Passivization of ditransitive verbs from the FSP point of view

Passivization of ditransitive verbs from the FSP point of view Gabriela Brůhová (Prague) ABSTRACT The present paper aims at contributing to the study of passivization of ditransitive complementation from the FSP point of view. English ditransitive verbs generally allow two passive constructions, i.e. the subject of a passive can in the active correspond either to Oi (Jack was sent a copy of the letter) or to Od (A copy of the letter was sent (to) Jack). As the passive usually serves as one of the means used to achieve the basic distribution of communicative dynamism, the choice of the subject and the object of the passive can be supposed to be motivated, respectively, by their thematic and rhematic function. Thus, the sentence is perspectived away from the subject and constitutes the quality scale. The aim of the paper is to verify this assumption and to find out whether a passive sentence with a ditransitive verb can also implement the presentation scale, i.e. whether the sentence can be perspectived towards the subject and introduce a phenomenon into discourse. Attention is paid to other potential factors that might play a role in the selection of the passive, namely the semantics of the verb and of the indirect object (whether the recipient is actual or intended), object deletion (i.e. omission of an object) and the expression of the by-agent. The analysis is based on examples obtained from the British National Corpus. KEYWORDS passive, ditransitive verbs, presentation scale, quality scale, object deletion, semantics of the verb, semantics of the indirect object 1. INTRODUCTION The present paper aims at contributing to the study of passivization of ditransitive complementation from the FSP point of view. English ditransitive verbs generally allow two passive constructions according to which object the passive subject corresponds to. The subject of the passive can in the active correspond either to the indirect object (Jack was sent a copy of the letter) or to the direct object (A copy of the letter was sent (to) Jack) (Quirk et al., 1985, 727). In accordance with Huddleston and Pullum (2002, 249) and Quirk et al. (1985, 1208), in the present paper I use the terms first and second passive for these two passive constructions, namely the term first passive where the subject corresponds to the first object in the active (Oi), and second passive where the subject corresponds to the second object in the active (Od). The terms also reflect the order of preference, since the first kind of passivization, the passive with a personal subject, appears to be far more common (cf. Quirk et al., 1985, 727, 1208; Huddleston and Pullum 2002, 249). The present paper focuses on the frequency of first and second passive and considers the factors influencing the selection of the particular construction. A major factor in the selection of the passive construction appears to be FSP, as the passive is OPEN ACCESS 28 LINGUISTICA PRAGENSIA 1/2015 a major device used to achieve the basic distribution of communicative dynamism (i.e. the object that carries a lower degree of communicative dynamism becomes the subject of the passive). In other words, the passive serves to achieve final placement of the rheme and preverbal placement of the theme. Passive predication is employed especially where it allows the theme of the utterance to become the grammatical subject. Thus, the sentence is perspectived away from the subject and constitutes the quality scale. The main aim of the paper is to verify this assumption and to find out whether a passive sentence with a ditransitive verb can also implement the presentation scale,1 i.e. whether the sentence can be perspectived towards the subject and introduce a phenomenon into discourse. Apart from FSP, other potential factors that might play a role in the selection of the passive, considered in the analysis, are the semantics of the verb and of the indirect object, and object deletion (i.e. omission of an object). 2. MATERIAL AND METHOD The analysis is carried out on the basis of 150 examples of finite passive constructions of five ditransitive verbs, i.e. 30 examples of each verb. The data for the analysis have been extracted from the British National Corpus. The selection of the five analysed verbs is based on the semantic verb classes defined by Gropen et al. (1989, 243ff.), Levin (1993, 45ff.) and Pinker (1989, 111). As one of the aims of the present paper is to find out whether a passive sentence with a ditransitive verb can also implement the presentation scale, I attempted to select prototypical ditransitive verbs which appear to be semantically disposed to perform the function of presentation, i.e. verbs that express or imply existence or appearance on the scene. The following five verbs have been selected for the purpose of the analysis: — verbs that inherently signify acts of giving: give, hand — verbs of future having (commitments that a person will have something at some point later): promise — verbs of obtaining: buy — verbs of future not having: deny The verb deny was chosen because it represents a special group of verbs of non-existence or disappearance (cf. Adam 2013, 132). These verbs do not actually present anything on the scene, but, conversely, something disappears from the scene. In this respect deny can be viewed as a direct counterpart of give. Apart from the semantics of the verb, these five verbs also differ in the semantics of indirect object, which was another reason why these particular verbs have been selected. Whereas the semantic role of the indirect object of give, hand and promise 1 The conception of the dynamic scales (presentation vs. quality scale) was introduced by Firbas (1992) and later elaborated by Svoboda (2005) and Chamonikolasová and Adam (2005). GABRIELA BRůHOVá 29 is an actual recipient, the indirect object of buy has the semantic role of intended recipient (i.e. beneficiary). This is manifested also syntactically, i.e. the indirect object of give, hand and promise alternates with the preposition to (to give sb. sth. vs. to give sth. to sb.), while the indirect object of the verb buy alternates with the preposition for (to buy sb. sth. vs. to buy sth. for sb.). The indirect object of deny alternates with the preposition to (to deny sb. sth. vs. to deny sth. to sb.) but its semantic role can be defined as ‘malefactive’, i.e. the opposite of the beneficiary Oi (Pinker 1989, 111; Haspelmath et al. 2010, 40), where the indirect object is not actually the recipient, but its opposite, since something is taken away from the referent of Oi. The present paper attempts to verify whether the semantic role of the indirect object may have an impact on the selection of the first or second passive. 3. ANALYSIS 3.1 DISTRIBUTION OF FIRST PASSIVE (INDIRECT OBJECT IN SUBJECT POSITION) VS. SECOND PASSIVE (DIRECT OBJECT IN SUBJECT POSITION) Looking at the overall results, Table 1 shows a slight preference of the first passive (54%) over the second, which is in accordance with the general assumption that the first passive is more common. However, if we look at the distribution of the first and second passive of each individual verb, the results are rather different. It seems that some verbs clearly tend to one type of the passive and it is probably not just a matter of FSP. This concerns the verbs promise, deny and buy. GIVE HAND PROMISE BUY DENY TOTAL % First passive 20 14 23 0 24 81 54% Second passive 10 16 7 30 6 69 46% Total 30 30 30 30 30 150 100% table 1: Distribution of first and second passive In the case of the verb promise the first passive (ex. 1) obviously prevails (23 instances), while the second passive is found only in seven instances (ex. 2); notably, in all seven cases the indirect object is omitted (mostly because a general human recipient is implied). Thus, the motivation for the choice of the second passive seems to be connected with object deletion (see section 3.4). The second verb that clearly tends to one type of passive is buy. Quite surprisingly, all 30 examples of the verb buy have the form of the second passive (i.e. the subject is the original Od), cf. (3). It should be pointed out that my results are in complete contradiction with Huddleston & Pullum (2002, 249), who point out that with the verbs alternating with the preposition for (i.e. verbs whose indirect object has the semantic role of beneficiary), “neither passive is completely acceptable, but to many speakers 30 LINGUISTICA PRAGENSIA 1/2015 the first passive would be possible” 2. Since the semantic role of Oi of the verb buy is a beneficiary (an intended recipient), it seems that it is the semantics of the indirect object that plays an important role in the choice of the passive. In addition, in all examples the indirect/prepositional object is omitted, so object deletion can be also viewed as an additional factor. The verb deny, on the other hand, prefers the first passive (ex. 4): 24 instances of the first passive against six instances of the second passive (ex. 5). The strong preference for the first passive appears to be associated again with the semantics (both of the Oi and the verb). It has been noted by Dušková (1988: 253) that a certain semantic class of verbs (namely, verbs of future not having) with the malefactive Oi prefer the first passive.3 (1) We have been promised around 80 000 tonnes. (2) This had been promised during Yeltsin’s visit to Germany in November. (3) Old and antique coins can be bought from specialist dealers like Spink and Son in London. (4) Meanwhile, opposition groups continued to protest that they were being denied access to the country’s media. (5) Conjugal rights were being denied him. As regards the object in the second passive, attention is also paid to the form of the object in the second passive, i.e. whether the recipient has the form of indirect or prepositional object. Our results imply that the prepositional object is far more common (ex. 6). The form of Oi is found only with three instances of deny; in all three cases the object is realized by a personal pronoun him (ex. 5) or them (ex. 7). This is in accordance with Quirk et al. (1985, 727), who point out that “instead of the retained indirect object […] the prepositional object is more usual.” (6) This form is in triplicate and must be handed to all employees leaving a firm. It shows their tax code number and details of their gross pay and tax deducted to date. (7) Similar fates befell pastoralists where large areas of seasonal pasture were denied them as in Kenya in both the Masai territory and Karamajong. 