1 The present paper goes back to a talk at the conference on Directions in English Language Studies (DELS) in Manchester in April 2006. We would like to thank the audience for a stimulating discussion. We are also grateful for the helpful comments provided by an anonymous reviewer. AAA - Arbeiten aus Anglistik und Amerikanistik Band 32 (2007) Heft 1 Gunter Narr Verlag Tübingen Ditransitive verbs in Indian English and British English: a corpus-linguistic study Sebastian Hoffmann and Joybrato Mukherjee Verb complementation has long been neglected as an area in which secondlanguage varieties of English deviate from native Englishes. The present paper focuses on Indian English as the largest institutionalised second-language variety of English and investigates differences between Indian English and British English at the level of ditransitive verbs and ditransitive verb complementation. By using various corpora, including large databases obtained from the World Wide Web, we show (1) that ditransitive verbs like GIVE are associated to different extents with individual complementation patterns in present-day Indian and British English, (2) that the range of verbs used in the basic ditransitive pattern (with two object noun phrases) is different between present-day Indian and British English, and (3) that the “new ditransitives” in Indian English do not represent cases of superstrate retention but rather genuinely innovative forms that Indian English users create on grounds of analogy. From a theoretical perspective, we argue that the concept of verbcomplementational profile is a useful framework for comparative studies of varieties of English. 1 1. Introduction: verb complementation in Indian English and British English The emergence of New Englishes around the world has led to an everincreasing interest in the description of differences between varieties of English at all linguistic levels. The largest institutionalised second-language Sebastian Hoffmann / Joybrato Mukherjee 6 2 Note, for example, that Mukherjee and Hoffmann (2006) observe that convey does not occur at all in the double-object construction in a 31-million-word corpus of Indian English newspaper texts, which raises the question whether this syntactic variant is really preferred by Indian users of English. variety of English is Indian English, the educated standard variant of which is regularly and competently used by at least 35 to 50 million speakers (cf. e.g. Crystal 2003). Research into Standard Indian English has spawned a vast literature over the past decades, including the work by Kachru (1983, 2005), Shastri (1988, 1992) and Mehrotra (1998). Yet, apart from some notable exceptions such as Shastri (1996), verb complementation in Indian English has so far been relatively neglected as a distinctive area of variety-specific norm formations. Some isolated observations on deviant verb-complementational patterns in Indian English are listed in the second edition of Nihalani et al.’s (2004) Indian English dictionary. For example, Nihalani et al. (2004: 55) note that convey tends to be complemented with an object noun phrase and a tophrase in British English (e.g. Please convey my best wishes to him) whereas the double-object construction is acceptable as well in Indian English (e.g. Please convey him my best wishes). However, the descriptions offered by Nihalani et al. (2004) are entirely based on personal observations rather than on representative corpus data and therefore have to be taken with a measure of caution. 2 Olavarría de Ersson and Shaw (2003), in contrast, have provided an empirically sound and systematic pilot study of the complementation patterns of provide, supply, pelt and other semantically related verbs in British and Indian English by using large web-derived databases, including three on-line Indian newspapers. They find out, for example, that pelt tends to be complemented with an object noun phrase and a withphrase in British English (e.g. they are pelting him with cans), while Indian English speakers prefer a reverse order of postverbal elements with an atphrase (e.g. they are pelting cans at him). Olavarría de Ersson and Shaw’s (2003) observations corroborate that verb-complementation is one of the structural features at the lexis-grammar interface that are significant in shaping local lexicogrammatical norms in varieties of English. Both at the descriptive and the methodological level, their study inspired us to compare ditransitive verbs and ditransitive complementation patterns in British and Indian English by using corpus data and web-derived data (cf. Mukherjee and Hoffmann 2006). Specifically, we not only compared the 1-million-word Indian and British components of the International Corpus of English (ICE; cf. Greenbaum 1996), but we also generated and analysed a large corpus of Indian English newspaper language, containing 31 million words from the national newspaper The Statesman. In this context, we introduced the concept of verb-complementational profile, which we believe is a useful frame- Ditransitive verbs in Indian English 7 work for the systematisation of differences in verb complementation between varieties of English. As visualised in Figure 1, the concept of verbcomplementational profile includes two related aspects: a) the range (and frequencies) of the patterns of an individual verb in a variety; b) the range of verbs with which an individual pattern is associated in a variety. individual verbs: GIVE individual patterns: S V [O i : NP] [O d : NP] basic ditransitive pattern other patterns Figure 1: Two complementary aspects of a verb-complementational profile - focus