Valency And Argument Structure Research Paper

View sample Valency And Argument Structure Research Paper. Browse other  research paper examples and check the list of research paper topics for more inspiration. If you need a religion research paper written according to all the academic standards, you can always turn to our experienced writers for help. This is how your paper can get an A! Feel free to contact our custom writing service for professional assistance. We offer high-quality assignments for reasonable rates.

The purpose of this research paper is to describe the valency and argument structure of the verb or predicate. The notion of argument is known to have been adapted from mathematics. In mathematics, a function may take one or more arguments, which will be used to determine the value of that function. In linguistics, argument is used as a technical term for syntactic elements that combine obligatorily with the verb (or the predicate as the verb is usually referred to in relation to arguments). For instance, the English verb die must take at least one syntactic element, e.g., the mayor, to form sentences, e.g., The mayor died. This syntactic element is taken to be the (sole) argument of the predicate die. The verb kill, on the other hand, requires two arguments to form sentences, one representing the killer (the man) and the other the person killed (the mayor), as in The man killed the mayor. The number of arguments that are required by a predicate is then specified in the argument structure of that predicate. Additional syntactic elements, e.g., in the office or in 1878, can also potentially combine with the predicate die, e.g., The mayor died in the office, The mayor died in 1878, or The mayor died in the office in 1878. But these are regarded only as optional, because their absence has no bearing upon the grammaticality of the ‘minimal’ sentence, The mayor died. Such optional syntactic elements are referred to as adjuncts. The distinction between arguments and adjuncts, albeit not always easy to make, is a useful one (see Comrie 1993 for discussion as to how arguments and adjuncts should be distinguished).

The argument structure also contains information concerning the relation between the predicate and its argument(s). For instance, in the sentence The man killed the mayor, the predicate kill and its arguments the man and the mayor have logical (or grammatical) relationships between them with the effect that the man is the actor (or the subject) who carried out the killing, and the mayor is the undergoer (or the direct object) who was affected by the actor’s killing. Adjuncts, e.g., in the office in The mayor died in the office, also bear logical (or grammatical) relationships, e.g., location (or oblique), with the predicate. Information on adjuncts, however, is not included in the argument structure of the predicate itself.

The term valency (or valence in the USA) is said to have been borrowed from chemistry into syntax. In chemistry, the valency of a chemical element is defined as the capacity of that element for combining with a fixed number of atoms of another element, for instance, two hydrogen atoms combining with an atom of oxygen in the case of water (i.e., H O). The notion of valency, when applied to verbs, refers to the number of arguments that are required for a given verb to form sentences. For example, the verb die is regarded as having a valency structure of a single argument. Thus, verbs such as die are characterized as one-place (or monovalent) verbs. The verb kill, on the other hand, is a two-place (or divalent) verb, because it requires two arguments.

The rest of the paper is concerned principally with illustrating how argument structure can be altered or manipulated with or without a concomitant change of valency. How languages may keep the valency of the derived verb from exceeding that of the basic verb by manipulating argument structure will also be illustrated. Finally, how valency-changing operations are handled in current grammatical theories will be discussed briefly.

1. Change Of Argument Structure With Change Of Valency

There are two main mechanisms through which the argument structure of the predicate can be altered, together with a change of valency: argument reduction, and argument addition. In argument reduction, one of the arguments of the basic predicate may be demoted to adjunct status. In argument addition, on the other hand, adjuncts may be promoted to argument status, or new arguments may be introduced into the argument structure. There are five important grammatical devices utilized in argument reduction or argument addition: passivization, antipassivization, noun incorporation, advancement, and causativization. The first three belong to argument reduction, whereas the last two fall under argument addition. (But for examples of noun incorporation consisting of argument reduction and addition, see Mithun 1984, and Sect. 2 for causativization involving both argument addition and reduction.)

1.1 Passivization

In the passive clause, one of the arguments of the basic two-place predicate (i.e., the actor) is demoted to an adjunct with the other argument (i.e., the undergoer) retaining its argument status (but being promoted to the subject of the passive clause). This operation is referred to as passivization (Siewierska 1984). The number of arguments of the basic predicate is thus reduced from two (i.e., the actor and undergoer) to one (i.e., the undergoer) in the corresponding passive clause. This is substantiated by the fact that it is possible to omit the newly created adjunct. This change of valency (or valency decreasing) is signaled on the verb either by a verbal affix or by a periphrastic element in the verb phrase. Korean provides a good example of the passive:

(1) Korean

cuin-i totuk-l l cap ss-ta

owner-NOM thief-ACC catch-PST-IND

‘The owner caught the thief.’

