosit’ o razrešenii, pomošči, sodejstvii, molčanii, proščenii, pomilovanii, poščade

‘ask-IPFV for permission, help, cooperation, silence, forgiveness, pardon, mercy’.

58 91% of the total occurrences in the RNC of spravljat’sja (IPFV) with the four ‘aboutness’ prepositions are with o+LOC. For spravit’sja (PFV), 90% are with o+LOC. For the purposes of this study, I consider the threshold for a preposition to be ‘disfavored’ with the object referent of a given verb to be 20% or less. 59 Novikova, I. (2003) “Preodelenie illjuzij (o romane Aleksandra Melixova Ljubov’ k otečeskim grobam). In Oktjabr’ (2003). (RNC) 60 Vysockij, V. (1968) Žizn’ bez sna. (RNC)

The collocation of prosit’/poprosit’ with an ‘aboutness’ PP represents a politeness strategy (where the illocutionary intent is to convey a directive speech act, in which the speaker attempts to get the hearer to do something) whereby the person doing the requesting is deflecting the pragmatic directness. The requested entity is portrayed semantically as a Topic (although it is actually a

Proposition), making it appear more peripheral than a direct request. English ask~ask for represents the same strategy: He asked my forgiveness (direct)~He asked for my forgiveness (indirect).

Evidence from RNC texts shows pro to be strongly disfavored with prosit’/poprosit’.61 I will discuss this in more detail in §5.8.

5.2.2 Other Activities Related to Asking/Requesting

In addition to prosit’/poprosit’, two other verbs describing direct requests are found almost exclusively with o+LOC: umoljat’/umolit’ ‘beg, beseech’, vzyvat’/vozzvat’ ‘appeal’, and xodatajstvovat’/poxodatajstvovat’ ‘solicit (for), petition (for)’.

Vzyvat’/vozzvat’ and umoljat’/umolit’ are similar to prosit’/poprosit’, but imply an attitude of urgency or emotional intensity toward the referent:

61 93% of the total occurrences in the RNC of prosit’ (IPFV) with the four ‘aboutness’ prepositions are with o+LOC. For poprosit’ (PFV), 87% the total occurrences are with o+LOC.

(7a) Tysjači umirajuščix ljudej vzyvajut o pomošči k revoljucionnomu narodu!62 ‘Thousands of dying persons are appealing to the revolutionary people for help!’

(7b) On kak by metalsja po doktoram i umoljal o skorejšej dezinfekcii.63 ‘He sort of rushed to various doctors and begged for the quickest possible disinfection.’

Xodatajstvovat’/poxodatajstvovat’ ‘petition (for)’ describe a formal request carried out on behalf of an entity or situation, the purpose of which is get the addressee to do something relevant to that entity/situation:

(8) Pervyj ėtaž institutskogo zdanija otdajut Gorzdravu, i Kramov xodatajstvuet o peredače laboratorii v Rybtrest.64 ‘They’ll give the first floor of the institute building to the Health Department, and Kramov is petitioning for the transfer of the laboratory to the Fishing Cartel.’

These requests are for action rather than physical entities, and are often directed toward a group of people or an institution. The verbs in (7a)-(8) imply subsequent actions on the part of the hearer, in contrast to sprašivat’/sprosit’ and rassprašivat’/rassprosit’, which are simply inquiries for information about a Topic.

The semantic role of an object referent of the verbs in (7a)-(8), on the other hand, is that of a Proposition.

62 Lyndina, Ė. (1974) “Vyrosli my v plameni”. In Sovetskij ėkran. (RNC) 63 Mandel’štam, O. (1930) Četvertaja proza. (RNC) 64 Kaverin, V. (1949-56) Otkrytaja kniga. (RNC)

5.3 Verbs of Cognition

The verbs of cognition in this study include znat’ ‘know-IPFV’, dumat’/podumat’ ‘think’, razmyšljat’/razmyslit’ ‘think, ponder’, sudit’ ‘judge-IPFV’, uznavat’/uznat’ ‘find out, learn (about)’, pomnit’ ‘remember-IPFV’, vspominat’/vspomnit’ ‘recall, recollect’, napominat’/napomnit’ ‘remind’, and zabyvat’/zabyt’/pozabyt’65 ‘forget’. These activities describe the awareness or mental manipulation of thoughts and ideas. Since most of these cognitive activities take place entirely within the mind of the agent, they do not instigate a signal path; no information is actually transmitted, since they involve only one person. Thus, they do not denote communication, only its cognitive manipulation. The exception to this lies with napominat’/napomnit’, which can be verbs of cognition or of speech.

Znat’ ‘know-IPFV’ denotes a state of awareness of the existence of an entity or situation, or knowledge about the nature, details, etc. of an entity or situation:

(9a) Papa i babuška znajut o vizite M.?66 ‘Papa and grandmother know about M’s visit.’

65 Zabyvat’ is imperfective, while zabyt’ and pozabyt’ are perfective. 66 Dubrovin, P. (1950) Annet. (RNC)

(9b) Esli by čelovek znal by pro ošibki, on by ix ispravil.67 ‘If a person would know about his mistakes, he would fix them.’

Dumat’/podumat’ denote the basic thought process which involves certain information—“to meditate, to centre one’s thoughts on smb. or smth.”

(Merzon/Pyatetskaya 1983: 72). The intensity of this ‘centering’ is nonspecific.

However, razmyšljat’/razmyslit’ ‘think, ponder’ denote a more intense mental focus than dumat’/podumat’, and which typically occurs over a longer period of time:

(10a) Pozvoni emu po telefonu, kak on nasčet tovarišča Lenina dumaet.68 ‘Call him on the telephone [to ask him] what he thinks about comrade Lenin.’

(10b) Ja zakuril sigaru, uselsja na divan i časa dva razmyšljal o pervom voprose.69 ‘I lit a cigar, sat down on the couch, and for two hours or so pondered over the first question.’

Evidence from RNC texts shows pro to be strongly disfavored with object referents of razmyšljat’.70

Sudit’ ‘judge-IPFV’ denotes the rendering of an opinion or judgment (either cognitively or communicatively) about an entity or situation:

67 Krapivin, V. (1962) Belyj ščenok iščet xozjaina. (RNC) 68 Kassil’, L. (1928-1931) Konduit i Švambranija. (RNC) 69 Apuxtin, A. (1891) Dnevnik Pavlika Dol’skogo. (RNC) 70 96% of the total occurrences in the RNC of razmyšljat’ with the four ‘aboutness’ prepositions are with o+LOC.

100

(11) On be bralsja sudit’ o damskix boleznjax, -- ėto bylo vne sfery ego kompetentnosti.71 ‘He would not undertake to judge ladies’ illnesses—this was outside the sphere of his expertise.’

Evidence from RNC texts shows pro to be strongly disfavored with object referents of sudit’.72 I will discuss both sudit’ and razmyšljat’/razmyslit’ in more detail in §5.8.

Uznavat’/uznat’ denote the discovery, learning, or realizing of information, usually concerning an event or state:

(12a) Vot čto napisal mne brat, kogda uznal ob isčeznovenii Sergeja Sergeeviča.73 ‘That’s what my brother wrote me when he found out about the disappearance of Sergej Sergejevič’.

A state or condition, the details of which are filled in over time, is expressed by the imperfective uznavat’:

(12b) Pro boevye dela ego tanka my uznavali so slov ėkipaža...74 ‘We were finding out [piecemeal] about the tank’s military achievements from the words of the crew.’

71 Veselickaja, L. (1891) Mimočka. (RNC) 72 98% of the total occurrences in the RNC of sudit’ with the four ‘aboutness’ prepositions are with o+LOC. 73 Bragin, V. (1971) V strane dremučix trav. (RNC) 74 Tolstoj, A. (1942-44) Russkij xarakter. (RNC)

101

Pomnit’ ‘remember, keep in mind-IPFV’ denotes the keeping of awareness/knowledge in mind, while vspominat’/vspomnit’ focus more on the recollecting or recalling of awareness/knowledge:

(13a) Pri vybore kostjuma nužno pomnit’ pro množestvo pravil, objazatel’nyx dlja pridanija emu priličnogo vida.75 ‘When choosing a suit, you need to keep in mind the multitude of rules necessary for giving it a decent appearance.’

(13b) Ona vspominala s blagodarnost’ju o svoem učitele.76 ‘She recalled her teacher with gratitude.’

Pomnit’ depicts more of a dynamic process than vspominat’/vspomnit’, which denote a more punctual type of mental recall.

Napominat’/napomnit’ ‘remind’ have two senses, one denoting a volitional, communicative activity (as in He reminded me about my promise), and the other, the non-volitional triggering of a mental or sensory association within an

Experiencer (He reminded me of my father). In this study, I refer to the speech sense of napominat’/napomnit’ as REMIND1, and the cognition sense as REMIND2:

(14a) No tebe ne sledujet napominat’ mne o moem dolge, bojarin.77

‘But you shouldn’t remind me about my debt, boyar.’ [REMIND1]

75 Stocakaja, L. (2004) “Mundir delovogo čeloveka”. In Biznes-žurnal, 2/13/2004. (RNC) 76 Čulkov, G. (1930) Gody stranstvij. (RNC) 77 Vasil’ev, B. (1996) Veščij Oleg. (RNC)

102

(14b) Ėtot nežnyj i strastnyj romans, ispolnennyj velikoj artistkoj, vdrug napomnil vsem ėtim ženščinam o pervoj ljubvi.78 ‘This tender and passionate romance, performed by the great artist, suddenly reminded all of these women of [their] first love.’ [REMIND2]

In the RNC (and in the datasets for the quantitative study in the present work), these verbs are rarely found in the REMIND2 sense when used with a pro PP. I will discuss them in more detail in §5.10.

The verbs zabyvat’/zabyt’pozabyt’ denote the nonvolitional losing of awareness or knowledge about something:

(15a) Vy ešče i pro Akademiju nauk ne zabyvajte.79 ‘Don’t forget about the Academy of Sciences.’

(15b) Čerez neskol’ko dnej v Rossii zabyli ob ėtom ubijstve, no ne zabyli vo dvorce.80 ‘After a few days in Russia, they forgot about this murder, but not in the palace.’

(15c) Ostavšis’ v živyx, druz’ja kak-to pozabyli o svoem železnom druge.81 ‘Left among the survivors, the friends somehow forgot about their iron friend [a vehicle].’

78 Kuprin, A. (1915) Jama. (RNC) 79 Sarnov, B. (1969) “Jura Krasikov tvorit čudesa”. In Pioner. (RNC) 80 Tolstoj, A. (1922) Xoždenie po mukam. Kniga pervaja: Sestry. (RNC) 81 Čepoveckij, E. (1989) Neposeda, Mjakiš i Netak. (RNC)

103

Zabyvat’/zabyt’pozabyt’ are verbs of low agency and low volition. Even when there is a conscious desire to forget something, the activity is only weakly volitional.

5.4 Verbs of Reading

The activity of reading a written text represents the decoding and mental absorption of the informational content of that text. An encoded message (topic) is subsequently decoded by a person possessing the skill to do so. Although

čitat’/pročitat’ can be used in certain contexts to denote the ability to read, or the physical act of reading (Ja čital do boli v glazax ‘I read until my eyes hurt’;

Mal’čik uže čitaet ‘The boy already reads’, i.e., he knows how to read), their typical use is to describe the attainment of awareness or knowledge about a topic by means of the printed word:

(16a) V kupė poezda v Ženevu Savinkov čital ob ubijstve velikogo knjazja Sergeja.82 ‘In the compartment of the train to Geneva, Savinkov read about [i.e., learned of] the murder of Grand Duke Sergei’.

(16b) Mnogim mne kažetsja, budet interesno pročitat’ pro krest’janskogo parnja, kotoryj stal trexzvezdnym generalom.83

82 Gul’, R. (1958) Azef. (RNC) 83 Sirkes, P. (1990-1999) Truba isxoda. (RNC)

104

‘It seems to me that for many people, it will be interesting to read about a peasant who became a three-star general.’

5.5 Verbs of Writing

The activity of writing (pisat’/napisat’) involves the encoding of textual information so that it may be subsequently decoded through reading. It can be considered an extension of speech:

Writing as an institution is subsequent to speech and seems merely to fix in linear script all the articulations that have already appeared orally. The attention given ... to phonetic writings seems to confirm that writing adds nothing to the phenomenon of speech other than the fixation that enables it to be conserved. Whence the conviction that writing is fixed speech. (Ricœur, P., 2008, 102-103).

Pisat’/napisat’ can denote the ability to write, or the physical act of writing, but they are most often employed to describe the recording and/or dissemination of written information:

(17a) Pišu Vam po povodu kandidatury I.E. Tamma na kafedru fiziki v Akademiju.84 ‘I’m writing to you about the candidacy of I.E. Tamma for the chair of physics at the Academy.’

(17b) Ja xoču skazat’, čto my ne možem xorošo napisat’ o kolxoze, esli naš impul’s—derevenskie pejzaži.85

84 Kapica, Tamm, Semenov (1998) Sbornik vospominanij i statej. (RNC) 85 Ginzburg, L. (1920-1943) Zapisnye knižki. Vospominanija. Ėsse. (RNC)

105

‘I want to say, we can’t write anything good about the collective farm if our impulse is toward country landscapes.’

The assumption in (17b) is that whatever will eventually be written about collective farms (whether positive or negative in attitude) will also be disseminated.

As with reading, writing is an acquired skill which is carried out intentionally. It is possible, and seems logical, that written information reflects a deeper or more serious consideration of a given topic than would be the case with spoken words. Not only does writing preserve information in a way that speech cannot (in the form of an ‘external’ memory), a written text is often intended for a larger audience than the spoken word is. Vox audita perit, litera scripta manet.86

Verbs of speech and of writing focus on the encoding and dissemination of information, while verbs of hearing and of reading focus on the decoding and mental processing of information.

5.6 Verbs of Hearing

87 The verbs in this category (slyšat’/uslyšat’ ‘hear’ and slyxat’ ‘hear (about)-IPFV’ ) describe the aural perception of sound. With an ‘aboutness’ PP however, they

86 ‘A heard voice perishes, but the written letter remains.’ 87 Slyxat’ is usually marked as colloquial, and is used only in the past tense.

106

focus specifically on the informational content of the spoken words which are the source of the information. They also entail that the information has become part of the hearer’s mental awareness; the hearer learns of (spoken) information:

(18a) My uže slyšali o vašix vystuplenijax na delijskoj konferencii.88 ‘We have already heard about your speeches at the Delhi conference.’

(18b) Starik Kolobov tol’ko axnul, kogda uslyšal pro novuju zateju.89 ‘The old man Kolobov only gasped when he heard about the new scheme.’

(18c) A kak tam... ne slyxat’ nasčet vtorogo fronta?90 ‘How could you not hear about the second front?’

Although grammatically, verbs of hearing are active in voice, there is a passive, unintentional, quality inherent in them; they lack voluntary participation.

This means that the hearer is semantically an Experiencer, not an Agent. The speaker of the information has not addressed the hearer directly; the hearer has

‘intercepted’ the message—a message which, it is safe to assume, was transmitted with the intention of reaching either someone besides the

Experiencer, or of reaching many Experiencers at once, as a sort of ‘public announcement’. And since the informational content of that message is exactly the same as it would be if it were transmitted to someone other than the

Experiencer, it is semantically a Topic.

88 Efremov, I. (1959-63) Lezvie britvy. (RNC) 89 Mamin-Sibirjak, D. (1895) Xleb. (RNC) 90 Gorbatov, B. (1945) Doroga na Berlin. (RNC)

107

5.7 Verbs of Emotional State (Psychological Verbs)

The verbs in this semantic category describe emotional states such as bespokoit’sja ‘be worried-IPFV’, volnovat’sja ‘be worried/agitated-IPFV’, žalet’

‘regret, be sorry for-IPFV’, skorbet’ ‘grieve, mourn-IPFV’, and toskovat’ ‘feel sad, languish-IPFV’. The object referent of the prepositional argument of these verbs codes the semantic role of Cause (which can be seen as a Topic of prior cognition). These verbs, like verbs of cognition, do not involve the communication of information. Because verbs of emotional state are so rarely attested in the RNC with ‘aboutness’ prepositions other than o+LOC, there were not enough examples with which to construct a valid quantitative study as part of this dissertation. Nevertheless, their overall importance warrants their inclusion in the qualitative discussion of this study.

(19a) Osmatrivaja nas, doktor vse vremja bespokoilsja o devuške.91 ‘Examining us, the doctor was worried the whole time about the girl.’

(19b) Očen’ volnujutsja v Minobrazovanija po povodu vozmožnoj fal’sifikacii rezul’tatov testov.92 ‘They’re very nervous in the Ministry of Education about the possible falsification of test results.’

(19c) I ja toskuju o pervyx našix otčajannyx nočax.93

91 Sidur, V. (1973-74) Pamjatnik sovremennomu sostojaniju. (RNC) 92 Šic, M. (2003) “Edinyj i nedelimyj.” In Rossijskaja gazeta. (RNC) 93 Xruckij, A. (2001) “Okajannye dni Ivana Aleksejeviča.” In Zvezda. (RNC)

108

‘And I feel miserable about our first desperate nights.’

(19d) V knige on pišet, čto mnogo raz žalel o vybrannom puti.94 ‘He writes in [his] book that he often regretted the path [he’d] chosen.’

(19a)-(19d) each portray an Experiencer feeling an emotion associated with the object referent of the ‘aboutness’ PP. Although, as with verbs of cognition, emotional states entail an awareness of the entity or situation represented by the object referent, the object referent of an emotional state does not constitute a message. Therefore, it is not semantically a Topic. Nor does not constitute a

Proposition, since the entity/situation to which it refers is not part of a request or supplicative act. For both Topic and Proposition, verbs associated with an object referent of an ‘aboutness’ PP must depict a volitional activity. The verbs in (19a)-

(19d), however, depict nonvolitional states. Thus, their object referents code the semantic role of Cause, and the individual experiencing the emotional state is semantically an Experiencer.

A state of awareness is inherent within emotional as well as cognitive states. However, in being worried about something, for example, the

Experiencer’s emotional state is (semantically, at least) determined by the

94 Peskov, V., and Strel’nikov, B. (1977) Zemlja za okeanom. (RNC)

109

external entity. An Experiencer does not initiate the state s/he is experiencing; it is a nonvolitional reaction to an external entity or situation.

Zabotit’sja/pozabotit’sja ‘look after, take care (of)’ depict action carried out in response to an emotion (i.e., an individual being troubled by an external entity or situation). The results of this action will either resolve the situation or benefit the external entity:

(20a) Siničke bol’še ne nado bylo zabotit’sja o nočlege.95 ‘The titmouse did not have to concern herself anymore about lodging for the night.’

(20b) “Russkoe bistro” zabotitsja o svoix klientax, odnako, ix vkusy ne vsegda sovpadajut s predložennym menju.96 ‘The “Russian bistro” takes care of its clients—however, their tastes don’t always coincide with the proposed menu.’

The semantic role of the object referents in (20a) and (20b) is that of Stimulus.

This role differs from that of Cause, in that it is associated with a volitional activity

(which is performed by an Agent) as opposed to a nonvolitional state (which is experienced by an Experiencer). I maintain that with the verbs razmyšljat’/razmyslit’ ‘ponder’, which I discussed in §5.3, an ‘aboutness’ object referent, as with zabotit’sja, is semantically a Stimulus for the activity.

95 Bianki, V. (1923-58) Lesnye byli i nebylicy. (RNC) 96 “Fast food! Kak mnogo v ėtom slove…” Reklamnyj mir (3/30/2000). (RNC)

110

As I mentioned in §5.2, spravljat’sja/spravit’sja ‘inquire, check (on)’ rarely occur with object referents of pro, yet an object referent with these verbs is semantically a Topic. The scarcity of pro with object referents of spravljat’sja/spravit’sja in the RNC appears to be an exception to the general pattern by which this preposition is seen to occur chiefly with verbs whose object referent is semantically a Topic.

5.8 A Few Other Verbs Used with ‘Aboutness’ Prepositions

In the course of completing the pilot project for the present study, I encountered several verbs associated with ‘aboutness’ PP’s which, although possessing certain characteristics of verbs of speech, are more complex. As with those verbs discussed in this chapter which occur primarily (or solely) with o+LOC in the RNC, if we examine the relationship between the activity and the object referent, a pattern can be discerned, one which allows us to extrapolate certain things about o+LOC and pro.

111

With svidetel’stvovat’ ‘witness, testify-IPFV’ (which is strongly favored with o+LOC object referents97), a entity or situation serves as evidence of a specific fact, condition, or characteristic associated with the object referent:

(21a) On svidetel’stvoval o mučitel’no dlinnoj čerede mnogoletnix zatjažnyx doždej.98 ‘He witnessed the excruciatingly long bouts of perennial rains.’

(21b) Nazyvalos’ ėto “otkrovenijami”, dolžno bylo svidetel’stvovat’ o mističeskoj sile Džonsa.99 ‘These were called “revelations”; they should bear witness to [Jim] Jones’ mystical power.’

In (21a), the Agent has personally observed, or can vouch for, the existence of extended periods of rain. In (21b), certain questionable revelations are presented as direct evidence of a characteristic of the referent (Jone’s mystical power). Witnessing/testifying does not constitute an assertive speech act (as would be the case, for example, with He heard/spoke/read about the excruciatingly long bouts of perennial rains), but rather, is constative100; the speaker is committing to the truth of the Proposition (the object referent).

97 99% of the total occurrences in the RNC of svidetel’stvovat’ with the four ‘aboutness’ prepositions are with o+LOC. 98 Kataev, V. (1975-77) Almaznyj moj venec. (RNC) 99 Vaxtin, B. (1978-80) Gibel’ Džonstauna. (RNC) 100 See Bach, K. “Speech Acts” Routledge Encyclopedia of Philosophy entry on website http://online.sfsu.edu/~kbach/spchacts.html. Viewed 4/7/2012.

112

The verb propovedovat’ ‘preach-IPFV’ appears to be oriented mainly toward the communication of information about a Topic, as with verbs of speech such as rasskazyvat’/rasskazat’:

(22) Ty že ne sobiraeš’sja za čas do svoej smerti propovedovat’ mne o pol’ze mašin?101 ‘You’re not going to preach to me about the benefit of the machines, right up to the hour of your death?’

The speaker in (22) regards the addressee’s continuous harping on the benefit of machines in a dystopian future society as either an attempt to get the speaker to accept the validity of the addressee’s assertions (a directive speech act), or the affirmation/insistence of the validity of a Proposition. In this way, although propovedovat’ appears to signify an assertive speech act, it is actually directive or constative in intent. The object referent is semantically a Proposition.

As speech acts, predupreždat’/predupredit’ ‘warn’ appear to represent an assertive illocutionary point, in which the speaker is representing a state of affairs as real, and as potentially harmful:

(23) Bolee togo, soglasno poslednim soobščenijam v mirovoj pečati, oni daže predupreždali amerikancev o teraktax, namečennyx na 11 sentjabrja.102 ‘Moreover, according to recent reports in the world press, they even warned Americans about the terrorist attacks planned for September 11th.’

101 Gluxovskij, D. (2005) Metro 2033. (RNC) 102 Umnov, A. “Neprikosnovennyj zapas”. In Afganskij perekrestok, 4/3/2003. (RNC)

113

The act of warning is somewhat similar to that of preaching. In both activities, the speaker asserts the validity of a given state of affairs, with the goal of getting the hearer to accept this validity as well. The purpose of the activity of warning is to get the hearer to accept the validity of a potentially harmful state of affairs, and take necessary action to avoid a negative outcome. Warn portrays a state of affairs as harmful, while preach typically portrays it as beneficial (though one can, of course, preach against a state of affairs considered to be harmful). With warn, as with preach, the speaker is either attempting to get the hearer to do something, or is affirming/insisting on the validity of a Proposition. As such, it is directive or constative in intent, and the ‘aboutness’ object referent of this activity is semantically a Proposition.

Rasporjažat’sja/rasporjadit’sja ‘order, make arrangements’ denote a directive for some further action on the part of the hearer:

(24) Prošu Vas vmešat’sja v ėto delo i rasporjadit’sja o sročnom otbore dlja nas kartin.103 ‘I am asking you to intervene in this matter and see to the immediate selection of pictures for us.’

These verbs are associated with directive speech acts, and their prepositional argument phrases convey Propositions—more specifically, the state of affairs which is the illocutionary point of the directive.

103 Kantor, Ju. “Real’nost’ i socrealizm: Ėrmitaž v 1917-1941 gg.” In Zvezda 2003, no. 5 (RNC)

114

As I discussed in §5.3, the verb sudit’ ‘judge-IPFV’ denotes the judging or assessing of an entity or situation represented by the ‘aboutness’ object referent.

As a speech act, judging can be seen as either assertive or expressive, the point of the latter being “to express a propositional attitude of the speaker about the state of affairs represented by the propositional content” (Vanderveken 1990:

106).