on ditransitive verbs and ditransitive verb complementation (Mukherjee and Hoffmann 2006: 149). Figure 1 visualises the concept of verb-complementational profile by focusing on ditransitive verbs and ditransitive verb complementation. For example, give is typically used in the basic ditransitive pattern with both objects realised as object noun phrases (e.g. John gave Mary a book), but it can also be used in other patterns, e.g. the to-phrase variant (e.g. John gave a book to Mary) and passive constructions (e.g. Mary was given a book [by John]). From a complementary perspective, each individual pattern can be used with specific verbs in a given variety. For example, while provide is not acceptable in the basic ditransitive pattern in British English, it is admissible - and frequently used - in this pattern in American English. We believe that all varieties of English are potentially characterised by differences between their verb-complementational profiles with regard to individual verbs and/ or entire verb classes. In the present paper, we start off from some of the findings on which we have reported elsewhere (cf. Mukherjee and Hoffmann 2006) and then go on to address the question as to whether the verb-complementational differences between present-day British English and Indian English can be explained by assuming a retention of “old” British English structures in Indian English, given that historically British English of the 18th and 19th century functioned as the actual input variety for the shaping of Indian English. This is reflected, for example, in a tendency for Indian English speakers to use lexical items that have a formal and/ or old-fashioned flavour in present-day British English (e.g. miscreant, thrice). Thus, we were interested in finding Sebastian Hoffmann / Joybrato Mukherjee 8 3 Examples (1) to (4) are taken from the 1-million-word British component of the International Corpus of English (ICE-GB). The pattern exemplified in (5) is a rare pattern that could only be found in a much larger corpus, the 100-million word British National Corpus (BNC). out whether superstrate retention is also a useful explicans for verbcomplementational differences between present-day British English and Indian English, or whether alternative explanations seem more plausible. The plan of the present paper is as follows. In section 2, we will briefly introduce the system of ditransitive complementation patterns that is at the heart of our syntactic analysis. We will then show that the verbcomplementational profile of Indian English diverges from British English with regard to ditransitive verbs: on the one hand, there are clear quantitative differences in the complementation of established ditransitive verbs like give; on the other hand, individual complementation patterns like the basic ditransitive pattern are used with a different range of verbs. In section 3, we will focus on the basic ditransitive pattern and discuss to what extent the “new ditransitives” that we find in present-day Indian English are truly new. In this context, we will combine the synchronic focus on present-day Indian and British English with a diachronic analysis of old and new ditransitives in the history of British English and report on the findings we have obtained from various synchronic and diachronic corpora. Finally, in section 4 we will offer some concluding remarks. We will argue in particular that the innovative extension of the basic ditransitive pattern to new verbs has always been a creative process in the history of native varieties of English, and that this creativity is now also exerted by non-native speakers of institutionalised second-language varieties like Indian English. 2. Differences between the verb-complementational profiles of Indian English and British English: focus on ditransitive verb complementation In a previous large-scale study of the British component of the International Corpus of English (ICE-GB), Mukherjee (2005) has provided a detailed quantitative and qualitative analysis of the complementation of ditransitive verbs in present-day British English. The analysis is based on a categorisation of all patterns with which a given ditransitive verb is attested into five types. These five types are described and exemplified in (1) to (5) by instances of the most frequent ditransitive verb in English, i.e. give. 3 Ditransitive verbs in Indian English 9 4 Note in this context that at this explorative stage of our strictly syntactic pattern analysis we do not distinguish between ‘free’ combinations of arguments (e.g. type I: John gave Mary a book; type II: John gave the book to Mary; type III: Give a reason) and pre-constructed and/ or idiomatic units (e.g. type I: John gave it a try; type II: The story gave rise to new (1a) I (S) GIVE [O i : NP] [O d : NP] (1b) On Tuesday members of Parliament gave the government their overwhelming support <ICE-GB: S2B-030 #54> (2a) II (S) GIVE [O d : NP] [O i : PP to ] (2b) I meant to give it to you earlier <ICE-GB: S1A-022 #176> (3a) III (S) GIVE [O d : NP] O i (3b) he wanted physical love and I couldn’t give that <ICE-GB: S1A-050 #184> (4a) IV (S) GIVE O i O d (4b) The other major point he raises is in addressing the question of ‘why give in the first place? ’ <ICE-GB: W1A-011 #94> (5a) V (S) GIVE [O i : NP] O d (5b) … well she did […] she did give Mrs <anonymised_last_name> didn’t she? <BNC: KE6 2368-70> Type I is the most basic type of ditransitive complementation with both objects realised as noun phrases. Type II is the pattern in which the indirect object is realised as a to-phrase and placed after the direct object. In type