1.Introduction

This paper extends the corpus-linguistic concept of local grammar (Hunston & Sinclair 2000Hunston, Susan, and John Sinclair 2000 “A Local Grammar of Evaluation.” In Evaluation in Text, ed. by Susan Hunston, and Geoff Thompson, 74–101. Oxford: Oxford University Press.; Barnbrook & Sinclair 2001Barnbrook, Geoff, and John Sinclair 2001 “Specialised Corpus, Local and Functional Grammars.” In Small Corpus Studies and ELT: Theory and Practice, ed. by Mohsen Ghadessy, Alex Henry, and Robert Roseberry, 237–276. Amsterdam: John Benjamins. ) to speech act studies (Austin 1962Austin, John 1962 How to Do Things with Words. Oxford: Oxford University Press.; Searle 1969Searle, John R. 1969 Speech Acts: An Essay in the Philosophy of Language. Cambridge: Cambridge University Press. ), focusing specifically on apologising in English. Briefly, local grammar in this paper refers to an approach to linguistic description which seeks to account for, both functionally and grammatically, specific areas of language in use (see Section 2 for more detail). Local grammar research has been shown to be influential and beneficial in several respects. First, differing from traditional or general grammars, each local grammar deals with one meaning or function only (Hunston 2002Hunston, Susan 2002 “Pattern Grammar, Language Teaching, and Linguistic Variation: Applications of a Corpus-Driven Grammar.” In Using Corpora to Explore Linguistic Variation, ed. by Randi Reppen, Susan Fitzmaurice, and Doug Biber, 167–183. Amsterdam: John Benjamins. , 178), thereby contributing to a more specific and adequate description of the targeted semantic or pragmatic phenomenon. Second, local grammars have pedagogical values, because, as will be shown in this study, they identify both formal and functional patterns of language in use and therefore facilitate the establishment of the repertoire of strategies that can be employed by EFL learners to express a meaning or to perform a function (see also Su 2017Su, Hang 2017 “Local Grammars of Speech Acts: An Exploratory Study.” Journal of Pragmatics 111: 72–83. , 2018Su, Hang 2018 “Thank Bloody God it’s Friday: A Local Grammar of Thanking.” Corpus Pragmatics 2(1): 83–105.). Moreover, local grammars have potential applications in natural language processing, which has been explored by, for example, Barnbrook (2002)Barnbrook, Geoff 2002 Defining Language: A Local Grammar of Definition Sentences. Amsterdam: John Benjamins. , Mason (2004)Mason, Oliver 2004 “Automatic Processing of Local Grammar Patterns.” In Proceedings of the 7th Annual Colloquium for the UK Special Interest Group for Computational Linguistics, ed. by Mark Lee, 166–171. Birmingham: University of Birmingham., and Bloom (2011)Bloom, Kenneth 2011 “Sentiment Analysis Based on Appraisal Theory and Functional Local Grammar.” Unpublished Ph.D. thesis, Illinois Institute of Technology, Chicago, US.. This paper will further show that local grammars are particularly useful for accounting for speech act realisations.

The emergence of local grammars raises the question as to whether or not it would be feasible to develop a set of local grammars to account for language used in social contexts. While Butler (2004Butler, Christopher 2004 “Corpus Studies and Functional Linguistic Theories.” Functions of Language 11 (2): 147–186. , 158) has argued that “rather than a single general grammar, we might end up with a set of local grammars for particular areas defined by their communicative functions in the discourse”, only a few studies have investigated empirically the possibility of doing so (e.g. Hunston & Sinclair 2000Hunston, Susan, and John Sinclair 2000 “A Local Grammar of Evaluation.” In Evaluation in Text, ed. by Susan Hunston, and Geoff Thompson, 74–101. Oxford: Oxford University Press.; Barnbrook 2002Barnbrook, Geoff 2002 Defining Language: A Local Grammar of Definition Sentences. Amsterdam: John Benjamins. ; Su 2017Su, Hang 2017 “Local Grammars of Speech Acts: An Exploratory Study.” Journal of Pragmatics 111: 72–83. ). Because each local grammar is a grammar of a discursive or pragmatic function and each speech act is concerned with one such function, research on local grammars of speech acts can be used as a heuristic to explore whether a local grammar approach to accounting for language in use would work. This study therefore presents a local grammar investigation into apologies in English, aiming to develop a local grammar of apology and, based on that, to further demonstrate the usefulness of a local grammar approach to advance speech act studies. Additionally, the relationship between local grammars, functional grammars and general grammars will also be discussed, with a view to further highlighting the value of local grammars in functional-pragmatic studies of language and discourse.

The remainder of this paper is organised into five sections. Section 2 offers background information about both local grammars and speech act theory. Section 3 introduces the data and methodology used for the current investigation. Section 4 presents our proposal for a local grammar of apology, followed by Section 5 in which the relationship between local grammars, functional grammars, and general grammars is discussed. Section 6 concludes this paper, arguing for the importance of using a local grammar approach to further investigate pragmatic functions in language description and pedagogy.

2.Local grammar and speech act theory

2.1Local grammar

Local grammar is an alternative approach to the description and theorising of language in use. It is situated within the field of what is now widely known as Corpus Linguistics (Sinclair 1991Sinclair, John 1991 Corpus, Concordance, Collocation. Oxford: Oxford University Press.; McEnery & Hardie 2012McEnery, Tony, and Andrew Hardie 2012 Corpus Linguistics: Method, Theory and Practice. Cambridge: Cambridge University Press.). Currently, local grammar has been used in (at least) three distinct but related senses, as discussed below.

Local grammar in one sense is used to account for sublanguage (Harris 1968Harris, Zellig 1968 Mathematical Structures of Languages. New York: John Wiley & Sons.), i.e. a subset of general language use that has particular syntactic, semantic or grammatical features and exhibits some form of closure (Kittredge & Lehrberger 1982Kittredge, Richard, and John Lehrberger (eds) 1982 Sublanguage: Studies of Language in Restricted Domains. Berlin & New York: Walter de Gruyter. ; Pearson 1998Pearson, Jennifer 1998 Terms in Context. Amsterdam: John Benjamins. ). One example of local grammar research in this tradition is Charrow et al. (1982)Charrow, Veda. R., Jo Ann Crandall, and Robert P. Charrow 1982 “Characteristics and Functions of Legal Language.” In Sublanguage: Studies of Language in Restricted Domains, ed. by Richard Kittredge, and John Lehrberger, 175–190. Berlin & New York: Walter de Gruyter., which argues that legal language should be regarded as a sublanguage and thus language used in legal contexts should be described in terms of a specialised grammar or local grammar. Second, local grammar has been employed to deal with epistemic or grammatical phenomena. For example, Brezina (2011)Brezina, Vaclav 2011 “Epistemic Markers in University Advisory Sessions: Towards a Local Grammar of Epistemicity.” Unpublished Ph.D. thesis, University of Auckland, New Zealand. adopts a local grammar approach to investigating epistemicity; Warren and Leung (2016)Warren, Martin, and Maggie Leung 2016 “Do Collocational Frameworks Have Local Grammars?” International Journal of Corpus Linguistics 21 (1): 1–27. study collocational frameworks from a local grammar perspective.