5.9 Occurrences of nasčet and po povodu in Non-Topic Semantic Roles

Searches of RNC texts created between 1890 and 2009 reveal that there are several occurrences of nasčet and po povodu with a several verbs of emotional state with which an ‘aboutness’ object referent has a non-Topic semantic role:

115

total RNC with with with with verb contexts, o+LOC pro nasčet po povodu 1890-2009 bespokoit’sja ≈ 5,800 ≈ 1,250 0 114 27 ‘be worried’ volnovat’sja ≈ 8,900 ≈ 320 1 18 29 ‘be worried, agitated’

Table 5.9 Occurrences of Object Referents of nasčet and po povodu in RNC, in Semantic Role of ‘Cause’

5.10 Summary

As I have demonstrated in this chapter, object referents with pro, nasčet, or po povodu rarely or never occur with certain verbs. Table 5.10 lists these verbs, along with the semantic role represented by an ’aboutness’ object referent of each verb:

116

Semantic Role of Type of Speech Act Verb Object Referent spravljat’sja/spravit’sja Topic assertive ‘inquire, check on’ prosit’/poprosit’ Proposition directive ‘request, ask for’ umoljat’/umolit’ Proposition directive ‘beg, beseech’ vzyvat’/vozzvat’ Proposition directive ‘appeal (for)’ xodatajstvovat’/poxodatajstvovat’ Proposition directive ‘solicit (for), appeal (for)’ svidetel’stvovat’ Proposition constative ‘witness, testify (to)’ propovedovat’ Proposition constative/directive ‘preach’ predupreždat’/predupredit’ Proposition constative/directive ‘warn’ rasporjažat’sja/rasporjadit’sja Proposition directive ‘order, make arrangements’ sudit’ ‘judge’ Proposition assertive/expressive napominat’/napomnit’ ‘remind’ Topic assertive (REMIND1) napominat’/napomnit’ ‘remind’ Phenomenon N/A (REMIND2) razmyšljat’/razmyslit’ Stimulus N/A ‘ponder’ zabotit’sja/pozabotit’sja ‘look after, take care (of)’ Stimulus N/A bespokoit’sja Cause N/A ‘be worried’ volnovat’sja Cause N/A ‘be worried, agitated’ toskovat’ Cause N/A ‘feel sad, languish’ žalet’ Cause N/A ‘regret, feel sad’

Table 5.10 Verbs that Favor Object Referents with o+LOC

117

Of the four ‘aboutness’ prepositions, only o+LOC is typically used with the verbs in Table 5.10 (as well as with all the remaining verbs discussed in this chapter).

Pro, nasčet, and po povodu, on the other hand, occur almost exclusively with verbs with which an ‘aboutness’ object referent is coded semantically as a Topic

(i.e., they express only an Area/Topic sense of ‘aboutness’). The situations in

Table 5.10 are larger in scope than those which involve the basic transmission and cognition of information (speaking, thinking, hearing, reading, writing, and asking about); they portray events/situations which involve a change of state in one or more of the participants in the discourse frame. The verbs spravljat’sja/spravit’sja ‘inquire, check on’, appear to be an exception to the general pattern, since, although they do not entail a change of state, and an

‘aboutness’ object referent of them is semantically a Topic, evidence from the

RNC shows these verbs to be disfavored with pro.

The verbs napominat’/napomnit’ ‘remind’ have two senses. The REMIND1 sense denotes a speech act—an assertive illocutionary point. In contrast,

REMIND2 denotes the nonagentive Phenomenon which is ‘triggered’ in the mind of an Experiencer. The fact that the distribution of the two senses of this verb in datasets with pro is heavily weighted toward the REMIND1 sense is further evidence that this preposition is favored with object referents whose semantic role is that of Topic.

118

Overall, the dichotomy between o+LOC and the other three prepositions shows that the overall semantic scope of o+LOC encompasses that of pro, nasčet, and po povodu, i.e., it functions as the unmarked ‘aboutness’ preposition.

Given the fact that all four prepositions share at least an Area/Topic semantic extension, it follows that the next task is to find the conditioning factor(s) for these prepositions within this shared context. A quantitative analysis of RNC data will help with this task.

119

Chapter 6: Quantitative Analysis of RNC Data: Results and Discussion

6.1 Introduction

I conducted the quantitative portion of the present study in order to ascertain, in the 253 datsets, whether and to what extent the factors pertaining to the OP—animacy, count/mass status, modification status, common/proper status, and deverbal status—and also the factor verbal aspect are statistically significant in the choice of one or other of the four ‘aboutness’ prepositions that are the focus of this study. As noted in Chapter 3, the given factors were found to be statistically significant within the dataset used in my pilot project (Kier

2009).

In the present study, in order to determine statistical significance, I performed a total of 1,174 individual tests on these datasets, using Fisher’s Exact

Test (two-tailed). Each test computed the exact probability of obtaining the distribution of values in 2x2 contingency tables, given the values in each cell.

The null hypothesis (H0) in each test states that the distribution of values in the cells can only be due to chance. If the resulting P-value is <0.05 (the accepted

120

threshold), the H0 can be rejected, and we can assert that this particular distribution of values cannot be due purely to chance. P-values of <0.05 are considered to be statistically significant.

I will discuss in this chapter the results of the quantitative tests (which reflect the independent variables pro, nasčet, and po povodu in comparison with o+LOC), and their significance in connection with the four prepositions in the study.

6.2 Tests Between Verbs Plus Individual Prepositions

In the first series of tests, the dependent variable consists of a verb plus the preposition o+LOC plus one of the five noun features, arranged in the factor groups (FG) Noun animacy, Count/mass noun status, Noun modification status,

Common/proper noun status, and Deverbal noun status. The independent variable consists of the same verb as with the dependent variable, plus pro, nasčet, or po povodu, plus each of the five noun factor groups. These five tests are carried out in each of the four time periods for each verb. This is illustrated in

Table 6.11, the arrangement of which is applicable in all verb categories in this study:

121

Dependent variable Independent variable dumat’ o + FG Noun animacy dumat’ pro + FG Noun animacy dumat’ o + FG Count/mass noun dumat’ pro + FG Count/mass noun dumat’ o + FG Noun modification dumat’ pro + FG Noun modification dumat’ o + FG Common/proper noun dumat’ pro + FG Common/proper noun dumat’ o + FG Deverbal noun dumat’ pro + FG Deverbal noun

Table 6.11 Arrangement of Tests for dumat’ with o+LOC (left) vs. pro (right) by FG

The distribution of values in each of the five tests performed on the verb skazat’, between o+LOC and pro, for Period A (1890–1930) are shown in Table

6.11. The numbers in the middle two columns represent the distribution of values of that factor group within the dataset for skazat’ with o+LOC or with pro.

In each cell they total 30 because each dataset (in this case) has a total of 30 tokens. For example, for the FG Noun animacy, the dataset for skazat’ + o+LOC

(Period A) had, in a total of 30 tokens, a distribution of 24 inanimate nouns and 6 animate nouns. This dataset was tested against the corresponding dataset for pro (i.e., skazat’ + pro) which had, in a total of 30 tokens, a distribution of 15 inanimate nouns and 15 animate nouns. The Fisher Exact Test between these two datasets gave a P-value of 0.0292, meaning that this particular distribution of animate and inanimate nouns cannot be due to chance alone. The pro dataset has a strong positive correlation with animate nouns.

122

Distribution of Distribution of FG values in o+LOC values in pro P-value tokens tokens inanimate: 24 inanimate: 15 Noun Animacy P=0.0292 animate: 6 animate: 15 count: 27 count: 28 Count/mass noun status P=1.0000 mass: 3 mass: 2 unmodified: 16 unmodified: 18 Noun modification P=0.7948 modified: 14 modified: 12 common: 29 common: 21 Common/proper noun status P=0.0122 proper: 1 proper: 9 non-deverbal: 23 non-deverbal: 29 Deverbal status of noun P=0.0523 deverbal: 7 deverbal: 1

Table 6.12 Distribution of Tests for skazat’ o+LOC and pro (Period A) by FG

As Table 6.12 shows, the P-values are statistically significant (<0.05) for both Noun animacy (P=0.0292) and Common/proper status (P=0.0122). The distribution of the two factors in each group shows a positive correlation between pro and animate nouns (15 inanimate, 15 animate) and between pro and proper nouns (21 common, 9 proper), in comparison with o+LOC (24-6, and 29-1, respectively). In other words, in comparison with o+LOC objects, pro objects show a stronger tendency to be animate and proper.

As for the remaining factor groups (Count/mass noun status, Noun modification status, and Deverbal noun status), the distribution of factors was not

123

statistically significant (although P=0.0523 for Deverbal noun status comes very close to the 0.05 threshold), which means that the distribution of these factors cannot be attributed to something other than chance. The P-value of 1.0000 for

Count/mass noun status shows that the distribution of count and mass nouns for this dataset is exactly the same for o+LOC as it is for pro (27-3 vs. 28-2), and that the comparative distribution of values in the two datasets can only be due to chance.

Overall, 635 individual tests were carried out between verbs with individual prepositions and noun features. Statistical significance was found in 75 instances. This is broken down by noun factor group and preposition in Table

6.13:

124

Statistically significant results Noun factor pro nasčet po povodu Totals Animate nouns 28 2 0 30 Noun animacy Inanimate nouns 0 0 4 4 Count nouns Count/mass noun 1 0 0 1 status Mass nouns 0 0 0 0 Unmodified nouns Modification status of 7 4 0 11 noun Modified nouns 0 0 0 0 Common nouns Common/proper noun 15 0 2 17 status Proper nouns 0 1 0 1 Non-deverbal noun Deverbal status of 11 0 0 11 noun Deverbal noun 0 0 0 0 Totals 62 7 6 75

Table 6.13 Distribution of Tokens in the First Series of Tests

Statistical significance (P=<0.05) means that there is a better-than-random possibility that the choice of the ‘aboutness’ preposition is correlated with the noun factor for a given verb. For example, it can be seen in Table 6.13 that in all

28 statistically significant results for pro in the factor group Noun animacy, there was a positive correlation between pro and animate nouns.

125

6.3 Tests Between Aspectually Paired Verbs Plus Prepositions

The second series of tests was performed on aspectually paired verbs, each with the same preposition (e.g., rasskazyvat’ + o~rasskazat’ + o ‘tell, relate-

IPFV~PFV’). In these tests, the dependent variable consisted of an imperfective verb plus a preposition plus each of the five noun factor groups. The independent variable is the perfective form of the verb in the dependent variable plus the same preposition. Five tests were carried out in each individual time period.

Table 6.14 illustrates an example of the set-up of the second type of test, which is applicable to all aspectually paired verbs in this study. The verbs in the table are čitat’~ pročitat’ ‘read-IPFV~PFV’:

Dependent variable Independent variable čitat’ o + FG Noun animacy pročitat’ o + FG Noun animacy čitat’ o + FG Count/mass noun pročitat’ o + FG Count/mass noun čitat’ o + FG Noun modification pročitat’ o + FG Noun modification čitat’ o + FG ‘Common/proper noun’ pročitat’ o + FG ‘Common/proper noun’ čitat’ o + FG Deverbal noun pročitat’ o + FG Deverbal noun

Table 6.14 Arrangement of Tests for čitat’/pročitat’ ‘read’

126

Table 6.15 shows the distribution of statistically significant results in the second series of tests:

o+LOC pro nasčet po povodu Noun factor PFV IPFV PFV IPFV PFV IPFV PFV IPFV Totals Animate nouns 0 0 1 1 1 1 0 0 4 Inanimate nouns 0 0 1 0 0 0 0 0 1 Count nouns 0 1 0 0 0 0 0 0 1 Mass nouns 0 0 0 0 0 0 0 0 0 Unmodified nouns 2 1 0 2 0 0 0 0 5 Modified nouns 1 2 0 0 0 0 0 0 3 Common nouns 1 0 1 1 0 0 0 1 4 Proper nouns 0 0 1 0 0 0 0 0 1 Non-deverbal nouns 0 1 0 0 0 0 0 0 1 Deverbal nouns 0 0 0 0 0 0 0 0 0 Totals 4 5 4 4 1 1 0 1 20

Table 6.15 Distribution of Statistically Significant Results in the Second Series of Tests (Indicating Positive Correlations)

127

6.4 Verbs of Speech

6.4.1 Verbs of Speech with Individual Prepositions

In the category Verbs of speech, 185 individual tests were carried out; statistical significance was found in 21 instances. In the tests between o+LOC and pro, the factor group Noun animacy was statistically significant in seven out of 21 tests; in all of these, pro has a positive correlation with animate nouns. The largest cluster occurs with the verbs rasskazyvat’ and rasskazat’ ‘tell, relate’

(Table 6.16):

rasskazyvat’ (IPFV) rasskazat’ (PFV) Period A P=0.0004 P=0.0251 Period B P=0.0470 P=0.0716 Period C P=0.1806 P=0.0153 Period D P=0.2092 P=0.0211

Table 6.16 Significant Results for FG Noun Animacy for rasskazyvat’/rasskazat’ ‘tell, relate’ with o+LOC vs. pro

The positive correlation between pro and animate nouns tapers off gradually over time with rasskazyvat’, while with rasskazat’, it remains more consistent. This indicates an increase (in the RNC) in the use of pro with inanimate noun objects with rasskazyvat’ over time.

128

Tests on the four other factor groups (Count/mass noun status, Noun modification, Common/proper noun status, and Deverbal noun status) showed scattered instances of significant results, but no clusters as with rasskazyvat’/rasskazat’ and animate nouns. With skazat’ ‘say-PFV’, pro has a positive correlation with animate nouns in Period A, which, in the same manner as with rasskazyvat’ ‘tell, relate-IPFV’, tapers off over time.

Overall, with verbs of speech, the significant results for pro indicate positive correlations with nouns which are either animate, unmodifed, proper, or non-deverbal.

In tests with o+LOC and nasčet with the FG Noun modification, there were three instances of statistical significance (skazat’ ‘say-PFV’ for Periods A and D, and govorit’ ‘say-IPFV’ Period D). All of these showed a positive correlation between nasčet and unmodified nouns.

6.4.2 Aspectually Paired Verbs of Speech

In this series of tests, 170 individual tests were carried out on the aspectual pairs govorit’/skazat’, govorit’/pogovorit’104, rasskazyvat’/rasskazat’, and dokladyvat’/doložit’. Four tests showed statistically significant results: two for

104 Govorit’ ‘speak, talk-IPFV’ can be paired with two perfectives, skazat’ and pogovorit’. Skazat’ is derived from an entirely different root than govorit’, while pogovorit’ is formed with the perfectivizing prefix po-. Since govorit’ can form an aspectual pair with either skazat’ or pogovorit’, I included both in this study.

129

Noun animacy (a positive correlation with skazat’ + nasčet in Period B, and with govorit’ + nasčet in Period D) and two for Modification (a positive correlation with govorit’ + pro in Period D). However, since each instance occurs in a different dataset, no trends are evident among these results.

Table 6.17 outlines the distribution of significant results in the Verbs of

Speech category, by noun factor group and preposition, in the first series of tests:

pro nasčet po povodu Total Animate nouns 7 0 0 7 Noun Animacy Inanimate nouns 0 0 0 0 Count/mass Count nouns 1 0 0 1 status Mass nouns 0 0 0 0 Modification Unmodifed nouns 3 3 0 6 status Modified nouns 0 0 0 0 Common/proper Common nouns 0 0 0 0 status Proper nouns 4 0 0 4 Non-deverbal 3 0 0 3 Deverbal status Deverbal 0 0 0 0 Total 18 3 0 21

Table 6.17 Distribution of Statistically Significant Results for Verbs of Speech, by FG and Preposition (First Series of Tests)

It can be seen that the majority of significant results for pro were with the factor group Noun animacy. In all such tests, pro had a positive correlation with

130

animate nouns. The second series of tests, however, had too few significant results to reveal evidence of any patterns.

6.5 Verbs of Writing

6.5.1 Verbs of Writing with Individual Prepositions

Sixty-five individual tests were carried out with the verbs pisat’/napisat’

‘write’. Pisat’ + pro showed a positive correlation with animate nouns in Periods

B and C, and with proper nouns in Period A. Pisat’ + po povodu showed a positive correlation with animate nouns in Period A, while napisat’ + po povodu showed a positive correlation with inanimate nouns in Periods A, B, and D, as well as a positive correlation with common nouns in Periods B and D.

Po povodu occurs relatively frequently in the datasets for the verbs of writing. In spite of a chronological gaps in these datasets (there were not enough tokens in Period C for pisat’ and napisat’), po povodu is better represented in this verb category than in any of the others; in fact, this is the sole category where po povodu has any significant results.

131

6.3.2 Aspectually Paired Verbs of Writing

This series of 50 individual tests with the aspectual pair pisat’/napisat’ shows only two instances of statistical significance: with o+LOC, napisat’ (PFV) showed a positive correlation with proper nouns in Period B, while pisat’ (IPFV) po povodu showed a positive correlation with proper nouns in Period D.

Table 6.18 outlines the distribution of significant results in the Verbs of

Writing category, by noun factor group and preposition, in the first series of tests:

pro po povodu Total Animate nouns 2 0 2 Noun Animacy Inanimate nouns 0 4 4 Count/mass Count nouns 0 0 0 status Mass nouns 0 0 0 Modification Unmodifed nouns 0 0 0 status Modified nouns 0 0 0 Common/proper Common nouns 1 0 1 status Proper nouns 0 2 2 Non-deverbal 0 0 0 Deverbal status Deverbal 0 0 0 Total 3 6 9

Table 6.18 Distribution of Statistically Significant Results for Verbs of Writing, by FG and Preposition (First Series of Tests)

132

As Table 6.18 shows, po povodu had a fair number of significant results with

Verbs of writing, which is surprising in light of the fact that this particular verb category is one of the smallest of the six in this study. There were no statistically significant results for datasets with nasčet for the category Verbs of writing. The second series of tests had too few significant results to reveal evidence of any patterns.

6.4 Verbs of Asking

6.4.1 Verbs of Asking with Individual Prepositions

In this category, the four verbs sprašivat’/sprosit’ ‘ask (about)’, and rassprašivat’/rassprosit’ ‘question, make inquiries (about)' were used in a total of

80 tests. Three results with the FG Noun animacy showed statistically significant positive correlations with animate nouns: sprašivat’ + pro in Period A, sprosit’ + pro in Period A, and rassprašivat’ + pro in Period C. There was also a positive correlation between sprašivat’ pro and proper nouns in Period A. In addition, low

P-values (close to being statistically significant) were shown in Period B

(P=0.0596) and C (P=0.0716) for sprosit’.

133

6.4.2 Aspectually Paired Verbs of Asking

Six tests out of a total of 110 showed statistical significance in this series.

For the aspectual pair sprašivat’/sprosit’ ‘ask (about)’, there was a positive correlation between sprašivat’ (IPFV) + pro and animate nouns in Period A. For the aspectual pair spravljat’sja/spravit’sja ‘ask (about)’ with o+LOC, there were three instances of statistically significant positive correlations: with spravljat’sja

(IPFV) and count nouns in Period A, modified nouns in Period A, and deverbal nouns in Period B. Additionally, there was one instance of statistically significant positive correlation between spravit’sja (PFV) and modified nouns in Period C.

Table 6.19 outlines the distribution of significant results in the Verbs of

Asking category, by noun factor group and preposition, in the first series of tests:

134

pro nasčet Total Animate nouns 3 0 3 Noun Animacy Inanimate nouns 0 0 0 Count/mass Count nouns 0 0 0 status Mass nouns 0 0 0 Modification Unmodifed nouns 0 0 0 status Modified nouns 0 0 0 Common/proper Common nouns 0 0 0 status Proper nouns 1 0 1 Non-deverbal 0 0 0 Deverbal status Deverbal 0 0 0 Total 4 0 4

Table 6.19 Distribution of Statistically Significant Results for Verbs of Asking, by FG and Preposition (First Series of Tests)

Table 6.20 outlines the distribution of significant results in the same category, by noun factor group and preposition, in the second series of tests:

135

o+LOC pro nasčet po povodu Noun factor PFV IPFV PFV IPFV PFV IPFV PFV IPFV Totals Animate nouns 0 1 0 1 0 0 0 0 2 Inanimate nouns 0 0 0 0 0 0 0 0 0 Count nouns 0 0 0 0 0 0 0 0 0 Mass nouns 0 0 0 0 0 0 0 0 0 Unmodified nouns 0 0 0 0 0 0 0 0 0 Modified nouns 2 1 0 0 0 0 0 0 3 Common nouns 0 0 0 0 0 0 0 0 0 Proper nouns 0 0 0 0 0 0 0 0 0 Non-deverbal nouns 0 0 0 0 0 0 0 0 0 Deverbal nouns 1 0 0 0 0 0 0 0 1 Totals 3 2 0 1 0 0 0 0 6

Table 6.20 Distribution of Statistically Significant Results for Verbs of Asking, by FG and Preposition (Second Series of Tests)

It can be seen in Table 6.19 that the majority of significant results in the first series of tests in this verb category are with pro and Noun animacy. There were no tests for po povodu because there were not enough datasets for quantitative testing of this preposition in this verb category.

136

6.5 Verbs of Cognition

6.5.1 Verbs of Cognition with Individual Prepositions

In this verb category, 235 tests were performed with the 13 verbs, resulting in 32 instances of statistical significance. As with verbs of speech, writing, and asking, the highest number of significant results in this category was with Noun animacy, with 14.

There were nine significant results with the verbs dumat’/podumat’ ‘think’ in tests between o+LOC and pro; seven of these were with podumat’ (PFV).

Podumat’ shows a positive correlation between pro and animate nouns in

Periods A through C; Period D is not quite significant (P=0.0840). Table 6.21 shows the P-values for the factor groups with statistical significance for podumat’:

podumat’ Common/proper Deverbal status of Noun animacy o~pro noun noun Period A 0.0002 1.0000 0.5558 Period B 0.0005 0.0273 0.1418 Period C <0.0001 <0.0001 0.0282 Period D 0.0840 0.3123 0.0257

Table 6.21 Statistically Significant Factor Groups in Tests with podumat’ o~pro, in Periods A-D

137

The verb znat’ ‘know-IPFV’ shows trends with Noun animacy and Deverbal status, in tests between o+LOC and pro (Table 6.22):

znat’ Common/proper Deverbal Noun animacy Modification of o~pro noun status of noun status of noun Period A 0.0523 0.7892 0.0237 0.0122 Period B 0.0003 1.0000 0.1028 0.0105 Period C 0.0056 0.0326 0.1945 0.0046 Period D 0.0008 0.7948 1.0000 0.4716

Table 6.22 Statistically Significant Factor Groups in Tests with znat’ o~pro, in Periods A-D

A positive correlation was shown between pro and animate nouns in Periods B through D, (as well as in Period A, which is close to being statistically significant).

In Periods A through C, pro had a positive correlation with non-deverbal nouns.

The preposition nasčet has datasets for Periods B, C, and D with znat’.

The results for the given collocation are shown in Table 6.23:

138

znat’ Noun Modification of Common/proper o~nasčet animacy noun status of noun Period B 0.0044 0.1333 0.0184 Period C 0.0521 0.0733 1.0000 Period D 0.0003 0.0022 0.1945

Table 6.23 Statistically Significant Factor Groups in Tests with znat’ o~nasčet, in Periods B, C, and D

Concerning the factor Noun animacy, nasčet showed a positive correlation with animate nouns in Periods B and D, and the P-value 0.0521 in Period C is close to being significant. For Modification status, znat’ + nasčet showed a statistically significant positive correlation with unmodified nouns in Period D, while the result for Period C (P=0.0733) is nearly significant.

In tests with vspominat’/vspomnit’ ‘remember’, there were only scattered results with statistical significance. However, there is one note of interest with vspominat’: a trend can be seen with Noun animacy in the form of a consistent increase in positive correlation between pro and animate nouns from Period A to

Period D, reflected in a steady decrease in the P-values (Table 6.24):

139

vspominat’ P-values for Noun o~pro Animacy tests Period A 1.0000 Period B 0.2487 Period C 0.0973 Period D 0.0021

Table 6.24 Increase in Positive Correlation between pro and Animate Nouns, Periods A-D

There were four instances of statistical significance with the verb uznat’

‘realize-PFV’ and one with uznavat’ ‘realize-IPFV’. With uznat’, pro showed a positive correlation with animate nouns in Periods B and C, and with uznavat’ in

Period A.

In the group zabyvat’/zabyt’/pozabyt’ ‘forget’ only two results showed statistical significance: in Period D, zabyvat’ showed a positive correlation between pro and animate nouns; in Period C, zabyt’ showed a positive correlation between pro and proper nouns.

140

6.5.2 Aspectually Paired Verbs of Cognition

This series of tests consisted of the aspectual pairs dumat’/podumat’

‘think’, napominat’/napomnit’ ‘remind’, zabyvat’/zabyt’105 ‘forget’, uznavat’/uznat’

‘realize’, and vspominat’/vspomnit’ ‘remember, recall’ in a total of 165 individual tests.106 Seven additional tests were carried out with the verbs napominat’ and napomnit’ in connection with the two possible meanings (speech and cognition) expressed by them.

Dumat’/podumat’ ‘think’ with pro showed significance in Periods B and C; a positive correlation between the perfective form podumat’ and animate nouns.

In the same two time periods, there is a positive correlation between podumat’ and proper nouns as well.

The aspectual pair uznavat’/uznat’ showed significance in one instance only, which revealed a positive correlation between modified nouns and the imperfective form uznavat’ with o+LOC.

The aspectual pair uznavat’/uznat’ showed significance in Period B, where the imperfective form uznavat’ with о+LOC showed a positive correlation with modified nouns.

105 Zabyt’ and pozabyt’ ‘forget’ are both perfectives. I tested the pair zabyvat’/zabyt’, but not zabyvat’/pozabyt’, since, although the first is imperfective and the second is perfective, they are not considered to be an aspectual pair. 106 Since the verb pomnit’ ‘remember-IPFV’ has no perfective counterpart, it was used only in the first series of tests.

141

6.5.3 The REMIND1 and REMIND2 Senses of Napominat’/napomnit’

This series of tests was carried out in order to analyze the distribution of the two senses expressed by the verbs napominat’ and napomnit’, which I discussed in Chapter 5. I have designated the speech sense of napominat’/napomnit’ as REMIND1, and the cognition sense as REMIND2.

The first type of test determined a P-value (using Fisher’s Exact Test) for the distribution of REMIND1 and REMIND2 between o+LOC and pro with each verb, in Period D (the only time period for which there are enough tokens of these verbs with pro). The values for napominat’ (IPFV) are shown in Table 6.25:

REMIND REMIND 1 2 Totals (speech) (cognition) napominat’ o+LOC 14 16 30 napominat’ pro 23 2 25107

Table 6.25 Distribution of REMIND1 (Speech) and REMIND2 (Cognition) Senses of napominat’ ‘remind-IPFV’ in Period D

Fisher’s Exact Test shows a statistically significant P-value of 0.0004 for the distribution shown in Table 6.25; napominat’ + pro has a strong positive

107 It can be seen that the dataset for napominat’ + pro has only 25 tokens, compared to 30 for napominat’ + o+LOC. However, the total number of tokens in datasets tested against each other, in this case and in all others does not need to match exactly.