III, the indirect object is not made explicit, in type IV both objects are deleted, and in type V it is the direct object that is not made explicit. Our system of five types is based on the assumption that even in those instances in which one or two object(s) are not made explicit, i.e. in types III to V, the verb give remains associated at the cognitive level with the ditransitive meaning ‘X causes Y to receive Z’ (X = agent; Y = recipient; Z = patient) in all instances and patterns (cf. Mukherjee and Hoffmann 2006: 151). Note that from each of the five basic patterns, it is possible to derive many structurally related patterns, e.g. patterns with specific elements in fronted positions, relative clause structures, participle constructions and passive constructions (e.g. type IP as the passive construction derived from the type-I pattern). A full list of the patterns of give is given in the Appendix. 4 Sebastian Hoffmann / Joybrato Mukherjee 10 speculations; type III: I don’t give a damn) within any one complementation type. It remains to be seen in future research whether differences in idiomatic usages of specific complementation types might be at the heart of divergent verb-complementational profiles of British and Indian English. 5 ARCHER includes 1.7 million words from various registers from the mid-17th century to the present and can thus be seen as complementing the Helsinki Corpus of English Texts (cf. Kytö 1991). As we intend to compare ditransitive verb complementation in present-day Indian English to the input variety of British English, it is useful to start off from a diachronic characterisation of the patterns of give in British English. It is well-known that in the history of English, the alternation between the double-object construction (type I) and the to-dative variant (type II), marked by different orders of the indirect object (O i ) and the direct object (O d ), was triggered by the loss of explicit dative case markings in the early Middle English period (cf. Denison 1993: 106). In fact, as Koopman (1990) observes, the double-object construction (with both objects case-marked and without a to-phrase) was used in Old English similarly frequently with both object orders. As McFadden (2002) notes, it is in the 14th century that a clear division of labour emerges between the two constructions, with the doubleobject construction (type I) exclusively used for the order ‘[O i ][O d ]’ and the todative construction (type II) clearly associated with the order ‘[O d ][O i ]’. Note, by the way, that the relevant literature is relatively agnostic about the development of the other basic patterns of give, especially the type-III pattern, which in present-day British English is the second-most frequent pattern of give (cf. Mukherjee 2005: 100). As British English entered the Indian subcontinent in the 17th century, after the East Indian Company had been granted a Royal charter in 1600, we were particularly interested in finding out whether the distribution of the by far most frequent patterns of give in present-day British English, i.e. type I, type II and type III, remained stable throughout more than three centuries of British colonisation of India (i.e. up to the Independence of India in 1947). Therefore, we analysed three randomly selected samples of 500 instances each of give from different 100-year periods in A Representative Corpus of Historical English Registers (ARCHER, cf. Biber et al. 1994): 5 1650-1749, 1750-1849, 1850-1949. Figure 2 provides an overview of the proportion of the complementation patterns I, II and III in these three 100-year periods. It also includes the proportion of the three patterns in ICE-GB, representing the present-day British English situation. Figure 2 shows that in present-day British English the most prototypical pattern - type I - is also the most frequent pattern in British English, accounting for well over a third of all instances of give. Type III (i.e. without explicit indirect object) is the second most common pattern, followed by type II (i.e. with toobject). The most interesting observation, however, is that the proportions remain relatively constant over the past 350 years. Although percentage Ditransitive verbs in Indian English 11 Type I Type III Type II other 0 5 10 15 20 25 30 35 40 45 1650-1749 1750-1849 1850-1949 ICE-GB values vary somewhat across time, the ranking of the three most common patterns stays the same for all the periods covered by the ARCHER and the ICE-GB data. Furthermore, none of the patterns exhibits a consistent increase or decrease over time. At least for the case of the most prototypical ditransitive verb give, then, a considerable level of stability in the system of verb complementation can thus be observed for British English from 1650 to the present, i.e. in the period of time when Indian English emerged. Figure 2: The most frequent complementation patterns of give in ARCHER and ICE- GB. Given the relative diachronic stability of the complementation patterns of give, it is not surprising that all of the patterns shown in examples (1) to (5), including the derivative patterns listed in the Appendix, are also attested in present-day Indian English. However, our exhaustive analysis and comparison of ICE-India and ICE-GB has shown that there are substantial differences in the frequency and distribution of the various give-patterns between the two varieties (cf. Mukherjee and Hoffmann 2006). Figure 3 focuses on the most frequent patterns; these patterns account for 90% of all instances of give in the two corpora. As Figure 3 reveals, the most significant difference is the much lower frequency