The development of and factors influencing double object construction preference of the ditransitive verbs envy and forgive

The development of and factors influencing double object construction preference of the ditransitive verbs envy and forgive Veronika Hlaváčková (Charles University, Prague)— Gabriela Brůhová (Charles University, Prague) VERONIKA HLAVÁČKOVÁ — GABRIELA BRŮHOVÁ ABSTRACT The paper investigates the double object constructions, viz. SVOiOd and SVOOprep clause patterns, of the ditransitive verbs envy and forgive. The syntactic and semantic specificity of the two verbs in question may indicate apossible future extinction of their ditransitive use. The present study aims to provide an extensive quantitative and qualitative analysis of the double object construction preference from both the diachronic and synchronic perspectives. Using acorpus sample of American English (COHA), the data reveal acomplex situation. While the double object constructions with envy prefer the indirect object clause pattern, there is anotable tendency of such constructions to gradually decline in frequency and give rise to the prepositional pattern. Forgive shows preference for the SVOOprep pattern. Nevertheless, it is the form of the objects that seems to play asignificant role in the double object construction preference. KEY WORDS argument structure, ditransitive verb, double object construction, object realization, syntactic change DOI https://doi.org/10.14712/18059635.2022.2.2 1 INTRODUCTION The present paper explores the development of object complementation of the ditransitive verbs envy and forgive. The verbs envy and forgive represent avery specific group of ditransitive verbs, displaying considerable differences from the prototypical ditransitive verbs. It is both the syntactic and semantic characteristics that differentiate the verbs envy and forgive from the ditransitive verbs such as give, send or offer, where the indirect object alternates with the prepositional object, and where the two objects exchange their position (exx 1 aand b). (1) a. She sent Paul [Oi] apresent [Od]. b. She sent apresent [Od] to Paul [Oprep]. (Quirk et al. 1985: 1211) The verbs envy and forgive also occur in two different clause patterns, viz. SVOiOd (exx 2 a, 3 a) and SVOOprep (exx 2 b, 3 b)1. However, the alternation of the two avail1 Although the term “double object construction” is introduced by Barss and Lasnik (1986) as referring strictly to the indirect object ditransitive pattern, for the purpose of the presOPEN ACCESS VERONIKA HLAVáČKOVá — GABRIELA BRůHOVá 199 able clause structures does not involve re-ordering of the objects, and their semantics substantially differs from the central sense of prototypical ditransitive verbs. (2) a. Ienvied him his freedom. b. Ienvied him for his freedom. (Huddleston and Pullum 2002: 297) (3) a. Ican’t forgive him his lies. b. Ican’t forgive him for his lies. (ibid.: 312) The specific nature of envy and forgive attracted the attention of several linguists, e.g. Goldberg (1995), Hunston and Francis (2000) or Colleman and De Clerck (2008, 2011), on whose 2008 study the present article is based. Consequently, the aim of this paper is to examine the postverbal preference in the double object constructions with envy and forgive, both from the diachronic and synchronic perspectives. The investigation concerns the development of the postverbal preference, and at the same time, it attempts to identify and elucidate other factors that might influence the preference of one or the other pattern, regardless of their time of origin. 2 THEORETICAL BACKGROUND 2.1 SYNTACTIC VIEWPOINT: THE TREATMENT OF THE SVOIOD AND SVOOPREP CLAUSE PATTERNS IN MAJOR ENGLISH GRAMMARS The present chapter summarises different approaches to ditransitive verbs in English major grammar books from the syntactic point of view in order to show the problematic classification of the verbs envy and forgive in terms of double object constructions, i.e. SVOiOd and SVOOprep. AComprehensive Grammar of the English Language by Quirk et al. (1985) “is most liberal in that it subsumes various types of formal realizations under this category and various verbs, even if they occur only in the prepositional construction” (Brůhová 2010: 19). Using the label “ditransitive complementation” for both structures, Quirk et al. (1985) further divide the class into two subcategories— clause patterns including either indirect object and direct object, or direct/indirect object and prepositional object (ibid.: 726–727). Ditransitive verbs are consequently categorised with respect to their possible complementation configurations (see Quirk et al. 1985: 1209–1210). Contrary to Quirk et al. (1985), Dušková et al. (2012) provide each sentence type with its own separate category and distinguish verbs that enter the SVOiOd pattern from those that do not, and therefore require the presence of apreposition (ibid.: 12.22.3, 12.22.4). Biber et al. (2021) include the verbs that allow at least one of the two double object sentence structures in one category, specifically calling them three-place verbs. Regarding ditransitivity, the authors note that “most ditransitive verbs also have ditransitive prepositional uses” (ibid.: 154). ent study, the label “double object construction” serves as an umbrella term that captures both types of verbal complementation involving two objects, viz. the SVOiOd and SVOOprep patterns. OPEN ACCESS 200 LINGUISTICA PRAGENSIA 2/2022 In their opinion, for averb to be considered ditransitive, it must be attested in the SVOiOd pattern, while verbs which occur only in the SVOdOprep pattern (e.g. remind sb of sth) are not considered ditransitive (ibid.). Huddleston and Pullum (2002), on the other hand, adopt aconsiderably different approach, as “[they] are the most strict and their group of verbs is the most limited” (Brůhová 2010: 19). Only those constructions are treated as ditransitive in which both objects are realized by noun phrases, i.e. core elements (see Huddleston and Pullum 2002: 216). As aresult, verbs occurring in the SVOdOprep pattern are considered monotransitive2 (ex 4). (4) Kim gave the key to Pat. (ibid.) It is noteworthy that while the presence of two objects would be sufficient for Quirk et al. (1985) to call the use of the verb ditransitive, it is the existence of its corresponding indirect object construction, i.e. Kim gave Pat the key (Huddleston and Pullum 2002: 216) that enables Biber et al. (2021) to regard ex 4 as ditransitive. Therefore, as far as the verbs envy and forgive are concerned, Quirk et al. (1985) and Biber et al. (2021) view their prepositional use as ditransitive, while Huddleston and Pullum (2002) regard the cases of envy and forgive with the preposition for as instances of monotransitive use. In the present study, we adhere to the interpretation of the SVOOprep pattern as proposed by Quirk et al. (1985), and thus consider the prepositional pattern ditransitive. 2.2 SEMANTIC VIEWPOINT: SEMANTIC CLASSIFICATION OF DITRANSITIVE VERBS Pinker (1989) establishes the thematic core that defines the ditransitive construction as “X causes Yto have Z” (ibid.: 110) and briefly comments on several verb classes that occur in the SVOiOd pattern. Naturally, the central position is occupied by “verbs of giving” (e.g. give, pay) and “verbs of sending” (e.g. send (ex 1 a), ship) that, except for the thematic core, also imply the means of transfer. Other verbal subdivisions are called “verbs of instantaneous causation of motion” (e.g. throw, toss), “illocutionary verbs of communication” (e.g. tell, write), including those that “[specify] an instrument of communication” (e.g. radio, telephone), “verbs of creation” (e.g. bake, sew), “verbs of obtaining” (e.g. get, steal), “verbs of future having” (e.g. offer, promise), and last, but in this case certainly not least, “verbs of future not having” (e.g. cost, refuse). The last-mentioned category is aconsiderably heterogeneous group in which one may also find the verbs envy and forgive (ibid.: 110–119)3. Hunston and Francis (2000: 89), on the other hand, subsume envy and forgive into 2 For amore detailed description of “core” and “non-core” elements see Huddleston and Pullum (2002: 216). 3 The thematic core defining the ditransitive construction does not necessarily imply the occurrence of arespective verb in the SVOiOd structure. Pinker emphasises the existence of verbs, such as donate, transport, release, admit, create, obtain, shout, entrust or select, whose meaning corresponds to the “X causes Yto have Z” sense but such verbs fail to dativize (Pinker 1989: 119). OPEN ACCESS VERONIKA HLAVáČKOVá — GABRIELA BRůHOVá 201 aclass of ditransitive verbs which they label “verbs concerned with feelings and attitudes”4. It should be pointed out that the inclusion of envy and forgive into ditransitive verbs is not generally agreed upon. Several linguists consider the ditransitive use of envy and forgive idiosyncratic5. Goldberg (1995), as one of the most prominent figures of the Construction Grammar, introduces the so-called “constructional polysemy”. Based on this terminology, constructions are associated with acentral meaning and other extended, related meanings. The central meaning of the ditransitive construction is defined by “agent [that] successfully causes recipient to receive patient” (ibid.: 32, 38). Nevertheless, none of the related meanings (see Goldberg 1995: 38) include envy or forgive. In Goldberg’s account, the two verbs in question are regarded as “positive exceptions to the semantic generalization” (ibid.: 133)6. (5) He forgave her her sins. (ibid.: 132) (6) He envied the prince his fortune. (ibid.) In exx 5 and 6 the transfer interpretation is not applicable, as “the subjects in these cases are not causal, and no reception is involved” (ibid.: 131–2). Citing