142

correlation (P=0.0004) with REMIND1 compared to o+LOC. The distribution in

Period D, of the PFV form napomnit’, is shown in Table 6.26:

REMIND REMIND 1 2 Totals (speech) (cognition) napomnit’ o+LOC 25 5 30 napomnit’ pro 30 0 30

Table 6.26 Distribution of REMIND1 (Speech) and REMIND2 (Cognition) Senses of napomnit’ ‘remind-PFV’ in Period D

Napomnit’ + pro shows a positive, though not quite statistically significant, correlation (P=0.0522) with the speech sense of this verb.

The second type of test involved the two verbs as aspectual pairs, with each of the two prepositions o+LOC and pro. For o+LOC, datasets from Periods

A through D were available. However, as I mentioned above, pro in collocation with napominat’ and napomnit’ has datasets only for Period D. Nevertheless, the results indicate that the distribution of REMIND1 and REMIND2 (with o+LOC) differs significantly according to aspect (Table 6.27):

143

napominat’ (IPFV) o+LOC napomnit’ (PFV) o+LOC Time REMIND REMIND REMIND REMIND P-value period 1 2 1 2 A 13 17 24 6 0.0073 B 14 16 27 3 0.0006 C 10 20 27 3 <0.0001 D 14 16 25 5 0.0061

Table 6.27 Fisher’s Exact Tests for napominat’ with o+LOC and napomnit’ with o+LOC (REMIND1 and REMIND2) in Periods A-D

However, this is not the case with pro in Period D (Table 6.28):

napominat’ (IPFV) pro napomnit’ (PFV) pro Time REMIND1 REMIND2 REMIND1 REMIND2 P-value period D 23 2 30 0 0.2020

Table 6.28 Fisher’s Exact Test for napominat’/napomnit’ + pro (REMIND1 and REMIND2) in Period D

As shown in Table 6.28, with pro there is no significant difference in the distribution of the two senses between imperfective and perfective (P=0.2020).

In contrast, as Table 6.19 shows, there is a significant difference in the case of o+LOC , which has remained fairly consistent over time.

144

Table 6.29 outlines the distribution of significant results in the Verbs of

Cognition category, by noun factor group and preposition, in the first series of tests:

pro nasčet Total Animate nouns 12 2 14 Noun Animacy Inanimate nouns 0 0 0 Count/mass Count nouns 0 0 0 status Mass nouns 0 0 0 Modification Unmodifed nouns 3 1 4 status Modified nouns 0 0 0 Common/proper Common nouns 0 0 0 status Proper nouns 6 1 7 Non-deverbal 7 0 7 Deverbal status Deverbal 0 0 0 Total 28 4 31

Table 6.29 Distribution of Statistically Significant Results for Verbs of Cognition, by FG and Preposition (First Series of Tests)

As is evident from Table 6.29, the tests showing statistically significant correlations between pro and the FG Noun animacy (all of which show a positive correlation with animate nouns) account for the majority of significant results for this preposition, within this verb category. There were no tests for po povodu because there were not enough datasets for quantitative testing of this

145

preposition in this verb category. The second series of tests with Verbs of cognition had too few significant results to reveal evidence of any patterns.

6.6 Verbs of Hearing

This semantic category consists of the aspectual pair slyšat’/uslyšat’

‘hear’, and slyxat’ ‘hear-IPFV’. As there are no datasets in this verb category with nasčet or po povodu, all tests were carried out for o+LOC and pro only.

6.6.1 Verbs of Hearing, with Individual Prepositions

Fifty-five individual tests were performed on the three verbs in this category. Statistical significance was found in six instances. These are clustered in Period B for imperfective slyšat’, and for imperfective slyxat’:

Unmodified Non-deverbal Animate nouns Proper nouns nouns nouns slyšat’ pro P=0.0009 P=0.0089 P=0.0303 P=0.0122 slyxat’ pro P=0.0251 P=0.0641 P=0.0419 P=0.2373

Table 6.30 Factor Groups with Statistical Significance (slyšat’ o~pro and slyxat’ o~pro) in Period B

146

Slyšat’ (IPFV) + pro shows a positive correlation in Period B with animate nouns, modified nouns, and proper nouns, and non-deverbal nouns. In this same period, slyxat’ shows a significant positive correlation between pro and animate nouns as well as proper nouns.

6.6.2 Aspectually Paired Verbs of Hearing

This series of 35 individual tests was performed on the aspectual pair slyšat’/uslyšat’. The tests revealed two instances of significance: with o+LOC, there was a positive correlation between the imperfective form slyšat’ and modified nouns in Period B, and pro had a positive correlation between slyšat’ and modified nouns as well, in Period C.

Table 6.31 outlines the distribution of significant results in the Verbs of

Hearing category, by noun factor group and preposition, in the first series of tests:

147

pro nasčet Total Animate nouns 2 0 2 Noun Animacy Inanimate nouns 0 0 0 Count/mass Count nouns 0 0 0 status Mass nouns 0 0 0 Modification Unmodifed nouns 1 0 1 status Modified nouns 0 0 0 Common/proper Common nouns 0 0 0 status Proper nouns 2 0 2 Non-deverbal 1 0 1 Deverbal status Deverbal 0 0 0 Total 6 0 6

Table 6.31 Distribution of Statistically Significant Results for Verbs of Hearing, by FG and Preposition (First Series of Tests)

There were no tests for po povodu because there were not enough datasets for quantitative testing of this preposition in this verb category. The second series of tests with Verbs of hearing had too few significant results to reveal evidence of any patterns.

6.7 Verbs of Reading

This category consists of the verbs čitat’ (IPFV) and pročitat’ (PFV) ‘read’.

As there are no tokens in this verb category with the prepositions nasčet or po povodu, all tests were carried out with the prepositions o+LOC and pro only.

148

6.7.1 Verbs of Reading, with Individual Prepositions

Thirty individual tests were performed on the two verbs in this category, showing three instances of statistical significance. In Period B, pro showed a positive correlation with proper nouns. In Periods C and D the perfective pročitat’

+ pro has a positive correlation with animate nouns in Periods C and D.

6.7.2 Aspectually Paired Verbs of Reading

Thirty individual tests were carried out on čitat’/pročitat’ as an aspectual pair, however, none of these tests resulted in statistical significance.

Table 6.32 outlines the distribution of significant results in the Verbs of

Reading category, by noun factor group and preposition, in the first series of tests:

149

pro nasčet Total Animate nouns 2 0 2 Noun Animacy Inanimate nouns 0 0 0 Count/mass Count nouns 0 0 0 status Mass nouns 0 0 0 Modification Unmodifed nouns 0 0 0 status Modified nouns 0 0 0 Common/proper Common nouns 0 0 0 status Proper nouns 1 0 1 Non-deverbal 0 0 0 Deverbal status Deverbal 0 0 0 Total 3 0 3

Table 6.32 Distribution of Statistically Significant Results for Verbs of Reading, by FG and Preposition (First Series of Tests)

The first series of tests was not carried out for datasets with po povodu because there were not enough for quantitative testing of this preposition in this verb category. Additionally, the second series of tests with Verbs of reading had too few significant results to reveal evidence of any patterns.

6.8 Summary

6.8.1 First Series of Tests (Verbs with Individual Prepositions)

Overall, the results of the first series of quantitative tests reveal several areas of note. The correlation between pro and animate nouns shows several trends over time, with some of the verbs of speech, cognition, and asking:

150

Speech Cognition Asking Reading Time rasskazyvat’ rasskazat’ podumat’ znat’ uznat’ sprosit’ pročitat’ Period (IPFV) (PFV) (PFV) (IPFV) (PFV) (PFV) (PFV) A 0.0004 0.0251 0.0002 0.0523 0.6707 0.0284 n/a B 0.0470 0.0716 0.0005 0.0003 0.0419 0.0596 n/a C 0.1806 0.0153 <0.0001 0.0056 0.0211 0.0716 0.0419 D 0.2092 0.0211 0.0840 0.0008 0.1455 0.4118 0.0015

Table 6.33 Positive Correlation between pro and Animate Nouns (Significant P-values in Red)

The P-values for rasskazyvat’ ‘tell, relate-IPFV’ increase with each time period, indicating that the positive correlation between pro and animate nouns has decreased over time; inanimate and animate nouns have shifted toward a more equal distribution. However, with rasskazat’ ‘tell, relate-PFV’, the positive correlation remains fairly consistent throughout. Podumat’ ‘think-PFV’ shows a consistently positive correlation which drops off somewhat by Period D. With znat’ ‘know-IPFV’, the correlation increases after Period A, holding steady up to the present time. Uznat’ starts out with a very high P-value (i.e., fairly equal distribution) followed by significance in Periods B and C. With sprosit’, there is a positive correlation between pro and animate nouns at the outset, which steadily decreases over time. Finally, Periods C and D of pročitat’ with pro show a

151

positive correlation with animates (Periods A and B did not have enough tokens for analysis).

A consistent negative correlation occurs between po povodu and animate nouns, with napisat’ ‘write-PFV’ (Table 6.34):

Time napisat’ Period (PFV) A 0.0204 B 0.0183 C n/a D 0.0335

Table 6.34 Positive Correlation between napisat’ + po povodu and Animate Nouns

This series of tests revealed more trends with the FG Noun animacy than with any of the other factor groups.

152

6.8.2 Tests on Aspectual Verb Pairs (Second Series of Tests)

The second series of tests showed a trend only with dumat’/podumat’ with pro, where a positive correlation between this preposition and the perfective form podumat’ occurs in Periods B and C.

6.9 Conclusions

6.9.1 Pro

The distribution of statistically significant results of tests between o+LOC and pro in each verb category is shown in Table 6.35:

Verb Noun Count/ Common/ Significant Modif. Deverbal category animacy mass proper results Speech 7 1 3 4 3 18 Cognition 12 0 3 6 7 28 Hearing 2 0 1 2 1 6 Asking 3 0 0 1 0 4 Reading 2 0 0 1 0 3 Writing 2 0 0 1 0 3 Totals: 28 1 7 15 11 62

Table 6.35 Distribution of Significant Results by Verb Category and FG, in Tests between o+LOC and pro

153

As shown in Table 6.35, within the population of statistically significant results, the majority of these results are for the FG Noun animacy, followed by

Common/proper status, Deverbal status, and Modification status of noun. Within these four factor groups, pro has a consistent positive correlation with animate, proper, unmodified, and non-deverbal nouns, in comparison with o+LOC. In contrast, the distribution of count and mass nouns in the datasets is, except for one instance, not significantly correlated with either o+LOC or pro.

6.9.1.1 Individuation

Several of the factors which figure prominently in the tests between o+LOC and pro coincide with certain semantic features regarded by Timberlake

(1975, 1977), Klenin (1980), Andersen (2001), and others, to be among those associated with noun objects whose referents are likely to be individuated.

Individuation is “the degree to which a given participant is characterized as a distinct entity or individual in the narrated event” (Timberlake, 1975: 124).

Generally speaking, the more individuated an entity is, the more prominent a role it has in the narrative. Features relevant to individuation form hierarchies of oppositional pairs:

154

(a) Proper Common (b) Human Non-human (c) Animate Inanimate

Table 6.36 Hierarchies of Noun Features Associated with Individuation (Andersen 2001: 31)

The three pairs of features in Table 6.36 are associated with the semantic type of the noun. Proper nouns are more likely to be individuated than common nouns, and so on. To this list can be added ‘Modification’ (Klenin 1980: 71), a feature associated with noun specificity. The referent of a modified noun is more specific

(individuated) than that of an unmodified noun.

In the first series of tests in the present study, in all significant instances, pro has a positive correlation with unmodified nouns. However, if a connection between pro and individuated noun objects were to obtain, then this correlation should be negative, not positive (i.e., the distribution with pro should be weighted toward modified noun objects).

155

6.9.2 Nasčet and po povodu

Nasčet is present only in datasets for verbs of speech, cognition, and asking, while po povodu is present only with verbs of speech and of writing.

Because nasčet was represented in so few datasets, the quantitative evidence is sparse, but what is available does show an association between nasčet and animate, unmodified nouns. Tests between o+LOC and nasčet with the verb znat’ (Periods B-D), show a positive correlation in B and D between nasčet and animate nouns, as well as a negative correlation with modified nouns.

Additionally, there is near-significance for these correlations in Period C

(P=0.0521 and P=0.0733, respectively).

Po povodu had statistically significant results only with verbs of writing. In the 30 tests between o+LOC and po povodu in this verb category, six results were statistically significant: four with Noun animacy (a negative correlation between po povodu and animate nouns), and two with Common/proper noun status (a negative correlation between po povodu and proper nouns) in comparison with o+LOC. This suggests that po povodu is associated with inanimate and common nouns.

156

Chapter 7: Co-occurrence of ‘Aboutness’ Prepositions in the Same Syntagma

7.1 Co-occurrence of o+LOC and pro PP’s in the Same Syntagma

In the RNC written corpus, there are several dozen sentences in which o+LOC and pro co-occur within the same syntagma (clause or complex sentence) in the same syntactic function. In some of these cases, the prepositional phrases have the same verb as head, i.e., appear in the same clause, as in (1a). In others, they occur in parallel slots in adjacent clauses, as in

(1b).

(1a) On znal i pro Sergeja Trubeckogo ... he-NOM know-M.PST.SG and about Sergej-ANIM.ACC Trubeckij-ANIM.ACC

108 i o Evgenii Trubeckom. and about Evgenij-LOC Trubeckij-LOC ‘He knew both about [pro] Sergej Trubeckoj and about [o] Evgenij Trubeckoj.’

(1b) Ty, master, pro den’gi ne dumaj ... you-NOM.SG master-NOM.SG about money-ACC.PL NEG think-IMP.SG

ty o rabote dumaj.109 you-NOM.SG about work-LOC.SG think-IMP.SG

108 Granin, D. (1987) Zubr. (RNC) 109 Mjasnikov, V. (2000) Vodka. (RNC)

157

‘Master, don’t think about [pro] the money—[just] think about [o] work.’

Example (1a) features two OPs with the same verb head, explicitly coordinated by the correlative conjunctions i… i… ‘both… and…’; their referents are separate entities that are related to each other topically (two members of the same family).

Example (1b) is a two-clause focus construction (a complex coordinate structure). Here the focal entities are the predicates, which convey two alternative hypothetical actions by the same subject; the strongest emphasis

(sentential stress) is on the juxtaposed prepositional phrases, which appear in the preverbal position.

The object referents in the co-occurrences can each refer to a separate entity (2a), or, less frequently, to the same entity (2b):

(2a) [O]ni pro sebja sovsem ne govorjat, a tol’ko They-NOM about self-ACC altogether not talk-PRS.3.PL but only

o sud’be rabočego naroda.110 about fate-LOC.SG working-M.GEN.SG people-GEN.SG ‘They did not talk at all about [pro] themselves, only about [o] the fate of the working people.’

(2b) O ljubvi, pro ljubov’ – no ne ljubov’ju že about love-LOC.SG about love-ACC.SG but not love-INS.SG just

pisat’.111

110 Gor’kij, M. (1928-35) Žizn’ Klima Samgina. (RNC) 111 Bitov, A. (2000) “Pjat’desjat let bez Platonova” In Zvezda 1/2001. (RNC)

158

WRITE-INF ‘[Write] about [o] love, about [pro] love—but don’t write by means of [sc. under the spell of] love.’

In (2b), the OPs are juxtaposed asyndetically (without any conjunction). As they refer to a single, albeit amorphous, entity, ‘love’, their co-occurrence seems, at first glance, to be tautological (for further discussion, see §7.3).

For brevity, I will use the term co-occurrence to refer to constructions like

(1a–b), in which different ‘aboutness’ prepositions are juxtaposed syntactically.

While co-occurrences constitute only a small portion of the corpus I utilized, they are extremely significant for the issues considered in this dissertation, as they can be viewed as “minimal pairs” on the syntactic level. As such, they bring the subtle semantic and/or pragmatic differences between the prepositions to the fore. Whatever the degree of semantic divergence between the PP’s may be, it cannot be extrapolated from any constituent within the clause other than the prepositions themselves.

The verbs in these co-occurrences are the same that are found with pro

(but not necessarily with o+LOC) in sentences which have a single PP—those denoting cognition, as in (1a-b), speaking, writing, hearing, reading, or asking.

This distribution is worthy of note because pro, being limited to the Area/Topic metaphorical extension, cannot be used with verbs denoting emotional states or

159

propositions (asking, appealing, petitioning, etc.). This constraint also obtains for pro in the co-occurrence sentences.

It is also important to note that the use of co-occurrences is a matter of authorial choice rather than rule; there is no syntactic or pragmatic constraint against repeating the same preposition in the same syntactic function within a single syntagma. In fact, constructions with conjoined, coordinate, or parallel o...o… or pro...pro… phrases are common in the corpus (see §7.4). I will call this kind of repetition preposition matching. I have not found any evidence that

Russians avoid o...o… or pro...pro… for the sake of “elegant variation”; indeed, examples occur in the writings of major literary figures. Thus it is necessary to seek other explanations for the phenomenon of co-occurrence.

In §7.2, I will survey the types of co-occurrences that are attested in the corpus. In §7.3, I will zero in on the semantic differences that are implied by co- occurrences. In §7.4, I will discuss instances of preposition matching in order to demonstrate that co-occurrences are not the result of any syntactic constraint.

7.2 Types of Relations in Co-occurrences

7.2.1 Co-occurrences in Coordinated Same-Clause PP’s

In the majority of co-occurrences, two or more PP’s with different referents appear as coordinated phrases within a single clause, i.e., under a single verb

160

head. The PP’s can be explicitly conjoined, as in (3a) (cf. also (1a) above, or connected asyndetically (without any conjunction), as in 3b):

(3a) [N]ačinaet govorit’ pro utrennjuju dostavku i begin-PRES.3.SG talk-INF about morning-F.ACC.SG delivery-ACC.SG and

o neispol’zovannyx rezervax.112 about unused-LOC.PL reserves-LOC.PL ‘He begins talking about [pro] the morning delivery and about [o] unused reserves.’

(3b) On rasskazal o krovi russkix, ... pro He-NOM tell-M.PST.SG about blood-LOC Russian-GEN.PL about

brata, vdavlennogo gusenicami brother-ANIM.ACC.SG press.in-M.PST.PASS.PTCP.SG tank.track-INST.PL

v pesok pustyni.113 in sand-ACC desert-GEN.SG ‘He told about the blood of Russians, about [his] brother who had been crushed in the desert sand by tank tracks.’

In (3a), Saša, a mail carrier, returns home to his wife in the evening, and tells her about the day’s morning delivery114 and about ‘untapped reserves’, probably in reference to the postal workforce. Since delivering mail each morning is part of a recurring routine, it is not likely that Saša is explaining to his wife what the morning delivery is (or the reason for its existence), but is simply recounting

112 Vasilenko, S. (1997-2000) Zvonkoe imja. (RNC) 113 German, Ju. (1961) Dorogoj moj čelovek. (RNC) 114 In Russia, mail is delivered twice a day—once in the morning and once in the evening.

161

individual events associated with it only on that particular day. Thus, the pro referent is being identified in connection to a particular, fixed reference, which is that specific day. In contrast, the ‘untapped reserves’ (the OP of o) are vaguer and more abstract in nature. They also are potential as opposed to actual; thus, their scope is unknown or undefined in comparison with Saša’s mail delivery on that particular morning (with pro)—an activity whose scope is discrete and defined.

In (3b), the speaker is recounting the horrors of the war, in which his own brother had been killed. ‘The blood of Russians’ (o+LOC) is a highly general referent which denotes the huge death toll of Russians in that conflict, of which his brother (pro) is a highly particular referent.

In addition to non-contrastive relations like (3a-b), the o+LOC and pro

PP’s may be conjoined in adversative (4a-b) relations.

(4a) Ja ne personal’no pro vas, a – voobšče o I-NOM NEG personally about you-ACC.PL but in.general about

štatskix, ob intelligentax.115 civilian-LOC.PL about intellectual-LOC.PL ‘I’m not (talking) about [pro] you personally, but in general about[o] civilians, about [o] intellectuals.’

115 Gor’kij, M. (1928-35) Žizn’ Klima Samgina. (RNC)

162

(4b) [G]ovorjat ne o byloj junosti, a pro talk-PRS.3.PL NEG about former-F.LOC.SG youth-LOC but about

Marusju.116 Marusja-ANIM.ACC ‘They’re talking not about [o] (their) former youth, but about [pro] Marusja.’

In (4a), an army officer is complaining to Samgin how the intellectuals of Russia, by antagonizing the tsar, the nobility, and God with their opposition to the war, have made the workers and the nobility hate the officers who are leading them.

The contrast here is between the highly particular referent of the pro OP

(Samgin, who, though an intellectual, is exempted from the speaker’s diatribe (ne personal’no pro vas ‘not personally about you’) and the highly general referent of the o OP’s (stereotypical civilians and intellectuals).

As with (4a), the pro object referent in (4b), the young girl Marusja is a named person, an entity that is presupposed in the discourse. This highly specific, indeed unique entity is being contrasted with an abstract category—

‘former youth’ (i.e., the general state of being young). In this context, Marusja is narrower in scope than the concept of youth.

In several co-occurrences, the referents of the co-occurring OP’s are in a whole-part or type-token relationship; that is, one of the referents is an

116 Žabotinskij, V. (1936) Pjatero. (RNC)

163

elaboration, specification, characterization, or example of the other. I will call this type explicative.

(5a) [B]ajal ja o vas starikam našim speak-M.PST.SG I about you-LOC.PL elder-DAT.PL our-DAT.PL

kazakam pro userdie vaše.117 Kazak-DAT.PL about diligence-ACC.SG your-ACC.PL ‘I spoke about [o] you to our Kazak elders, about [pro] your zeal.’

(5b) [E]mu xotelos’ i dal’še rasskazyvat’. Rasskazyvat’ o he-DAT want-N.PST.SG and further tell-INF tell-INF about

sebe i o svoej nelepoj smešnoj self-LOC and about own-F.LOG.SG ridiculous-F.LOC.SG comical-F.LOC.SG

žizni, pro to, gde on rodilsja ... life-LOC.SG about that-N.ACC.SG where he-NOM be.born-M.PST.SG.REFL

pro otca s mater’ju, pro staršuju about father-ANIM.ACC.SG with mother-F.INS.SG about older-F.ACC.SG

sestru, pro ee muža.118 sister-ACC.SG about her-ACC husband-ANIM.ACC.SG ‘He wanted to tell him more. Tell him about himself and his ridiculous, comical life, about where he was born ... about (his) father and mother, about his older sister, about her husband.’

117 Šiškov, V. (1934-39) Emel’jan Pugačev. (RNC) 118 Dombrovskij, Ju. (1978) Fakul’tet nenužnykh veščej. (RNC)

164

In (5a), userdie ‘zeal, diligence’ describes an identifying characteristic of the addressee (the eighteenth century rebel leader Emeljan Pugačev). The narrator

P’janov has told the Kazak elders about Pugačev, and more specifically, about his zeal as a military leader. The speaker presents this particular trait (with pro) as a characteristic of Pugachev as a whole person.

In (5b), Kornilov is engaged in an animated conversation with the priest,

Father Andrej, about the Gospel of John. Kornilov finds himself wanting to open up about himself (o sebe). The pro referents—Kornilov’s birthplace, his parents, his sister, and her husband—are personal details fitting into the larger scope of his life as a whole.

7.3 Semantic Opposition between o+LOC and pro in the Co-occurrences

To get clues to the nature of the opposition between o and pro in co- occurrences, we can look at several cases in the corpus where the OP’s of the two prepositions are closely related or even identical. One such example can be seen in (6a), where a former criminal intends talk about himself in two different ways—more generally or holistically (i.e., about his life as a whole) and more specifically (i.e., about specific events in his life).

(6a) Šatilov ... xotel o sebe rasskazyvat’, ne tol’ko

165

Šatilov-NOM want-M.PST.SG about self-LOC tell-INF not only

pro grabeži i edinstvennoe ubijstvo: on about robbery-ACC.PL and single-N.ACC.SG murder-ACC.SG he-NOM

xotel povedat’ vsju svoju žizn’.119 want-M.PST.SG relate-INF whole-F.ACC.SG own-F.ACC.SG life-ACC.SG ‘Šatilov wanted to tell about [o] himself, not only about [pro] [his] robberies and [his] one murder; he wanted to recount his entire life.’

(6b) [P]rosto i xorošo govorili ... o “Nežnom I[osife]”, simply and well speak-PST.PL about gentle-M.LOC.SG Iosif-LOC

o nem, i pro ego “xuliganskij” period.120 about he-LOC and about his ruffian-M.ACC.SG period-ACC.SG ‘They spoke simply and nicely ... about “Gentle Iosif”—about him, and about his “ruffian” period.’

In (6a), the conjoined OP’s in the pro phrase both refer to events in the life of

Šatilov; both are quasi-deverbal nouns (cf. grabit’ ‘rob’ and ubit’ ‘kill’), i.e., denote actions or activities, even though they are not formed like the productive type of deverbal noun. The pro PP is set off by the conjunction ne tol’ko ‘not only’, which leads up to the clause ‘he wanted to recount his entire life’ (an implicit correlative

‘but also’). The topics that the objects of pro convey—robberies and a murder that Šatilov has committed—are more limited in scope than the one conveyed by the preceding object of o+LOC—the person who was involved in these events.

119 Makanin, V. (1996-96) Andegraund, ili geroj našego vremeni. (RNC) 120 Švarsalon, V. (1908-10) Dnevnikovye zapisi. (RNC)

166

They represent a few among countless other actions or activities performed by that agent. In other words, o sebe denotes Šatilov as a complex entity of broader scope than that of a selected few criminal activities (‘marked’ events in his life).

Presumably, in the narrated event, Šatilov wishes to mitigate the crimes by placing them within the larger context of his life as a whole.