Third, ‘local’ in local grammar indicates the restriction of linguistic description to one specific semantic or pragmatic domain (Hunston & Su 2017Hunston, Susan, and Hang Su 2017 “Patterns, Constructions and Local Grammar: A Case Study of Evaluation.” Applied Linguistics 1–28. ). Local grammar used in this sense, which has gained most popularity in the linguistic community so far, has been extensively elaborated and exemplified in the work of Sinclair and his colleagues (e.g. Hunston & Sinclair 2000Hunston, Susan, and John Sinclair 2000 “A Local Grammar of Evaluation.” In Evaluation in Text, ed. by Susan Hunston, and Geoff Thompson, 74–101. Oxford: Oxford University Press.; Barnbrook & Sinclair 2001Barnbrook, Geoff, and John Sinclair 2001 “Specialised Corpus, Local and Functional Grammars.” In Small Corpus Studies and ELT: Theory and Practice, ed. by Mohsen Ghadessy, Alex Henry, and Robert Roseberry, 237–276. Amsterdam: John Benjamins. ; Barnbrook 2002Barnbrook, Geoff 2002 Defining Language: A Local Grammar of Definition Sentences. Amsterdam: John Benjamins. ; Hunston 2002Hunston, Susan 2002 “Pattern Grammar, Language Teaching, and Linguistic Variation: Applications of a Corpus-Driven Grammar.” In Using Corpora to Explore Linguistic Variation, ed. by Randi Reppen, Susan Fitzmaurice, and Doug Biber, 167–183. Amsterdam: John Benjamins. , 2003Hunston, Susan 2003 “Frame, Phrase or Function: A Comparison of Frame Semantics and Local Grammars.” In Corpus Linguistics 2003, ed. by Dawn Archer, Paul Rayson, Andrew Wilson, and Tony McEnery, 342–358. University Centre for Computer Corpus Research on Language, University of Lancaster, UK., 2011). In the Sinclairian tradition, local grammar is considered to be useful for dealing with all areas of language use, including those which general grammars could cope quite easily with. This is different from Gross’ (1993)Gross, Maurice 1993 “Local Grammars and Their Representation by Finite Automata.” In Data, Description, Discourse: Papers on the English Language in Honour of John Sinclair, ed. by Michael Hoey, 26–38. London: HarperCollins. discussion of local grammar, where a local grammar is designed to deal with those highly specialised expressions (e.g. numbers, dates) which general grammatical analyses may not have adequately accounted for. Moreover, another significant difference between the Sinclairian tradition of local grammar and the abovementioned two types of research lies in their view of the role phraseology plays in the description of language in use: phraseology in the Sinclairian tradition is considered to be central whereas in the other two types (though Martin and Warren’s (2016) study is an exception) it is peripheral to linguistic description and explanation. Given that corpus studies have shown that there is a phraseological tendency11.This phraseological tendency, or in Sinclair’s term, ‘the idiom principle’, means that “a language user has available to him a large number of semi-preconstructed phrases that constitute single choices, even though they might appear to be analysable into segments” (Sinclair 1991Sinclair, John 1991 Corpus, Concordance, Collocation. Oxford: Oxford University Press., 110). of language in use (Sinclair 1991Sinclair, John 1991 Corpus, Concordance, Collocation. Oxford: Oxford University Press., 2004Sinclair, John 2004 Trust the Text: Language, Corpus and Discourse. London: Routledge.; Hunston & Francis 1998Hunston, Susan, and Gill Francis 1998 “Verbs Observed: A Corpus-Driven Pedagogic Grammar.” Applied Linguistics 19 (1): 45–72. , 2000Hunston, Susan, and Gill Francis 2000 Pattern Grammar: A Corpus-Driven Approach to the Lexical Grammar of English. Amsterdam: John Benjamins. ; Hanks 2013Hanks, Patrick 2013 Lexical Analysis. Cambridge, MA: The MIT Press. ), it is arguable that the Sinclairian tradition of local grammar research would be more appropriate and useful for linguistic or pragmatic investigation. This paper thus draws on the Sinclairian concept of local grammar.

The defining features of local grammar in this tradition are described as follows. First, each local grammar deals with one meaning or function only. Second, local grammar takes into account the functions language fulfils in social contexts, therefore each local grammar is in essence a functional account of language use. More notably, the functional elements used in a local grammar analysis are proposed within the specific context of that chosen meaning or function (for example, the present study uses terms such as Apologiser, Apologising, Apologisee to analyse apology expressions), thereby contributing to the transparency of the description.

