Transitivity from a Cognitive Perspective
2
“Transitivity in Russian from a Cognitive Perspective”[*]
by Laura A. Janda, University of North Carolina at Chapel Hill
0.0 Transitivity and Russian syntax
This paper will explore the variations in form observed in expressing transitivity in Russian. Given the prototypical transitive construction, it is possible to examine its relationship to many similar constructions, thus deriving a radial category of constructions that express transitivity to a varying degree. The semantic core of transitivity is the transfer of energy to an object. Transfer of energy to an object requires a verb and an object, but does not specify the way in which the object must be expressed. As we shall see, the verb and the object are indeed the most crucial elements in expressing transitivity, which can be present even when there is no subject or when the object is expressed in a non-prototypical way.
A rich Slavic morphological heritage has endowed Russian with a plethora of syntactic constructions that encode interactions among participants in the action of a verb. Cognitive linguistics has heretofore focused primarily on the semantic variety among transitive expressions, usually under the assumption that transitivity is syntactically unitary and discrete. In other words, most researchers have focused on the semantic variation available within the confines of a single (prototypical) transitive construction. Russian syntax offers us the opportunity to explore a complex network of over a dozen interrelated syntactic constructions that express transitivity. By examining these constructions, we can obtain an accurate picture of the true complexity of transitivity and analyze the relationship between semantics and syntax in expressing transitivity. In so doing, we can construct a radial network for Russian transitive constructions that reveals a hierarchy of constructions and specific constraints on the types of constructions that can express transitivity. Ultimately this radial network can provide a basis for cross-linguistic comparison of transitivity constructions.
This article will commence with an overview of what the community of cognitive linguists can provide us on the topic of transitivity (1.0). After a brief summary of relevant facts of Russian syntax (2.0), I will proceed to the analysis of transitivity constructions in Russian (3.0), first identifying the prototypical transitive construction (3.1), and then examining elaborations on that construction related via the addition of items to the construction (3.2), alterations in the case-marking of items in the construction (3.3), and removal of items from the construction (3.3). The resulting family of constructions will be then gathered into a radial network (3.4). In the conclusion (4.0) I will discuss the advantages derivable from this type of analysis.
1.0 Primacy of semantics
Cognitive linguistics is predicated on the assumption that language is structured through symbolic relations between semantic structures and their expression, be it syntactic, phonological, etc. (Croft 2001: 203-240, Langacker 1987: 76-96). In other words, the relationship between meaning and form is the primary force motivating linguistic phenomena. For cognitive linguists, meaning plays the more crucial role in this relationship, and meaning itself is grounded in human embodied experience. There will be no attempt to challenge any of these assumptions in this paper, but it does appear that in the case of transitivity, focus on the semantics of transitivity has thus far largely excluded discussion of the expression of transitivity. We will depart from this trend by focusing on the various syntactic forms that transitivity can take in Russian. The advantage of this undertaking is that it will facilitate a more nuanced view of transitivity, as a linguistic category that encompasses several constructions, making it possible to determine which syntactic elements play more or less important roles in expressing transitivity. Thus we can see both the “parts” of the transitivity “whole”, as well as how those “parts” interact, rather than seeing a single, unanalyzable “whole”.
Transitivity is itself a highly salient phenomenon, well-rooted in human experience. Langacker’s (2000: 13-14) famous “billiard-ball model” demonstrates the way in which folk models of the world motivate the conceptualization of nouns as representing discrete objects and verbs as representing energy, thus providing the basic elements that participate in transitivity. The grammatical expression of transitivity is the projection of an “abstract story” of agency and causation onto grammatical structure (Turner 1996: 140-168). Dowty (1991) identifies the properties of the “Proto-Agent” (volition, sentience, causation, movement, independence) and the “Proto-Patient” (change of state, incremental, affectedness, stationary, dependent) in the transitivity relationship. Langacker’s (1991: 285) “role archetypes” for agent and patient are very similar: “The archetype agent is a person who volitionally initiates physical activity resulting, through physical contact, in the transfer of energy to an external object. Its polar opposite is an archetypal patient, an inanimate object that absorbs the energy transmitted via externally initiated physical contact and thereby undergoes an internal change of state.” For Langacker (2000: 69), “[a]n essential component of transitivity is the interaction of participants.” Langacker (1991: 285-6) combines these concepts in the “canonical event model”, which contains an agent that produces an effect on a patient in a setting seen by a viewer; thus “the canonical event model represents the normal observation of a prototypical action.” Langacker’s canonical event serves as a prototype both of transitive events and of events in general, and it will serve as the starting point for our discussion of transitivity in Russian. However, Langacker (cf. especially 1991: 293-329) directs his discussion to the encoding of various types of semantic structures with transitive constructions, without exploring the possibility of syntactic variation in the transitivity construction. Furthermore, Langacker (1991: 321) assumes that transitivity entails the presence of both participants: “Subjects occur in both transitive and intransitive clauses, whereas only the former have direct objects – or to put it another way, a direct object presupposes a subject, but not conversely.”