of the type-I pattern in ICE-India, which comes along with a markedly higher frequency of the type-II pattern and the type-III pattern, the latter being the most frequent pattern in ICE-India. Given the relative diachronic stability of give-patterns in British English displayed in Figure 2, this divergence of Indian English from the British English input variety is certainly striking. In particular, it clearly emerges that the complementational divergence in present-day Indian English is not a replication of an earlier stage of British English and, thus, not a symptom of superstrate retention. Sebastian Hoffmann / Joybrato Mukherjee 12 Type III Type I Type II Type IP Type IIIP Type IIP Type IIIPb Other 0 5 10 15 20 25 30 35 40 ICE-India ICE-GB Figure 3: Complementation of give in ICE-India and ICE-GB - focus on the most frequent patterns (Mukherjee and Hoffmann 2006: 152). From a cognitive point of view, differences in the complementation of give between British and Indian English are particularly interesting because give is a key verb in English with regard to both semantics and language acquisition (cf. Beermann 2001, Ninio 1999). Against this background, the much lower frequency of the type-I pattern of give in Indian English is quite remarkable. In Olavarría de Ersson and Shaw’s (2003: 159) opinion, this difference may at least partly be a reflection of cultural differences between Northern European and South Asian cultures. Thus the former cultures’ emphasis on subjectivism would favour the overt encoding of the recipient role - as for example in the type-I pattern - whereas the South Asian’s view of the individual as a part or a small object in a larger whole would be more conducive to the choice of a type-III pattern, where the recipient is not made explicit. A more language-internal explanation of the observed differences would be that give fulfils a larger range of communicative functions in Indian English. For example, Indian English speakers tend to use give in many contexts where in British English one would expect other verbs (e.g. give a complaint instead of make a complaint) and, more specifically, in collocations that are unusual in British English but which are linked to the type-II pattern (e.g. give provocation to someone instead of provoke someone). However, these are only tentative suggestions and a detailed explanation of the quantitative Ditransitive verbs in Indian English 13 6 A detailed description of the various steps involved in the compilation, reformatting and annotation of The Statesman Archive is provided in Mukherjee and Hoffmann (2006). differences between Indian English and British English must remain beyond the scope of the present paper. 3. “New ditransitives” and old ditransitives in Indian English and British English The verb-complementational profile of Indian English also differs from British English with regard to the range of verbs attested in individual ditransitive patterns. From a construction-grammar point of view, the type-I pattern is of particular interest because it represents a basic argument structure construction that is linked to the cognitive event of TRANSFER (cf. Goldberg 1995, 2006). From a syntactic point of view, it is the occurrence of a verb in the type-I pattern that defines the verb as a member of the class of ditransitive verbs. For a description of the verb-complementational profiles of Indian and British English, it is therefore interesting to find out whether the range of verbs that are used in the type-I pattern differs between the two varieties. Specifically, we were interested in identifying verbs to which the type-I pattern is extended in Indian English, i.e. “new ditransitives”. To this end, we made use of a large web-derived corpus of Standard Indian English, containing 31 million words from the on-line archive of the national Indian newspaper The Statesman. After converting the web pages to a format that can be searched with the help of standard corpus tools or Perl scripts, the corpus was part-of-speech tagged using the Constraint Grammar Parser of English (ENGCG, Karlsson et al. 1995). 6 This step made it possible to search for grammatical patterns (e.g. any verb that is immediately followed by two noun phrases) and consequently allowed us to also identify verbs that occur only sporadically in the basic type-I pattern. Table 1 gives an overview of ditransitive verbs in The Statesman Archive that usually do not occur in the type-I pattern in present-day British English. Some illustrative examples are given in (6) to (10). Sebastian Hoffmann / Joybrato Mukherjee 14 7 With 26 instances in the Statesman Archive, gift is particularly interesting as a new Indian English ditransitive. Type-I pattern uses of this verb are indeed also found in low frequen- Table 1: “New ditransitives” in The Statesman Archive. Verb n Verb n ADVISE 10 NOTIFY 1 BRIEF 1 PRESENT 18 CONFER 3 PRINT 1 DESPATCH/ DISPATCH 1 PROVIDE 217 EXPLAIN 2 PUT 2 FATHER 1 REMIND 4 GIFT 26 ROB 4 IMPART 8 SUBMIT 1 INFORM 4 SUPPLY 15 INTIMATE 1 THREATEN 1 (6) I have advised him some technical changes like using both hands while stopping the ball. <The Statesman 2004-03-26> (7) She said she wanted to gift him a dream. <The Statesman 2003-02-17> (8) … teachers should study at least five times more than the students to be able to impart them the correct knowledge and wisdom. <The Statesman 2004-11- 13> (9) The employee is also required to inform the appointing authority the amount of monthly instalment … <The Statesman 2004-08-1> (10) I put him a question as to whether he had an auspicious time for … <The Statesman 2003-08-09> A number of the verbs listed in Table 1 are of course also found with a type-I complementation pattern in native varieties other than British English. Thus, as already noted, provide occurs quite commonly in the basic ditransitive pattern in American English (cf. Quirk et al. 1985: 1210, Mukherjee 2001: 299) and this is also increasingly the case for supply and present. However, a type-I use of the remaining verbs must clearly be considered marginal at best in any native variety of English - a fact that is supported by various native-speaker informants who considered none of the sentences shown in (6) to (10) to be acceptable in British English. Given that at least some of the verbs listed in Table 1 recur too frequently to be interpreted as errors or nonce uses (e.g. gift, advise and impart), a much more likely interpretation is that they are indeed “new distransitives” in Indian English. 7 Ditransitive verbs in Indian English 15 cies in British English. However, their occurrence is virtually limited to the domain of sports commentaries (e.g. gift the player the ball) while no such register-specific restriction appears to operate in Indian English. In fact, the BNC provides only a single type-I pattern occurrence of gift outside of the sports domain: (i) By conceding on the timetable, O’Neill gifted the Republicans an important advantage, helping them by drastically curtailing the protracted and damaging interplay of pluralist forces that would have otherwise taken place over the budget. (BNC: EAY: 1342) However, this sentence appears to be a metaphorical extension of the standard use in sports commentaries, which is therefore clearly distinct from the much more general Indian English uses illustrated by example (7). What sentences (6) to (10) therefore show is that Indian English speakers extend the ditransitive construction to new verbs in a creative way as is typical of native speakers (e.g. with new verbs like email someone something, fax someone something). In construction-grammar terminology (cf. Goldberg 1995: 50), the participant roles of the verbs in (6) to (10) are “fused”, as it were, with the argument roles of the ditransitive construction so that these verbs now refer to TRANSFER events like established ditransitive verbs. The examples illustrate that new fusions in Indian English only occur in contexts in which semantically closely related classic ditransitive verbs could be used as well: (6) advise ~ offer, (7) gift ~ give, (8) impart ~ teach, (9) inform ~ tell, and (10) put ~ ask. We are fully aware that the label “new ditransitives” for verbs that are used in the basic ditransitive pattern in Indian English but not standardly in British English is potentially misleading since our comparison of the two varieties has so far been based on synchronic data alone. A key question in this context is whether “new ditransitives” like the ones listed in Table 1 and exemplified in (6) to (10) are really the result of post-colonial divergence in Indian English or whether we are instead looking at the verb-complementational profile of 18th or 19th century British English which has been preserved by its Indian speakers in its original form (i.e. in terms of superstrate retention). In other words, the question is: could it be the case that the “new ditransitives” in Indian English are actually old ditransitives in British English? Since most of the verbs shown in Table 1 are fairly infrequent, we needed a much larger corpus than ARCHER in order to test this hypothesis. We therefore made use of the online Gutenberg Archive (cf. <http: / / www. gutenberg.org/ >) and compiled a corpus of approximately 23.5 million words of both fiction and non-fiction texts covering a time-span from the second half of the 17th century to the 19th century. The overwhelming majority of texts were written by British authors, but a small number of American texts were also included. Again, the data was part-of-speech tagged, which made it possible to run the same scripts that were devised for the retrieval of type-I ditransitive constructions in The Statesman Archive. This approach allowed Sebastian Hoffmann / Joybrato Mukherjee 16 us to achieve two goals at the same time: on the one hand, our search retrieved historical instances of the “new” Indian ditransitives - if they exist. On the other hand, our retrieval method would find type-I verbs that are no longer used with this pattern in present-day British English. As was shown in section 2, the ARCHER data for the verb give reveal a high level of stability in the frequency and distribution of the complementation patterns of the most prototypical ditransitive verb over several centuries. We therefore expected to find no relatively frequent verbs whose complementational profile had radically changed since the 17th century. In fact, our analysis of the Gutenberg texts corroborates the relative stability of the complementation system of high-frequency ditransitive verbs over the past 350 years. Of particular interest, however, are those verbs that occur relatively infrequently in the type-I pattern. Table 2 provides a list of such verbs in the Gutenberg texts. Table 2: Verbs with type-I complementation pattern in the Gutenberg texts. Verb n Verb n ABATE/ BATE 5 KILL 1 ADDRESS 8 KISS 1 BANISH 1 LEAD 4 BESPEAK 1 LEARN 4 BESTOW 1 METE 1 BREATHE 1 MOUNT 1 CARRY 7 NUDGE 1 COUNT 1 PREVENT 1 DEPRIVE 1 PURVEY 2 DIGHT 2 PUT 8 DISPUTE 1 RECOMMEND 7 DISTIL 1 RESTORE 2 ENJOIN 1 REVIVE 1 EQUIP 1 SAY 3 FRANK 1 SPOIL 1 HOLD 2 TEND 1 INFORM 1 VOTE 2 Sample uses of some of the verbs listed in Table 2 verbs are given in (11) to (15). Ditransitive verbs in Indian English 17 8 For a more detailed discussion of the issues involved in using low-frequency corpus data, see e.g. Hoffmann (2004: passim, and in particular 196-8). (11) I had never encountered the great man at whose feet poor Dawling had most submissively sat and who had addressed him his most destructive sniffs … (Henry James: Glasses) (12) … and very often I would go and carry them some ears of barley, or a handful of rice, and feed them out of my hand. (Daniel Defoe: Robinson Crusoe) (13) “… I have often the satisfaction of hearing the publican, the baker, and sometimes even the parish-clerk, petitioning my housekeeper … to inform him the exact time by Master Humphrey’s clock.” (Charles Dickens: Master Humphrey’s Clock) (14) “I tell thee” said De Bracy, “that I mean to purvey me a wife after the fashion of the tribe of Benjamin (Walter Scott: Ivanhoe) (15) You ask me to recommend you some books for your perusal. (Elizabeth C. Gaskell: Life of Charlotte Brontë, Vol. I) A small number of verbs shown in Table 2 are clearly obsolete - for example dight (‘make ready’, ‘prepare’) or learn (as synonym of teach, a sense which the OED labels as “now vulgar”). The large majority of verbs in Table 2, however, are still current today in British English - but of course not with a type-I complementation pattern. The lists in Tables 1 and 2 show very little overlap - in fact, only two verbs (inform and put, both shown in boldface in Table 2) are found in both tables. While it is certainly true that comparisons of low-frequency phenomena across corpora have to be taken with a measure of caution, our data are nonetheless insightful: they do not support the view that the “new ditransitives” in Indian English are remnants of an earlier stage of the verbcomplementational profile of English. 8 In the light of our data, the superstrate retention hypothesis thus seems to be an unlikely explanation. Instead, we believe that both our rare present-day Indian English and our historical British English type-I pattern uses are in fact testimony to creative language use which allows speakers of native and non-native varieties of English alike - on grounds of semantic analogy - to extend the ditransitive situation schema to new verbs. Many of these type-I pattern uses clearly have no lasting effect on the verb-complementational profile of English - they are rare occurrences whose context-specific use is licensed by the systematic link between the basic argument structure, verbalising the cognitive event of TRANSFER, and the lexical meaning of verbs whose participant roles can be fused with the arguments of the ditransitive construction. Some of these uses, however, may be repeatedly employed by certain groups of speakers and may, thus, for example lead to a variety-specific extension of the pattern. This is no Sebastian Hoffmann / Joybrato Mukherjee 18 9 In this context, it is also worth noting that the verb provide does not feature in Table 2. This suggests that its relatively common type-I pattern use in American and Indian English is indeed of fairly recent origin. However, a brief look at other historical sources such as the OED quotations database reveals that language users of earlier centuries occasionally extended the type-I pattern to provide, too. As a case in point, consider example (ii): (ii) Prouide me ynke and paper, and I will write. (1581; in Confer. III. 1584, Riv) This again underscores our point that the mechanism of creative extension is in principle available to all speakers of a language. However, we cannot at the moment offer any additional insight as to what causes a construction to become fully acceptable in some varieties while it remains highly marked in others. doubt the case for the verb gift, which is particularly frequent in Indian English - although, as mentioned above, not in principle restricted to this variety. 9 Finally, our data suggest that this type of extension may also be a reflection of idiolectal preferences. It is interesting to see, for example, that a disproportionally large number of “new” type-I pattern uses in the Gutenberg texts stem from Henry James. Among these, we find six instances of address - see example (11) above - and three instances of the combination to put somebody a question. 4. Concluding remarks In the present paper, we hope to have shown that verb complementation in general and ditransitive verb complementation in particular is an area where different varieties of English are marked by diverging preferences and structural options. More specifically, there are clear and identifiable differences between British English and Indian English with regard to the range of verbs to which the basic ditransitive pattern, the type-I pattern, is extended by way of analogy. The “new ditransitives” in Indian English that we identified by using various corpora, including large web-derived databases, are not reflections of older British English forms, but represent creative innovations introduced by Indian users of English. With regard to institutionalised secondlanguage varieties of English, it seems necessary to complement nativespeaker creativity as a potential source of language change (cf. Jones and Singh 2005: 18ff.) with non-native speakers’ creative force. In the light of the “new ditransitives” that we identified, it is obvious that the extension of the type-I pattern to novel verbs is not restricted to native varieties of English but can also be found in second-language varieties like Indian English. This process is based on what we wish to call nativised semantico-structural analogy: [Nativised semantico-structural analogy is] a process by means