The feasibility of developing local grammars to account for particular meanings or functions has been explored in a few studies. For example, the pioneering work by Barnbrook (1995Barnbrook, Geoff 1995 “The Language of Definition.” Unpublished Ph.D. thesis, University of Birmingham, Birmingham, UK., 2002Barnbrook, Geoff 2002 Defining Language: A Local Grammar of Definition Sentences. Amsterdam: John Benjamins. ) and Barnbrook and Sinclair (1995Barnbrook, Geoff, and John Sinclair 1995 “Parsing Cobuild Entries.” In The Language of Definition: The Formalization of Dictionary Definitions for Natural Language Processing, ed. by John Sinclair, Martin Hoelter, and Carol Peters, 13–58. Luxemburg: European Commission., 2001Barnbrook, Geoff, and John Sinclair 2001 “Specialised Corpus, Local and Functional Grammars.” In Small Corpus Studies and ELT: Theory and Practice, ed. by Mohsen Ghadessy, Alex Henry, and Robert Roseberry, 237–276. Amsterdam: John Benjamins. ) contributed to a local grammar of definition; and Su (2017)Su, Hang 2017 “Local Grammars of Speech Acts: An Exploratory Study.” Journal of Pragmatics 111: 72–83. recently built a local grammar of request. Most notably, local grammar in the Sinclairian tradition has been applied to study the discourse function of evaluation. Hunston and Sinclair (2000)Hunston, Susan, and John Sinclair 2000 “A Local Grammar of Evaluation.” In Evaluation in Text, ed. by Susan Hunston, and Geoff Thompson, 74–101. Oxford: Oxford University Press. first brought together the concept of local grammar and that of evaluation, demonstrating the possibility of developing a local grammar of evaluation (see Hunston & Su 2017Hunston, Susan, and Hang Su 2017 “Patterns, Constructions and Local Grammar: A Case Study of Evaluation.” Applied Linguistics 1–28. for a recent update) and further offering theoretical and methodological insights for subsequent local grammar research. Drawing on the Appraisal framework (Martin & White 2005Martin, Jim, and Peter White 2005 The Language of Evaluation: Appraisal in English. New York: Palgrave Macmillan. ), studies such as Hunston (2003)Hunston, Susan 2003 “Frame, Phrase or Function: A Comparison of Frame Semantics and Local Grammars.” In Corpus Linguistics 2003, ed. by Dawn Archer, Paul Rayson, Andrew Wilson, and Tony McEnery, 342–358. University Centre for Computer Corpus Research on Language, University of Lancaster, UK. and Bednarek (2008)Bednarek, Monika 2008 Emotion Talk Across Corpora. New York: Palgrave Macmillan. have contributed substantially to building a local grammar of Affect, and Su (2015)Su, Hang 2015 “Judgement and Adjective Complementation Patterns: A Corpus Study.” Unpublished Ph.D. thesis, University of Birmingham, Birmingham, UK. has attempted to construct a local grammar of Judgement. In general, these studies have shown that, compared with general grammars, local grammars can provide a more specific and precise description of the chosen semantic or pragmatic phenomenon, which indicates both the wide applicability of local grammars and the significance of local grammar research.

2.2Speech act theory

Speech acts generally refer to the fact that in saying something we are also doing something (Austin 1962Austin, John 1962 How to Do Things with Words. Oxford: Oxford University Press.; Searle 1969Searle, John R. 1969 Speech Acts: An Essay in the Philosophy of Language. Cambridge: Cambridge University Press. ). Speech act theory has attracted much attention since its emergence. Studies have addressed issues including but not limited to the following: the classification of speech acts (Searle 1976Searle, John R. 1976 “A Classification of Illocutionary Acts.” Language in Society 5: 1–23. ), speech acts in conversations (Geis 1995Geis, Michael L. 1995 Speech Acts and Conversational Interaction. Cambridge: Cambridge University Press. ), speech acts in pedagogic contexts (Achiba 2003Achiba, Machiko 2003 Learning to Request in a Second Language: A Study of Child Interlanguage Pragmatics. Toronto & Sydney: Multilingual Matters Ltd. ; Al-Gahtani & Roever 2012Al-Gahtani, Saad, and Carsten Roever 2012 “Proficiency and Sequential Organization of L2 Requests.” Applied Linguistics 33 (1): 42–65. ), and in digital contexts such as Twitter (Page 2014Page, Ruth 2014 “Saying ‘Sorry’: Corporate Apologies Posted on Twitter.” Journal of Pragmatics 62: 30–45. ) and blogs (Lutzky & Kehoe 2017aLutzky, Ursula, and Andrew Kehoe 2017a “Oops, I didn’t Mean to Be So Flippant: A Corpus Pragmatic Analysis of Apologies in Blog Data.” Journal of Pragmatics 116: 27–36. , bLutzky, Ursula, and Andrew Kehoe 2017b “ I Apologise for My Poor Blogging: Searching for Apologies in the Birmingham Blog Corpus.” Corpus Pragmatics 1 (1): 37–56. ). Recently, there has been an increasing interest in using corpus methods to investigate speech acts (e.g. Wichmann 2004Wichmann, Anne 2004 “The Intonation of Please-Requests: A Corpus-Based Study.” Journal of Pragmatics 36: 1521–1549. ; Adolphs 2008Adolphs, Svenja 2008 Corpus and Context: Investigating Pragmatic Functions in Spoken Discourse. Amsterdam: John Benjamins. ; Cheng 2010Cheng, Stephanie 2010 “A Corpus-Based Approach to the Study of Speech Act of Thanking.” Concentric: Studies in Linguistics 36 (2): 257–274.; Jautz 2013Jautz, Sabine 2013 Thanking Formulae in English: Explorations Across Varieties and Genres. Amsterdam: John Benjamins. ; Flöck & Geluykens 2015Flöck, Ilka, and Ronald Geluykens 2015 “Speech Acts in Corpus Pragmatics: A Quantitative Contrastive Study of Directives in Spontaneous and Elicited Discourse.” In Yearbook of Corpus Linguistics and Pragmatics: Current Approaches to Discourse and Translation Studies, ed. by Jesús Romero-Trillo, 7–37. Heidelberg: Springer. ; Su 2017Su, Hang 2017 “Local Grammars of Speech Acts: An Exploratory Study.” Journal of Pragmatics 111: 72–83. ), among other pragmatic phenomena.

Corpus techniques can offer ample opportunities to further speech act studies. The most obvious one may be that corpus tools enable us to search and extract a considerable number of authentic speech act instances, which allows us to describe more thoroughly the realisations of speech acts. In addition, corpus investigation can also provide information about the context in which language is used (see also Adolphs 2008Adolphs, Svenja 2008 Corpus and Context: Investigating Pragmatic Functions in Spoken Discourse. Amsterdam: John Benjamins. ). While it is commonly assumed that corpus searches can only give us the narrower syntagmatic co-text, Baker et al. (2008)Baker, Paul, Costas Gabrielatos, Majid KhosraviNik, Michal Krzyzanowski, Tony McEnery, and Ruth Wodak 2008 “A Useful Methodological Synergy? Combining Critical Discourse Analysis and Corpus Linguistics to Examine Discourses of Refugees and Asylum Seekers in the UK Press.” Discourse & Society 19 (3): 273–306. argued that:

The examination of expanded concordances (or whole texts when needed) can help the analyst infer the contextual elements in order to sufficiently recreate the context (Brown and Yule 1982Brown, Gillian, and George Yule 1982 Discourse Analysis. Cambridge: Cambridge University Press.). During language communication, addressees do not need to take the full context into account, as according to the principle of local interpretation, addressees need not construct a context more complex than that needed for interpretation (Brown and Yule 1982Brown, Gillian, and George Yule 1982 Discourse Analysis. Cambridge: Cambridge University Press.). In turn, the co-text provided by the (expanded) concordances helps in ‘limiting the interpretation’ to what is contextually appropriate or plausible (Brown and Yule 1982Brown, Gillian, and George Yule 1982 Discourse Analysis. Cambridge: Cambridge University Press.: 59).(Baker et al. 2008Baker, Paul, Costas Gabrielatos, Majid KhosraviNik, Michal Krzyzanowski, Tony McEnery, and Ruth Wodak 2008 “A Useful Methodological Synergy? Combining Critical Discourse Analysis and Corpus Linguistics to Examine Discourses of Refugees and Asylum Seekers in the UK Press.” Discourse & Society 19 (3): 273–306. , 279)