their respective entries in the OED, Goldberg suggests that their syntactic association to prototypical ditransitive verbs relates to their former subsenses whose meaning was that of the central sense of the ditransitive verbs. Thus, it seems that the data suggest aditransitive use that has remained “frozen” even after the respective semantic changes (ibid.). In addition, Goldberg notes that the presence of apreposition in the SVOOprep structure serves as aclearer marker of the syntactic function and semantic role of the object. Goldberg claims that “it seems reasonable that syntactic change should tend toward patterns that are more transparent to the speaker” (ibid.: 132). As a result, “it would be natural for odd cases of ditransitives involving forgive and envy to drop out of use” (ibid.). She argues that the degree to which she and her students accept the ditransitive use of envy and forgive is rather low. She finds archaic-sounding sentences with forgive and envy (exx 7 a and 8 a) much more acceptable than modern-sounding sentences (exx 7 b and 8 b), while none of the sentences appear to be acceptable for her students. 4 For additional verb classes see Hunston and Francis (2000: 88–89). 5 The treatment of envy and forgive in the SVOiOd clause pattern is considerably heterogeneous. In some cases, the two verbs in question are not mentioned at all (see Wierzbicka 1988). In the canonical work of lexicalists, the semantic load of the indirect object pattern is disregarded in favour of an exhaustive outline that encapsulates all possible clause pattern configurations of each respective verb (see Levin 1993). 6 Croft (2003: 58) also argues for the idiosyncratic reading of the ditransitive use of the verbs envy and forgive, recognising such constructions as verb-specific constructions rather than verb-class-specific constructions that comprise prototypical verbs associated with the sense of giving. OPEN ACCESS 202 LINGUISTICA PRAGENSIA 2/2022 (7) a. She forgave him his sins. b. ?*She forgave him his goof. (ibid.) (8) a. She envied him his vast fortune. b. ?*She envied him his extensive stock portfolio. (ibid.) Although Goldberg admits that the current ditransitive use of envy and forgive suggests the possibility of speakers learning the idiosyncratic construction “on an instance-by-instance basis as idioms”, she expects the verbs envy and forgive to cease to appear in the SVOiOd clause pattern. 2.3 LAYERING, THE DEVELOPMENT OF DOUBLE OBJECT CONSTRUCTION PREFERENCE AND THE DITRANSITIVE STATUS OF ENVY AND FORGIVE When two constructions of similar meaning co-exist, such asituation can be described as layering, aterm which may also be applied to double object constructions (Hopper 1991: 22). Since the indirect object ditransitive construction was awell-established pattern in the period of Old English, while the more recent SVOOprep clause structure was not as frequent at the time, the subsequent loss of endings, which English underwent, has put the prepositional pattern into the foreground (Bybee 2015: 173–4). Colleman and De Clerck (2011) explore the later stages of this development, specifically the indirect object ditransitive construction that seems to gradually limit its use to fewer verbs. Colleman and De Clerck note that although the range of verbs operating as ditransitives has been recently enriched by verbs of communication that specify the necessary instrument, such as radio or fax, the set of verbs allowing the ditransitive complementation has been considerably narrowed down during the last 300 years. Verbs which no longer enter the SVOiOd clause pattern include the following verb classes: “verbs of banishment” (e.g. banish, discharge), “verbs of ‘pure benefaction’” (e.g. open) and several “communication verbs” whose semantic load involves the manner of communication (e.g. shout, mumble). Additionally, special attention is given to “verbs concerned with feelings and attitudes” (e.g. begrudge, excuse, envy, forgive), which, although not as frequently as they used to do, still show signs of grammatical acceptability in the ditransitive use (ibid.: 190–199). The continuous decline of the verbs envy and forgive occurring in the indirect object ditransitive structure is explored in the study by Colleman and De Clerck (2008), analysing the data from the Corpus of Late Modern English Texts and the British National Corpus. Establishing four distinct periods, Colleman and De Clerck find asignificant drop in the SVOiOd use of the verb forgive over all four periods, while such atendency with envy is found only in the first three periods. Nevertheless, “their occurrence cannot (yet) be labelled as near-obsolete: at least in written British English, they still occur with arespectable frequency” (ibid.: 210). What is also necessary to emphasise is the fact that Colleman and De Clerck’s findings have revealed that the frequency of verbs such as refuse and order, which are traditionally included in the accounts of ditransitive verbs, is lower than that of envy or forgive. (9) Further back, Judie had envied Anne her college boyfriend. (ibid.: 197) (10) ‘Oh, shut up, Jay, forgive amother her blind spots.’ (ibid.: 198) OPEN ACCESS VERONIKA HLAVáČKOVá — GABRIELA BRůHOVá 203 Furthermore, the authors note that there seems to be no indication that the ditransitive use is restricted to only archaic-sounding direct objects (see exx 9 and 10), thus disputing Goldberg’s prediction of near extinction as well as the assertion of lexical limitations regarding envy and forgive (ibid.: 190–210). Goldberg’s etymological elucidation regarding the ‘frozen’ ditransitive use of envy and forgive is also challenged. Croft (2012: 389) argues against such interpretation, claiming that the loss of the ditransitive meaning should have hindered the ongoing existence of the construction and made the ditransitive use grammatically unacceptable. Afurther argument against the ‘frozen’ interpretation lies in cross-linguistic evidence that confirms the possibility of envy and forgive occurring in the SVOiOd clause pattern with no trace of aformer subsense related to the act of giving (Colleman and De Clerck 2008: 199–200). To exemplify, Czech ditransitive construction accepts not only verbs denoting the transfer of possession but also verbs that “express an action that is directed towards someone, whether the recipient is aware of such action or not” (Hlaváčková 20217: 31), such as věřit (‘believe’), ukrást (‘steal’) and Czech equivalents of envy and forgive (Dvořák 2017). The postverbal complementation of envy and forgive, which refer to mental actions, is comprised of obligatory arguments: the addressee and the patient (Lopatková et al. 2020). Additionally, Colleman and De Clerck (2008) provide an exhaustive outline of available semantic extensions based on the multidimensional approach that tie the semantic load of envy and forgive together with other ditransitive verbs. “[W]eshall argue that ditransitive clauses involving envy and forgive instantiate such acombination of extensions from several dimensions, too. They can be described as combining (i) ametaphorical extension from material to abstract transfers with (ii) ashift in direction from atransfer towards the indirect object to atransfer away from the indirect object and/or (iii) an extension from the actual causation of apossessional transfer to an attitude towards such atransfer.” (Colleman and De Clerck 2008: 202) Regarding the first point, i.e. “the metaphorical extension from material to abstract transfers”, the data have revealed that although both abstract and concrete entities may occur in the position of direct object in the valency frame of envy, the verb forgive is associated with abstract direct objects only (ibid.: 202–203). The second point concerning the reversal of transfer direction, in which “the indirect object referent [acts] as its source rather as its target” (ibid.) can be already inferred from Pinker’s “verbs of future not having” or Goldberg’s later introduction of averb class that expresses arelated ditransitive meaning “X causes Yto lose Z” and contains, for instance, the verb cost (Goldberg 2002: 333). Colleman and De Clerck find that the meaning of envy also includes the subject’s wish to become the possessor of the affected object; thus, the subject becomes the presupposed target of the transfer. Forgive exhibits bidirectionality; “the indirect object refer7 The main source for the present paper is the unpublished diploma thesis Complementation of the ditransitive verbs envy and forgive (Hlaváčková 2021). OPEN ACCESS 204 LINGUISTICA PRAGENSIA 2/2022 ent ‘receives’ forgiveness and thereby ‘loses’ its burden” (Colleman and De Clerck 2008: 205–206). Finally, the last point elaborates on the established verb categories allowing the ditransitive use (specifically those listed by Goldberg), claiming that the attitudinal aspect of the verbs envy and forgive, i.e. a“feeling or attitude towards an (actual or potential) possessive relationship between indirect and direct object” (Colleman and De Clerck 2008: 206), should not be ahindrance for their ditransitive interpretation. Envy, on the one hand, demonstrates anegative attitude, since the subject referent wishes either for the possessor to lose the patient referent or to become the possessor themself, whereas forgive can be considered positive in terms of its attitudinal aspect (ibid.). The present paper also treats the verb envy and forgive as ditransitive due to their occurrence in the SVOiOd pattern and the fact that even when one of the objects is omitted, it is always recoverable from the context. Additionally, the historical development discussed above is in accordance with such interpretation. 3 MATERIAL AND METHODOLOGY The material for the present analysis was drawn from the Corpus of Historical American English (COHA). With its size nearing 500 million tokens, the corpus provides an extensive selection of more than 100 000 texts of four distinct genres— fiction and TV/movies, magazine, newspaper and non-fiction— that demonstrates the use of language from the 1820s-2010s period (Davies 2010). For the purpose of this study, the 200-year-long span is further divided into four distinct 50-year-long periods to enable mapping of ongoing diachronic processes. The retrieved sample thus contains 400 instances of envy and forgive used in one of the double object construction patterns, i.e. 200 instances of each verb from which 50 instances represent one of the established periods. Due to query limitations of the COHA interface, it was necessary to collect all instances of the lemmas envy and forgive, randomise the sample in R Studio8 and select relevant data, viz. the SVOiOd or SVOOprep clause structures with envy and forgive (see exx 2 and 3), manually. The in-depth analysis is partially based on Colleman and De Clerck’s (2008) study, as it focuses on the diachronic development of the two investigated verbs, but synchronic aspects are also taken into account. Not only does the present study investigate the ratio between the two available clause patterns and how it changes over time but also identifies various factors that may play acrucial role for the speaker to opt for one of the double object constructions, specifically the formal aspect of objects complementing the verbal event. Therefore, the analysis is both quantitative and qualitative. Since the present analysis investigates the two competing patterns, i.e. SVOiOd and SVOOprep, of the verbs envy and forgive, we focus only on those sentences where both objects are overtly expressed and which allow the alternation with the other type of the double object construction. In other words, we excluded those instances which do not permit either the insertion or deletion of the preposition for. Thus, SVOiOd 8 With the seed set to ‘42’. OPEN ACCESS VERONIKA HLAVáČKOVá — GABRIELA BRůHOVá 205 clause structures that involve that-clauses in the position of direct object have been removed from the analysis, as they cannot be transformed into the SVOOprep pattern (ex 11). Similarly, instances of objects in the preverbal positions are also excluded to offer amore cohesive and restricted study that is liberated from the possibility of other operating factors as much as possible (ex 12). The aforementioned criterion led to the exclusion of sentences where one of the objects was omitted (exx 13 and 14), as they provide no elucidation on the double object construction preference. It is worth noting that instances with one of the objects omitted were found to be far more frequent than the double object constructions in which both objects are explicitly expressed. The omitted participant is usually derivable from the context. (11) They envied Dolly that her husband loved her so fiercely after four years of married life, and no children either.9 (12) Iwas once old in sin, for which God forgive me! (13) You are wrong, you are unkind; but you love me, and Iforgive you. (14) […] Iregretted my lost youth when Ionly envy the delights of losing it. The greatest complication arose with instances that showed signs of syntactic ambiguity. Those involve agerund taking the position of the patient (ex 15). In such cases, the object (you in ex 15) can be regarded either as “the agent of the gerund action (thus implying the S-V-Opattern) or as the recipient of the main clause verbal event” (Hlaváčková 2021: 50). To complicate the matter further, if the position of the indirect object is occupied by the 3rd person singular feminine her (ex 16), the pronoun “can be either considered apossessive personal pronoun or the dative case of the pronoun she” (ibid.). Nevertheless, both readings are possible, and it would seem ill-advised to disregard instances of this kind. Additionally, Dušková (1991: 66) introduces apo koinou constructions, in which “acoreferentially identical participant operates in two propositions in different semantic roles, e.g. He saw Charles coming (He saw Charles + Charles was coming)” (ibid.). Such interpretation is also taken into consideration. As aresult, the verbal complementation of envy and forgive by agerund is incorporated into the analysis. (15) Ican not forgive you letting me carry all that water for afainting fit— and there was no fainting fit! (16) “First, Tom,” he pursues, “be to yourself afriend; second, forget the error of your mother, and forgive her sending you here; […].” The last point concerns isolated uses of double object constructions and fixed expressions, both of which were not included in the present study. The data have revealed aminor but notable existence of SVOOprep structures that exchange the position of the object and make use of prepositions such as about or in to introduce