(2) Korean

totuk-i (cuin-eke) cap-hi ss-ta

thief-NOM (owner-by) catch-PASS-PST-IND

‘The thief was caught (by the owner).’

[Abbreviations used to gloss the data: ABS = Absolutive; ACC = Accusative; APASS = Antipassive; APPL = Applicative; ASP = Aspect; CS = Causative Suffix; DAT = Dative; DO = Direct Object; ERG = Ergative; F = Feminine; FUT = Future; IND = Indicative; INTR = Intransitive; IO = Indirect Object; LOC = Locative; M = Masculine; NFUT = Non-Future; NOM = Nominative; PASS = Passive; PERF = Perfective; PRS = Present; PST = Past; TR = Transitive.]

1.2 Antipassivization

Antipassivization is often regarded as the counterpart of passivization in languages with the ergative– absolutive case system. But it should not be inferred from this that the passive is never found in languages with the ergative–absolutive system; both the passive and the antipassive may occur in the same language. In the antipassive, the actor argument of the basic two-place predicate remains the (sole) argument of the corresponding antipassive (albeit marked by the absolutive case), with the undergoer argument demoted to adjunct status or, optionally, omitted. If and when expressed, however, the undergoer is marked accordingly by an oblique case. This suggests that the antipassive is as much intransitive or detransitivized as is the passive: a decrease in valency from two (i.e., the actor and undergoer) to one (i.e., the actor) argument. The verb in the antipassive also bears an affix as a marker of valency decreasing. Dyirbal provides a good example of the antipassive:

(3) Dyirbal

balan dyugumbil ba gul yara gu

F-ABS woman-ABS M-ERG man-ERG



‘The man is hitting the woman.’

(4) Dyirbal

bayi yara bagun dyugumbil-gu

M-ABS man-ABS F-DAT woman-DAT

balgal a-nyu


‘The man is hitting the woman.’

1.3 Noun Incorporation

Noun incorporation is another type of valency-changing operation found in some languages of the world, whereby one of the arguments of the predicate becomes part of the predicate, thereby losing its argument status (Mithun 1984). The resulting predicate may function as a single word. In Chukchi, for instance, vowels of incorporated nouns must participate in word-internal vowel harmony in conjunction with host verbs:

(5) Chukchi

tumg-e na-nt wat n kupre-n

friends-ERG set-TR net-ABS

‘The friends set the net.’

(6) Chukchi

tumg t kopra-nt wat-g at

friend-NOM net-set-INTR

‘The friends set nets.’

Note that, with the undergoer incorporated into the predicate, the actor argument of the basic two-place predicate in (5) loses its ergative marking, and is instead marked by the nominative case in (6). In Chukchi, noun incorporation also motivates the verb to shed its transitive marking in favor of the in- transitive marking. These changes indicate that noun incorporation involves reduction of two arguments of the basic predicate to one (i.e., valency decreasing).

1.4 Advancement Of Adjuncts To Arguments

So far, argument reduction has been illustrated. In many languages, it is also possible to do the opposite, that is, to increase the valency of the basic verb by promoting an adjunct to argument status. In order to signal such promotion or advancement, the verb is normally marked by a so-called applicative affix. In Kinyarwanda, for instance, instrumental adjuncts can be promoted optionally to argument or direct object status. Compare the following two sentences:

(7) Kinyarwanda

umwaalımu a-ra-andik-a ıbaruwa

teacher he-PRS-write-ASP letter



‘The teacher is writing a letter with the pen.’

(8) Kinyarwanda

umwaalımu a-ra-andik-iish-a ıbaruwa

teacher he-PRS-write-APPL-ASP letter



‘The teacher is writing a letter with the pen.’

The instrumental adjunct in (7) is promoted to direct object in (8), evidence for which can be adduced from the fact that the instrumental nominal, which is marked by the instrumental prefix in (7), appears as a bare nominal (i.e., no case marking) in (8). This advancement must also be registered on the verb by the applicative suffix-iish, as in (8). Both the old direct object and the newly created direct object in (8) exhibit grammatical properties of objecthood. This suggests that the valency of the derived verb in (8) is greater by one argument than that of the basic verb in (7) (see Sect. 3). In (7), there are two arguments (i.e., teacher, and letter) and one adjunct (i.e., pen). In (8), on the other hand, there are three arguments (i.e., teacher, letter, and pen). Moreover, the argument structure of the predicate in (7) specifies the relation between the actor and the undergoer, and the argument structure of the derived predicate in (8) the relation between the actor, the undergoer, and the instrument. To put it differently, (7) without the instrumental phrase n’ııkaramu, will remain grammatical, whereas (8), with ıkaramu deleted, will be rendered ungrammatical.

1.5 Causativization

In common with advancement of adjuncts, causativization adds an argument to the argument structure of the basic predicate. However, it does so not by converting an existing adjunct into an argument, but by introducing a completely new argument into the argument structure. This new argument is the causer argument. Thus, if the basic verb is a one-place verb, the corresponding causative verb will be a two-place verb (the newly added causer argument and the sole argument of the original basic verb). If the basic verb is a two-place verb, the corresponding causative verb will be a three-place verb (three arguments including the causer argument). This is illustrated by the following pair of Turkish sentences. As opposed to the basic predicate in (9), the causative predicate in (10), with the causative suffix -t, has the causer argument, disci.

(9) Turkish

mudur mektub-u imzala-dı

director letter-DO sign-PST

‘The director signed the letter.’

(10) Turkish

disci mektub-u mudur-e imzala-t-tı

dentist letter-DO director-IO sign-CS-PST

‘The dentist made the director sign the letter.’

The valency of the causative verb thus exceeds that of the basic verb by one argument as can be neatly expressed in:


[n: number of arguments]

The formula in (11) indicates that causativization is a case of argument addition, with the status of the argument(s) of the basic predicate remaining intact. This may indeed be true of one-place verbs; causativization only increases the valency of the basic one-place verb from one to two arguments. Causativization of two or three-place (or trivalent) verbs, on the other hand, would give rise to three or four arguments, respectively, more than many languages can cope with—regardless of whether verbs are basic or derived. Indeed, there is much cross-linguistic variation in causativization of two-or three-place verbs. For instance, there are languages which derive causative verbs only from one-place verbs (Song 2001, Dixon and Aikhenvald 1997, Dixon 2000). Moreover, in many languages, argument structure is manipulated in such a way that the valency of the derived verb is prevented from exceeding that of the basic verb (Comrie 1989, Song 1996, Dixon 2000).

2. Change Of Argument Structure Without Change Of Valency

There are languages which introduce a new argument into the argument structure of the causative verb, only if one of the arguments of the basic twoor three-place verb is demoted to an adjunct, or omitted completely. In other words, causativization of the basic twoor three-place verb in such languages results in no increase in valency, contrary to the formula in (11), bur rather the valency of the derived causative verb is equal to that of the basic verb. In Babungo, for instance, when the causative suffix -s is added to twoplace verbs, the direct object of the basic verb must be omitted or expressed as an adjunct. Thus, the direct object in (12) appears as an optional prepositional phrase in (13).

(12) Babungo

nwe fee z3

he fear-PERF snake

‘He was afraid of a snake.’

(13) Babungo

me fes nwe (n5 z5)

I fear-CS-PERF him (with snake)

‘I frightened him with a snake.’

This phenomenon suggests that, regardless of whether the predicate is basic or derived, languages allow only a certain number of arguments per predicate (i.e., two or three) (Song 1996, Kemmer and Verhagen 1994 from a cognitive–linguistic perspective). Other devices drawn upon in causativization for this purpose include antipassivization and noun incorporation. The net effect of the use of these devices, in conjunction with causativization, is no change of valency, although there may be a change of argument structure between the basic predicate and the derived predicate: addition of the causer argument offset by demotion to adjunct status (or omission) of one of the arguments of the basic predicate.