We agree and would further concur with Rühlemann’s (2011Rühlemann, Christoph 2011 “Corpus-Based Pragmatics II: Quantitative Studies.” In Foundations of Pragmatics, ed. by Wolfram Bublitz, and Neal R. Norrick, 629–656. Berlin: Walter de Gruyter. , 630) argument that corpora can provide “the analyst with illustrative examples that are not only attested and, in this sense, authentic but also embedded in their co-texts, thus giving some evidence of the context in which they were used”. This is crucial for pragmatic investigation, in particular for speech act studies, since speech acts are highly context-sensitive.

This study adopts a corpus-based approach to investigate apologies in English. Apology is defined here as an act performed by an apologiser, who has done something annoying or damaging or violated accepted social norms, to restore equilibrium and social harmony (cf. Aijmer 1996Aijmer, Karin 1996 Conversational Routines in English. London & New York: Longman.; Bella 2014Bella, Spyridoula 2014 “A Contrastive Study of Apologies Performed by Greek Native Speakers and English Learners of Greek as a Foreign Language.” Pragmatics 24 (4): 679–713. ). The reasons for focusing specifically on apology are, first, that apology is ubiquitous and is important for maintaining interpersonal rapport. This is consistent with Goffman’s (1971)Goffman, Erving 1971 Relations in Public: Microstudies of the Public Order. London: Penguin. argument that apologising is like an ‘everyday ritual’, helping to establish and strengthen the social bonds between individuals. Second, apologies have been shown to be realised by more or less fixed, recurring patterns. Aijmer (1996Aijmer, Karin 1996 Conversational Routines in English. London & New York: Longman., 84), for example, notes that “[a]pologies are generally made up of a small repertoire of relatively fixed expressions representing verbs (apologize, excuse, pardon), adjectives (sorry, afraid) and nouns (pardon) and their expansions, modifications, etc.”. Deutschmann (2003)Deutschmann, Mats 2003 “Apologising in British English.” Unpublished Ph.D. thesis, Umeå University, Sweden. further shows that key terms including afraid, apologize/se, apology, excuse, forgive, pardon, regret and sorry are routinely used to make apologies and that these key terms are useful for the identification of apology expressions in corpora. Since “[c]omputerized searches for specific speech acts can only be undertaken if the speech act tends to occur in routinized forms, with recurrent phrases and or [sic] with standard Illocutionary Force Indicating Devices (IFIDs)” (Taavitsainen & Jucker 2008Taavitsainen, Irma, and Andreas Jucker 2008 “Speech Acts Now and Then: Towards a Pragmatic History of English.” In Speech Acts in the History of English, ed. by Andreas Jucker, and Irma Taavitsainen, 1–23. Amsterdam: John Benjamins. , 10), the observation that realisations of apologies are highly conventionalised further justifies our choice of taking apologising as a starting point to explore the feasibility of developing a set of local grammars to account for speech acts.

While apologies have been extensively investigated, most studies have been situated within research on interlanguage pragmatics (e.g. Trosborg 1995Trosborg, Anna 1995 Interlanguage Pragmatics: Requests, Complaints and Apologies. Berlin: Mouton de Gruyter. ; Al-Gahtani & Roever 2012Al-Gahtani, Saad, and Carsten Roever 2012 “Proficiency and Sequential Organization of L2 Requests.” Applied Linguistics 33 (1): 42–65. ; Cheng 2017Cheng, Dongmei 2017 ““Communication is a Two-Way Street”: Instructors’ Perceptions of Student Apologies.” Pragmatics 27 (1): 1–32. ) or cross-cultural studies (e.g. Blum-Kulka et al. 1989Blum-Kulka, Shoshana, Juliane House, and Gabriele Kasper (eds.) 1989 Cross-Cultural Pragmatics: Requests and Apologies. Norwood, NJ: Ablex.; Kondo 2010Kondo, Sachiko 2010 “Apologies: Raising Learners’ Cross-Cultural Awareness.” In Speech Act Performance: Theoretical, Empirical and Methodological Issues, ed. by Alicia Martinez-Flor, and Esther Usó-Juan, 145–162. Amsterdam: John Benjamins. ; Bella 2014Bella, Spyridoula 2014 “A Contrastive Study of Apologies Performed by Greek Native Speakers and English Learners of Greek as a Foreign Language.” Pragmatics 24 (4): 679–713. ). There appears to be no study to date which has attempted to account for the realisations of apologies (and realisations of other types of speech acts) both functionally and grammatically. By ‘functionally and grammatically’, we mean that the elements used in the description should not only reflect the function of corresponding formal elements, but also be comparable to traditional grammatical elements, i.e. the elements used can in a way be seen as analogous to elements used in traditional grammatical analyses (‘Apologiser’ used in the local grammar analysis below, for example, can be seen as analogous to ‘Subject’ used in general grammars). It will be shown in this study that local grammars are able to bring together functional and grammatical analysis and that local grammar descriptions can contribute to a more transparent and comprehensive account of speech act realisations (see also Su 2017Su, Hang 2017 “Local Grammars of Speech Acts: An Exploratory Study.” Journal of Pragmatics 111: 72–83. , 2018Su, Hang 2018 “Thank Bloody God it’s Friday: A Local Grammar of Thanking.” Corpus Pragmatics 2(1): 83–105.).

3.Corpus and data retrieval

The corpus used in this study is compiled of transcripts of the first seven seasons of the sitcom The Big Bang Theory.22.The transcripts are contributed by the fans and openly available to the public. Using the transcripts as our data therefore does not constitute any infringement on copyright. This is preferred over transcripts of spontaneous conversations occurring in real contexts, because scripted discourse tends to be less ambiguous, in terms of pragmatic functions, than naturally occurring discourse. Although it cannot be demonstrated that TV dialogue exactly mimics naturally occurring conversation (cf. Rey 2001Rey, Jennifer M. 2001 “Changing Gender Roles in Popular Culture: Dialogue in Star Trek Episodes from 1966 to 1993.” In Variation in English: Multi-Dimensional Studies, ed. by Susan Conrad, and Doug Biber, 138–155. London: Longman.; Quaglio 2009Quaglio, Paulo 2009 Television Dialogue: The Sitcom Friends vs. Natural Conversation. Amsterdam: John Benjamins. ), it is certainly designed to be natural-sounding. In addition, the use of transcripts of TV dialogue can be considered adequate for this study, since its primary aim is to demonstrate the feasibility of using a local grammar approach to further speech act studies, rather than to investigate how apologies are typically made in face-to-face conversations. A contrastive investigation into apologies made in scripted and unscripted discourses would be worthwhile though.