Croft, Goldberg, and Taylor maintain focus on transitivity as the syntactic encoding of a relationship between an agent and a patient. Croft (2001: 136) asserts that relationships such as transitivity must be identified according to their semantics, and that syntax follows secondarily. Although Croft does explore various means of encoding transitivity, his is a typological study aimed at sorting out the different conceptualizations involved in accusative vs. ergative marking, and he does not take up the issue of variation in transitivity marking within a given language. Goldberg (1995: 101-119) assumes that the transitive construction is a “unitary structure” related to other similar units such as ditransitive, caused-motion, intransitive, etc. She admits that “it may be more felicitous to assign a family of related meanings to it” (Goldberg 1995: 118), but does not recognize variation in syntactic form within transitivity. Taylor (2002: 415-426) allows some variation within transitivity: he admits that it is often hard to distinguish between participants and circumstances, that construal plays a role, and that participants can sometimes be omitted. Unfortunately Taylor’s discussion is limited to English, but at least there is a recognition that transitivity may not be syntactically monolithic. Talmy (2000: 92-3) foregrounds the role of construal in transitivity, by recognizing that transitivity is dependent upon construal and that similar semantic structures may be encoded differently.
As we shall see, Russian provides us with opportunities to vary the transitivity construction and delete items, even including the subject participant, so we will find it necessary to amend and extend Langacker’s model. Talmy’s model of construal will also facilitate insights into the ways in which similar constructions may be related.
2.0 Relevant facts about Russian syntax
The presence of a complex system of grammatical cases facilitates relatively free word order in Russian. It also means that grammatical roles are quite clear, obviating the need for “dummy” subjects: if a nominative subject is present, a verb agrees with it (according to some subset of person, number, and gender, depending upon tense), but if there is no nominative subject, a verb will have default (neuter singular) agreement. All of the non-nominative cases share some of the burden of expressing transitivity, so it is worth giving a brief overview of their meanings before looking at the constructions they appear in. The characterizations below are by no means comprehensive, but they do represent a significant departure from traditional accounts of Russian case. They target only the case meanings relevant to transitive constructions; a fuller account of Russian case meaning (from which these characterizations are derived) can be found in Janda & Clancy 2002.
Accusative: This is the case of destinations, be they of motion, purpose, or merely of the energy of an action, as we see in the prototypical direct object, which is marked with the accusative.
Dative: The dative has three overlapping roles, as the receiver of an object, the experiencer of an event, or a competitor which may be on an equal footing with or submit to the superiority of a nominative subject. Crucial to the dative is the ability to react, to serve as a potential subject.
Instrumental: An instrumental entity is a peripheral accessory to something else, be it a verb or a noun. When associated with a verb, the instrumental marks a conduit for the action of the verb, the means by which the action takes place, which can be either an instrument or the agent of an action. When associated with a noun, the instrumental can identify a label through which another item is accessed.
Genitive: The genitive is usually a mental address for something else. It is the point of origin, or source of another item, or conversely the goal of an item. The genitive can also mark the whole of which another item is a part, or can just be a reference point for locating the other item. To generalize: “[t]he genitive is a backgrounded item… that yields focus of attention to something else… which exists or maneuvers in its proximity” (Janda & Clancy 2002: 112).