3.2 PRESENTATION VS. QUALITY SCALE Previous studies of presentation and quality scales have shown that in a majority of cases the passive implements the Quality Scale, i.e. it perspectives the sentence away from the subject (cf. Firbas 1992; Adam 2013; Dušková 1999; Dušková forthcoming). 2 Note that they mention a different verb, namely order (?Sue was ordered a copy. *A copy was ordered Sue). 3 Although Dušková (1988: 253) does not explicitly state that the preference for the first passive is associated with the semantics of the verb and indirect object, she provides an example with the verb charge, which is also one of the verbs of future not having. GABRIELA BRůHOVá 31 This has certainly been proved in my analysis, as the quality scale (ex. 8) accounts for 93% of all examples (see Table 2). (8) She feared John might not be at home but he was, and was very excited because he had been promised a job. However, Firbas (1992, 63–64) points out that even passives of ditransitive verbs are semantically disposed to perform the presentative function. My analysis has revealed eleven instances (i.e. 7%) of the presentation scale, i.e. the communication is perspectived towards the subject and it introduces a new phenomenon into discourse (exx. 9–11). Note that in ex. 10 the rhematic function of the subject is reinforced by the focalizer only. When determining the dynamic semantic function of the verb, particular attention was paid to the context-dependence / independence of the clause elements, as for a verb to perform the presentation function, the subject must be context-independent (then it has the function of a phenomenon presented). Thus, in the instances of P-Scale the subject is context-independent4 and introduces a new phenomenon into discourse (similarly to typical verbs of existence or appearance). The verb has the Pr-function and the Oi represents the Setting.5 Thus, the analysis has proved that even passives of ditransitive verbs may perform the presentative function, since the semantic component of some ditransitive verbs expresses or implies the appearance on the scene, and since a new element is brought on the scene through an unexpressed external agency (the by-agent is not expressed). Verb GIVE HAND PROMISE BUY DENY TOTAL % Quality Scale 1st passive 20 14 23 — 24 81 93% 2st passive 8 15 5 27 3 58 Presentation Scale 1st passive — — — — — — — 2st passive 2 1 2 3 3 11 7% Total 30 30 30 30 30 150 100% table 2: Presentation vs. Quality Scale (9) At common law no protection was given to those reporting speeches made in Parliament. (10) Only religious instruction was denied him. (11) I’d now like to talk about places where stamps can be bought. 4 Note that in the case of rhematic subjects, the use of the passive results in a deviation from the basic distribution of CD. 5 In prototypical Presentation sentences with intransitive verbs the setting is realized by an adverbial, as in A girl came into the room (Firbas 1966, 243). It has been pointed out by Adam (2012, 74) that in ditransitive constructions the setting is represented by the indirect object (it expresses the goal of the action). 32 LINGUISTICA PRAGENSIA 1/2015 From the syntactic point of view, it should be pointed out that in all eleven presentation sentences the postverbal complementation is relatively simple, i.e. there is usually only the indirect object, no adverbial, and the by-agent is not expressed (exx. 9,10,11). Occasionally, even the indirect object is omitted and hence, the setting is not expressed (ex. 11). Thus, the analysis has confirmed both Adam’s assumption that the presentation scale predominantly occurs in sentences where the verb has simple complementation6 and Firbas’s assumption that a presentation scale involving a passive verb form is mainly found in passive sentences without an expressed agent. The following example (ex. 12) can be seen as a sentence implementing a special type of presentation scale, namely the so called Extended Presentation Scale (cf. Chamonikolasová and Adam 2005, 62). The structure of the Extended Presentation Scale resembles the structure of the Presentation Scale, differing only in that it also contains a Specification. Thus, the sentence displays a “double rheme”, i.e. one constituted by the subject (i.e. the phenomenon presented) and the other by a specification (i.e. the adverbial) (12) With stocks running perilously low, the Czechoslovak government on Oct. 6 introduced a fuel rationing system whereby a maximum 25 litres could be bought at existing prices, with purchases over that limit charged at prices one-third higher (all petrol was to be charged at the higher price from Nov. 1). As follows from Table 2, the results of my analysis imply that the presentation scale is confined to the second passive only (where the subject corresponds to the original direct object, viz. it is the inanimate entity that takes the subject position). Nevertheless, our results should be viewed as tentative; it would require analysis of more examples to verify whether the presentation scale may also be implemented in the first passive. 3.3 EXPRESSION OF THE BY-AGENT Since the primary motivation for the use of the passive is to leave the agent unexpressed,7 it was expected that in a majority of cases the by-agent would be omitted. Quirk et al. (1985, 164ff.) even claim that “the omission of the by-agent has in fact been described as the unmarked state of affairs for the passive voice in general”. The assumption has been confirmed, as 89% of all examples represent agentless passives. The agent is expressed in 16 examples (11%) and in all 16 examples, the by-agent was invariably context-independent and thus performs the function of the rheme or its component, which demonstrates the primary function of the passive; namely, the passive enables to place the rhematic elements in the final position8 (ex. 13). 6 “[f]rom the point of view of syntax, the more complex is the valency of the verb, the more likely it will perform the Q-function, and vice-versa…” (Adam 2013, 165). 7 The reason for the suppression of the agent may be that “the agent is unknown, redundant, or irrelevant (i.e. of particularly low information value)” (Biber et al. 1999, 938) or a general human agent is implied. 8 According to previous studies context-dependent by-agents may also be found (cf. Dušková 2005b); however, they did not occur in the present analysis. GABRIELA BRůHOVá 33 (13) Nicola had whispered to several people that she had been promised Jane Pargeter’s job by Blufton. GIVE HAND PROMISE BUY DENY TOTAL % Agentive 0 3 3 10 0 16 11 Agentless 30 27 27 20 30 134 89 Total 30 30 30 30 30 150 100 table 3: The ratio of agentive vs agentless passives Example 14 demonstrates an unusual position of the by-agent, as it precedes the prepositional object. The ordering is most likely influenced by the degree of communicative dynamism of the postverbal elements. The by-agent is a component of the rheme and the object in the final position has the function of the rheme proper. (14) In the event, the problem of how to streamline Whitehall was handed by the electorate to that old Cabinet Office hand Harold Wilson. The results show that the verb that tends to express the by-agent most frequently is the verb buy (ex. 15). The by-agent is expressed in one third of all instances. The reason might be sought in the specificity of the by-agent. While with the other verbs (like give, hand) the agent is omitted because it is irrelevant, redundant or because a general human agent is implied, the agent of buy is often more specific and must be expressed; it functions either as the rheme proper or a component of the rheme. (15) En route, the slavers make an early sale when Ian and two other prisoners are bought by the master of a galley ship. 3.4 OBJECT DELETION The last aspect to be analysed is object deletion, i.e. omission of one of the objects. A previous study of mine (Brůhová 2012) has revealed that in the active the righthand participants (i.e. the objects) can be omitted, although semantically the ditransitive construction always includes three participants. The analysis of passive constructions has shown that an object can be omitted even in the passive.9 As Table 4 shows, object deletion was found only in the second passive (41 instances, 27%), where the indirect object (usually the animate recipient) is omitted. This is in accordance with the generally accepted assumption that indirect object is more easily omitted than direct object. The object of the passive verb is omitted under similar conditions as in the active, namely the referent of the object 9 It should be pointed out that I regard ditransitivity as a stable lexicosemantic property of the ditransitive verb, and suggest that these instances of object omission should also be considered as examples of ditransitivity. 