of which nonnative speakers of English as a second language are licensed to introduce new forms and structures into the English language because corresponding Ditransitive verbs in Indian English 19 semantic and formal templates already exist in the English language system. (Mukherjee and Hoffmann 2006: 166f.) With regard to “new ditransitives”, an analogy is drawn between the meaning of established ditransitive verbs such as give on the one hand and the similar semantics of other verbs like address and gift on the other so that the same ditransitive construction as a formal and semantic template can be used. The availability of this extension of the ditransitive construction is part of the common core of English grammar, as shown in Figure 4. However, the output of this creative process differs from variety to variety, e.g. Indian English and British English. Additionally, there may be idiosyncratic usages which may obscure variety-specific trends (e.g. in the case of Henry James’s writing). Common core: Nativised semantico-structural analogy GIVE ‘X causes Y to receive Z’ other verbs [S: NP] V [O: NP] [O: NP] Variety 1 (Indian English) Speaker 1 Variety 2 (British English) Speaker 2 Variety n Speaker n Figure 4: Extending the type-I pattern to novel verbs - common core, varieties and individual speakers. The tendency in Indian English to extend the ditransitive construction to verbs that are semantically closely related to established ditransitive verbs in British English ties in well with observations in other areas of corpus-based research into contact varieties of English world-wide. In identifying morphosyntactic features that are shared by geographically distant varieties of English, Sand (2005: 459), for example, suggests that second-language speakers of English may find themselves in a situation in which “language contact magnifies ‘problem areas’ or ‘weak spots’ in the grammatical system of English and intensifies a development away from these”. Note, for example, that from a semantic point of view it is rather implausible that the verb address, which is closely related to tell, is not used in the ditransitive construction in British English. The extension of the ditransitive construction to address in Indian English, as exemplified in (11), thus indicates a ‘rationalis- Sebastian Hoffmann / Joybrato Mukherjee 20 10 This idea is closely related to the concept of ‘pre-determined breaking points’ in a grammatical system, that is areas which are characterised by forms and rules which are highly arbitrary and non-transparent and/ or only very partially systematic, and hence particularly liable to simplification and rationalisation in dialect contact, language contact or the emergence of new standards (Christian Mair, personal communication). ation’, as it were, of the verb-complementational system. 10 It might well be - and this is certainly a speculative claim at this stage - that it is the non-native second-language speaker who, to a greater extent than the native speaker, simplifies and increases transparency and systematicity in the grammatical system of English in general and the verb-complementational system in particular. References Beermann, D. (2001). “Verb semantics and double object constructions.” In: M. van Oostendorp & E. Anagnostopoulou (eds.). Progress in Grammar: Articles at the 20 th Anniversary of the Comparison of Grammatical Models Group in Tilburg. Available at <http: / / www.roquade.nl/ meertens/ progressingrammar/ beermann.pdf>. Biber, D., E. Finegan & D. Atkinson (1994). “ARCHER and its challenges: Compiling and exploring A Representative Corpus of Historical English Registers.” In: U. Fries, G. Tottie & P. Schneider (eds.). Creating and Using English Language Corpora. Amsterdam: Rodopi. 1-14. Crystal, D. (2003). The Cambridge Encyclopedia of the English Language. Cambridge: Cambridge University Press. Denison, D. (1993). English Historical Syntax: Verbal Constructions. London: Longman. Goldberg, A.E. (1995). Constructions: A Construction Grammar Approach to Argument Structure. Chicago, IL: University of Chicago Press. Goldberg, A.E. (2006). Constructions at Work: The Nature of Generalization in Language. Oxford: Oxford University Press. Greenbaum, S. (ed.) (1996). Comparing English Worldwide: The International Corpus of English. Oxford: Clarendon. Hoffmann, S. (2004). “Are low-frequency complex prepositions grammaticalized? On the limits of corpus data - and the importance of intuition.” In: H. Lindquist & Chr. Mair (eds.) Corpus Approaches to Grammaticalization in English. Amsterdam & Philadelphia: Benjamins. 171-210. Jones, M.C. & I. Singh (2005). Exploring Language Change. London: Routledge. Kachru, B.B. (1983). The Indianization of English: The English Language in India. Delhi: Oxford University Press. Kachru, B.B. (2005). Asian Englishes: Beyond the Canon. Hong Kong: Hong Kong University Press. Karlsson, F., A. Voutilanen, J. Heikkilä & A. Anttila (eds.) (1995). Constraint Grammar: A Language-independent System for Parsing Unrestricted Text. Berlin: Mouton de Gruyter. Koopman, W. (1990). Word Order in Old English. Amsterdam: University of Amsterdam, PhD thesis. Ditransitive verbs in Indian English 21 Kytö, M. (1991). Manual to the Diachronic Part of the Helsinki Corpus of English Texts. Helsinki: University of Helsinki, Department of English. McFadden, T. (2002). “The rise of the to-dative in Middle English.” Syntactic Effects of Morphological Change, ed. D. Lightfoot. Oxford: Oxford University Press. 107-123. Mehrotra, R.R. (1998). Indian English: Texts and Interpretation. Amsterdam: John Benjamins. Mukherjee, J. (2001). “Principles of pattern selection: a corpus-based case study.” Journal of English Linguistics 29(4). 