A similar opposition can be seen in (6b). ‘Gentle Iosif’ (Nežnyj Iosif) is the title of a short story by Mixail Kuzmin, recollected by Vera Švarsalon in a 1909 diary entry describing her conversation with friends about this newly-published work. ‘Gentle Iosif’ in (6b) refers to the title, while o nem denotes the character himself; Švarsalon mentions a particular episode from the story, concerning

Iosif’s “ruffian period”. O+LOC denotes the character Iosif himself, against which this particular episode in his life (with pro) is presented as being narrower in scope than the totality of the character.

This type of opposition is very similar to several other that of (6a), in which

Šatilov wants to talk about himself as a cohesive whole (o sebe), and to avoid mentioning specific events in his life which are narrower in scope (pro). It is also quite similar to the opposition found in (5b), in which Kornilov wants to tell the priest about himself (Kornilov) and his life, which comprises details that are narrower in scope, such as his place of birth, and his relationships with his family members (pro).

167

In several co-occurrences, the o+LOC and pro objects are either identical or co-referential (i.e., anaphoric). As noted in 7.1, in such “minimal pair” contexts the two prepositions bear the entire burden of a semantic contrast. For example, in (7a) the contrasting manner in which the rich and poor think about God is entirely conveyed by means of the opposition between pro and o+LOC, respectively:

(7a) [B]ogatye i žirnye pro Boga ne vspominajut, ... rich-NOM.PL and fat about God-ANIM-ACC not remember-PRS.3.PL

O Boge tol’ko bednye, bol’nye da nesčastnye About God-LOC only poor-NOM.PL sick-NOM.PL and unhappy-NOM.PL

vspominajut.121 remember-3.PRS.PL ‘The rich and fat don’t remember about [pro] God ... It’s only the poor, sick, and unhappy who remember about [o] God.’

The background to the story in (7a) is as follows: during a train trip, the main character is talking with his companion about a wealthy political figure who, despite his prominent participation in religious events in his hometown, had accepted a large bribe earlier that year, just before Easter. Overhearing this, a third passenger offers his own opinion on the matter in the following passage

(7b), which provides the larger context for (7a):

121 Volodarskij, Ė. (1997) Dnevnik samoubijcy. (RNC)

168

(7b) Interesting, is this man now praying for his sins in church, or is he out drinking and carousing with women? You’ll see, when you grab him by the throat, he’ll remember about God [pro Boga] right away. He’ll give you a line about how his demons had led him astray, and that darkness had come over him, that he wasn’t thinking about what he was doing ... Maybe somewhere there are [genuine believers], but I’ve never met one. I’ve met more that remember about God [pro Boga] only when something unusually bad happens to them. ... I tell you, the rich and fat don’t remember about God [pro Boga], because they think it’s their birthright to be in the place where they are; they’re so special, everything is just handed to them... Only the poor, sick, and unhappy remember about God [o Boge] [emphasis mine—AJK].

Evidently, the reported speaker means that the ‘rich and fat’ show awareness of

God only when difficult circumstances strike them—and even then only temporarily, until things return to normal; by contrast, the poor, sick, and unhappy, think of God continually, not merely as a fleeting topic.

The several instances of ‘about God’ with pro in (7a) and (7b) depict Him as being of limited scope for the wealthy political figure, and for rich people in general. Conversely, ‘about God’ with o+LOC shows Him as having a broad, or even unlimited, scope, i.e., a constant, and consistent reality, through good and bad times. The o~pro opposition in this example can be compared to that of

(6a), in which Šatilov wants to talk about himself (o+LOC) as a ‘big picture’, not just about isolated fragments of it. He wants his life as a whole, in comparison to

169

the criminal acts he has committed, to be ‘unmarked’. In both examples, the o+LOC referent is presented as being of broader scope than the pro referents.

(8a) A Xeminguėj govorit počti vsegda otkryto o but Hemingway-NOM talk-PRS.3.SG almost always openly about

sebe pro sebja, xotja razgovor vedet self-LOC about self-ACC though conversation-NOM.SG carry.on-PRS.3.SG

budto by sovsem o drugix.122 as.if altogether about other-LOC.PL ‘But Hemingway talks almost always openly about [o] himself, [and] about [pro] himself, even though he is carrying on a conversation as if it were about [o] other people altogether.’

(8b) O ljubvi, pro ljubov’ – no ne ljubov’ju že about love-LOC.SG about love-ACC.SG but not love-INS.SG just

pisat’.123 WRITE-INF ‘About love, about love—just don’t write by means of [i.e., under the spell of] love.’

This type of construction, which is extremely rare in the RNC, shows o+LOC and pro in a context in which they cannot be synonymous with each other. There must be some sort of semantic or pragmatic contrast between them; otherwise,

122 Ėfros, A. (1975-87) Professija: režisser. 123 Bitov, A. (2000) “Pjat’desjat let bez Platonova”. In Zvezda 1/2001. (RNC)

170

the use of both prepositions would be redundant (which is improbable, as both examples come from edited texts).

In (8a), where Ernest Hemingway is talking “about himself and about himself,” the writer is portrayed, as it were, as having a duplex ‘self’ or two

‘selves’ that are different from each other in some way.124 The task here is to determine which aspects of ‘self’ are being contrasted in this example.

Hemingway is portrayed as being fond of talking openly about who he is (o sebe), and—if we posit that pro limits the scope of its referent—he is also fond of talking about himself as ‘Papa Hemingway’, i.e., about his ‘image’. By the same token, in (8b), one can write about love as a concept (o ljubvi), or about love for a specific person or thing (pro ljubov’). This is especially clear if we posit that example (6a) is analogous to (8a) and (8b): in (6a), o sebe ‘about oneself’ denotes the full complexity of Šatilov, presented in contrast with several incidents in his personal history. In the same kind of opposition, Hemingway talks about himself (o sebe) as a whole, as well as about himself (pro sebja) in connection with various events associated with him.

In the present section, I have discussed co-occurrences which feature OP referents whose oppositional relationship consists of an entity portrayed holistically, (with pro), in contrast to an entity portrayed as complex or analyzed

124 Although the fixed phrase pro sebja ‘to oneself’, which reflects the obsolete meaning of pro, still exists in the modern Russian lexicon (and comprises the majority of attestations of this phrase in the RNC), it can also denote ‘about oneself’. A reading of ‘to oneself’ would not make sense, given that Hemingway is conversing with others.

171

(with o+LOC). With this oppositional model, the same entity can portrayed in two different contexts—as a single instantiation (or limited number of instantiations) of itself (with pro) of narrow scope, occurring within the same entity, presented as complex and of broad scope. In §7.31, I will discuss co-occurrences which feature an opposition between an animate entity and event-like entities (or entities which are part of events) linked to that animate entity. These reflect an explicative relationship between the referents, albeit one in which one entity is shown in opposition to another entity which is not within, but outside, the first entity.

7.3.1 Explicative Relationships between o+LOC and pro Object Referents

As I have mentioned in §7.21, several co-occurrences show an explicative relationship between the referents of the o+LOC and pro objects. In cases like

(9), where one or more o+LOC referents coordinate with a pro referent; the former represent details or specifications of the latter:

(9) [J]a stremilas’ uvidet’ Annu Andreevnu, čtoby I-NOM long-F.PST.SG see-INF Anna-ACC Andrejeva-ACC in.order

rasskazat’ ej pro Mandel’štamov, ob ix tell-INF she-DAT about Mandel’štam-ANIM.ACC.PL about their-LOC

vozvraščenii v Moskvu.125

125 Gerštejn, Ė. (1985-2002) Lišnjaja ljubov’. (RNC)

172

return-LOC.SG to Moscow-ACC ‘I longed to see Anna Andrejevna, in order tell her about the Mandel’štams, about their return to Moscow.’

(10) Kul’čickij ... rassprašivaet pro Ljubjanku126 , pro Kul’čickij-NOM ask-PRES.3.SG about Lubyanka-ACC about

Abakumova, opjat’ o moix s nim Abakumov-ANIM.ACC again about my-LOC.PL with he-M.INS.SG

vzaimootnošenijax.127 relationship-LOC.PL ‘Kul’čickij ... is asking about [pro] the Lubyanka, about [pro] Abakumov, [and] again about [o] my relations with him [sc. Abakumov].’

In (9), the narrator wants to tell Anna Andrejevna (Axmatova) about the return of the poet Osip Mandel’štam and his wife to Moscow (ob ix vozvraščenii v

Moskvu). In first telling Anna about the Mandel’štams (pro Mandel’štamov), the narrator is not analyzing the Mandel’štams—whom Anna already knows—but is presenting them as a holistic topic before proceeding to a newsworthy detail about them, a process/event which has implications beyond that of simply a physical action. In initially identifying the Mandel’štams, the author uses pro, but, in predicating something about them rather than just naming them as a topic, she uses o+LOC.

126 The Moscow building which formerly housed KGB headquarters. 127 Okunevskaja, T. (1998) Tat’janin den’. (RNC)

173

In (10), the referent of the first pro OP is identified (the Lubyanka) as the place where Abakumov is being held. The prosecutor, Colonel Kul’čickij, is not asking specifically about the Lubyanka as an edifice; he is interested in it as the place where the referent of the second pro OP, Abakumov, is located. As both the Lubyanka and Abakumov are entities, they can be presented as holistic topics. This is not the case with the referent of the o+LOC OP (o moix s nim vzaimootnošenijax ‘about my relations with him’) which, as a series of intermittent activities, are more complex. The narrator’s (actress Tatiana Okunevskaja’s) relations with Abakumov denoted by the deverbal noun vzaimootnošenija, are more complex, and they represent recurring events by which she and Abakumov have become linked. During Okunevskaja’s interrogation (which does not take place in the Lubyanka), Kul’čickij enters her cell several times in order to question her about her relationship with Abakumov, who is being held in the

Lubyanka. Being afraid, she speaks with a stutter, necessitating several interviews before Kul’čickij can piece together her complete story—hence the use of opjat’ ‘again’ (‘Kul’čickij is asking ... again about my relations with him

[Abakumov]’).

In both (9) and (10), the o+LOC referent is a deverbal noun (or ‘event noun’), a noun derived from a verb. Although deverbal nouns function primarily as nouns (by which they establish the existence of something), they are, in fact, embedded predicates. A deverbal noun can refer to a specific event (such as

174

vozvraščenie ‘return’ in (9)), which it can present as a deverbal noun can refer to the depicted event as a natural end point (as with (9), which depicts the

Mandel’štams’ return as an accomplishment), or as an unfolding process (as with vzaimootnošenija ‘relations’ in (10)).

In (9), the o+LOC referent (‘their return to Moscow’) represents a predicate

(‘They returned (to Moscow’)). In (10), the o+LOC referent (‘my relations with him’) denotes a multiplicity of recurring events. The use of the plural form of this noun stresses the perdurative nature of this relationship: the two men have met a number of times; Kul’čickij is interested in what took place during these encounters. These are more complex and of broader scope than the pro referents because they involve more than one participant: in (9), the referent includes ‘they’ and ‘Moscow’, while in (10), it includes ‘I’ (the narrator) and

‘Abakumov’, as arguments of the deverbal nouns. In the given passages, pro is used with entities rather than predications, and is therefore narrower in scope.

This association between a referent of an o+LOC OP and predications can also be seen in (11). While the o+LOC referent here is not a deverbal noun as with (9) and (10), it denotes multiple events:

(11) Na odnoj govorili o trudovyx podvigax on one-F.LOC.SG speak-PST.PL about labor-LOC.PL feat-LOC.PL

šaxterov na Donbasse, na vtoroj – pro miner-GEN.PL on Donbass-LOC on second-F.LOC.SG about

175

kolxoz.128 kolkhoz-ACC.SG ‘On the one hand, they spoke about [o] the labor feats of the miners in the Donbass region, [and] on the other hand, about [pro] the kolkhoz (collective farm).’

The feats of labor carried out by the Donbass miners represent a series of related actions or events (predications), while a kolkhoz (a collective farm) is a single, discrete entity.

In (12), similarly to (9) and (10), the o+LOC referent is part of a complex noun phrase. Although the noun is not deverbal, the first adjective in the noun phrase, as the present active participle of a verb, makes the noun phrase, as a unit, a predication:

(12) Razve možno pisat’ pro živogo poėta129 really may write-INF about living-M.ANIM.ACC.SG poet-ANIM.ACC.SG

o nadvigajuščejsja na nego about impend-F.PRS.ACT.PTCP.SG.REFL on he-ANIM.ACC

besprosvetnoj noči?130 impenetrable-F.LOC.SG night-LOC.SG ‘Can one really write about a living poet, about the impenetrable night impending upon him?

128 Karabaš, A. (2002) Byt’ Vasey Strelnikoff. (RNC) 129 There is no comma separating the two PP’s in this example. Even so, the two PP’s are coordinate to each other, and both are subordinate to the verb. 130 Gerštejn, Ė. (1985-1999) Vblizi poėta. (RNC)

176

The background to the passage in (12) is as follows: Ėmma Gerštejn is recounting how Anna Axmatova had written a poem around the time of the arrest, exile, and death of Osip Mandel’štam in 1938. Gerštejn notes that although it was widely believed that Axmatova had completed the poem before

Mandel’štam’s death in December of that year, certain lines were most likely written after his death. She asserts that certain lines refer specifically to him.

The o+LOC PP in this example is a complex event, virtually a narrative in itself, in which the night (i.e., Mandel’štam’s fate) is metaphorically moving toward the poet, to eventually engulf him. The second noun phrase describes what is happening to the first (the ‘living poet’, i.e., Mandel’štam). Whereas pro identifies ‘the living poet’ as a discrete, holistic entity, the ‘impenetrable night impending upon him’ is a temporally complex situation involving two participants—the night (i.e., death), and the ‘living poet’. This is another example of co-occurrence in which the o+LOC referent denotes a temporally complex situation, while the pro referent is presented more holistically, in that identified as an entire entity.

7.4 Preposition Matching

As noted in §7.1, in addition to co-occurrences of o+LOC and pro, there are also numerous cases in the RNC of what I have called preposition matching,

177

in which the same preposition recurs in coordinated phrases within a single clause. Clearly, in these constructions, no contrast can be conveyed by the prepositions alone. For present purposes, the main importance of these examples is that they prove that co-occurrence is not a rule but an authorial choice; that is, they show that there is no constraint against the same preposition recurring in parallel structures.

While my main focus was on the contrastive examples, I also wanted to get an approximate sense of the kinds of contexts in which preposition matching was more typical than co-occurrence. Accordingly, I randomly selected thirty examples each of o...o… and of pro...pro… sequences from the RNC, written between 1890 and 2009 (the same time frame I used elsewhere in the dissertation).

In the sample, the majority of cases of o...o… (28/30) and pro… pro…

(29/30) appeared in coordinate constructions, and in particular in lists of miscellaneous topics presented as non-overlapping items, as in (13a–b).

(13a) Ja rasskazyvaju ej o svoix vstrečax s I-nom tell-PRS.1.SG she-DAT about own-LOC.PL meeting-LOC.PL with

ljud’mi, o cirke, o kino, o people-INS.PL about circus-LOC.SG about cinema-LOC about

zarubežnyx poezdkax.131

131 Nikulin, Ju. (1979) Moje ljubimoe kino. (RNC)

178

foreign-LOC.PL trip-LOC.PL ‘I tell her about [o] my meetings with people, about [o] the circus, about [o] the cinema, about [o] foreign trips.’

(13b) Mamočka ... uspela vsem rasskazat’ pro naš mother-NOM manage-F.PST.SG all-DAT.PL tell-INF about our-ACC

spektakl’ i pro naši kartiny ...132 performance-ACC.SG and about our-ACC.PL picture-ACC.PL ‘Mom managed to tell everyone about [pro] our show and about [pro] our tableaux vivant.’

Both of these examples show a connective relationship between the OP’s. No single OP referent can be said to stand out from the others in any way. Though topically related to each other to a certain degree, the entities in each example have a neutral or unmarked relationship with each other.

This also seems to be true for the remaining example of pro… pro…, which appears in a focus construction (14):

(14) To est’, pro pravitel’stvo možno, a vot that be-PRS.3.SG about government-ACC.SG one.may but PC

pro Kreml’ tak skazat’ nel’zja.133 about Kremlin-ACC so say-INF must.not ‘That is, you can [speak that way] about [pro] the government, but you can’t speak that way about [pro] the Kremlin.’

132 Novickaja, V. (1912) Xorošo žit’ na svete! (RNC) 133 Dorenko, S. (2003) “Levye sily—perezagruzka”. In Zavtra, 8/13/2003. (RNC)

179

Here the two PP’s are presented as simple alternatives in focus of contrast. The only difference between the OP referents in (14) is the fact that one expresses a certain criticism, detailed in the prior context, about the first but not about the second. If the writer had opted for co-occurrence rather than preposition matching, he would perhaps have treated the OP referents as entities of different levels, not as simple items of comparison from the same natural class.

In the remaining two examples of o...o…, the matched prepositions govern

OP’s whose referents are being equated with each other. In (15), a complex subordinating comparative construction, the manner in which the subject of the sentence does not think about God (i.e., takes God for granted) is identical to the manner in which he thinks about the salient properties of summer and winter; he takes all three for granted as ambient facts.

(15) Matvej ... privyk ne dumat’ o boge, kak ne Matvej-NOM used not think-INF about God-LOC like not

dumal letom o teple a think-M.PST.SG summer-INS.SG about warmth-LOC and

zimoju o snege i xolode.134 winter-INS.SG about snow-LOC and cold-LOC ‘Matvej was used to not thinking about [o] God, just as he didn’t think

134 Gor’kij, M. (1910) Žizn’ Matveja Kožemjakina. (RNC)

180

about [o] warmth in the summer or about [o] snow and cold in the winter.’

If the author (Maksim Gor’kij) had switched prepositions here, he would have implied some kind of contrast, which would have directly undermined the equivalence he was trying to establish.

The same explanation probably applies to (16), where Nina Berberova, a leading writer of the twentieth century, discusses the high regard that the British journalist (and secret agent) Sir Bruce Lockhart had for Raymond Robins, head of the American Red Cross in Moscow during the 1917 Revolution.

(16) Lokkart pišet o Robinse kak o Lockhart-NOM write-PRS.3.SG about Robins-LOC as about

talantlivom živom sobesednike ...135 talented-M.LOC.SG lively-M.LOC.SG interlocutor-LOC.SG ‘Lockhart writes about [o] Robins as (about [o]) a talented, lively interlocutor.’

Here, the noun-complement PP o… sobesednike is presented as a characterization (or categorization) of the main PP o Robinse. The conjunction kak ‘like, as’ does not indicate the manner in which Lockhart is writing; rather, it is the pivot (marker) of the comparison that Lockhart is said to make between

Robins (comparee) and the type of a skilled orator (standard of comparison).

135 Berberova, N. (1978-80) Železnaja ženščina. (RNC)

181

The matched PP’s in (16) can be with contrasted with a case of co- occurrence involving the same noun-complement construction (17). Here, the mismatch of prepositions is appropriate, as the equivalence between the token referent (the main PP) and the type (the complement) is purely ironic:

(17) Čto ž ėto tovarišč Berezkin, a mne pro what PC this comrade-NOM.SG Berezkin-NOM but I-DAT about

vas govorili kak o čeloveke xrabrom you-ACC.PL speak-PST.PL like about person-LOC.SG brave-M.LOC.SG

spokojnom.136 calm-M.LOC.SG ‘What’s this, Comrade Berezkin? And they told me that you were a brave and coolheaded person [literally, they spoke me about [pro] you as (about [o]) a brave and coolheaded person’].

In this example, the represented speaker is a Red Army division commander chastising Major Berezkin for his “timidity” after two incidents in which the latter deliberately took his men out of the line of heavy German fire in order to prevent needless casualties. His comparison between Berezkin and the standard—a

(hypothetical) brave, cool person—is unmistakably sarcastic; the implicature is obviously that Berezkin does not belong to the category of brave, cool people. If the author had matched the prepositions, it would not have changed the basic message, but it would not have highlighted the incongruity between Berezkin and

136 Grossman, V. (1960) Žizn’ i sud’ba. (RNC)

182

his prior reputation as effectively. (Note that the author of (17) was a highly regarded writer of serious fiction, Vasilij Grossman; thus there is every reason to assume that he was using the contrastive prepositions to convey a subtle pragmatic nuance.) The nature of the opposition in (17) is that of a discrete entity (Berezkin) being contrasted with a more analyzed entity (the ideal type of officer for combat).

While the pragmatic differences between (16) and (17) are suggestive, there are too few examples of the complement construction in the sample to determine conclusively whether or not the referents of this type of sentence are more likely to be equated than contrasted. This question is worthy of future research.

7.5 Summary

In the co-occurrences, the contrast between o+LOC and pro highlights specific properties within their object referents. The referent of a pro OP tends to be presented as having coherent, defined boundaries; in contrast, the referent of an o+LOC OP tends to be presented as broader in scope, less-defined in its boundaries, and to be process- or event-oriented. Thus, when the referent is not an entity but a proposition or a situation/event which unfolds or develops over time (e.g., deverbal or quasi-deverbal nouns)—there is a decided preference for

183

o+LOC. This suggests that when there is co-occurrence of a pro and an o+LOC referent (i.e., when the two prepositions are in contrast), the pro referent represents an entity of narrower scope than that of the o+LOC referent.

The use of the opposition to indicate or emphasize a pragmatic contrast between co-occurring pro and o+LOC object referents is summarized in Table

7.37:

pro object referent o+LOC object referent narrow in scope broad in scope concrete abstract holistic analyzing/detailing actual potential specific generic determined entities propositions, processes, events

Table 7.37 Opposition of Co-occurring pro and o+LOC Object Referents

Although not every co-occurrence manifests the relationship between the o+LOC and pro referents within them which I have discussed here, there is enough evidence to show that such a relationship does indeed exist. What seems at first glance to be a purely random choice turns out to be strongly motivated—not constrained at 100%, but valid in the majority of cases.

184

Chapter 8: Summary and Conclusions

8.1 Object of the Investigation

8.1.1 Semantic Variation between o+LOC and pro

The main issue addressed in this dissertation is the variation between the

‘aboutness’ prepositions o+LOC and pro, as well as the less common po povodu and nasčet in standard modern Russian between 1890 and the present, as represented by the Russian National Corpus. The given research is, as far as I am aware, the only existing study to address this issue both qualitatively and quantitatively. As such, I hope it will be of use for the study of prepositional

‘aboutness’ in general.

Although o+LOC and pro are frequent in Russian, there have been few previous studies of the factors motivating the variation, and none have paid any attention to the contexts in which one of the prepositions tends to occur instead of another. This dissertation is the first work to examine in depth the semantic nuances of the individual prepositions and their co-occurrence patterns—that is, to what extent they collocate with different semantic classes of verbs, and to what extent they govern nouns with different semantic (individuation) properties.

185

My goal has been to offer a cogent explanation for these tendencies, primarily on the basis of cognitive semantics, and thus to fill in some of the gaps in the previous literature on the subject.

8.2 Stylistic vs. Semantic Factors

For the most part, the explanations offered for the variation in the previous studies have been purely stylistic, involving a supposed dichotomy between ‘colloquial’ pro and neutral o+LOC. For example, on the assumption that o+LOC and pro fufilled the same essential function(s), Vinogradov

(1947/2001), the foremost Soviet linguist of the mid-twentieth century, believed that the use of pro was generally declining, and that the preposition would soon disappear altogether in favor of o+LOC. However, the fact that over six decades later both o and pro are still alive and well suggests that there is enough of a difference between the two prepositions that speakers still feel the need for both.

Therefore, as a preliminary to formulating a viable semantic hypothesis, I had to demonstrate that the labelling of pro as merely a colloquial synonym of o+LOC is to a large degree inaccurate and has for many years been a non- explanation for the variation. As evidence for this, I showed, inter alia, that pro is often found in non-colloquial contexts, such as with verb participles, and that pro and o+LOC can co-occur in adjacent clauses in the same syntactic function.

186

Moreover, the co-occurrence sentences that I examined in Chapter 7 can only be understood if the opposition between the two prepositions involved semantic nuances rather than stylistic differences.

8.3 Main Results of the Quantitative Study

I performed two series of quantitative tests on the datasets in this study; I verified the statistical significance of the results by means of Fisher’s Exact test.

The first series measured the distribution of five noun features between o+LOC and pro datasets with corresponding verbs (govorit’ o~govorit’ pro, etc.). The second series measured the distribution of five noun features between individual prepositions with each member of aspectually paired verbs (govorit’ o~skazat’ o, etc.).

In the first series of tests, the datasets show no clear evidence of diachronic change. For several verbs (such as podumat’ ‘think-PERF’, znat’

‘know-IMP’, rasskazat’ ‘tell-PERF’), there are significant results for the same noun feature in several time periods; however, there are no trends which encompass all verbs (or all verbs within any given semantic category).

The majority of significant results in the first series of tests are with the factor groups Animacy status, Modification status, Proper/Common noun status, and Deverbal status. There is a definite tendency for pro to govern animate,

187

modified, proper, and non-deverbal nouns. In other words, the statistically significant results in the first series of tests indicate that the referents of pro objects are more likely than those of o+LOC to be animate, non-deverbal, proper, or unmodified than those of o+LOC objects.

The results of the second series of tests were not conclusive. There were far fewer significant results than in the first series of tests, and the distribution of these results does not show any diachronic trends. In other words, the aspect of the verb does not appear to be a significant factor in the choice of ‘aboutness’ preposition.

8.4. Main Results of the Qualitative Study

Important clues to understanding the semantics of the different

‘aboutness’ prepositions come from the classes of verbs with which they typically co-occur. The main class of verbs that govern pro are those which convey acts of communication or cognition; pro introduces the semantic Topic of the communication or cognition. By contrast, the objects of o+LOC can refer not only to the Topic of communication or cognition, but also to the Cause of a nonvolitional state (an emotional/psychological reaction) or the Stimulus of a volitional activity (with zabotit’sja ‘take care of-IPFV’ and razmyšljat’/razmyslit’

‘ponder’). Moreover, the Topic introduced by o+LOC does not have to be an

188

entity; it can also be a proposition (svidetel’stvovat’ ‘witness-IPFV’, propovedovat’

‘preach-IPFV’), sudit’ ‘judge-IPFV’, and razmyšljat’/razmyslit’ ‘ponder’), or a future action (with directive speech act verbs such as prosit’ ‘ask (for)’, xodatajstvovat’

‘petition-IPFV’, and predupreždat’/predupredit’ ‘warn’. In the last case, the verbs that denote directive speech acts (prosit’ o doroge ‘ask about the road’, i.e., ‘ask to be given directions’) represent an attempt to cause a future action involving the OP referent. In contrast, as a Topic, a pro object referent can be used only as part of an assertive point, in which “a speaker represents as actual a state of affairs” (Vanderveken (1990-91: 105).

Thus, pro can be considered a hyponym of o+LOC, since it occupies a portion of the overall semantic field of o+LOC, but has no additional senses which o+LOC does not. Table 8.38 lists verbs with which the object referent of an ‘aboutness’ preposition codes a semantic role other than Topic, along with the type of speech act represented by an ‘aboutness’ argument phrase. The verbs in

Table 8.38 occur very rarely with object referents of pro, nasčet, or po povodu:

189

Semantic Role of Type of Speech Verb Object Referent act spravljat’sja/spravit’sja Topic assertive ‘inquire, check on’ prosit’/poprosit’ Proposition directive ‘request, ask for’ umoljat’/umolit’ Proposition directive ‘beg, beseech’ vzyvat’/vozzvat’ Proposition directive ‘appeal (for)’ xodatajstvovat’/poxodatajstvovat’ Proposition directive ‘solicit (for), appeal (for)’ svidetel’stvovat’ Proposition constative ‘witness, testify (to)’ propovedovat’ Proposition constative/directive ‘preach’ predupreždat’/predupredit’ Proposition constative/directive ‘warn’ rasporjažat’sja/rasporjadit’sja Proposition directive ‘order, make arrangements’ sudit’ ‘judge’ Proposition assertive/expressive napominat’/napomnit’ ‘remind’ Topic assertive (REMIND1) napominat’/napomnit’ ‘remind’ Phenomenon N/A (REMIND2) razmyšljat’/razmyslit’ Stimulus N/A ‘ponder’ zabotit’sja/pozabotit’sja ‘look after, take care (of)’ Stimulus N/A bespokoit’sja Cause N/A ‘be worried’ volnovat’sja Cause N/A ‘be worried, agitated’ toskovat’ Cause N/A ‘feel sad, languish’ žalet’ Cause N/A ‘regret, feel sad’

Table 8.38 Verbs, Semantic Roles of Object Referents, and Speech Acts

190

The opposition of Topic and Non-topic is also evident in the use of these prepositions with the verb napominat’ ‘remind-IPFV’, which can be either a verb of speech, by which it describes the communication of a topic (REMIND1) or of cognition, by which it describes the nonvolitional causation of a mental recollection (REMIND2). The distribution of the speech sense and the cognition sense in the pro dataset (Pd. D) for this verb is heavily weighted toward the speech sense compared to that of o+LOC; the P-value is a statistically significant

0.0004. This is not surprising, considering that the semantic role of an object referent of pro must be that of Topic; an object referent of REMIND2 is not a Topic, but a Phenomenon.137

However, pro can also be used in PP’s with verbs of cognition that portray the OP referent as a cognized Topic, either volitionally (znat’ ‘know-IPFV’, dumat’/podumat’ ‘think’) or nonvolitionally (zabyvat’/zabyt’/pozabyt’ ‘forget’, vspominat’/vspomnit’ ‘remember’, etc.). One exception is the aspective pair of cognitive verbs razmyšljat’/razmyslit’ ‘ponder’, which rarely occur with pro. They describe a depth of thought more intense and reflective than that of dumat’/podumat’ ‘think’. An entity/situation over which someone ponders is therefore likely to bring about a change of state in that person; it motivates a

137 See “Unified Verb Index”, University of Colorado. Web site https://framenet2.icsi.berkeley.edu/fnReports/data/frame/Evoking.xml. Viewed 12/18/2012.

191

more complex internal discussion than something which is simply being thought about. I maintain that razmyšljat’/razmyslit’ are disfavored with pro because the

‘aboutness’ object referent of these verbs codes the semantic role of Stimulus.

The other area of opposition between o+LOC and pro is with emotional/psychological verbs (bespokoit’sja ‘be anxious, upset’, volnovat’sja ‘be worried, agitated’, etc.). An object referent of these verbs codes the semantic role of Cause or Stimulus of a (nonvolitional) state in an Experiencer. With these verbs, as with the verbs mentioned above in which the object referent is not a

Topic, pro is strongly disfavored in RNC texts throughout the four time periods represented in this study.

As far as nasčet and po povodu and their relation to o+LOC and pro, the observation that both are attested in the RNC with certain verbs of emotional state such as bespokoit’sja ‘be worried’ and volnovat’sja ‘be worried/agitated’

(though not in large numbers) suggests that the semantic field of these two prepositions is larger than that of pro, but smaller than that of o+LOC.

In sum, both the quantitative and qualitative analyses that I have carried out on the four ‘aboutness’ prepositions indicate that the referents of pro objects can code the semantic role of Topic, but rarely occurs in other roles such as

Cause, Stimulus, or Proposition. By contrast, the referents of o+LOC objects can convey the semantic roles of Topic, Proposition, Cause, or Stimulus, while the

192

referents of nasčet and po povodu objects can convey the semantic role of Topic,

Cause, or Stimulus.

The verbs associated either mostly or exclusively with object referents of o+LOC tend to be those which denote speech acts which are more complex in function than those associated with pro, and in some cases, denote several speech acts (e.g., propovedovat’ ‘preach-IPFV’ and predupreždat’/predupredit’

‘warn’ can denote an assertive and/or a constative illocutionary point). On the other hand, verbs associated with object referents of pro can only denote an assertive speech act (e.g. talking, telling, inquiring, hearing (about), reading, writing, and cognition), in which the object referent codes the semantic role of

Topic.

The variation between o+LOC and pro seen in the co-occurrence sentences is context sensitive. O+LOC is unmarked when it occurs alone, however, it can be seen as having a contrastive meaning to that of pro when the two co-occur, and is thus marked within that particular context.

The research in this work has involved more than just the four ‘aboutness’ prepositions themselves; it has encompassed a study of the semantics of other constituents, such as nouns and verbs, as well. To reiterate, “You shall know the meaning of a word by the company it keeps” (J.R. Firth, cited in Cruse 2011:

215). Certainly, this has proven to be true in the data used in this study.

193

8.2 Perspectives for Further Research

This study of four ‘aboutness’ prepositions in Russian has encompassed several different areas that reveal possibilities for future research. For example, regarding speech acts/illocutionary points, it can be seen that certain verbs appear to represent more than one illocutionary point (such as propovedovat’

‘preach-IMPV’ and predupreždat’/predupredit’ ‘warn’, which can be directive or constative), depending on the speaker’s intention. Further research on these verbs, may yield more information about the semantics of o+LOC and pro.

Sample datasets of these verbs may contain a certain percentage of directives vs. constatives, for example. Regarding the factor groups for the noun objects in the present study, it would be worthwhile to conduct further quantitative study using other factors such as grammatical number (singular~plural), grammatical polarity (of either the noun object or the verb), or to include deverbal nouns of other derivational types in the factor group Deverbal status.

194

References

Aksakov, K. (1880) Sočinenija filologičeskija. Čast’ 2. Opyt russkoj grammatiki. Moskva: Universitetskaja tip.

Avanesov, R. and Sidorov, V. (1945) Očerk grammatiki russkogo literaturnogo jazyka. Čast’ 1: Fonetika i morfologija. Moskva: Gos. učebno-pedagog. izd-vo Narkomprosa RSFSR.

Barsov, A. (1981) Rossijskaja grammatika Antona Alekseeviča Barsova. Editor, B. Uspenskij. Moskva: Izd-vo Moskovskogo universiteta. (Originally published in 1788)

Bondarenko, V. (1961) Predlogi v sovremennom russkom jazyke. Moskva: Gos. učebno-pedagog. izd-vo.

Buslaev, O. (1858) Opyt istoričeskoj grammatiki russkago jazyka. Č. 2; Sintaksis. Moskva: Universitetskaja tip.

Case, animacy, and semantic roles (2011). Kittilä, S., Västi, K., and Ylikoski, J., eds. Amsterdam: John Benjamins Publishing Co.

Čerkasova, E. (1967) Perexod polnoznačnyx slov v predlogi. Moskva: Nauka.

Corbett, Greville. 2008. Determining Morphosyntactic Feature Values: the Case of Case. In Case and Grammatical Relations: Papers in Honour of Bernard Comrie, ed. Greville G. Corbett and Michael Noonan, 1-34. Amsterdam: John Benjamins.

195

Cruse, A. (2011) Meaning in Language: an Introduction to Semantics and Pragmatics. Oxford: Oxford University Press.

Dal’, V. (1863-66) Tolkovyj slovar’ živago velikorusskago jazyka. Izd. 1-e. Website www.slovari.yandex.ru.

------(1882) Tolkovyj slovar’ živago velikorusskago jazyka. Izd. 2-e. S.- Peterburg: Izd. M.O. Vol’fa.

------(1903-1909) Tolkovyj slovar’ živago velikorusskago jazyka. Izd. 3. I.A. Baudouin de Courtenay, ed. Sankt-Peterburg: Izd. T-va M.O. Vol’f.

Davidson, D. (1978) “What metaphors mean”. In Critical inquiry 5: 31-47.

Drevnerusskie berestjanye gramoty. Website http://gramoty.ru. Viewed on 2/4/2012.

Dervin, R. (1993) “Dividing up physical and mental space into conceptual categories by means of English prepositions.” In Zelinsky-Wibbelt.

Dobroljubov, N. (1859) “Čto takoe oblomovščina?” On website Lib.ru: Klassika http://az.lib.ru/d/dobroljubow_n_a/text_0022.shtml. Viewed on 3/5/2012.

Evgen’eva, A., ed. (1981-84) Slovar’ russkogo jazyka v 4 tomax. Izd. 2, ispr. i dop. Moskva: Russkij jazyk.

Greč, N. (1827) Praktičeskaja grammatika russkogo jazyka. Sanktpeterburg: Tip. Imper. Sanktpeterburgskago vospitatel’nago doma.

196

------(1830) Prostrannaja russkaja grammatika. T. 1. Izd. 2-e. Sankt-Peterburg: N. Greč.

------(1830) Načal’nyja pravila russkoj grammatiki. Izd. 2-e. Sankt-Peterburg: N. Greč.

------(1845) Kratkaja russkaja grammatika. 10-e izd. Sankt-Peterburg: Tip. V. Gaspera.

Jackendoff, R. (1973) "The Base Rules for Prepositional Phrases." In S. Anderson and P. Kiparsky, eds., A Festschrift for Morris Halle. New York: Holt, Rinehart and Winston, Inc., pp. 345-356.

------(1983) Semantics and cognition. Cambridge, Mass.: MIT Press.

Jakobson, R. (1936) “Contribution to the general theory of case: general meanings of the Russian cases.” In Russian and Slavic Grammar: Studies, 1931-1981, ed. Linda R. Waugh and Morris Halle, Ch. 6, [59]- 103. Berlin: Mouton.

Kier, A. (2009) “The prepositions pro and o in modern Russian: a corpus analysis.” Unpublished manuscript.

Kurganov, N. (1793) Pismovnik. Kniga 1: Grammatika. St.-P.: Imperatorskaja Akademija nauk. 5. izd., pt. 1.

Kuznecov, S., ed. (2000) Bol’šoj tolkovyj slovar’ russkogo jazyka. Sankt- Peterburg: Norint.

Lomonosov, M. (1755) Rossiīskaja grammatika Mixajla Lomonosova. Sankt

197

Peterburg: Imper. Akademija nauk.

Lunt, H. (1969) Old Church Slavonic glossary. Corrected reprint. [Boston, MA]: Harvard University, Dept. of Slavic Languages and Literatures.

Lyons, J. (1977) Semantics. Cambridge: Cambridge Univ. Press.

Merzon, S. and Pyatetskaya, S. (1983) Russian verbs in speech. Moscow: Russky Yazyk.

Milroy, L. and Gordon, M. (2003) Sociolinguistics: method and interpretation. Malden, Mass.: Blackwell.

Opyt obščesravnitel’noj grammatiki russkago jazyka (1853) Izd. 2-e. Sankt- Peterburg: Tip. Imper. Akademii nauk.

Ogienko, I., ed. (1914) Slovar’ nepravil’nyx, trudnyx, i somnitel’nyx slov, sinonimov i vyraženij v russkoj rěči. Izd. 3-e, značitel’no dop. i soveršenno perer. Kīev: Knigoizd-vo I.I. Samonenko.

Ožegov, S., ed. (1986) Slovar’ russkogo jazyka. Izd. 18. Moskva.

Palevskaja, M. (1964) Sinonimy v russkom jazyke. Moskva: Prosveščenie.

Peškovskij, A. (2001) Russkij sintaksis v naučnom osveščenii. 8-e izd., dop. Moskva: Jazyki slavjanskoj kul’tury.

Potebnja, A. (1941) Iz zapisok po russkoj grammatike. v. 4: Glagol. Mestoimenie. Čislitel’noe. Predlog. Moskva: Izd-vo Akademii nauk SSSR.

198

Ricœur, P. (2008) From text to action. London: Continuum.

Russian National Corpus [RNC]. Website http://ruscorpora.ru/en/index.html. Viewed 3/1/2011.

Russkaja grammatika, Akademija nauk SSSR, Institut russkogo jazyka. http://www.rusgram.narod.ru/1121-1146.html, viewed 9/1/2012.

Šaxmatov, A. (2001) Sintaksis russkogo jazyka. 3rd ed. Moskva: URSS.

Sazonova, I. (1963) Leksika i frazeologia sovremennogo russkogo literaturnogo jazyka. Moskva: Izd-vo lit-ry na inostrannyx jazykax.

The Semantics of Prepositions: From Mental Processing to Natural Language Processing (1993). Editor, Cornelia Zelinsky-Wibbelt. Berlin: Mouton de Gruyter.

Skoblikova, E. (1971) Soglasovanie i upravlenie v russkom jazyke. Moskva: Prosveščenie.

Slovar’ Akademii Rossijskoj (1789-1794). Sankt-Peterburg: Imper. Akademija nauk. Čast’ 4; M-R (1793). Website www.runivers.ru.

Slovar’ Akademii Rossijskoj. Čast’ 5 (1822). SPb: Tip. V. Plavil’ščikov. Facsimile ed., Izd-vo Universiteta Odense, 1971.

Slovar’ russkogo jazyka XI-XVII vv. (1987). Vyp. 12 (O–Oparnyj). Moskva: Nauka.

------(1995). Vyp. 20 (Prisvoenie–Pročnutisja). Moskva: Nauka.

199

Slovar’ russkogo jazyka XVIII veka. Vyp. 14. (2004) Sankt-Peterburg: Nauka. Website http://feb-web.ru/feb/sl18/slov-abc/. Viewed 2/12/2012.

Sovremennyj literaturnyj russkij jazyk (1988). Pod redakciej P.A. Lekanta. Izd. 2., ispr. Moskva: Vysšaja škola.

Stojan, P., ed. (1916) Malyj tolkovyj slovar’. 3-e izd. Petrograd: N.Ja. Ogolbin.

Švedova, N., ed. (2007) Tolkovyj slovar’ russkogo jazyka: s vključeniem svedenij o proisxoždenii slov. Moskva: Azbuovnik.

Tagliamonte, S. (2006) Analysing Sociolinguistic Variation. Cambridge: Cambridge University Press.

Talmy, L. (1978) Figure and ground in complex sentences. In Universals of human language, ed. Joseph H. Greenberg 4: Syntax, 627–49. Stanford, Calif.

Taylor, J. (1993) “Prepositions: Patterns of Polysemization and Strategies of Disambiguation.” In Zelinsky-Wibbelt.

Timberlake, A. (1975) “Hierachies in the Genitive of Negation.” Slavic and East European Journal: 19. 123-139.

------(2004) A Reference Grammar of Russian. Cambridge: Press Syndicate of the University of Cambridge.

Unified Verb Index. University of Colorado. Website https://framenet2.icsi.berkeley.edu. Viewed 6/18/2012.

200

Ušakov, D., ed. (1935-1940) Bol’šoj tolkovyj slovar’ sovremennogo russkogo jazyka. Website http://ushdict.narod.ru. Viewed 1/10/12.

Vanderveken, D. (1990-91) Meaning and speech acts. Cambridge: Cambridge University Press.

VassarStats: Website for Statistical Computation. Website http://vassarstats.net/index.html. Viewed on 1/1/12.

Vinogradov, V. (2001) Russkij jazyk: grammatičeskoe učenie o slove. 4-e izd. Pod red. G.A. Zolotovoj. Moskva: Russkij jazyk. (1st edition published in 1947).

Vostokov, A. (1845) Sokrašennaja russkaja grammatika. Izd. 4-e. Moskva: Universitetskaja tip.

------(1874) Russkaja grammatika. Izd. 12-e. Sankt-Peterburg: D.F. Fedorov.

Worth, D. (1998) “Vtoroj vinitel’nyj v sovremennom russkom literaturnom jazyke.” In Liki jazyka: k 45-letiju naučnoj dejatel’nosti E.A. Zemskoj, 7-13. Moscow: Nasledie.

Zelinsky-Wibbelt, C. (1993) “Interpreting and translating prepositions: A cognitively based formalization.” In The Semantics of Prepositions: From Mental Processing to Natural Language Processing, ed. Cornelia Zelinsky-Wibbelt, [351]-390. Berlin: Mouton de Gruyter.

201

Appendix A: Results of Fisher’s Exact Tests on Datasets

1. Noun Factors (Statistically significant results are in bold red)

Verbs of Speech (185 tests)

boltat’ o~pro (D) dolozit’ o~pro (D) molčat’ o~pro (D) tolkovat’ o~pro (D) animacy 0.0878 animacy 0.0092138 animacy 0.2326 animacy 1.0000 ct./mass 1.0000 ct./mass 1.0000 ct./mass 0.1028 ct./mass 0.6942 modif. 0.0018139 modif. 0.7310 modif. 0.0539 modif. 0.5326 com./prop. 0.0274140 com./prop. 0.0633 com./prop. 0.7065 com./prop. 1.0000 deverbal 0.5402 deverbal 0.0183141 deverbal 1.0000 deverbal 0.3938

govorit’ o~pro (A) govorit’ o~pro (B) govorit’ o~pro (C) govorit’ o~pro (D) animacy 0.5321 animacy 0.2668 animacy 1.0000 animacy 0.1432 ct./mass 0.0575 ct./mass 0.1028 ct./mass 1.0000 ct./mass 0.1945 modif. 0.1154 modif. 0.4118 modif. 1.0000 modif. 0.6010 com./prop. 1.0000 com./prop. 0.4716 com./prop. 1.0000 com./prop. 0.1945 deverbal 0.0523 deverbal 0.3533 deverbal 0.1945 deverbal 1.0000

govorit’ o~nasčet (A) govorit’ o~nasčet (B) govorit’ o~nasčet (C) govorit’ o~nasčet (D) animacy animacy 0.3995 animacy 0.2314 animacy 0.1825 (N/A) ct./mass ct./mass 0.1550 ct./mass 0.3491 ct./mass 0.4940

138 In tests between o and pro for the factor group ‘Animacy Status of Noun’, all significant P-values show an association between between pro and animate nouns. 139 In tests between o and pro for the factor group ‘Modification Status of Noun’, all significant P-values show an association between pro and unmodified nouns. 140 In tests between o and pro for the factor group ‘Proper/Common Status of Noun’, all significant P-values show an association between pro and proper nouns. 141 In tests between o and pro for the factor group ‘Deverbal Status of Noun’, all significant P-values show an association between pro and non-deverbal nouns.

202

modif. modif. 1.0000 modif. 0.0603 modif. 0.0058142 com./prop. com./prop. 0.6299 com./prop. 1.0000 com./prop. 0.1432 deverbal deverbal 1.0000 deverbal 0.6963 deverbal 1.0000

govorit’ o~po povodu govorit’ o~po povodu govorit’ o~po povodu govorit’ o~po povodu

(A) (B) (C) (D) animacy 1.0000 animacy 1.0000 animacy animacy 0.6852 ct./mass 0.1696 ct./mass 0.1167 ct./mass ct./mass 0.3806 modif. 0.0581 modif. 0.0521 modif. (N/A) modif. 0.1842 com./prop. 1.0000 com./prop. 0.6868 com./prop. com./prop. 0.2933 deverbal 1.0000 deverbal 0.4893 deverbal deverbal 0.1091

pogovorit’ o~pro (A) pogovorit’ o~pro (B) pogovorit’ o~pro (C) pogovorit’ o~pro (D) animacy animacy animacy 0.7292 animacy 1.0000 ct./mass ct./mass ct./mass 0.6359 ct./mass 1.0000 modif. (N/A) modif. (N/A) modif. 0.7688 modif. 0.2092 com./prop. com./prop. com./prop. 0.0329 com./prop. 1.0000 deverbal deverbal deverbal 0.5153 deverbal 1.0000

pogovorit’ o~nasčet pogovorit’ o~nasčet pogovorit’ o~nasčet pogovorit’ o~nasčet

(A) (B) (C) (D) animacy animacy 0.1630 animacy 0.7292 animacy 0.0675 ct./mass ct./mass 0.5418 ct./mass 1.0000 ct./mass 1.0000 modif. (N/A) modif. 0.1043 modif. 0.7688 modif. 0.6974 com./prop. com./prop. 0.6299 com./prop. 1.0000 com./prop. 0.3811 deverbal deverbal 0.5183 deverbal 1.0000 deverbal 0.6411

skazat’ o~pro (A) skazat’ o~pro (B) skazat’ o~pro (C) skazat’ o~pro (D) animacy 0.0292 animacy 0.0539 animacy 0.1581 animacy 0.2668 ct./mass 1.0000 ct./mass 1.0000 ct./mass 1.0000 ct./mass 1.0000 modif. 0.7948 modif. 0.0292 modif. 0.1102 modif. 0.0033 com./prop. 1.0000 com./prop. 0.1028 com./prop. 0.2990 com./prop. 0.2990 deverbal 0.0523 deverbal 0.0522 deverbal 0.1124 deverbal 0.3533

142 In tests between o and nasčet for the factor group ‘Modification Status of Noun’, all significant P-values show an association between nasčet and unmodified nouns.

203

skazat’ o~nasčet (A) skazat’ o~nasčet (B) skazat’ o~nasčet (C) skazat’ o~nasčet (D) animacy 1.0000 animacy 0.2918 animacy 0.5184 animacy 1.0000 ct./mass 1.0000 ct./mass 1.0000 ct./mass 0.5613 ct./mass 0.7065 modif. 0.0309 modif. 0.3496 modif. 0.1563 modif. 0.0182 com./prop. 0.4948 com./prop. 0.1141 com./prop. 0.6804 com./prop. 0.4716 deverbal 0.1638 deverbal 0.1470 deverbal 0.1336 deverbal 1.0000

rasskazyvat’ o~pro rasskazyvat’ o~pro rasskazyvat’ o~pro rasskazyvat’ o~pro

(A) (B) (C) (D) animacy 0.0004 animacy 0.0470 animacy 0.1806 animacy 0.2092 ct./mass 1.0000 ct./mass 1.0000 ct./mass 0.1028 ct./mass 1.0000 modif. 0.1205 modif. 0.3015 modif. 1.0000 modif. 0.1964 com./prop. 0.0797 com./prop. 0.0523 com./prop. 1.0000 com./prop. 0.6707 deverbal 0.1455 deverbal 0.2373 deverbal 0.1945 deverbal 0.0046

rasskazat’ o~pro (A) rasskazat’ o~pro (B) rasskazat’ o~pro (C) rasskazat’ o~pro (D) animacy 0.0251 animacy 0.0716 animacy 0.0153 animacy 0.0211 ct./mass 0.0237143 ct./mass 0.6120 ct./mass 0.6120 ct./mass 1.0000 modif. 0.5675 modif. 0.2949 modif. 0.1188 modif. 0.2993 com./prop. 0.0257 com./prop. 0.1028 com./prop. 1.0000 com./prop. 0.7065 deverbal 0.0797 deverbal 0.0522 deverbal 0.0105 deverbal 0.1124

Verbs of Cognition (235 tests)

dumat’ o~pro (A) dumat’ o~pro (B) dumat’ o~pro (C) dumat’ o~pro (D) animacy 0.0191 animacy 0.5321 animacy 0.7847 animacy 0.2326 ct./mass 1.0000 ct./mass 1.0000 ct./mass 0.2542 ct./mass 0.0523 modif. 1.0000 modif. 0.5889 modif. 1.0000 modif. 0.7611 com./prop. 1.0000 com./prop. 1.0000 com./prop. 1.0000 com./prop. 0.1945 deverbal 0.0237 deverbal 0.4915 deverbal 0.4915 deverbal 1.0000

podumat’ o~pro (A) podumat’ o~pro (B) podumat’ o~pro (C) podumat’ o~pro (D) animacy 0.0002 animacy 0.0005 animacy <0.0001 animacy 0.0840 ct./mass 1.0000 ct./mass 0.3878 ct./mass 0.2100 ct./mass 0.3533 modif. 0.2686 modif. 1.0000 modif. 0.2515 modif. 1.0000 com./prop. 1.0000 com./prop. 0.0273 com./prop. <0.0001 com./prop. 0.3123

143 This result shows an association between pro and count nouns.

204

deverbal 0.5558 deverbal 0.1418 deverbal 0.0282 deverbal 0.0257

zabyvat’ o~pro (A) zabyvat’ o~pro (B) zabyvat’ o~pro (C) zabyvat’ o~pro (D) animacy animacy 0.1646 animacy 0.7065 animacy 0.0048 ct./mass ct./mass 0.2453 ct./mass 1.0000 ct./mass 0.6707 modif. (N/A) modif. 1.0000 modif. 0.6042 modif. 0.2789 com./prop. com./prop. 0.6457 com./prop. 1.0000 com./prop. 0.6707 deverbal deverbal 1.0000 deverbal 0.3533 deverbal 0.1028

zabyt’ o~pro (A) zabyt’ o~pro (B) zabyt’ o~pro (C) zabyt’ o~pro (D) animacy 1.0000 animacy 0.7480 animacy 1.0000 animacy 1.0000 ct./mass 0.7065 ct./mass 0.4716 ct./mass 0.4238 ct./mass 0.7065 modif. 1.0000 modif. 0.2326 modif. 1.0000 modif. 0.1872 com./prop. 0.7065 com./prop. 0.2373 com./prop. 0.0237 com./prop. 0.6707 deverbal 0.2990 deverbal 1.0000 deverbal 0.2373 deverbal 1.0000

pozabyt’ o~pro (A) pozabyt’ o~pro (B) pozabyt’ o~pro (A) pozabyt’ o~pro (A) animacy 0.2227 animacy 0.7325 animacy 0.6918 animacy 0.7847 ct./mass 0.5586 ct./mass 0.0771 ct./mass 1.0000 ct./mass 0.1124 modif. 0.2600 modif. 0.1888 modif. 0.3620 modif. 0.7847 com./prop. 0.5586 com./prop. 1.0000 com./prop. 0.2814 com./prop. 1.0000 deverbal 0.6034 deverbal 0.2847 deverbal 0.6400 deverbal 1.0000

znat’ o~pro (A) znat’ o~pro (B) znat’ o~pro (C) znat’ o~pro (D) animacy 0.0523 animacy 0.0003 animacy 0.0056 animacy 0.0008 ct./mass 0.3533 ct./mass 0.6120 ct./mass 1.0000 ct./mass 0.4915 modif. 0.7892 modif. 1.0000 modif. 0.0326 modif. 0.7948 com./prop. 0.0237 com./prop. 0.1028 com./prop. 0.1945 com./prop. 1.0000 deverbal 0.0122 deverbal 0.0105 deverbal 0.0046 deverbal 0.4716

znat’ o~nasčet (A) znat’ o~nasčet (B) znat’ o~nasčet (C) znat’ o~nasčet (D) animacy animacy 0.0044144 animacy 0.0521 animacy 0.0003 ct./mass (N/A) ct./mass 0.1916 ct./mass 1.0000 ct./mass 1.0000 modif. modif. 0.1333 modif. 0.0733 modif. 0.0022

144 In tests between o and nasčet for the factor group ‘Animacy Status of Noun’, all significant P-values show an association between nasčet and animate nouns.

205

com./prop. com./prop. 0.0184145 com./prop. 1.0000 com./prop. 0.1945 deverbal deverbal 0.1638 deverbal 0.0831 deverbal 0.2542

vspominat’ o~pro (A) vspominat’ o~pro (B) vspominat’ o~pro (C) vspominat’ o~pro (D) animacy 1.0000 animacy 0.2487 animacy 0.0973 animacy 0.0021 ct./mass 0.5213 ct./mass 0.4993 ct./mass 1.0000 ct./mass 1.0000 modif. 0.1351 modif. 1.0000 modif. 0.0307 modif. 0.1188 com./prop. 0.4682 com./prop. 0.3243 com./prop. 0.0745 com./prop. 0.0257 deverbal 1.0000 deverbal 0.4993 deverbal 0.0525 deverbal 0.0105

vspomnit’ o~pro (A) vspomnit’ o~pro (B) vspomnit’ o~pro (C) vspomnit’ o~pro (D) animacy 0.7925 animacy 0.7611 animacy 0.4220 animacy 0.7480 ct./mass 1.0000 ct./mass 0.1945 ct./mass 1.0000 ct./mass 1.0000 modif. 0.7925 modif. 0.1102 modif. 0.1872 modif. 0.1154 com./prop. 1.0000 com./prop. 0.4238 com./prop. 0.7480 com./prop. 1.0000 deverbal 1.0000 deverbal 0.6120 deverbal 0.1124 deverbal 1.0000

pomnit’ o~pro (A) pomnit’ o~pro (B) pomnit’ o~pro (C) pomnit’ o~pro (D) animacy animacy 1.0000 animacy 1.0000 animacy 1.0000 ct./mass ct./mass 0.3843 ct./mass 0.5059 ct./mass 0.3533 modif. (N/A) modif. 0.7542 modif. 1.0000 modif. 0.3015 com./prop. com./prop. 1.0000 com./prop. 1.0000 com./prop. 1.0000 deverbal deverbal 0.5402 deverbal 1.0000 deverbal 1.0000

uznavat’ o~pro (A) uznavat’ o~pro (B) uznavat’ o~pro (C) uznavat’ o~pro (D) animacy animacy animacy animacy 0.0360 ct./mass ct./mass ct./mass ct./mass 0.2847 modif. (N/A) modif. (N/A) modif. (N/A) modif. 0.3517 com./prop. com./prop. com./prop. com./prop. 1.0000 deverbal deverbal deverbal deverbal 1.0000

uznat’ o~pro (A) uznat’ o~pro (B) uznat’ o~pro (C) uznat’ o~pro (D) animacy 0.6707 animacy 0.0419 animacy 0.0211 animacy 0.1455 ct./mass 1.0000 ct./mass 0.4915 ct./mass 1.0000 ct./mass 1.0000

145 In tests between o and nasčet for the factor group ‘Common/Proper Status of Noun’, all significant P- values show an association between nasčet and proper nouns.

206

modif. 1.0000 modif. 0.0470 modif. 0.1799 modif. 0.2668 com./prop. 1.0000 com./prop. 0.6707 com./prop. 0.0419 com./prop. 0.2092 deverbal 0.0840 deverbal 0.1028 deverbal 0.1042 deverbal 0.6707

uznat’ o~nasčet (A) o~pro (B) o~pro (C) o~pro (D) animacy animacy animacy animacy 1.0000 ct./mass ct./mass ct./mass ct./mass 0.4948 modif. (N/A) modif. (N/A) modif. (N/A) modif. 0.2769 com./prop. com./prop. com./prop. com./prop. 1.0000 deverbal deverbal deverbal deverbal 1.0000

napominat’ o~pro (A) napominat’ o~pro (B) napominat’ o~pro (C) napominat’ o~pro (D) animacy animacy animacy animacy 0.6894 ct./mass ct./mass ct./mass ct./mass 1.0000 modif. (N/A) modif. (N/A) modif. (N/A) modif. 0.2993 com./prop. com./prop. com./prop. com./prop. 1.0000 deverbal deverbal deverbal deverbal 0.6779

napomnit’ o~pro (A) napomnit’ o~pro (B) napomnit’ o~pro (C) napomnit’ o~pro (D) animacy animacy animacy animacy 1.0000 ct./mass ct./mass ct./mass ct./mass 0.6707 modif. (N/A) modif. (N/A) modif. (N/A) modif. 0.1060 com./prop. com./prop. com./prop. com./prop. 0.6120 deverbal deverbal deverbal deverbal 1.0000

Verbs of Hearing (55 tests)

slyšat’ o~pro (A) slyšat’ o~pro (B) slyšat’ o~pro (C) slyšat’ o~pro (D) animacy 0.4946 animacy 0.0009 animacy 0.7925 animacy 0.4296 ct./mass 1.0000 ct./mass 0.2373 ct./mass 1.0000 ct./mass 0.4238 modif. 0.7759 modif. 0.0089 modif. 0.0641 modif. 0.2949 com./prop. 1.0000 com./prop. 0.0303 com./prop. 0.5520 com./prop. 0.2326 deverbal 1.0000 deverbal 0.0122 deverbal 0.1028 deverbal 0.2373

uslyšat’ o~pro (A) uslyšat’ o~pro (B) uslyšat’ o~pro (C) uslyšat’ o~pro (D) animacy (N/A) animacy 0.6963 animacy 0.1093 animacy 1.0000

207

ct./mass ct./mass 0.6244 ct./mass 0.6345 ct./mass 1.0000 modif. modif. 0.5451 modif. 0.7543 modif. 0.7892 com./prop. com./prop. 0.6918 com./prop. 0.1647 com./prop. 0.7611 deverbal deverbal 0.1418 deverbal 0.0638 deverbal 0.2542

slyxat’ o~pro (A) slyxat’ o~pro (B) slyxat’ o~pro (C) slyxat’ o~pro (D) animacy 0.4117 animacy 0.0251 animacy 1.0000 animacy 0.7892 ct./mass 1.0000 ct./mass 1.0000 ct./mass 0.4993 ct./mass 1.0000 modif. 1.0000 modif. 0.0641 modif. 0.7865 modif. 0.2949 com./prop. 0.2675 com./prop. 0.0419 com./prop. 0.7720 com./prop. 0.3817 deverbal 0.3673 deverbal 0.2373 deverbal 0.1238 deverbal 0.0522

Verbs of Asking (80 tests)

sprašivat’ o~pro (A) sprašivat’ o~pro (B) sprašivat’ o~pro (C) sprašivat’ o~pro (D) animacy 0.0003 animacy 0.7710 animacy 0.1872 animacy 0.4296 ct./mass 0.6818 ct./mass 0.7065 ct./mass 0.6120 ct./mass 0.1028 modif. 0.5673 modif. 0.5520 modif. 1.0000 modif. 0.3334 com./prop. 0.0072 com./prop. 0.5321 com./prop. 0.5520 com./prop. 0.1806 deverbal 0.1203 deverbal 0.2373 deverbal 0.1945 deverbal 0.3533

sprašivat’ o~nasčet sprašivat’ o~nasčet sprašivat’ o~nasčet sprašivat’ o~nasčet

(A) (B) (C) (D) animacy animacy animacy animacy 0.2326 ct./mass ct./mass ct./mass ct./mass 0.6514 modif. (N/A) modif. (N/A) modif. (N/A) modif. 0.7004 com./prop. com./prop. com./prop. com./prop. 1.0000 deverbal deverbal deverbal deverbal 0.6514

sprosit’ o~pro (A) sprosit’ o~pro (B) sprosit’ o~pro (C) sprosit’ o~pro (D) animacy 0.7710 animacy 0.0596 animacy 0.0716 animacy 0.4118 ct./mass 0.7065 ct./mass 1.0000 ct./mass 0.3533 ct./mass 0.4716 modif. 0.5520 modif. 0.5675 modif. 0.1042 modif. 0.7306 com./prop. 0.5321 com./prop. 0.2326 com./prop. 0.1455 com./prop. 1.0000 deverbal 0.2373 deverbal 0.4238 deverbal 0.6120 deverbal 0.1124

sprosit’ o~nasčet (A) sprosit’ o~nasčet (B) sprosit’ o~nasčet (C) sprosit’ o~nasčet (D)

208

animacy animacy animacy 1.0000 animacy 0.5321 ct./mass ct./mass ct./mass 1.0000 ct./mass 0.1806 modif. (N/A) modif. (N/A) modif. 0.2368 modif. 1.0000 com./prop. com./prop. com./prop. 0.4238 com./prop. 1.0000 deverbal deverbal deverbal 1.0000 deverbal 0.6707

rassprašivat’ o ~pro rassprašivat’ o ~pro rassprašivat’ o ~pro rassprašivat’ o ~pro

(A) (B) (C) (D) animacy 0.5104 animacy 0.5321 animacy 0.0223 animacy 0.7892 ct./mass 0.7306 ct./mass 1.0000 ct./mass 1.0000 ct./mass 1.0000 modif. 0.5843 modif. 1.0000 modif. 0.7768 modif. 1.0000 com./prop. 1.0000 com./prop. 0.7306 com./prop. 0.4429 com./prop. 0.7306 deverbal 0.0776 deverbal 0.1945 deverbal 0.4444 deverbal 0.2373

Verbs of Writing (65 tests)

pisat’ o~pro (A) pisat’ o~pro (B) pisat’ o~pro (C) pisat’ o~pro (D) animacy 0.3266 animacy 0.0101 animacy 0.0292 animacy 0.1702 ct./mass 1.0000 ct./mass 0.7065 ct./mass 1.0000 ct./mass 0.1124 modif. 0.5602 modif. 1.0000 modif. 0.1702 modif. 0.4348 com./prop. 0.0124 com./prop. 0.2092 com./prop. 1.0000 com./prop. 0.7306 deverbal 1.0000 deverbal 0.3533 deverbal 1.0000 deverbal 0.6120

pisat’ o~po povodu pisat’ o~po povodu pisat’ o~po povodu pisat’ o~po povodu

(A) (B) (C) (D) animacy 0.0328146 animacy 0.3585 animacy animacy 0.1178 ct./mass 1.0000 ct./mass 0.2003 ct./mass ct./mass 0.1238 modif. 0.4130 modif. 0.7875 modif. (N/A) modif. 1.0000 com./prop. 1.0000 com./prop. 0.6749 com./prop. com./prop. 0.2995 deverbal 0.3819 deverbal 1.0000 deverbal deverbal 0.4512

napisat’ o~pro (A) napisat’ o~pro (B) napisat’ o~pro (C) napisat’ o~pro (D) animacy animacy 1.0000 animacy 0.2326 animacy 0.7892 ct./mass ct./mass 0.2403 ct./mass 1.0000 ct./mass 1.0000 (N/A) modif. modif. 1.0000 modif. 0.6058 modif. 0.3985 com./prop. com./prop. 0.3990 com./prop. 1.0000 com./prop. 0.1455

146 In tests between o and po povodu for the factor group ‘Animacy Status of Noun’, all significant P-values show an association between po povodu and inanimate nouns.

209

deverbal deverbal 0.6151 deverbal 0.0522 deverbal 0.2373

napisat’ o~po povodu napisat’ o~popovodu napisat’ o~popovodu napisat’ o~popovodu

(A) (B) (C) (D) animacy 0.0204 animacy 0.0183 animacy animacy 0.0335 ct./mass 0.3162 ct./mass 1.0000 ct./mass ct./mass 1.0000 modif. 0.2900 modif. 0.3227 modif. (N/A) modif. 0.5412 com./prop. 0.1267 com./prop. 0.0091147 com./prop. com./prop. 0.0338 deverbal 0.5511 deverbal 0.0883 deverbal deverbal 0.2331

Verbs of Reading (30 tests)

čitat’ o~pro (A) čitat’ o~pro (B) čitat’ o~pro (C) čitat’ o~pro (D) animacy 0.5882 animacy 0.0840 animacy 0.1799 animacy 0.7611 ct./mass 0.5920 ct./mass 1.0000 ct./mass 0.6120 ct./mass 1.0000 modif. 0.7740 modif. 0.2882 modif. 1.0000 modif. 0.4348 com./prop. 1.0000 com./prop. 0.0122 com./prop. 1.0000 com./prop. 1.0000 deverbal 0.3585 deverbal 0.0797 deverbal 0.6120 deverbal 0.6707

pročitat’ o~pro (A) pročitat’ o~pro (B) pročitat’ o~pro (C) pročitat’ o~pro (D) animacy animacy animacy 0.0419 animacy 0.0015 ct./mass ct./mass ct./mass 1.0000 ct./mass 1.0000 modif. (N/A) modif. (N/A) modif. 0.7158 modif. 0.4348 com./prop. com./prop. com./prop. 0.3608 com./prop. 0.7306 deverbal deverbal deverbal 0.5590 deverbal 0.1945

147 In tests between o and po povodu for the factor group ‘Common/Proper Status of Noun’, all significant P- values show an association between po povodu and common nouns.

210

2. Verb aspect

Verbs of Speech (170 tests)

govorit’~skazat’ govorit’~skazat’ govorit’~skazat’ govorit’~skazat’

o+LOC (A) o+LOC (B) o+LOC (C) o+LOC (D) animacy 1.0000 animacy 1.0000 animacy 1.0000 animacy 0.7480 ct./mass 0.5750 ct./mass 0.2542 ct./mass 1.0000 ct./mass 0.7065 modif. 0.2882 modif. 0.1102 modif. 1.0000 modif. 0.6292 com./prop. 0.3533 com./prop. 0.6120 com./prop. 1.0000 com./prop. 0.6120 deverbal 0.7480 deverbal 1.0000 deverbal 1.0000 deverbal 0.6707

govorit’~skazat’ govorit’~skazat’ govorit’~skazat’ govorit’~skazat’

pro (A) pro (B) pro (C) pro (D) animacy 0.1102 animacy 0.7948 animacy 0.2668 animacy 1.0000 ct./mass 0.6120 ct./mass 1.0000 ct./mass 0.6120 ct./mass 0.6120 modif. 0.6042 modif. 0.0539 modif. 0.2882 modif. 0.0292148 com./prop. 0.3604 com./prop. 1.0000 com./prop. 1.0000 com./prop. 0.7480 deverbal 1.0000 deverbal 1.0000 deverbal 1.0000 deverbal 1.0000

govorit’~skazat’ govorit’~skazat’ govorit’~skazat’ govorit’~skazat’

nasčet (A) nasčet (B) nasčet (C) nasčet (D) animacy animacy 0.0284149 animacy 0.0985 animacy 0.5467 ct./mass ct./mass 1.0000 ct./mass 0.6466 ct./mass 0.4940 modif. (N/A) modif. 1.0000 modif. 0.4642 modif. 0.4895 com./prop. com./prop. 1.0000 com./prop. 0.6094 com./prop. 1.0000 deverbal deverbal 0.4483 deverbal 0.2065 deverbal 1.0000

govorit’~skazat’ govorit’~skazat’ govorit’~skazat’ govorit’~skazat’

po povodu (A) po povodu (B) po povodu (C) po povodu (D) animacy animacy animacy 0.5688 animacy 1.0000 ct./mass ct./mass ct./mass 1.0000 ct./mass 0.2168 modif. (N/A) modif. (N/A) modif. 0.4270 modif. 0.2527 com./prop. com./prop. com./prop. 0.6309 com./prop. 0.6804 deverbal deverbal deverbal 0.4633 deverbal 0.2489

govorit’~pogovorit’ govorit’~pogovorit’ govorit’~pogovorit’ govorit’~pogovorit’

148 This result shows an association between o+LOC with the imp. form govorit’ and modified nouns. 149 This result shows an association between nasčet with the pf. form skazat’ and animate nouns.

211

o+LOC (A) o+LOC (B) o+LOC (C) o+LOC (D) animacy 0.4238 animacy 1.0000 animacy 1.0000 animacy 0.5321 ct./mass 0.3817 ct./mass 1.0000 ct./mass 1.0000 ct./mass 1.0000 modif. 0.1154 modif. 0.7787 modif. 1.0000 modif. 0.0876 com./prop. 0.3533 com./prop. 1.0000 com./prop. 1.0000 com./prop. 0.1945 deverbal 0.7065 deverbal 0.3533 deverbal 0.4238 deverbal 1.0000

govorit’~pogovorit’ govorit’~pogovorit’ govorit’~pogovorit’ govorit’~pogovorit’

pro (A) pro (B) pro (C) pro (D) animacy animacy animacy 1.0000 animacy 0.3985 ct./mass ct./mass ct./mass 1.0000 ct./mass 0.3533 modif. (N/A) modif. (N/A) modif. 0.5636 modif. 0.1872 com./prop. com./prop. com./prop. 0.1899 com./prop. 1.0000 deverbal deverbal deverbal 1.0000 deverbal 1.0000

govorit’~pogovorit’ govorit’~pogovorit’ govorit’~pogovorit’ govorit’~pogovorit’

nasčet (A) nascet (B) nasčet (C) nasčet (D) animacy animacy 0.0730 animacy 1.0000 animacy 0.0436150 ct./mass ct./mass 1.0000 ct./mass 0.6599 ct./mass 0.6948 modif. (N/A) modif. 1.0000 modif. 0.2075 modif. 1.0000 com./prop. com./prop. 1.0000 com./prop. 1.0000 com./prop. 0.3416 deverbal deverbal 1.0000 deverbal 1.0000 deverbal 1.0000

rasskazyvat’~ rasskazyvat’~ rasskazyvat’~ rasskazyvat’~

rasskazat’ o+LOC (A) rasskazat’ o+LOC (B) rasskazat’ o+LOC (C) rasskazat’ o+LOC (D) animacy 0.4238 animacy 1.0000 animacy 0.2990 animacy 0.6707 ct./mass 0.4716 ct./mass 1.0000 ct./mass 0.4716 ct./mass 1.0000 modif. 0.2668 modif. 0.7961 modif. 0.7961 modif. 0.7948 com./prop. 1.0000 com./prop. 1.0000 com./prop. 1.0000 com./prop. 1.0000 deverbal 1.0000 deverbal 0.4716 deverbal 1.0000 deverbal 0.2092

rasskazyvat’~ rasskazyvat’~ rasskazyvat’~ rasskazyvat’~

rasskazat’ pro (A) rasskazat’ pro (B) rasskazat’ (C) rasskazat’ (D) animacy 1.0000 animacy 0.7925 animacy 0.6707 animacy 1.0000 ct./mass 0.4915 ct./mass 1.0000 ct./mass 1.0000 ct./mass 0.6707 modif. 0.0379151 modif. 0.7892 modif. 0.4296 modif. 1.0000 com./prop. 1.0000 com./prop. 1.0000 com./prop. 1.0000 com./prop. 1.0000 deverbal 1.0000 deverbal 1.0000 deverbal 1.0000 deverbal 1.0000

150 This result shows an association between the imp. form govorit’ and animate nouns. 151 This result indicates an association between the pf. form rasskazat’ and modified nouns.

212

dokladyvat’~doložit’ dokladyvat’~doložit’ dokladyvat’~doložit’ dokladyvat’~doložit’

o+LOC (A) o+LOC (B) o+LOC (C) o+LOC (D) animacy 1.0000 animacy 1.0000 animacy 1.0000 animacy 1.0000 ct./mass 0.1269 ct./mass 1.0000 ct./mass 0.7065 ct./mass 1.0000 modif. 0.7875 modif. 0.6042 modif. 1.0000 modif. 1.0000 com./prop. 1.0000 com./prop. 1.0000 com./prop. 1.0000 com./prop. 1.0000 deverbal 1.0000 deverbal 0.1581 deverbal 0.7787 deverbal 0.5796

Verbs of Cognition (165 tests)

dumat’~podumat’ dumat’~podumat’ dumat’~podumat’ dumat’~podumat’

o+LOC (A) o+LOC (B) o+LOC (C) o+LOC (D) animacy 0.5520 animacy 0.5321 animacy 0.1042 animacy 1.0000 ct./mass 0.6707 ct./mass 0.0522 ct./mass 0.4238 ct./mass 0.1945 modif. 0.1702 modif. 0.7892 modif. 0.2789 modif. 0.2516 com./prop. 1.0000 com./prop. 1.0000 com./prop. 0.1124 com./prop. 1.0000 deverbal 0.7306 deverbal 0.4238 deverbal 0.2542 deverbal 0.3334

dumat’~podumat’ dumat’~podumat’ dumat’~podumat’ dumat’~podumat’

pro (A) pro (B) pro (C) pro (D) animacy 0.2307 animacy 0.0005152 animacy 0.0025 animacy 0.7892 ct./mass 0.3059 ct./mass 0.3835 ct./mass 0.1167 ct./mass 0.3533 modif. 1.0000 modif. 1.0000 modif. 1.0000 modif. 0.7787 com./prop. 1.0000 com./prop. 0.0055153 com./prop. 0.0018 com./prop. 1.0000 deverbal 1.0000 deverbal 1.0000 deverbal 1.0000 deverbal 0.1945

napominat’~ napominat’~ napominat’~ napominat’~

napomnit’ o+LOC (A) napomnit’ o+LOC (B) napomnit’ o+LOC (C) napomnit’ o+LOC (D) animacy 1.0000 animacy 0.0522 animacy 0.6120 animacy 1.0000 ct./mass 1.0000 ct./mass 0.6707 ct./mass 0.5062 ct./mass 1.0000 modif. 1.0000 modif. 0.4389 modif. 0.1954 modif. 0.4296 com./prop. 0.3533 com./prop. 0.6120 com./prop. 0.4915 com./prop. 0.6120 deverbal 0.1455 deverbal 1.0000 deverbal 0.1806 deverbal 0.6707

152 The significant results in Pds. B and C between dumat’~podumat’ with o+LOC for this factor group show an association between the pf. form podumat’ and animate nouns. 153 The significant results in Pds. B and C between dumat’~podumat’ with o+LOC for this factor group show an association between the pf. form podumat’ and proper nouns.

213

napominat’~ napominat’~ napominat’~ napominat’~

napomnit’ pro (A) napomnit’ pro (B) napomnit’ pro (C) napomnit’ pro (D) animacy animacy animacy animacy 0.6894 ct./mass ct./mass ct./mass ct./mass 0.6499 modif. (N/A) modif. (N/A) modif. (N/A) modif. 1.0000 com./prop. com./prop. com./prop. com./prop. 1.0000 deverbal deverbal deverbal deverbal 1.0000

zabyvat’~zabyt’ zabyvat’~zabyt’ zabyvat’~zabyt’ zabyvat’~zabyt’

o+LOC (A) o+LOC (B) o+LOC (C) o+LOC (D) animacy 1.0000 animacy 0.2990 animacy 0.4716 animacy 0.2542 ct./mass 0.7065 ct./mass 0.4716 ct./mass 1.0000 ct./mass 0.4238 modif. 0.1969 modif. 1.0000 modif. 0.6042 modif. 0.7961 com./prop. 1.0000 com./prop. 1.0000 com./prop. 1.0000 com./prop. 1.0000 deverbal 0.5062 deverbal 1.0000 deverbal 1.0000 deverbal 1.0000

zabyvat’~zabyt’ zabyvat’~zabyt’ zabyvat’~zabyt’ zabyvat’~zabyt’

pro (A) pro (B) pro (C) pro (D) animacy animacy 0.5104 animacy 1.0000 animacy 0.0840 ct./mass ct./mass 0.2453 ct./mass 1.0000 ct./mass 1.0000 modif. (N/A) modif. 0.4075 modif. 1.0000 modif. 1.0000 com./prop. com./prop. 0.0816 com./prop. 0.2542 com./prop. 1.0000 deverbal deverbal 1.0000 deverbal 1.0000 deverbal 0.3533

uznavat’~uznat’ uznavat’~uznat’ uznavat’~uznat’ uznavat’~uznat’

o+LOC (A) o+LOC (B) o+LOC (C) o+LOC (D) animacy 1.0000 animacy 0.4328 animacy 1.0000 animacy 1.0000 ct./mass 0.7065 ct./mass 1.0000 ct./mass 0.1124 ct./mass 0.3533 modif. 0.7948 modif. 0.0470154 modif. 1.0000 modif. 0.6042 com./prop. 1.0000 com./prop. 1.0000 com./prop. 1.0000 com./prop. 1.0000 deverbal 0.1581 deverbal 1.0000 deverbal 0.7847 deverbal 1.0000

uznavat’~uznat’ uznavat’~uznat’ uznavat’~uznat’ uznavat’~uznat’

pro (A) pro (B) pro (C) pro (D) animacy animacy animacy animacy 1.0000 (N/A) (N/A) (N/A) ct./mass ct./mass ct./mass ct./mass 1.0000

154 This result shows an association between the imp. form uznavat’ and modified nouns.

214

modif. modif. modif. modif. 0.5131 com./prop. com./prop. com./prop. com./prop. 0.2847 deverbal deverbal deverbal deverbal 0.6244

vspominat’~ vspominat’~ vspominat’~ vspominat’~

vspomnit’ o+LOC (A) vspomnit’ o+LOC (B) vspomnit’ o+LOC (C) vspomnit’ o+LOC (D) animacy 0.0693 animacy 1.0000 animacy 0.7710 animacy 0.4238 ct./mass 1.0000 ct./mass 0.4238 ct./mass 1.0000 ct./mass 0.3533 modif. 0.1205 modif. 0.0641 modif. 1.0000 modif. 0.0858 com./prop. 0.7065 com./prop. 1.0000 com./prop. 0.4238 com./prop. 0.6120 deverbal 1.0000 deverbal 1.0000 deverbal 1.0000 deverbal 0.1455

vspominat’~ vspominat’~ vspominat’~ vspominat’~

vspomnit’ pro (A) vspomnit’ pro (B) vspomnit’ pro (C) vspomnit’ pro (D) animacy 1.0000 animacy 0.0786 animacy 1.0000 animacy 0.1702 ct./mass 0.5213 ct./mass 1.0000 ct./mass 1.0000 ct./mass 0.1945 modif. 0.5544 modif. 0.7865 modif. 0.5544 modif. 1.0000 com./prop. 0.2654 com./prop. 0.0308155 com./prop. 1.0000 com./prop. 0.0797 deverbal 1.0000 deverbal 0.5855 deverbal 1.0000 deverbal 0.2373

Verbs of Hearing (35 tests)

slyšat’~uslyšat’ slyšat’~uslyšat’ slyšat’~uslyšat’ slyšat’~uslyšat’

o+LOC (A) o+LOC (B) o+LOC (C) o+LOC (D) animacy 0.2185 animacy 1.0000 animacy 0.1432 animacy 1.0000 ct./mass 1.0000 ct./mass 1.0000 ct./mass 1.0000 ct./mass 1.0000 modif. 0.1864 modif. 0.0191156 modif. 1.0000 modif. 0.6042 com./prop. 0.5399 com./prop. 0.4716 com./prop. 0.7306 com./prop. 1.0000 deverbal 0.5191 deverbal 0.3604 deverbal 0.7611 deverbal 0.4716

slyšat’~uslyšat’ slyšat’~uslyšat’ slyšat’~uslyšat’ slyšat’~uslyšat’

pro (A) pro (B) pro (C) pro (D) animacy animacy 0.1305 animacy 0.7779 animacy 0.2882 ct./mass (N/A) ct./mass 0.1356 ct./mass 1.0000 ct./mass 0.0522 modif. modif. 0.5216 modif. 0.0246157 modif. 1.0000

155 This result shows an association between the pf. form vspomnit’ and proper nouns. 156 This result shows an association between the imp. form slyšat’ and modified nouns. 157 This result shows an association between the imp. form slyšat’ and modified nouns.

215

com./prop. com./prop. 0.0922 com./prop. 1.0000 com./prop. 0.7787 deverbal deverbal 1.0000 deverbal 1.0000 deverbal 0.4915

Verbs of Asking (110 tests)

sprašivat’~sprosit’ sprašivat’~sprosit’ sprašivat’~sprosit’ sprašivat’~sprosit’

o+LOC (A) o+LOC (B) o+LOC (C) o+LOC (D) animacy 0.3334 animacy 1.0000 animacy 0.2092 animacy 0.7787 ct./mass 1.0000 ct./mass 0.7065 ct./mass 1.0000 ct./mass 0.4716 modif. 0.2668 modif. 1.0000 modif. 0.5520 modif. 0.7611 com./prop. 1.0000 com./prop. 1.0000 com./prop. 1.0000 com./prop. 0.7065 deverbal 0.6707 deverbal 0.7065 deverbal 0.7065 deverbal 1.0000

sprašivat’~sprosit’ sprašivat’~sprosit’ sprašivat’~sprosit’ sprašivat’~sprosit’

pro (A) pro (B) pro (C) pro (D) animacy 0.0392158 animacy 0.1872 animacy 0.4328 animacy 0.7948 ct./mass 1.0000 ct./mass 1.0000 ct./mass 1.0000 ct./mass 0.1028 modif. 0.3715 modif. 1.0000 modif. 0.4716 modif. 1.0000 com./prop. 0.3412 com./prop. 0.7787 com./prop. 0.7710 com./prop. 0.5321 deverbal 0.4894 deverbal 0.4915 deverbal 1.0000 deverbal 1.0000

sprašivat’~sprosit’ sprašivat’~sprosit’ sprašivat’~sprosit’ sprašivat’~sprosit’

nasčet (A) nasčet (B) nasčet (C) nasčet (D) animacy animacy animacy animacy 1.0000 ct./mass ct./mass ct./mass ct./mass 0.3994 modif. (N/A) modif. (N/A) modif. (N/A) modif. 1.0000 com./prop. com./prop. com./prop. com./prop. 1.0000 deverbal deverbal deverbal deverbal 0.3003

rassprašivat’~ rassprašivat’~ rassprašivat’~ rassprašivat’~ rassprosit’ o+LOC rassprosit’ o+LOC rassprosit’ o+LOC rassprosit’ o+LOC (A) (B) (C) (D) animacy 1.0000 animacy 1.0000 animacy 0.2990 animacy 1.0000 ct./mass 1.0000 ct./mass 1.0000 ct./mass 1.0000 ct./mass 1.0000 modif. 0.5646 modif. 0.2092 modif. 0.5959 modif. 1.0000 com./prop. 1.0000 com./prop. 1.0000 com./prop. 0.7306 com./prop. 0.7306 deverbal 0.3168 deverbal 1.0000 deverbal 0.2373 deverbal 1.0000

158 This result shows an association between the imp. form sprašivat’ and animate nouns.

216

rassprašivat’~ rassprašivat’~ rassprašivat’~ rassprašivat’~

rassprosit’ pro (A) rassprosit’ pro (B) rassprosit’ pro (C) rassprosit’ pro (D) animacy animacy animacy animacy 1.0000 ct./mass ct./mass ct./mass ct./mass 1.0000 modif. (N/A) modif. (N/A) modif. (N/A) modif. 0.6806 com./prop. com./prop. com./prop. com./prop. 1.0000 deverbal deverbal deverbal deverbal 0.3023

spravljat’sja~ spravljat’sja~ spravljat’sja~ spravljat’sja~ spravit’sja o+LOC (A) spravit’sja o+LOC (B) spravit’sja o+LOC (C) spravit’sja o+LOC (D) animacy 1.0000 animacy 0.7480 animacy 1.0000 animacy 0.4716 ct./mass 0.0255159 ct./mass 1.0000 ct./mass 0.5061 ct./mass 1.0000 modif. 0.0139160 modif. 0.2326 modif. 0.0280161 modif. 0.5589 com./prop. 0.6851 com./prop. 1.0000 com./prop. 1.0000 com./prop. 1.0000 deverbal 0.7766 deverbal 0.0509 deverbal 1.0000 deverbal 1.0000

prosit’~poprosit’ prosit’~poprosit’ prosit’~poprosit’ prosit’~poprosit’

o+LOC (A) o+LOC (B) o+LOC (C) o+LOC (D) animacy 1.0000 animacy 1.0000 animacy 0.2128 animacy 1.0000 ct./mass 0.6943 ct./mass 1.0000 ct./mass 0.3319 ct./mass 1.0000 modif. 0.2170 modif. 0.0319162 modif. 0.3238 modif. 1.0000 com./prop. 1.0000 com./prop. 1.0000 com./prop. 0.0864 com./prop. 1.0000 deverbal 0.4993 deverbal 0.5358 deverbal 0.7466 deverbal 1.0000

Verbs of Writing (50 tests)

pisat’~napisat’ pisat’~napisat’ pisat’~napisat’ pisat’~napisat’

o+LOC (A) o+LOC (B) o+LOC (C) o+LOC (D) animacy 1.0000 animacy 0.0716 animacy 1.0000 animacy 0.5675 ct./mass 1.0000 ct./mass 0.7065 ct./mass 0.6707 ct./mass 0.6707 modif. 0.4296 modif. 1.0000 modif. 0.6058 modif. 0.4348 com./prop. 0.3533 com./prop. 0.0391163 com./prop. 1.0000 com./prop. 0.5062

159 This result shows an association between the pf. form spravit’sja and mass nouns. 160 This result shows an association between the imp. form spravljat’sja and modified nouns. 161 This result shows an association between the imp. form spravljat’sja and unmodified nouns. 162 This result shows an association between the imp. form prosit’ and unmodified nouns. 163 This result shows an association between the pf. form napisat’ and proper nouns.

217

deverbal 0.4238 deverbal 0.4161 deverbal 0.4238 deverbal 1.0000

pisat’~napisat’ pisat’~napisat’ pisat’~napisat’ pisat’~napisat’

pro (A) pro (B) pro (C) pro (D) animacy animacy 0.5957 animacy 0.2949 animacy 1.0000 ct./mass ct./mass 1.0000 ct./mass 0.4716 ct./mass 0.4238 modif. (N/A) modif. 1.0000 modif. 0.1702 modif. 0.3985 com./prop. com./prop. 0.3533 com./prop. 0.4238 com./prop. 0.7306 deverbal deverbal 1.0000 deverbal 1.0000 deverbal 1.0000

pisat’~napisat’ pisat’~napisat’ pisat’~napisat’ pisat’~napisat’

po povodu (A) po povodu (B) po povodu (C) po povodu (D) animacy 1.0000 animacy 1.0000 animacy animacy 1.0000 ct./mass 0.3041 ct./mass 0.5377 ct./mass ct./mass 0.4524 modif. 1.0000 modif. 0.7281 modif. (N/A) modif. 0.3532 com./prop. 0.3333 com./prop. 1.0000 com./prop. com./prop. 0.0244164 deverbal 0.2755 deverbal 0.2319 deverbal deverbal 0.7075

Verbs of Reading (30 tests)

čitat’~pročitat’ čitat’~pročitat’ čitat’~pročitat’ čitat’~pročitat’

o+LOC (A) o+LOC (B) o+LOC (C) o+LOC (D) animacy 0.1956 animacy 0.7065 animacy 0.3334 animacy 0.4716 ct./mass 0.5473 ct./mass 0.4238 ct./mass 0.3533 ct./mass 1.0000 modif. 1.0000 modif. 0.1799 modif. 0.7948 modif. 1.0000 com./prop. 0.1254 com./prop. 1.0000 com./prop. 1.0000 com./prop. 0.6707 deverbal 1.0000 deverbal 0.7611 deverbal 1.0000 deverbal 1.0000

čitat’~pročitat’ čitat’~pročitat’ čitat’~pročitat’ čitat’~pročitat’

pro (A) pro (B) pro (C) pro (D) animacy animacy animacy 1.0000 animacy 0.1102 ct./mass ct./mass ct./mass 0.3533 ct./mass 1.0000 modif. (N/A) modif. (N/A) modif. 0.4783 modif. 0.0698 com./prop. com./prop. com./prop. 1.0000 com./prop. 0.6707 deverbal deverbal deverbal 1.0000 deverbal 1.0000

164 This result shows an association between the pf. form napisat’ and common nouns.

218

Appendix B: Tallies of Distribution of Noun Features

Codes:

Animacy status: Count/mass status: Modification status: i: inanimate noun c: count noun u: unmodified noun a: animate noun m: mass noun #: modified noun

Common/proper status: Deverbal status: o: common noun n: non-deverbal noun p: proper noun d: deverbal noun

Verbs of Speech

boltat’ boltat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 29 c: 24 u: 30 o: 30 n: 29 a: 1 m: 6 #: 0 p: 0 d: 1

boltat’ + pro (14 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 11 c: 11 u: 9 o: 11 n: 13 a: 3 m: 3 #: 5 p: 3 d: 1

dokladyvat’ dokladyvat’ + o (26 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status

219

i: 22 c: 20 u: 14 o: 24 n: 20 a: 4 m: 6 #: 12 p: 2 d: 6

dokladyvat’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 28 c: 26 u: 18 o: 29 n: 24 a: 2 m: 4 #: 12 p: 1 d: 6

dokladyvat’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 29 c: 27 u: 16 o: 30 n: 22 a: 1 m: 3 #: 14 p: 0 d: 8

dokladyvat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 27 c: 26 u: 18 o: 29 n: 22 a: 3 m: 4 #: 12 p: 1 d: 8

doložit’ doložit’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 26 c: 28 u: 18 o: 28 n: 22 a: 4 m: 2 #: 12 p: 2 d: 8

doložit’ + o ( tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 28 c: 25 u: 15 o: 29 n: 18 a: 2 m: 5 #: 15 p: 1 d: 12

doložit’ + o ( tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 30 c: 25 u: 16 o: 30 n: 20 a: 0 m: 5 #: 14 p: 0 d: 10

doložit’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 27 c: 26 u: 17 o: 29 n: 19

220

a: 3 m: 4 #: 13 p: 1 d: 11

doložit’ + pro (12 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 6 c: 11 u: 8 o: 9 n: 12 a: 6 m: 1 #: 4 p: 3 d: 0

govorit’ govorit’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 20 u: 21 o: 26 n: 25 a: 5 m: 10 #: 9 p: 4 d: 5

govorit’ + pro (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 22 c: 27 u: 15 o: 25 n: 29 a: 8 m: 3 #: 15 p: 5 d: 1

govorit’ + po povodu (16 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 13 c: 14 u: 6 o: 14 n: 12 a: 3 m: 2 #: 10 p: 2 d: 4

govorit’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 23 c: 24 u: 22 o: 27 n: 26 a: 7 m: 6 #: 8 p: 3 d: 4

govorit’ + pro (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 18 c: 29 u: 16 o: 24 n: 29 a: 12 m: 1 #: 4 p: 6 d: 1

govorit’ + nasčet (13 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status

221

i: 12 c: 13 u: 9 o: 11 n: 12 a: 1 m: 0 #: 4 p: 2 d: 1

govorit’ + po povodu (24 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 19 c: 23 u: 11 o: 20 n: 19 a: 5 m: 1 #: 13 p: 4 d: 5

govorit’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 24 c: 28 u: 16 o: 27 n: 25 a: 6 m: 2 #: 14 p: 3 d: 5

govorit’ + pro (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 23 c: 27 u: 16 o: 28 n: 29 a: 7 m: 3 #: 14 p: 2 d: 1

govorit’ + nasčet (18 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 17 c: 15 u: 15 o: 17 n: 16 a: 1 m: 3 #: 3 p: 1 d: 2

govorit’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 25 u: 19 o: 29 n: 28 a: 5 m: 5 #: 21 p: 1 d: 2

govorit’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 19 c: 29 u: 15 o: 25 n: 29 a: 11 m: 1 #: 15 p: 5 d: 1

govorit’ + nasčet (20 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 13 c: 15 u: 17 o: 16 n: 18 a: 7 m: 5 #: 3 p: 4 d: 2

222

govorit’ + po povodu (21 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 19 c: 20 u: 14 o: 18 n: 16 a: 2 m: 1 #: 7 p: 3 d: 5

molčat’ molčat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 29 u: 16 o: 25 n: 27 a: 5 m: 1 #: 14 p: 5 d: 3

molčat’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 20 c: 24 u: 24 o: 27 n: 28 a: 10 m: 6 #: 6 p: 3 d: 2

pogovorit’ pogovorit’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 28 c: 24 u: 14 o: 29 n: 27 a: 2 m: 6 #: 16 p: 1 d: 3

pogovorit’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 23 c: 27 u: 20 o: 27 n: 29 a: 7 m: 3 #: 10 p: 3 d: 1

pogovorit’ + nasčet (13 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 7 c: 13 u: 10 o: 11 n: 12 a: 6 m: 0 #: 3 p: 2 d: 1

pogovorit’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 24 c: 28 u: 17 o: 27 n: 28 a: 6 m: 2 #: 13 p: 3 d: 2

223

pogovorit’ + pro (19 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 14 c: 17 u: 12 o: 15 n: 19 a: 5 m: 2 #: 7 p: 4 d: 0

pogovorit’ + nasčet (10 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 10 c: 10 u: 6 o: 9 n: 9 a: 0 m: 0 #: 4 p: 1 d: 1

pogovorit’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 22 c: 26 u: 26 o: 25 n: 27 a: 8 m: 4 #: 4 p: 5 d: 3

pogovorit’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 23 c: 26 u: 21 o: 24 n: 28 a: 7 m: 4 #: 9 p: 6 d: 2

pogovorit’ + nasčet (20 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 19 c: 17 u: 16 o: 19 n: 19 a: 1 m: 3 #: 4 p: 1 d: 1

rasskazat’ rasskazat’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 24 u: 7 o: 29 n: 21 a: 5 m: 6 #: 23 p: 1 d: 9

rasskazat’ + pro (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 16 c: 30 u: 10 o: 22 n: 28 a: 14 m: 0 #: 20 p: 8 d: 2

224

rasskazat’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 26 c: 29 u: 15 o: 29 n: 24 a: 4 m: 1 #: 15 p: 1 d: 6

rasskazat’ + pro (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 19 c: 27 u: 20 o: 24 n: 30 a: 11 m: 3 #: 10 p: 6 d: 0

rasskazat’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 27 c: 27 u: 13 o: 27 n: 22 a: 3 m: 3 #: 17 p: 3 d: 8

rasskazat’ + pro (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 18 c: 29 u: 20 o: 27 n: 30 a: 12 m: 1 #: 10 p: 3 d: 0

rasskazat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 28 c: 27 u: 14 o: 27 n: 26 a: 2 m: 3 #: 16 p: 3 d: 4

rasskazat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 20 c: 26 u: 19 o: 25 n: 30 a: 10 m: 4 #: 11 p: 5 d: 0

rasskazyvat’ rasskazyvat’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 28 c: 27 u: 12 o: 28 n: 22 a: 2 m: 3 #: 18 p: 2 d: 8

225

rasskazyvat’ + pro (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 15 c: 28 u: 19 o: 22 n: 28 a: 15 m: 2 #: 11 p: 8 d: 2

rasskazyvat’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 28 u: 13 o: 29 n: 27 a: 5 m: 2 #: 17 p: 1 d: 3

rasskazyvat’ + pro (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 17 c: 27 u: 18 o: 23 n: 30 a: 13 m: 3 #: 12 p: 7 d: 0

rasskazyvat’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 23 c: 24 u: 15 o: 27 n: 25 a: 7 m: 6 #: 15 p: 3 d: 5

rasskazyvat’ + pro (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 22 c: 29 u: 16 o: 27 n: 29 a: 8 m: 1 #: 14 p: 3 d: 1

rasskazyvat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 26 c: 28 u: 12 o: 28 n: 21 a: 4 m: 2 #: 18 p: 2 d: 9

rasskazyvat’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 21 c: 28 u: 18 o: 26 n: 30 a: 9 m: 2 #: 12 p: 4 d: 0

skazat’

226

skazat’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 24 c: 27 u: 16 o: 29 n: 23 a: 6 m: 3 #: 14 p: 1 d: 7

skazat’ + pro (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 15 c: 28 u: 18 o: 21 n: 29 a: 15 m: 2 #: 12 p: 9 d: 1

skazat’ + nasčet (12 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 10 c: 11 u: 11 o: 11 n: 12 a: 2 m: 1 #: 1 p: 1 d: 0

skazat’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 24 c: 28 u: 15 o: 29 n: 25 a: 6 m: 2 #: 15 p: 1 d: 5

skazat’ + pro (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 16 c: 29 u: 24 o: 24 n: 30 a: 14 m: 1 #: 6 p: 6 d: 0

skazat’ + nasčet (16 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 10 c: 15 u: 11 o: 13 n: 16 a: 6 m: 1 #: 5 p: 3 d: 0

skazat’ + po povodu (15 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 13 c: 14 u: 9 o: 14 n: 10 a: 2 m: 1 #: 6 p: 1 d: 5

skazat’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 24 c: 29 u: 15 o: 27 n: 26

227

a: 6 m: 1 #: 15 p: 3 d: 4

skazat’ + pro (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 18 c: 29 u: 21 o: 23 n: 30 a: 12 m: 1 #: 9 p: 7 d: 0

skazat’ + nasčet (21 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 15 c: 19 u: 15 o: 18 n: 21 a: 6 m: 2 #: 6 p: 3 d: 0

skazat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 23 c: 27 u: 12 o: 27 n: 26 a: 7 m: 3 #: 18 p: 3 d: 4

skazat’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 18 c: 27 u: 24 o: 23 n: 29 a: 12 m: 3 #: 6 p: 7 d: 1

skazat’ + nasčet (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 22 c: 25 u: 22 o: 24 n: 27 a: 8 m: 5 #: 8 p: 6 d: 3

skazat’ + po povodu (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 27 c: 24 u: 14 o: 27 n: 27 a: 3 m: 6 #: 16 p: 3 d: 3

tolkovat’ tolkovat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 23 c: 24 u: 17 o: 27 n: 24 a: 7 m: 6 #: 13 p: 3 d: 6

228

tolkovat’ + pro (16 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 12 c: 14 u: 11 o: 14 n: 15 a: 4 m: 2 #: 5 p: 2 d: 1

Verbs of Writing napisat’ napisat’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 24 c: 28 u: 20 o: 26 n: 25 a: 6 m: 2 #: 10 p: 4 d: 5

napisat’ + po povodu (11 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 11 c: 8 u: 4 o: 10 n: 11 a: 0 m: 3 #: 7 p: 1 d: 0

napisat’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 19 c: 27 u: 18 o: 18 n: 27 a: 11 m: 3 #: 12 p: 12 d: 3

napisat’ + pro (26 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 16 c: 26 u: 16 o: 19 n: 25 a: 10 m: 0 #: 10 p: 7 d: 1

napisat’ + po povodu (12 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 12 c: 11 u: 5 o: 12 n: 8 a: 0 m: 1 #: 7 p: 0 d: 4

napisat’ + o (30 tokens) – Pd. C

229

anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 28 u: 14 o: 26 n: 25 a: 5 m: 2 #: 16 p: 4 d: 5

napisat’ + pro (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 20 c: 27 u: 17 o: 26 n: 30 a: 10 m: 3 #: 13 p: 4 d: 0

napisat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 20 c: 28 u: 19 o: 23 n: 27 a: 10 m: 2 #: 11 p: 7 d: 3

napisat’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 18 c: 28 u: 23 o: 28 n: 30 a: 12 m: 2 #: 7 p: 2 d: 0

napisat’ + po povodu (19 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 18 c: 18 u: 13 o: 19 n: 14 a: 1 m: 1 #: 6 p: 0 d: 5

pisat’ pisat’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 24 c: 27 u: 16 o: 29 n: 28 a: 6 m: 3 #: 14 p: 1 d: 2

pisat’ + pro (20 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 13 c: 17 u: 13 o: 14 n: 19 a: 7 m: 3 #: 7 p: 6 d: 1

pisat’ + po povodu (22 tokens) – Pd. A

230

anim. status count/mass modific. comm./prop. deverb. status i: 22 c: 20 u: 9 o: 22 n: 18 a: 0 m: 2 #: 13 p: 0 d: 4

pisat’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 26 c: 25 u: 18 o: 26 n: 26 a: 4 m: 5 #: 12 p: 4 d: 4

pisat’ + pro (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 16 c: 29 u: 19 o: 21 n: 29 a: 4 m: 1 #: 11 p: 9 d: 1

pisat’ + po povodu (26 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 25 u: 14 o: 24 n: 22 a: 1 m: 1 #: 12 p: 2 d: 4 pisat’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 24 c: 26 u: 17 o: 25 n: 28 a: 6 m: 4 #: 13 p: 5 d: 2

pisat’ + pro (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 15 c: 24 u: 23 o: 25 n: 29 a: 15 m: 6 #: 7 p: 5 d: 1

pisat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 23 c: 26 u: 15 o: 26 n: 27 a: 7 m: 4 #: 15 p: 4 d: 3

pisat’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 17 c: 25 u: 19 o: 24 n: 29

231

a: 13 m: 5 #: 11 p: 6 d: 1

pisat’ + po povodu (23 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 22 c: 23 u: 12 o: 17 n: 19 a: 1 m: 0 #: 11 p: 6 d: 4