The corpus of The Big Bang Theory compiled (henceforward TBBT) comprises 159 texts and has 485,602 tokens. The corpus has been uploaded to Sketch Engine (Kilgarriff et al. 2004Kilgarriff, Adam, Pavel Rychlý, Pavel Smrz, and David Tugwell 2004 “The Sketch Engine.” In Proceedings of EURALEX 2004, ed. by Geoffrey Williams, and Sandra Vessier, 105–116. Retrieved from: http://​euralex​.org​/publications​/the​-sketch​-engine/.), the program through which instances of apologies are retrieved.

Speech act studies using corpus investigation techniques have usually taken conventionalised forms as the starting point to search and identify speech act instances. For example, Jucker and Taavitsainen (2008)Jucker, Andreas, and Irma Taavitsainen 2008 “Apologies in the History of English: Routinized and Lexicalized Expressions of Responsibility and Regret.” In Speech Acts in the History of English, ed. by Andreas Jucker, and Irma Taavitsainen, 229–443. Amsterdam: John Benjamins. take conventionalised forms of apology such as sorry, I regret as the point of departure to perform a diachronic and contrastive speech act analysis of Renaissance data; Jautz (2013)Jautz, Sabine 2013 Thanking Formulae in English: Explorations Across Varieties and Genres. Amsterdam: John Benjamins. similarly uses a list of linguistic forms (e.g. thank, appreciate) to investigate gratitude expressions across language varieties and genres; and in another study, Su (2017)Su, Hang 2017 “Local Grammars of Speech Acts: An Exploratory Study.” Journal of Pragmatics 111: 72–83. , starting with a set of conventionalised forms of request, further demonstrates that the use of conventionalised forms can indeed facilitate a corpus investigation into speech acts.

The method of simply starting with conventionalised forms to identify speech act instances in corpora is not without problems, however. Putting aside that not all illocutionary forces in English are realised syntactically or lexicalised (Vanderveken 2001Vanderveken, Daniel 2001 “Universal Grammar and Speech Act Theory.” In Essays in Speech Act Theory, ed. by Daniel Vanderveken, and Susumu Kubo, 25–62. Amsterdam: John Benjamins. , 30), “there is no one-to-one correspondence between linguistic features and speech acts” (Garcia 2015Garcia, Paula 2015 “Speech Acts: A Synchronic Perspective.” In Corpus Pragmatics: A Handbook ed. by Karin Aijmer, and Christoph Rühlemann, 29–51. Cambridge: Cambridge University Press., 47). For example, the conventionalised form of requests like ‘can you …’ may as well just be used to ask a question, as in can you drive; and similarly, sorry in he says sorry does not have the illocutionary force of apologising either. This suggests that the retrieved instances have to be manually examined so as to make sure that all instances to be analysed have the targeted illocutionary force. The methodological implication here is that it is necessary to combine automated searches and manual examination when identifying speech act instances (cf. Rühlemann & Aijmer 2015Rühlemann, Christoph, and Karin Aijmer 2015 “Corpus Pragmatics: Laying the Foundations.” In Corpus Pragmatics: A Handbook, ed. by Karin Aijmer, and Christoph Rühlemann, 1–26. Cambridge: Cambridge University Press. , 13–14).

The next question, then, is what forms usually occur in apology expressions. Drawing on insights from previous investigations into apology (e.g. Blum-Kulka & Olshtain 1989Blum-Kulka, Shoshana, Juliane House, and Gabriele Kasper (eds.) 1989 Cross-Cultural Pragmatics: Requests and Apologies. Norwood, NJ: Ablex.; Aijmer 1996Aijmer, Karin 1996 Conversational Routines in English. London & New York: Longman.; Deutschmann 2003Deutschmann, Mats 2003 “Apologising in British English.” Unpublished Ph.D. thesis, Umeå University, Sweden.; Jucker & Taavitsainen 2008Jucker, Andreas, and Irma Taavitsainen 2008 “Apologies in the History of English: Routinized and Lexicalized Expressions of Responsibility and Regret.” In Speech Acts in the History of English, ed. by Andreas Jucker, and Irma Taavitsainen, 229–443. Amsterdam: John Benjamins. ), the following lexical items have been used to search and extract instances of apology in TBBT; the variants and relevant quantitative information are given in Table 1.

Table 1 lists the apology items and their variants that are to be examined in this study. The much higher frequency of SORRY and EXCUSE confirms that apology expressions are indeed highly conventionalised and are routinely expressed by idiomatic expressions such as (I’m) sorry, excuse me (see Section 4). At this point, it should be acknowledged that this is obviously not a complete list of linguistic resources that can be used to express apologies. We can, nevertheless, be fairly certain that the items listed in Table 1 should enable us to identify most apology expressions in naturally occurring discourses,33.This is confirmed by our manual check of three randomly selected episodes (Episode 5 in Season 2, 3, and 4 respectively). because, as noted above, these items have been shown to be the conventional realisations of apologies.

4.A local grammar of apology

As discussed in Section 2.1, local grammar analyses each discourse unit using a term that is directly related to its pragmatic function in social contexts. Based on a pilot study, we propose the functional elements shown in Table 2 for a local grammar analysis of apology. The identification of functional elements is fundamental for local grammars, because these are what make local grammars more specific and transparent and therefore distinguish them from traditional or general grammars. The working principle behind the proposal of these functional elements can be described in terms of “Occam’s razor”, which means that no more terms should be proposed than necessary.

The following presents the detailed local grammar analyses of apology expressions. According to their similarities and the degree of complexity, the analyses are divided into 6 sets, resulting in 14 specific local grammar patterns in total. The first set of analyses is straightforward, apology typically being expressed by idiomatic expressions such as (I’m) sorry, excuse me, and (my) apologies (Table 3a). Two variants of this pattern involve cases where the offense/reason for an apology is clearly specified, hence the label ‘Specification’ (Table 3b), and where the Apologisee is included (Table 3c).

In the second set of analyses, there is an element, usually a link verb (e.g. be, want to), that links the Apologiser and the action of apologising; this element is labelled ‘Hinge’. These instances instantiate the local grammar pattern Apologiser + Hinge + Apologising, as shown in Table 4a. A variant of this pattern is where the Apologiser increases the degree of sincerity of the apology made by using adverbs such as truly, really etc.; these are labelled ‘Intensifier’, as shown in Table 4b.

The third set of analyses is similar to the second one; the difference lies in whether the offense/reason for an apology is explicitly expressed (see also Table 3b). Instances in this category typically construe apology as Apologiser + Hinge + Apologising + Specification (Table 5a). A variant of this pattern is instantiated by those instances where there is an Intensifier, as shown in Table 5b.

It is worth noting that previous studies have also investigated types of offense categories (e.g. Holmes 1990Holmes, Janet 1990 “Apologies in New Zealand English.” Language in Society 19 (2): 155–199. ; Aijmer 1996Aijmer, Karin 1996 Conversational Routines in English. London & New York: Longman.). Deutschmann (2003Deutschmann, Mats 2003 “Apologising in British English.” Unpublished Ph.D. thesis, Umeå University, Sweden., 64) probably proposed the most comprehensive taxonomy of offenses, including accidents, mistake and misunderstandings, breach of expectations, lack of consideration, talk offenses, social gaffes, requests, hearing offenses, and offenses involving breach of consensus. These are all glossed as ‘Specification’ in this study, because one may not only apologise for an offense, but may also apologise by giving a reason (cf. Jucker & Taavitsainen 2008Jucker, Andreas, and Irma Taavitsainen 2008 “Apologies in the History of English: Routinized and Lexicalized Expressions of Responsibility and Regret.” In Speech Acts in the History of English, ed. by Andreas Jucker, and Irma Taavitsainen, 229–443. Amsterdam: John Benjamins. , 238), as in Sorry I’m late. The fact that ‘Specification’ covers both offense types and various reasons further indicates that functional labels used in the present local grammar analysis have achieved an appropriate level of granularity.

Local grammar patterns in the fourth set of analyses are instantiated by instances containing performative verbs of apology – apologize/se and regret. In these instances, there is usually no link verb and, consequently, the label ‘Hinge’ is not needed in the analysis. Apology in these instances is typically construed as Apologiser + Apologising (Table 6a); two variants are cases where elements labelled Specification and Intensifier are present (Tables 6b and 6c).

The fifth set is a minority, comprising only one instance. The peculiarity of this set is that both Apologiser and Apologisee are explicitly expressed; the local grammar pattern realised is Apologiser + Hinge + Apologising + Apologisee, as shown in Table 7.

The last set of analyses is quite different from those discussed above, which leads to the identification of another strategy of apologising, that is, to seek forgiveness from the offended (see also Jucker & Taavitsainen 2008Jucker, Andreas, and Irma Taavitsainen 2008 “Apologies in the History of English: Routinized and Lexicalized Expressions of Responsibility and Regret.” In Speech Acts in the History of English, ed. by Andreas Jucker, and Irma Taavitsainen, 229–443. Amsterdam: John Benjamins. ). Apology terms such as forgive and excuse are used for this purpose. The prototypical pattern in this category is Forgiveness-seeking (Table 8a); variants of this pattern include Forgiveness-seeking + Apologiser + Specification (Table 8b), and Apologisee + Hinge + Forgiveness-seeking + Apologiser (Table 8c).

Based on the above analyses, local grammar patterns of apologies are summarised and the quantitative information for each pattern is given in Table 9. We assume that the majority of apology expressions found in any naturally occurring texts can be analysed using these patterns. This assumption is made on two grounds. First, our manual check of three randomly selected episodes (Episode 5 in Season 2, 3, and 4 respectively) shows that searching those items listed in Table 1 enables us to identify all apology expressions therein. Although this does not allow us to claim that we can extract all apology expressions in a given corpus by searching these pre-determined items, this suggests that searching them can indeed yield a high hit-rate of apology expressions. Second, these patterns are not thought-up, but are generalised by analysing all the instances which contain those items and have the illocutionary force of apologising in TBBT. At this point, it should be noted, however, that the frequency of each pattern may vary from register to register or from genre to genre.

Overall, the local grammar analyses suggest two typical strategies for apologising, i.e. making apologies and seeking forgiveness; and the quantitative information indicates that the former is the prototypical way to apologise. Further, Figure 1 shows that the local grammar patterns identified in the first three sets of analyses account for 96% of all apology instances found in TBBT. This suggests that not only the formal realisations of apology are conventionalised, but also its functional patterns, which offers additional support to the observation that apology expressions are routinised.

Furthermore, combinations of different patterns are possible in face-to-face conversations. For example, the instance ‘I’m sorry, will you forgive me?’ combines the patterns Apologiser + Hinge + Apologising and Hinge + Apologisee + Forgiveness-seeking + Apologiser. The selection and combination of different patterns, as discussed in Bella (2014Bella, Spyridoula 2014 “A Contrastive Study of Apologies Performed by Greek Native Speakers and English Learners of Greek as a Foreign Language.” Pragmatics 24 (4): 679–713. , 682), depends on various factors, “such as the severity of the offense and the degrees of power and social distance existing between interlocutors”. Accordingly, the combination and corresponding complexity of apology patterns has two implications: one is that the more complex the pattern, the severer the offense (cf. Drew et al. 2016Drew, Paul, Alexa Hepburn, Piera Margutti, and Renata Galatolo (eds) 2016 “Special Issue on Apologies in Discourse.” Discourse Processes 53 (1–2): 1–131. , 2); and the other is that the more complex the pattern, the sincerer the apology.

The method for developing the local grammar of apology is replicable for building local grammars of other speech acts, in particular those whose realisations are also highly conventionalised. Take thanking as an example. Studies have shown that thank you and thanks are most frequently used to express gratitude (e.g. Aijmer 1996Aijmer, Karin 1996 Conversational Routines in English. London & New York: Longman.; Cheng 2010Cheng, Stephanie 2010 “A Corpus-Based Approach to the Study of Speech Act of Thanking.” Concentric: Studies in Linguistics 36 (2): 257–274.; Jautz 2013Jautz, Sabine 2013 Thanking Formulae in English: Explorations Across Varieties and Genres. Amsterdam: John Benjamins. ), which means that they can be used as key terms to search and identify gratitude expressions in corpora. The retrieved instances can then be analysed using functional terms that are designed for a local grammar of thanking. For the purpose of illustration, a set of sample analyses is given in Table 10 (see Su 2018Su, Hang 2018 “Thank Bloody God it’s Friday: A Local Grammar of Thanking.” Corpus Pragmatics 2(1): 83–105. for a more detailed discussion).