34 LINGUISTICA PRAGENSIA 1/2015 is either recoverable from the context (contextual deletion) (ex. 16) or it refers to the general human recipient (indefinite deletion) (ex. 17). Object deletion can be regarded as an additional factor influencing the choice of a passive construction, as whenever the recipient/Oi is omitted, it triggers off the second passive. Verb GIVE HAND PROMISE BUY DENY TOTAL % 1st passive Indefinite deletion — — — — —— Contextual deletion — — — — —— 2st passive Indefinite deletion 1 — 5 20 — 26 63% Contextual deletion 2 — 2 10 1 15 37% Total 3 0 7 30 1 41 100% table 4: Indefinite vs Contextual deletion (16) Any notice required to be given under this Article by the Company to a Member or by a Member to the Company shall be given or served either personally or by sending it by first class post to the registered office of the Company. (17) Old and antique coins can be bought from specialist dealers like Spink and Son in London. It is worth noting that although the omission of the indirect object is possible, there are great differences in the behaviour of the five analysed verbs. The verbs allow object deletion to a different extent, e.g. while the verb hand revealed no instance of object deletion, with the verb buy the recipient (the Oi/Oprep) is omitted in all 30 examples. We have already seen that the verb buy behaves differently from the other verbs (see Section 3.3). Thus, while the object is in all cases omitted, the by-agent is very often expressed. An explanation for this may be seen in the fact that in most cases the Oi/Oprep (i.e. the recipient) and the by-agent are identical10 (exx. 15 and 18). In other words, it seems that in the examples where the expressed by-agent and the recipient would be identical, the Oi/Oprep is typically omitted because the referent can be recovered from the context (contextual deletion) and the by-agent expressed. (18) Church Times was bought by the Canterbury Press. 10 A similar situation can be found in the active, e.g. He bought (himself) a new car. The indirect object (the recipient) himself is usually omitted because it is identical with the subject he (the agent). GABRIELA BRůHOVá 35 4. CONCLUSIONS The present paper studied passivization of ditransitive complementation focusing on the frequency of the first and second passive and on the factors influencing the selection of the particular construction. The major factor in the selection of the passive construction appears to be FSP. The analysis has verified the assumption that in most instances the sentence is perspectived away from the subject and constitutes the quality scale. Nevertheless, we hope to have demonstrated that even ditransitive verbs in the passive may perform the presentative function; in other words, passives of ditransitive verbs are semantically disposed to perform the presentative function. For a verb to perform the presentative function the subject must be context-independent, as it introduces a new phenomenon into discourse; the verb has the Pr-function and the Oi, which is the goal of the action, represents the Setting. In all presentation sentences the by-agent was not expressed; thus, a new phenomenon appears on the scene through some unexpressed external agency. Besides FSP, the paper has identified a further factor that might play a role in the selection of the passive, semantics of the verb and of the indirect object. The analysis has shown that the verb of obtaining buy, whose indirect object has the semantic role of beneficiary, clearly prefers the second passive, while the verb of future not having deny, whose indirect object has the semantic role of malefactive, tends to take the second passive. The last factor that was considered is object deletion. Omission of an object was found only in the second passive, where the indirect object was omitted. Object deletion may be regarded as an additional factor, as the omission of the indirect object triggers off the second passive. REFERENCES Adam, M. (2013) Presentation Sentences (Syntax, Semantics and FSP). Brno: Masaryk University — Faculty of Education. Allerton, D. J. (1982) Valency and the English Verb. New York: Academic Press. Biber, D. et al. (1999) Longman Grammar of Spoken and Written English. London: Longman. Brůhová, G. (2010) Syntactic, semantic and FSP aspects of ditransitive complementation: a study of give, lend, send, offer and show. Ph.D. Thesis. Prague: Charles University. Brůhová, G. (2012) Object deletion in ditransitive verbs and its semantic consequences. In Malá,M. and P. Šaldová (eds) A Centenary of English Studies at Charles University: from Mathesius to Present-day Linguistics, 123–133. Faculty of Arts — Charles University in Prague. Brůhová, G. (2014) Ditransitive complementation from the FSP point of view. Prague Studies in English XXVI, Acta Universitatis Carolinae — Philologica 3/2013, 61–83. Prague: Charles University in Prague — The Karolinum Press. Chamonikolasová, J., Adam, M. (2005) The presentation scale in the theory of functional sentence perspective. In Čermák, J. et al. (eds) Patterns — A Festschrift for Libuše Dušková, 59–69. Faculty of Arts — Charles University in Prague. Dušková, L. (1999a) Basic distribution of communicative dynamism vs. nonlinear indication of functional sentence perspective. Travaux du Cercle linguistique de Prague. NS. Prague Linguistic Circle Papers 3. 249–261. Ed.