Verbs of Cognition dumat’ dumat’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 21 c: 26 u: 23 o: 24 n: 24 a: 9 m: 4 #: 7 p: 6 d: 6 dumat’ + pro (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 11 c: 25 u: 23 o: 23 n: 30 a: 19 m: 5 #: 7 p: 7 d: 0

dumat’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 25 u: 18 o: 26 n: 28 a: 5 m: 5 #: 12 p: 4 d: 2

dumat’ + pro (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 22 c: 25 u: 21 o: 27 n: 30 a: 8 m: 5 #: 9 p: 3 d: 0

dumat’ + nasčet (17 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 14 c: 16 u: 13 o: 16 n: 14 a: 3 m: 1 #: 4 p: 1 d: 3

232

dumat’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 21 c: 28 u: 22 o: 26 n: 28 a: 9 m: 2 #: 8 p: 4 d: 2

dumat’ + pro (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 19 c: 24 u: 22 o: 27 n: 30 a: 11 m: 6 #: 8 p: 3 d: 0

dumat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 23 u: 24 o: 29 n: 26 a: 5 m: 7 #: 6 p: 1 d: 4

dumat’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 20 c: 29 u: 22 o: 25 n: 25 a: 10 m: 1 #: 8 p: 5 d: 5

napominat’ napominat’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 26 c: 25 u: 15 o: 29 n: 28 a: 4 m: 5 #: 15 p: 1 d: 2

napominat’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 30 c: 26 u: 13 o: 29 n: 25 a: 0 m: 4 #: 17 p: 1 d: 5

napominat’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 29 c: 23 u: 11 o: 30 n: 27 a: 1 m: 7 #: 19 p: 0 d: 3

233

napominat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 27 c: 27 u: 10 o: 29 n: 26 a: 3 m: 3 #: 20 p: 1 d: 4

napominat’ + pro (25 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 21 c: 22 u: 16 o: 25 n: 23 a: 4 m: 3 #: 9 p: 0 d: 2

napomnit’ napomnit’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 26 c: 25 u: 15 o: 26 n: 23 a: 4 m: 5 #: 15 p: 4 d: 7

napomnit’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 28 u: 17 o: 27 n: 26 a: 5 m: 2 #: 13 p: 3 d: 4

napomnit’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 27 c: 26 u: 17 o: 28 n: 22 a: 3 m: 4 #: 13 p: 2 d: 8

napomnit’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 26 c: 26 u: 14 o: 27 n: 28 a: 4 m: 4 #: 16 p: 3 d: 2

napomnit’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 27 c: 28 u: 19 o: 29 n: 27 a: 3 m: 2 #: 11 p: 1 d: 3

podumat’

234

podumat’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 24 c: 28 u: 17 o: 25 n: 26 a: 6 m: 2 #: 13 p: 5 d: 4

podumat’ + pro (10 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 1 c: 10 u: 8 o: 8 n: 10 a: 9 m: 0 #: 2 p: 2 d: 0

podumat’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 22 c: 30 u: 20 o: 25 n: 25 a: 8 m: 0 #: 10 p: 5 d: 5

podumat’ + pro (19 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 4 c: 18 u: 13 o: 10 n: 19 a: 15 m: 1 #: 16 p: 9 d: 0

podumat’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 27 c: 25 u: 17 o: 30 n: 24 a: 3 m: 5 #: 13 p: 0 d: 6

podumat’ + pro (24 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 5 c: 23 u: 18 o: 12 n: 24 a: 19 m: 1 #: 6 p: 12 d: 0

podumat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 29 u: 19 o: 29 n: 22 a: 5 m: 1 #: 11 p: 1 d: 8

podumat’ + pro (30 tokens) – Pd. D

235

anim. status count/mass modific. comm./prop. deverb. status i: 18 c: 26 u: 20 o: 25 n: 29 a: 12 m: 4 #: 10 p: 5 d: 1

podumat’ + nasčet (20 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 15 c: 20 u: 16 o: 16 n: 16 a: 5 m: 0 #: 4 p: 4 d: 4

pomnit’ pomnit’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 27 u: 18 o: 25 n: 27 a: 5 m: 3 #: 12 p: 5 d: 3

pomnit’ + pro (15 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 13 c: 12 u: 8 o: 12 n: 15 a: 2 m: 3 #: 7 p: 3 d: 0

pomnit’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 22 c: 28 u: 16 o: 28 n: 28 a: 8 m: 2 #: 14 p: 2 d: 2

pomnit’ + pro (21 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 16 c: 21 u: 12 o: 19 n: 20 a: 5 m: 0 #: 9 p: 2 d: 1

pomnit’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 23 c: 26 u: 13 o: 25 n: 28 a: 7 m: 4 #: 17 p: 5 d: 2

pomnit’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status

236

i: 23 c: 29 u: 18 o: 25 n: 28 a: 7 m: 1 #: 12 p: 5 d: 2

pozabyt’ pozabyt’ + o (26 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 17 c: 22 u: 17 o: 22 n: 23 a: 9 m: 4 #: 9 p: 4 d: 3

pozabyt’ + pro (10 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 9 c: 10 u: 4 o: 10 n: 8 a: 1 m: 0 #: 6 p: 0 d: 2

pozabyt’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 22 c: 23 u: 17 o: 28 n: 26 a: 8 m: 7 #: 13 p: 2 d: 4

pozabyt’ + pro (15 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 10 c: 15 u: 12 o: 14 n: 15 a: 5 m: 0 #: 3 p: 1 d: 0

pozabyt’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 24 c: 27 u: 14 o: 26 n: 26 a: 6 m: 3 #: 16 p: 4 d: 4

pozabyt’ + pro (17 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 15 c: 16 u: 11 o: 17 n: 16 a: 2 m: 1 #: 6 p: 0 d: 1

pozabyt’ + o (30 tokens) – Pd. D

237

anim. status count/mass modific. comm./prop. deverb. status i: 19 c: 26 u: 21 o: 27 n: 26 a: 11 m: 4 #: 9 p: 3 d: 4

pozabyt’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 21 c: 30 u: 19 o: 26 n: 27 a: 9 m: 0 #: 11 p: 4 d: 3

uznat’ uznat’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 28 c: 27 u: 18 o: 29 n: 18 a: 2 m: 3 #: 12 p: 1 d: 12

uznat’ + pro (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 26 c: 28 u: 18 o: 28 n: 25 a: 4 m: 2 #: 12 p: 2 d: 5

uznat’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 28 c: 30 u: 25 o: 28 n: 24 a: 2 m: 0 #: 5 p: 2 d: 6

uznat’ + pro (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 21 c: 28 u: 17 o: 26 n: 29 a: 9 m: 2 #: 13 p: 4 d: 1

uznat’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 28 c: 30 u: 16 o: 28 n: 21 a: 2 m: 0 #: 14 p: 2 d: 9

238

uznat’ + pro (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 20 c: 30 u: 22 o: 21 n: 27 a: 10 m: 0 #: 8 p: 9 d: 3

uznat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 28 c: 29 u: 18 o: 29 n: 26 a: 2 m: 1 #: 12 p: 1 d: 4

uznat’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 23 c: 30 u: 23 o: 26 n: 28 a: 7 m: 0 #: 7 p: 4 d: 2

uznat’ + nasčet (12 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 11 c: 11 u: 10 o: 12 n: 11 a: 1 m: 1 #: 2 p: 0 d: 1

uznavat’ uznavat’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 28 c: 25 u: 16 o: 30 n: 24 a: 2 m: 5 #: 14 p: 0 d: 6

uznavat’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 29 u: 17 o: 28 n: 23 a: 5 m: 1 #: 13 p: 2 d: 7

uznavat’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 28 c: 26 u: 15 o: 28 n: 19 a: 2 m: 4 #: 15 p: 2 d: 11

uznavat’ + o (30 tokens) – Pd. D

239

anim. status count/mass modific. comm./prop. deverb. status i: 29 c: 26 u: 15 o: 30 n: 23 a: 1 m: 4 #: 15 p: 0 d: 7

uznavat’ + pro (15 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 11 c: 15 u: 5 o: 15 n: 11 a: 4 m: 0 #: 10 p: 0 d: 4

vspominat’ vspominat’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 18 c: 28 u: 12 o: 25 n: 27 a: 12 m: 2 #: 18 p: 5 d: 3

vspominat’ + pro (18 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 11 c: 18 u: 12 o: 13 n: 16 a: 7 m: 0 #: 6 p: 5 d: 2

vspominat’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 22 c: 28 u: 14 o: 25 n: 28 a: 8 m: 2 #: 16 p: 5 d: 2

vspominat’ + pro (23 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 13 c: 23 u: 11 o: 16 n: 23 a: 10 m: 0 #: 12 p: 7 d: 0

vspominat’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 23 c: 29 u: 15 o: 28 n: 25 a: 7 m: 1 #: 15 p: 2 d: 5

vspominat’ + pro (28 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status

240

i: 15 c: 27 u: 22 o: 21 n: 28 a: 13 m: 1 #: 6 p: 7 d: 0

vspominat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 28 c: 29 u: 13 o: 29 n: 23 a: 2 m: 1 #: 17 p: 1 d: 7

vspominat’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 17 c: 29 u: 20 o: 22 n: 30 a: 13 m: 1 #: 10 p: 8 d: 0

vspomnit’ vspomnit’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 17 c: 27 u: 19 o: 27 n: 28 a: 13 m: 3 #: 11 p: 3 d: 2

vspomnit’ + pro (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 19 c: 28 u: 17 o: 26 n: 27 a: 11 m: 2 #: 13 p: 4 d: 3

vspomnit’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 22 c: 25 u: 22 o: 25 n: 27 a: 8 m: 5 #: 8 p: 5 d: 3

vspomnit’ + pro (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 24 c: 29 u: 21 o: 23 n: 30 a: 6 m: 1 #: 9 p: 7 d: 0

vspomnit’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 21 c: 28 u: 15 o: 25 n: 26

241

a: 9 m: 2 #: 15 p: 5 d: 4

vspomnit’ + pro (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 17 c: 29 u: 21 o: 23 n: 30 a: 13 m: 1 #: 9 p: 7 d: 0

vspomnit’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 26 u: 14 o: 27 n: 28 a: 5 m: 4 #: 16 p: 3 d: 2

vspomnit’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 23 c: 25 u: 21 o: 28 n: 27 a: 7 m: 5 #: 9 p: 2 d: 3

zabyt’ zabyt’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 23 c: 27 u: 20 o: 25 n: 23 a: 7 m: 3 #: 10 p: 5 d: 7

zabyt’ + pro (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 22 c: 25 u: 20 o: 27 n: 27 a: 8 m: 5 #: 10 p: 3 d: 3

zabyt’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 23 c: 24 u: 25 o: 27 n: 28 a: 7 m: 6 #: 5 p: 3 d: 2

zabyt’ + pro (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 27 u: 20 o: 30 n: 29 a: 5 m: 3 #: 10 p: 0 d: 1

242

zabyt’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 24 c: 28 u: 18 o: 30 n: 25 a: 6 m: 2 #: 12 p: 0 d: 5

zabyt’ + pro (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 24 c: 27 u: 17 o: 24 n: 30 a: 6 m: 3 #: 13 p: 6 d: 0

zabyt’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 24 c: 25 u: 15 o: 28 n: 25 a: 6 m: 5 #: 15 p: 2 d: 5

zabyt’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 27 u: 21 o: 26 n: 26 a: 5 m: 3 #: 9 p: 4 d: 4

zabyvat’ zabyvat’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 22 c: 25 u: 17 o: 25 n: 26 a: 8 m: 5 #: 13 p: 5 d: 4

zabyvat’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 27 c: 27 u: 17 o: 28 n: 29 a: 3 m: 3 #: 13 p: 2 d: 1

zabyvat’ + pro (24 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 18 c: 24 u: 13 o: 21 n: 23 a: 6 m: 0 #: 11 p: 3 d: 1

243

zabyvat’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 28 c: 28 u: 17 o: 28 n: 24 a: 2 m: 2 #: 13 p: 2 d: 6

zabyvat’ + pro (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 27 u: 18 o: 28 n: 29 a: 5 m: 3 #: 12 p: 2 d: 1

zabyvat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 28 c: 28 u: 17 o: 28 n: 24 a: 2 m: 2 #: 13 p: 2 d: 6

zabyvat’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 18 c: 26 u: 22 o: 26 n: 29 a: 12 m: 4 #: 8 p: 4 d: 1

znat’ znat’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 29 c: 29 u: 9 o: 30 n: 21 a: 1 m: 1 #: 21 p: 0 d: 9

znat’ + pro (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 23 c: 26 u: 12 o: 24 n: 29 a: 7 m: 4 #: 18 p: 6 d: 1

znat’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 30 c: 29 u: 20 o: 29 n: 23 a: 0 m: 1 #: 10 p: 1 d: 7

244

znat’ + pro (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 19 c: 27 u: 20 o: 24 n: 30 a: 11 m: 3 #: 10 p: 6 d: 0

znat’ + nasčet (12 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 8 c: 10 u: 11 o: 8 n: 12 a: 4 m: 2 #: 1 p: 4 d: 0

znat’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 29 c: 28 u: 14 o: 29 n: 22 a: 1 m: 2 #: 16 p: 1 d: 8

znat’ + pro (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 20 c: 27 u: 23 o: 25 n: 30 a: 10 m: 3 #: 7 p: 5 d: 0

znat’ + nasčet (11 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 8 c: 10 u: 6 o: 11 n: 11 a: 3 m: 1 #: 5 p: 0 d: 0

znat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 30 c: 28 u: 14 o: 29 n: 24 a: 0 m: 2 #: 16 p: 1 d: 6

znat’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 20 c: 30 u: 12 o: 28 n: 27 a: 10 m: 0 #: 18 p: 2 d: 3

znat’ + nasčet (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 19 c: 27 u: 26 o: 25 n: 28

245

a: 11 m: 3 #: 4 p: 5 d: 2

Verbs of Reading

čitat’

čitat’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 19 c: 29 u: 22 o: 22 n: 26 a: 11 m: 1 #: 8 p: 8 d: 4

čitat’ + pro (26 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 14 c: 24 u: 18 o: 19 n: 25 a: 12 m: 2 #: 8 p: 7 d: 1

čitat’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 28 u: 16 o: 29 n: 22 a: 5 m: 2 #: 14 p: 1 d: 8

čitat’ + pro (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 18 c: 28 u: 21 o: 21 n: 28 a: 12 m: 2 #: 9 p: 9 d: 2

čitat’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 22 c: 29 u: 16 o: 27 n: 27 a: 8 m: 1 #: 14 p: 3 d: 3

čitat’ + pro (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 16 c: 27 u: 17 o: 26 n: 29 a: 14 m: 3 #: 13 p: 4 d: 1

246

čitat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 24 c: 28 u: 15 o: 28 n: 26 a: 6 m: 2 #: 15 p: 2 d: 4

čitat’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 22 c: 28 u: 11 o: 27 n: 28 a: 8 m: 2 #: 19 p: 3 d: 2

pročitat’ pročitat’ + o (18 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 15 c: 16 u: 14 o: 17 n: 15 a: 3 m: 2 #: 4 p: 1 d: 3

pročitat’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 27 c: 25 u: 22 o: 28 n: 24 a: 3 m: 5 #: 8 p: 2 d: 6

pročitat’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 26 c: 26 u: 18 o: 26 n: 26 a: 4 m: 4 #: 12 p: 4 d: 4

pročitat’ + pro (11 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 6 c: 10 u: 8 o: 8 n: 11 a: 5 m: 1 #: 3 p: 3 d: 0

pročitat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 27 c: 27 u: 15 o: 26 n: 25 a: 3 m: 3 #: 15 p: 4 d: 5

247

pročitat’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 15 c: 28 u: 19 o: 24 n: 29 a: 15 m: 2 #:11 p: 6 d: 1

Verbs of Hearing

slyšat’ slyšat’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 28 u: 13 o: 24 n: 25 a: 5 m: 2 #: 17 p: 6 d: 5

slyšat’ + pro (22 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 16 c: 21 u: 8 o: 18 n: 19 a: 6 m: 1 #: 14 p: 4 d: 3

slyšat’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 26 c: 27 u: 11 o: 27 n: 21 a: 4 m: 3 #: 19 p:3 d: 9

slyšat’ + pro (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 13 c: 30 u: 22 o: 19 n: 29 a: 17 m: 0 #: 8 p: 11 d: 1

slyšat’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 19 c: 28 u: 22 o: 24 n: 24 a: 11 m: 2 #: 8 p: 6 d: 6

248

slyšat’ + pro (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 17 c: 29 u: 13 o: 21 n: 29 a: 13 m: 1 #: 17 p: 9 d: 1

slyšat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 20 c: 28 u: 15 o: 25 n: 27 a: 10 m: 2 #: 15 p: 5 d: 3

slyšat’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 16 c: 25 u: 20 o: 20 n: 30 a: 14 m: 5 #: 10 p: 10 d: 0

slyxat’ slyxat’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 17 c: 30 u: 18 o: 21 n: 26 a: 13 m: 0 #: 12 p: 9 d: 4

slyxat’ + pro (24 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 10 c: 24 u: 14 o: 13 n: 23 a: 14 m: 0 #: 10 p: 11 d: 1

slyxat’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 29 u: 14 o: 28 n: 27 a: 5 m: 1 #: 16 p: 2 d: 3

slyxat’ + pro (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 16 c: 28 u: 22 o: 21 n: 30 a: 14 m: 2 #: 8 p: 9 d: 0

249

slyxat’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 15 c: 28 u: 17 o: 19 n: 26 a: 15 m: 2 #: 13 p: 11 d: 4

slyxat’ + pro (23 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 11 c: 23 u: 12 o: 16 n: 23 a: 12 m: 0 #: 11 p: 7 d: 0

slyxat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 18 c: 27 u: 20 o: 24 n: 25 a: 12 m: 3 #: 10 p: 6 d: 5

slyxat’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 20 c: 28 u: 15 o: 20 n: 30 a: 10 m: 2 #: 15 p: 10 d: 0

uslyšat’ uslyšat’ + o (27 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 18 c: 26 u: 17 o: 19 n: 20 a: 9 m: 1 #: 10 p: 8 d: 7

uslyšat’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 27 u: 21 o: 24 n: 25 a: 5 m: 3 #: 9 p: 6 d: 5

uslyšat’ + pro (18 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 16 c: 16 u: 11 o: 16 n: 18 a: 2 m: 2 #: 7 p: 2 d: 0

uslyšat’ + o (30 tokens) – Pd. C

250

anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 27 u: 21 o: 26 n: 22 a: 5 m: 3 #: 9 p: 4 d: 8

uslyšat’ + pro (21 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 13 c: 21 u: 16 o: 14 n: 20 a: 8 m: 0 #: 5 p: 7 d: 1

uslyšat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 20 c: 27 u: 18 o: 24 n: 24 a: 10 m: 3 #: 12 p: 6 d: 6

uslyšat’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 21 c: 30 u: 20 o: 22 n: 28 a: 9 m: 0 #: 10 p: 8 d: 2

Verbs of Asking

rassprašivat’ rassprašivat’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 26 u: 13 o: 28 n: 29 a: 5 m: 4 #: 17 p: 2 d: 1

rassprašivat’ + pro (24 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 18 c: 18 u: 13 o: 22 n: 19 a: 6 m: 6 #: 11 p: 2 d: 5

rassprašivat’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 28 u: 21 o: 24 n: 25

251

a: 5 m: 2 #: 9 p: 6 d: 5

rassprašivat’ + pro (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 22 c: 28 u: 20 o: 26 n: 29 a: 8 m: 2 #: 10 p: 4 d: 1

rassprašivat’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 27 c: 28 u: 20 o: 27 n: 30 a: 3 m: 2 #: 10 p: 3 d: 0

rassprašivat’ + pro (24 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 15 c: 22 u: 17 o: 19 n: 23 a: 9 m: 2 #: 7 p: 5 d: 1

rassprašivat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 20 c: 29 u: 25 o: 26 n: 27 a: 10 m: 1 #: 5 p: 4 d: 3

rassprašivat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 18 c: 28 u: 25 o: 24 n: 30 a: 12 m: 2 #: 5 p: 6 d: 0

rassprosit’ rassprosit’ + o (20 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 16 c: 17 u: 11 o: 18 n: 17 a: 4 m: 3 #: 9 p: 2 d: 3

rassprosit’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 17 c: 28 u: 26 o: 23 n: 26 a: 13 m: 2 #: 4 p: 7 d: 4

252

rassprosit’ + o (29 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 23 c: 28 u: 17 o: 26 n: 27 a: 6 m: 1 #: 12 p: 3 d: 2

rassprosit’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 20 c: 28 u: 20 o: 24 n: 27 a: 10 m: 2 #: 10 p: 6 d: 3

rassprosit’ + pro (13 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 8 c: 12 u: 10 o: 10 n: 12 a: 5 m: 1 #: 3 p: 3 d: 1

sprašivat’ sprašivat’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 23 c: 26 u: 18 o: 28 n: 26 a: 7 m: 4 #: 12 p: 2 d: 4

sprašivat’ + pro (24 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 6 c: 22 u: 17 o: 15 n: 24 a: 18 m: 2 #: 7 p: 9 d: 0

sprašivat’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 23 c: 25 u: 21 o: 25 n: 27 a: 7 m: 5 #: 9 p: 5 d: 3

sprašivat’ + pro (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 21 c: 27 u: 24 o: 22 n: 30 a: 9 m: 3 #: 6 p: 8 d: 0

253

sprašivat’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 21 c: 27 u: 24 o: 24 n: 25 a: 9 m: 3 #: 6 p: 6 d: 5

sprašivat’ + pro (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 15 c: 29 u: 24 o: 21 n: 29 a: 15 m: 1 #: 6 p: 9 d: 1

sprašivat’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 20 c: 24 u: 22 o: 27 n: 26 a:10 m: 6 #: 8 p: 3 d: 4

sprašivat’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 16 c: 29 u: 26 o: 22 n: 29 a:14 m: 1 #: 4 p: 8 d: 1

sprašivat’ + nasčet (11 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 10 c: 10 u: 9 o: 10 n: 9 a: 1 m: 1 #: 2 p: 1 d: 2

sprosit’ sprosit’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 26 c: 25 u: 23 o: 27 n: 28 a: 4 m: 5 #: 7 p: 3 d: 2

sprosit’ + pro (24 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 14 c: 23 u: 13 o: 19 n: 22 a: 10 m: 1 #: 11 p: 5 d: 2

254

sprosit’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 23 c: 27 u: 20 o: 25 n: 25 a: 7 m: 3 #: 10 p: 5 d: 5

sprosit’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 15 c: 28 u: 23 o: 20 n: 28 a: 15 m: 2 #: 7 p: 10 d: 2

sprosit’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 26 c: 26 u: 21 o: 28 n: 27 a: 4 m: 4 #: 9 p: 2 d: 3

sprosit’ + pro (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 19 c: 29 u: 27 o: 23 n: 29 a: 11 m: 1 #: 3 p: 7 d: 1

sprosit’ + nasčet (13 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 11 c: 12 u: 12 o: 11 n: 13 a: 2 m: 1 #: 1 p: 2 d: 0

sprosit’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 22 c: 27 u: 24 o: 25 n: 26 a: 8 m: 3 #: 6 p: 5 d: 4

sprosit’ + pro (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 18 c: 24 u: 26 o: 25 n: 30 a: 12 m: 6 #: 4 p: 5 d: 0

sprosit’ + nasčet (29 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status

255

i: 24 c: 21 u: 23 o: 27 n: 27 a: 5 m: 8 #: 6 p: 2 d: 2

spravljat’sja spravljat’sja + o (27 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 24 c: 14 u: 14 o: 25 n: 15 a: 3 m: 16 #: 13 p: 2 d: 12

spravljat’sja + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 23 c: 22 u: 20 o: 28 n: 21 a: 7 m: 8 #: 10 p: 2 d: 9

spravljat’sja + o (22 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 15 c: 12 u: 19 o: 19 n: 12 a: 7 m: 10 #: 3 p: 3 d: 10

spravljat’sja + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 24 c: 18 u: 21 o: 29 n: 15 a: 6 m:12 #: 9 p: 1 d: 15

spravit’sja spravit’sja + o (23 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 20 c: 18 u: 20 o: 21 n: 15 a: 3 m: 5 #: 3 p: 2 d: 9

spravit’sja + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 25 c: 23 u: 25 o: 27 n: 23 a: 5 m: 7 #: 5 p: 3 d: 7

spravit’sja + o (16 tokens) – Pd. C

256

anim. status count/mass modific. comm./prop. deverb. status i: 11 c: 11 u: 8 o: 14 n: 10 a: 5 m: 5 #: 8 p: 2 d: 6

spravit’sja + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 27 c: 18 u: 18 o: 29 n: 15 a: 3 m: 12 #: 12 p: 1 d: 15

prosit’ prosit’ + o (30 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 29 c: 21 u: 25 o: 30 n: 15 a: 1 m: 9 #: 5 p: 0 d: 15

prosit’ + o (30 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 30 c: 22 u: 22 o: 30 n: 18 a: 0 m: 8 #: 8 p: 0 d: 12

prosit’ + o (30 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 29 c: 13 u: 21 o: 30 n: 18 a: 1 m: 17 #: 9 p: 0 d: 12

prosit’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 30 c: 19 u: 21 o: 30 n: 22 a: 0 m: 11 #: 9 p: 0 d: 8

poprosit’ poprosit’ + o (11 tokens) – Pd. A anim. status count/mass modific. comm./prop. deverb. status i: 11 c: 9 u: 7 o: 11 n: 7 a: 0 m: 2 #:4 p: 0 d: 4

257

poprosit’ + o (18 tokens) – Pd. B anim. status count/mass modific. comm./prop. deverb. status i: 18 c: 14 u: 7 o: 18 n: 13 a: 0 m: 4 #:11 p: 0 d: 5

poprosit’ + o (13 tokens) – Pd. C anim. status count/mass modific. comm./prop. deverb. status i: 11 c: 8 u: 7 o: 11 n: 7 a: 2 m: 5 #: 6 p: 2 d: 6

poprosit’ + o (30 tokens) – Pd. D anim. status count/mass modific. comm./prop. deverb. status i: 29 c: 18 u: 21 o: 29 n: 23 a: 1 m: 12 #: 9 p: 1 d: 7

258