The above has presented a local grammar of apology and a partial local grammar of thanking, which, together with the study of requests (Su 2017Su, Hang 2017 “Local Grammars of Speech Acts: An Exploratory Study.” Journal of Pragmatics 111: 72–83. ), should have amassed sufficient evidence to show the possibility and feasibility of developing a set of local grammars to account more adequately for speech acts in general. It has to be pointed out that to fully develop local grammars of speech acts is a challenging task. The main challenge relates to the identification of speech act instances (see also Garcia 2015Garcia, Paula 2015 “Speech Acts: A Synchronic Perspective.” In Corpus Pragmatics: A Handbook ed. by Karin Aijmer, and Christoph Rühlemann, 29–51. Cambridge: Cambridge University Press.; Su 2017Su, Hang 2017 “Local Grammars of Speech Acts: An Exploratory Study.” Journal of Pragmatics 111: 72–83. ). As shown above, the analysis focused primarily on a set of conventionalised forms of apologies and, consequently, those instances which do not contain these forms have been left undetected. This implies that the local grammar developed is not entirely complete, which, however, is inevitable due to the fact that it is very difficult to (semi-)automatically detect all possible speech act realisations in naturally occurring texts (cf. Kohnen 2008Kohnen, Thomas 2008 “Tracing Directives Through Text and Time: Towards a Methodology of a Corpus-Based Diachronic Speech-Act Analysis.” In Speech Acts in the History of English, ed. by Andreas Jucker, and Irma Taavitsainen, 295–310. Amsterdam: John Benjamins. ). This would point to the significance of devising alternative methods which are robust and efficient to extract more exhaustively and reliably speech act instances in corpora.

5.Local grammar, functional grammar, and general grammar

While local grammars involve the mapping of functional elements onto formal elements, they are not just simply adding functional or semantic labels to the corresponding formal elements; rather, local grammars are function-oriented and represent an alternative approach, as opposed to general grammars, to linguistic description and explanation. Hence, it is necessary to discuss further the relationship between local grammars, functional grammars, and general grammars.

One defining feature of local grammars is that the functional labels used in a local grammar analysis are based on “the function of the sentence, not the words it has in it” (Hunston 2003Hunston, Susan 2003 “Frame, Phrase or Function: A Comparison of Frame Semantics and Local Grammars.” In Corpus Linguistics 2003, ed. by Dawn Archer, Paul Rayson, Andrew Wilson, and Tony McEnery, 342–358. University Centre for Computer Corpus Research on Language, University of Lancaster, UK., 345). The resulting description “is ‘functional’ in a different way from the tradition of functional grammar” (Hunston & Sinclair 2000Hunston, Susan, and John Sinclair 2000 “A Local Grammar of Evaluation.” In Evaluation in Text, ed. by Susan Hunston, and Geoff Thompson, 74–101. Oxford: Oxford University Press., 79), because what is meant by ‘functional’ from a local grammar perspective is “a grammar that would label each element of an analysed unit in terms that [are] related directly to its discourse function” (Hunston 2011Hunston, Susan 2011 Corpus Approaches to Evaluation: Phraseology and Evaluative Language. London: Routledge., 142). The local grammar of definition (Barnbrook 2002Barnbrook, Geoff 2002 Defining Language: A Local Grammar of Definition Sentences. Amsterdam: John Benjamins. ), for example, uses functional labels such as Definiens (the content of a definition) and Definiendum (the word being defined) to analyse definition sentences in the specific context of ‘defining’.

In contrast, few traditional functional grammars have described instances of one particular meaning or function within its specific context. Hallidayan systemic functional linguistics (Halliday & Matthiessen 2004Halliday, Michael A. K., and Christian Matthiessen 2004 An Introduction to Functional Grammar, 3rd edition. London: Arnold.) may be considered the currently most influential approach to functional grammars, but it cannot provide an adequate or a ‘real’ functional account of one meaning or function. Take apology as an example. Table 11 presents the systemic functional analyses of an instance of apology.

Put simply, Transitivity is concerned with the construal of ideational experiences through language, Mood with interpersonal aspects of language use, and Thematic analysis with textual organisation. Although the three types of systemic functional analyses are adequate in their own ways, it has to be pointed out that neither the Transitivity, nor the Mood or the Thematic analyses can straightforwardly reveal the function of this instance as an apology. This calls for alternative functional analyses; and local grammar analysis represents one candidate (Table 12).

Apologiser and Apologising relate directly to the function of I and apologise (see Section 4 for more examples). It appears that traditional functional grammars, represented here by Halliday’s SFL, and local grammars capture different aspects of the overall grammatical picture. Halliday’s SFL captures the regularities of language use in more ‘general’ terms, whereas the fact that local grammars use more context-specific and transparent terms to analyse instances associated with a chosen meaning or function makes their descriptions ‘local’ or specialised. Nevertheless, “[t]he loss in generalizability is compensated for by the gains in qualities such as accuracy, transparency, cumulative coverage” (Hunston & Thompson 2000Hunston, Susan, and Geoff Thompson (eds) 2000 Evaluation in Text. Oxford: Oxford University Press., 74). Specifically, a local grammar description is accurate because each semantic or pragmatic phenomenon is accounted for within its specific context; it is transparent because each discourse or utterance unit is analysed using a term that is directly related to its discursive or pragmatic function; and once local grammars of different meanings or functions have been developed, cumulative and generalised descriptions of language use can be achieved. Seen in this light, it is plausible to argue that the relationship between local grammars and traditional functional grammars is one of ‘complementarity’, which, using Halliday’s (2006Halliday, Michael A. K. 2006 “Afterwords.” In System and Corpus: Exploring Connections, ed. by Geoff Thompson, and Susan Hunston, 293–299. London: Equinox., 297) words, means that “each highlights different aspects of the total grammatical picture” and that the relation between these aspects is not one of ‘either … or’ but one of ‘both … and’ (Halliday 2008Halliday, Michael A. K. 2008 Complementarities in Language. Beijing: The Commercial Press., 36).

The next issue worth discussing is the relationship between local grammars and general grammars. Take the instance I apologise as an example again. Its analysis using traditional grammatical elements is shown in Table 13. Although the analysis might be a simplistic rendering of general grammars, it should be sufficient to indicate that, like the systemic functional analyses discussed above, general grammar descriptions cannot reveal the pragmatic function of the corresponding linguistic form either.