3. Doubling Of Grammatical Relations

One theoretical issue in argument addition concerns the question of whether or not identical coding of nominals entails doubling of grammatical relations. In some languages, two nominals in the causative construction may be coded identically, for instance. In Kinyarwanda causatives, both the actor argument (or subject) and the undergoer argument (or direct object) of the basic predicate are coded identically as objects. Comrie (1989) interprets such identical coding to be indicative of doubling of the direct object. This position, however, has been challenged by Polinsky (1994), who demonstrates that the two identicallycoded nominals in Kinyarwanda causatives behave differently in terms of syntactic properties. In fact, only one of the identically-coded ‘direct objects’ is syntactically a direct object, the other being an indirect object, identical coding notwithstanding. Kozinsky and Polinsky (1993) describe more or less the same situation in Korean and Dutch causatives. Similarly, the old direct object and the newly created direct object in the Kinyarwanda applicative sentence (8) are bare nominals (that is, coded identically). Moreover, both nominals exhibit grammatical properties of objecthood. This also led to a controversy as to whether or not (8), as opposed to (7), has two direct objects, i.e., doubling of the direct object. Gary and Keenan (1977) take the view that (8) has two direct objects. Perlmutter and Postal (1983), on the other hand, argue that (8) does not have two direct objects, but rather one direct object and one indirect object. Further evidence against doubling of direct object in Kinyarwanda applicative constructions is provided by Dryer (1983) and Polinsky (1994).

4. Theoretical Approaches To Valency-Changing Operations

Current grammatical theories have proposed different approaches to representation of valency-changing operations; more than one approach may also be possible within one and the same grammatical theory (Brown and Miller 1996). These approaches can be characterized broadly as lexical or syntactic. Lexical Functional Grammar regards valency-changing operations as productive lexical processes giving rise to alternative assignments of grammatical functions (or grammatical relations) to arguments. In mainstream Government and Binding Theory, valency-changing operations are treated as syntactic. In causativization, for instance, the basic verb undergoes movement to combine with the causative affix at an abstract syntactic level. This movement is then claimed to comply with the same principles that are obeyed by other movements in syntax. Relational Grammar invokes revaluation rules to handle valency-changing operations. These revaluation rules are regarded as syntactic, because they operate on grammatical relations between initial and noninitial strata (or different syntactic levels) of the relational structure.


  1. Brown K, Miller J (eds.) 1996 Concise Encyclopedia of Syntactic Theories. Pergamon, Oxford, UK
  2. Comrie B 1989 Language Universals and Linguistic Typology. Blackwell, Oxford, UK
  3. Comrie B 1993 Argument structure. In: Jacobs J, von Stechow A, Sternefeld W, Vennemann T (eds.) Syntax: An International Handbook of Contemporary Research. Walter de Gruyter, Berlin
  4. Dixon R M W 2000 A typology of causatives: Form, syntax and meaning. In: Dixon R M W, Aikhenvald A Y (eds.) Valency Changing: Case Studies in Transitivity. Cambridge University Press, Cambridge, UK
  5. Dixon R M W, Aikhenvald A Y 1997 A typology of argumentdetermined constructions. In: Bybee J, Haiman J, Thompson S A (eds.) Essays on Language Function and Language Type. John Benjamins, Amsterdam, The Netherlands
  6. Dryer M S 1983 Indirect objects in Kinyarwanda revisited. In: Perlmutter D M (ed.) Studies in Relational Grammar 1. University of Chicago Press, Chicago
  7. Gary J O, Keenan E L 1977 On collapsing grammatical relations in Universal Grammar. In: Cole P, Sadock J (eds.) Syntax and Semantics 8: Grammatical Relations. Academic Press, New York
  8. Kemmer S, Verhagen A 1994 The grammar of causatives and the conceptual structure of events. Cognitive Linguistics 5: 115–56
  9. Kozinsky I, Polinsky M 1993 Causee and patient in the causative of transitive: Coding conflict or doubling of grammatical relations? In: Comrie B, Polinsky M (eds.) Causatives and Transitivity. John Benjamins, Amsterdam, The Netherlands
  10. Mithun M 1984 The evolution of noun incorporation Language 64: 847–94
  11. Perlmutter D M, Postal P M 1983 Some proposed laws of basic clause structure. In: Perlmutter D M (ed.) Studies in Relational Grammar 1. University of Chicago Press, Chicago
  12. Polinsky M 1994 Double objects in causatives: Towards a study of coding conflict. Studies in Language 19: 129–221
  13. Siewierska A 1984 Passive: A Comparative Linguistic Analysis. Croom Helm, London
  14. Song J J 1996 Causatives and Causation: A Universal-Typological Perspective. Addison Wesley Longman, London
  15. Song J J 2001 Linguistic Typology: Morphology and Syntax. Pearson Education, Harlow, UK
Economics Of Value Of Life Research Paper
Studies Of Utility And Subjective Probability Research Paper


Always on-time


100% Confidentiality
Special offer! Get discount 10% for the first order. Promo code: cd1a428655