295-314. Mukherjee, J. (2005). English Ditransitive Verbs: Aspects of Theory, Description and a Usage-based Model. Amsterdam: Rodopi. Mukherjee, J. & S. Hoffmann (2006). “Describing verb-complementational profiles of New Englishes: a pilot study of Indian English.” English World-Wide 27(2). 147-173. Nihalani, P., R.K. Tongue, P. Hosali & J. Crowther (2004). Indian and British English: A Handbook of Usage and Pronunciation. Second edition. Delhi: Oxford University Press. Ninio, A. (1999). “Pathbreaking verbs in syntactic development and the question of prototypical transitivity.” Journal of Child Language 26. 619-654. Olavarría de Ersson, E.O. & P. Shaw (2003). “Verb complementation patterns in Indian Standard English.” English World-Wide 24(2). 137-161. Quirk, R., S. Greenbaum, G. Leech & J. Svartvik (1985). A Comprehensive Grammar of the English Language. London: Longman. Sand, A. (2005). “The effects of language contact on the morpho-syntax of English.” In: L. Moessner & C.M. Schmidt (eds.). Anglistentag Aachen 2004: Proceedings. Trier: Wissenschaftlicher Verlag Trier. 449-460. Schneider, E.W. (2001). “Rev. Douglas Biber et al., Longman grammar of spoken and written English (Harlow: Longman, 1999).” English World-Wide 22(1). 137-143. Shastri, S.V. (1988). “The Kolhapur Corpus of Indian English and work done on its basis so far.” ICAME Journal 12. 15-26. Shastri, S.V. (1992). “Opaque and transparent features of Indian English.” In: G. Leitner (ed.). New Directions in English Language Corpora: Methodology, Results, Software Developments. 263-275. Shastri, S.V. (1996). “Using computer corpora in the description of language with special reference to complementation in Indian English.” In: R. Baumgardner (ed.). South Asian English: Structure, Use, and Users,. Urbana, IL: University of Illinois Press. 70-81. Corpora and electronic text collections used The British National Corpus (BNC), <http: / / www.natcorp.ox.ac.uk/ > The Gutenberg Archive, <http: / / www.gutenberg.org/ > A Representative Corpus of Historical English Registers (ARCHER) The Statesman, <http: / / www.thestatesman.net/ > Sebastian Hoffmann / Joybrato Mukherjee 22 Appendix The pattern formulas are based on the following notational conventions: - […] obligatory element; - […(…)] obligatory element with a specific form/ function; - (…) optional element; - O i / O d clause element which is not part of the lexicogrammatical pattern at the level of syntactic surface structure. Type-I patterns I (S) GIVE [O i : NP] [O d : NP] (a) Give me a warning next time <ICE-GB: S1A-091 #313> I a (S) GIVE [O d : NP] [O i : NP] (b) ‘He’s my dog. You gave him me.’ <ICE-GB: W2F-001 #107-108> I b [O d : NP antecedent ] (rel. pron.) [S] GIVE [O i : NP] (c) Those batteries that you gave me lasted an hour <ICE-GB: S1A-085 #132> I c [O i : NP antecedent ] (rel. pron.) [S] GIVE [O d : NP] (d) Uh an American lady that I gave a lecture uhm on architecture […] <ICE-GB: S2A-024 #34> I d [O d : NP fronted ] [S] GIVE [O i : NP] (e) Somebody she gave me <ICE-GB: S1A-043 #69> IP [S < O i active] BE given [O d : NP] (by-agent) (f) One was that I was being given the opportunity to uhm learn and develop uhm physical skills <ICE-GB: S1A-001 #34> IP b [O d : NP antecedent ] (rel. pron.) [S < O i ] BE given (by-agent) (g) And uh this the letter that the UN Secretary General has been given uh by John McCarthy <ICE-GB: S2A-008 #166> Type-II patterns II (S) GIVE [O d : NP] [O i : PP to ] (h) Well perhaps you ‘re giving too much to other people <ICE-GB: S1A-067 #288> II a (S) GIVE [O d : NP] [O i : PP for ] (i) The defendant had been ordered to attend at Leeds Crown Court to give evidence for the prosecution in a fraud trial <ICE-GB: W2B-020 #68> Ditransitive verbs in Indian English 23 II b [O d : NP antecedent ] (rel. pron.) [S] GIVE [O i : PP to ] (j) One of the great things about by-elections […] is of course this extra bit of power that they give to a voter <ICE-GB: S1B-029 #60> II c (S) GIVE [O i : PP to ] [O d : NP] (k) It was the Queen of course who gave to Norman Schwarzkopf the knighthood that makes him now Sir Norman <ICE-GB: S2A-019 #9> IIP [S < O d active] BE given [O i : NP to ] (by-agent) (l) more time should have been given to sanctions <ICE-GB: S2B-018 #93> IIP b [antecedent] co (S < O d ) co (BE) given [O i : NP to ] (by-agent) (m) I’ll leave you with the final message that was given to the world leaders at the summit <ICE-GB: S2B-022 #148> Type-III patterns III (S) GIVE [O d : NP] O i (n) he wanted physical love and I couldn’t give that <ICE-GB: S1A-050 #184> III b [O d : NP antecedent ] (rel. pron.) [S] GIVE O i (o) The sermon he gave when his daughter was married <ICE-GB: S1A-053 #186> IIIP [S < O d active] BE given O i (by-agent) (p) More specific implementation details are given at the end of the report <ICE-GB: W1A-005 #5> IIIP b [antecedent] co (S < O d ) co (BE) given O i (by-agent) (q) […] it also is of relevance when considering the evidence given by Mr Holt because there is a clear conflict […] <ICE-GB: S2A-068 #40> Type-IV patterns IV (S) GIVE O i O d (r) If you give and take when there’s that close bodily contact it’s great <ICE-GB: S1A-003 #146> (s) IVP given O i O d (by-agent) The run rate here as given by our scorer immediately and it’s two point four one <ICE-India: S2A-016 #12> Sebastian Hoffmann / Joybrato Mukherjee 24 Type-V pattern V (S) GIVE [O i : NP] O d (t) … well she did […] she did give Mrs <anonymised_last_name> didn’t she? <BNC: KE6 2368-70> Sebastian Hoffmann Lancaster University Joybrato Mukherjee Universiät Gießen