This then raises the question as to whether a local grammar approach would work better than general grammars to account for language in use. Barnbrook and Sinclair (2001)Barnbrook, Geoff, and John Sinclair 2001 “Specialised Corpus, Local and Functional Grammars.” In Small Corpus Studies and ELT: Theory and Practice, ed. by Mohsen Ghadessy, Alex Henry, and Robert Roseberry, 237–276. Amsterdam: John Benjamins. argued that it does.

Experiment will tell us whether the definition grammar is always superior to the general grammar, or whether there are some conditions where it is better to ignore the potential of some sentences as definitions. The likelihood is that such a specialised grammar will outperform a general grammar, and that raises some interesting questions for the future of grammars.(Barnbrook & Sinclair 2001Barnbrook, Geoff, and John Sinclair 2001 “Specialised Corpus, Local and Functional Grammars.” In Small Corpus Studies and ELT: Theory and Practice, ed. by Mohsen Ghadessy, Alex Henry, and Robert Roseberry, 237–276. Amsterdam: John Benjamins. , 273; emphasis added)

Clearly, in Barnbrook and Sinclair’s view, local grammar descriptions may work better, compared with general grammar descriptions, to explain how language is used. They further note that the reason “why a local grammar may be able to produce a more satisfactory analysis than a general one is that it has advance information of the communicative function of the sentence” (Barnbrook & Sinclair 2001Barnbrook, Geoff, and John Sinclair 2001 “Specialised Corpus, Local and Functional Grammars.” In Small Corpus Studies and ELT: Theory and Practice, ed. by Mohsen Ghadessy, Alex Henry, and Robert Roseberry, 237–276. Amsterdam: John Benjamins. , 249).

While we subscribe to the view that local grammars are better at capturing the pragmatic aspects of language used in interactive contexts, we would not argue that local grammars can ‘outperform’ general grammars. General grammars still have their indispensable value in language description and pedagogy, as has been shown in the history of linguistic research and education. What we argue, then, is that local grammars are an alternative approach to linguistic description and explanation, supplementing general grammars, and most notably, that local grammars are particularly useful for accounting for pragmatic functions,44. Sinclair (2010)Sinclair, John 2010 “Defining the Definiendum.” In A Way with Words: Recent Advances in Lexical Theory and Analysis, ed. by Gilles-Maurice de Schryver, 37–47. Kampala: Menha Publishers. suggested local grammars of words, exemplifying his suggestion with a local grammar of the word sever. This raises the possibility that local grammars may also be useful for accounting for pragmatic markers. as exemplified in the present study.

6.Conclusion

This study has reported on an investigation into a local grammar of apology, demonstrating the usefulness of a local grammar approach to studying speech acts. It has identified 14 local grammar patterns of apology which can be used to analyse most apology expressions found in any corpora, though quantitative features of each pattern may vary according to contexts. Two advantages of local grammars are especially worth recapitulating. First, local grammars use context-specific functional elements to analyse corresponding formal elements; the resulting description is therefore transparent and function-oriented. Second, compared with general grammars, local grammars are simpler in that each local grammar deals with one meaning or function only. In the case of speech acts, each local grammar accounts for one speech act type. Although this might imply a loss of generalisability of the description, this is compensated for by the gains of cumulative coverage achieved by a set of local grammars, as noted earlier.

The potential of local grammars in functional and pragmatic studies has not yet been fully exploited; consequently, more explorations into local grammar research are expected. Further investigation into local grammars of other speech acts would be particularly valuable and desirable. As Stubbs (2014)Stubbs, Michael 2014 “Searle and Sinclair on Communicative Acts: A Sketch of a Research Problem.” In The Functional Perspectives on Language and Discourse: Applications and Implications, ed. by María de los Ángeles Gómez González, Francisco José Ruiz de Mendoza Ibáñez, Francisco Gonzálvez García, and Angela Downing, 243–260. Amsterdam: John Benjamins. puts it, functional-pragmatic studies of language and discourse (e.g. speech act studies) need to be grounded in more corpus-based data, so as to investigate more thoroughly the phenomenon under examination, whereas corpus studies need to take into consideration the functions language fulfils in social contexts, so that findings of such studies can be strengthened by social rationale. Provided each local grammar deals with one particular area of language use and each speech act is concerned with one particular communicative function, research on local grammars of speech acts can offer important insights into the issue of how functional-pragmatic and corpus approaches to linguistic description and explanation can be reconciled. This would not only contribute to research into local grammars and speech acts, but also to corpus linguistics, pragmatics, and corpus pragmatics in general.

From a pedagogical perspective, research on local grammars of speech acts can greatly facilitate the EFL teaching and learning of how to perform speech acts appropriately. Usó-Juan (2010Usó-Juan, Esther 2010 “Requests: A Sociopragmatic Approach.” In Speech Act Performance: Theoretical, Empirical and Methodological Issues, ed. by Alicia Martinez-Flor, and Esther Usó-Juan, 237–256. Amsterdam: John Benjamins. , 237), for example, notes that “[r]esearch on the use of requests suggests that many learners have problems in performing this speech act in sociopragmatically appropriate ways”. Although Usó-Juan’s (2010)Usó-Juan, Esther 2010 “Requests: A Sociopragmatic Approach.” In Speech Act Performance: Theoretical, Empirical and Methodological Issues, ed. by Alicia Martinez-Flor, and Esther Usó-Juan, 237–256. Amsterdam: John Benjamins. discussion focuses on request, it is likely that EFL learners may also have difficulty in performing other speech acts appropriately, for example apology (Cheng 2017Cheng, Dongmei 2017 ““Communication is a Two-Way Street”: Instructors’ Perceptions of Student Apologies.” Pragmatics 27 (1): 1–32. ). The reason for this may be that there is no transparent and comprehensive description of speech act realisations available to learners. One solution, as suggested by Aijmer (1996)Aijmer, Karin 1996 Conversational Routines in English. London & New York: Longman., is to explore how to describe to learners the routines and their functional elements that are associated with one particular speech act. As shown in the present study, local grammars are a useful way to describe the routines of speech acts, identifying both formal variations and their corresponding functional patterns, which enriches the repertoire of strategies that can be employed by EFL learners to perform specific speech acts. Furthermore, the quantitative information obtained via corpus investigation can be used to inform EFL learners of the typical way(s) to perform a given speech act. This can contribute substantially to improving EFL learners’ pragmatic competence, indicating that research on local grammars has potentially valuable pedagogical applications.