the recipient (ex 17). It is also fixed phrases that were not taken into consideration (specifically those 9 Unless indicated otherwise, all following examples are taken from the COHA. OPEN ACCESS 206 LINGUISTICA PRAGENSIA 2/2022 with direct object expressed by sins/trespasses/debts, ex 18), as their repetitive nature would impact the sample under investigation and provide little to no information on syntactic preference from the diachronic perspective. (17) What could abigwig bishop, albeit his oldest friend, envy in acountry parson? (18) Forgive us our sins as we forgive those who sin against us. Regarding methodology, each relevant instance was categorised based on the postverbal complementation, i.e. SVOiOd or SVOOprep. We assessed the double object construction preference and the way it changes over time with respect to the four individual periods; any shifts in frequency were considered markers of apossible syntactic change. The analysis took into account not only the diachronic aspect of the postverbal preference of envy and forgive but also possible syntactic factors affecting the choice of the two patterns. Therefore, we determined means of expressing the objects and identified semantic features associated with the respective object referents in attempt to find significant relations between the available variables. In the case of notable correlations, we conducted the chi-square test to determine the degree of significance. 4 ANALYSIS 4.1 DIACHRONIC VIEWPOINT: DOUBLE OBJECT CONSTRUCTIONS WITH ENVY AND FORGIVE AND THE COMPLEMENTATION PREFERENCE The instances of envy in the COHA have brought about several interesting findings. Overall, there is anotable preference for the indirect object ditransitive structure (91% of instances) over the prepositional one (9% of instances). Nevertheless, amore detailed investigation reveals that such preference is undergoing achange as time progresses. 1820–1869 1870–1919 1920–1969 1970–2019 Total SVOiOd49 24.5% 49 24.5% 45 22.5% 39 19.5% 182 91% SVOOprep 1 0.5% 1 0.5% 5 2.5% 11 5.5% 18 9% Total 50 25% 50 25% 50 25% 50 25% 200 100% Table 1:Number of occurrences of the SVOiOd or the SVOOprep patterns with envy attested in the four established periods As is evident from Table 1, the first half of the inspected period is determined by an almost strictly exclusive inclination to the SVOiOd pattern. But it is the latter stages that show the gradual tendency towards of the SVOOprep clause structure. Therefore, the data suggest an ongoing syntactic change that is progressively altering the complementation preference of the verb envy. The use of forgive in double object constructions reflects arather contrary trend (see Table 2). First, the majority of instances involve the preposition for (72% of inOPEN ACCESS VERONIKA HLAVáČKOVá — GABRIELA BRůHOVá 213 Goldberg, A.E. (1995) Constructions: AConstruction Grammar Approach to Argument Structure. Chicago: The University of Chicago Press. Goldberg, A.E. (2002) Surface Generalizations: An Alternative to Alternations. Cognitive Linguistics 13/4, 327–356. Available at https://www.researchgate.net/ publication/228880162 [last accessed 30July 2021]. Hlaváčková, V. (2021) Complementation of the ditransitive verbs envy and forgive. [Unpublished diploma thesis]. Prague: Charles University. Hopper, P.J. (1991) On some principles of grammatizication. In: Traugott, E.C., and B.Heine (eds) Approaches to Grammaticalization, Volume I.Theoretical and methodological issues, 17–36. Amsterdam: John Benjamins. Huddleston, R. and G.K.Pullum (2002) The Cambridge Grammar of the English Language. Cambridge: Cambridge University Press. Hunston, S. and G.Francis (2000) Pattern Grammar: ACorpus-Driven Approach to the Lexical Grammar of English. Amsterdam/ Philadelphia: John Benjamins. Levin, B. (1993) English Verb Classes and Alternations: Apreliminary investigation. Chicago: The University of Chicago Press. Lopatková, M., V.Kettnerová, A.Vernerová, E.Bejček, and Z.Žabokrtský (2020): VALLEX 4.0. LINDAT/CLARIAH-CZ Digital Library at the Institute of Formal and Applied Linguistics, Faculty of Mathematics and Physics, Charles University. Available at https://hdl.handle.net/11234/1-3524 [last accessed 30July 2021] Pinker, S. (1989). Learnability and Cognition: The acquisition of argument structure. Cambridge, Mass.: The MIT Press. Quirk R., S.Greenbaum, G.Leech and J.Svartvik (1985) AComprehensive Grammar of the English Language. London: Longman. Wierzbicka, A. (1988) The Semantics of Grammar. Amsterdam: John Benjamins. Veronika Hlaváčková Department of English Language and ELT Methodology Faculty of Arts, Charles University Nám. Jana Palacha 2, 11638 Praha ORCID ID: 0000-0002-7169-8421 [email protected] Gabriela Brůhová Department of English Language and ELT Methodology Faculty of Arts, Charles University Nám. Jana Palacha 2, 11638 Praha ORCID ID: 0000-0001-5014-7965 [email protected] OPEN ACCESS