Locative: The locative marks a place in space, time, or another metaphorical domain.
The cases and their meanings play an important role in demonstrating the relationship between semantics and syntax in the expression of transitivity in Russian.
3.0 A radial network of transitive constructions
After identifying the prototypical transitive construction in Russian, I will move on to variations of that construction and discussions of how these variants are related to the prototype and how well they express transitivity by signaling a transfer of energy to an object. Variation will be explored along three parameters: addition of items, alteration in case marking of items, and removal of items. Once all the variants have been presented, it will be possible to posit a network of related constructions.
3.1 The prototypical transitive construction: N+V+A
Let’s assume that the prototypical transitive construction for Russian is a clause containing a nominative subject, a verb, and an accusative direct object, which we can represent as N+V+A (though word order may vary), as in:
(1) Девушка сшила юбку.
[Girl-N sewed skirt-A.]
‘The girl sewed a skirt.’
where the girl serves as the agent and a skirt serves as the patient. This construction instantiates the canonical event model, and its participants conform to Langacker’s definitions of role archetypes. In this construction prototypical transitive verbs focus energy on a patient, they involve “doing something to X”. The prototypical transitive construction is clearly distinct from the prototypical intransitive construction which contains only a nominative subject and an intransitive verb (often describing a state or emotion), N+V, as in:
(2) Мальчик спит.
[Boy-N sleeps.]
‘The boy is sleeping.’
The prototypical transitive construction, N+V+A, will serve as the point of departure for our analysis of transitivity. As we look at variants of this construction and examine their expression of transitivity, it is important to remember that even the N+V+A construction does not always entail a transfer of energy to an object. Here is an illustration of an intransitive clause with N+V+A structure, where the verb does not transfer energy to an object and the accusative marks instead the duration of the action:
(3) Девушка работала всю ночь.
[Girl-N worked all night-A.]
‘The girl worked all night.’
3.2 Adding items to N+V+A
It is possible to add more items to a transitive N+V+A clause, and this strategy does not disturb the capacity of the clause to signal a transfer of energy to an object. The items that can be added to N+V+A include: a prepositional phrase, producing N+V+A+P+G/L/A/D/I; a Dative participant, producing N+V+A+D; and an Instrumental participant, producing N+V+A+I. We will examine each of these possibilities in turn.
N+V+A+P+G/L/A/D/I: The addition of a prepositional phrase to N+V+A does no harm to transitivity, regardless of what case follows the preposition. All of the following examples, illustrating the use of prepositions governing the genitive, locative, accusative, dative, and instrumental cases, illustrate the transfer of energy to an object:
(4) Девушка сшила юбку из старого пальто.
[Girl-N sewed skirt-A from old coat-G.]
‘The girl sewed a skirt from an old coat.’
(5) Девушка сшила юбку на машинке.
[Girl-N sewed skirt-A on machine-L.]
‘The girl sewed a skirt on a machine.’
(6) Девушка принесла юбку в школу.
[Girl-N brought skirt-A to school-A.]
‘The girl brought a skirt to school.’
(7) Девушка купила юбку к лету.
[Girl-N bought skirt-A for summer-D.]
‘The girl bought a skirt for the summer.’
(8) Девушка сшила юбку с прикладкой
[Girl-N sewed skirt-A with lining-I.]
‘The girl sewed a skirt with a lining.’
In (4)-(8) the transitive relationship between the girl and the skirt is every bit as strong as in the prototypical example (1).
N+V+A+D: Making the clause ditransitive by adding a dative participant likewise does not decrease the impact of the agent on the patient, and thus does not negatively affect the transfer of energy to an object, as in:
(9) Людмила Путина сшила мужу костюм.
[Ljudmila Putina-N sewed husband-D suit-A.]
‘Ljudmila Putina sewed her husband a suit.’
Again, the transitivity interaction between Ljudmila Putina and a suit is equivalent to that observed in (1).
N+V+A+I: The addition of an instrumental participant does not detract from transitivity either; rather this construction instantiates what Langacker (2000: 38) terms “a canonical transitive clause that profiles a full action chain”, since it specifies an instrument as well, illustrated here: