Introduction
Language change is inevitable. If it wasn't, English learners would all be trying to sound like King Alfred. There is never a period of stability in language and the only languages which have reached a kind of equilibrium are those like Latin where are there are no longer any native speakers. The pressure for change on English is particularly high because of its global status and the diversity of contexts in which it operates. In 2006 David Graddol (p. 101) stated that in 2010 two billion people would be learning English. The size of the figures involved makes it impossible to verify whether this prediction was accurate but Graddol's most recent publication (Reference Graddol2010: 68) states that up to 350 million people may speak English in India alone. Obviously, most of these English users around the world speak it as a second language. Consequently, any discussion of change in modern English must take into account the input of those who have had to learn English. The purpose of this article is to present examples of learner language which demonstrate principles and mechanisms of language change through the much-discussed phenomenon of language play.
There is a great deal of resistance to crediting learners with an active role in the development of English. Learners are seen as on the receiving end of change when it comes and their attempts to be innovative are dismissed as uninformed. Native speakers have a licence for change which is withdrawn for learners. Prodromou (Reference Promodrou2007) has tested these different perceptions empirically. He investigated the notion of creative idiomaticity whereby variations on fixed idioms are used, often for humorous effect, as in it's raining kittens and puppies. Prodromou's hypothesis was that listeners would judge the appropriateness of creative idiomaticity on the linguistic status of the speaker. He asked correspondents to respond to the sentence I bumped into some information in the library. The phrasal verb is marked here because it does not take a human object. Half of the correspondents were told that this sentence was from a native speaker, half were told that it was from a learner. Predictably, the correspondents who thought this was first language use were far more accepting. Promodrou (p. 22) noted the contradiction.
The simultaneous existence of collocational norms and the potential for violating these norms is a defining feature of L1 competence and, at the same time (paradoxically), it seems to be…one of the defining features of the limits of ‘non-native’ competence, even at advanced level.
The lay person maintains a strict first / second-language speaker dichotomy that does not entertain the concept of change originating from learners. Innovation has a ceiling effect on learner production.
Much depends on what is deemed to be language change. Innovation and change are used interchangeably here when some commentators (e.g., Britain, Reference Britain2008) distinguish between them in terms of their transitory and permanent impact respectively on the language. A full discussion of the nature and motivation for change remains beyond the scope of this article. Change can be defined and approached variously but the argument made here is that it can originate from learners' attempts at language play. As demonstrated by Prodromou's colourful examples, high proficiency is characterised by the ability to manipulate form in order to make production more personalised, striking and meaningful. No matter that much of this usage will be idiosyncratic and ephemeral, it still illustrates the potential of learners to influence the direction of English. Change cannot be prescribed, it is hardly predictable, but it has its source in the growing acceptance of usage by the larger speech community. Once learners are adopted into this community, their isolated attempts at language play can be considered as part of the phenomenon which powers change.
The ludic dimension of language has received considerable attention. In a book length study, Cook (Reference Cook2000) claims that language play is instinctive as part of the expression of self-identity and construction of a suitable persona. Cook provides many examples of child language play but studies drawing on learner language are much rarer. O'Keeffe et al. (Reference O'Keeffe, McCarthy and Carter2007: 190) identify creativity in a corpus of second language emails in the way that two respondents merge written and spoken registers for humour and solidarity; Tin et al. (Reference Tin, Manara and Ragawanti2010) discuss novel word combinations such as humble to the blackboard. Both studies are informative but small scale and the field of enquiry needs more empirical input. This article illustrates from a learner corpus that language is a game which learners can play on equal terms with native speakers. In any game there is an element of sportsmanship. By bending the rules, competitors can improve their own chances in the game as well as contribute to a wider reevaluation of the rules. This is a risky endeavour for naturally the game is refereed and not all violations of the rules are tolerated. As is shown, the most successful attempts to circumvent the rules still keep within the spirit if not the law of the game.
Data and discussion
The learner data all come from the International Corpus of Learner Language (ICLE) (Granger, Reference Granger2003). The ICLE is a 2.5 million word corpus of academic essays on a set number of topics, typically argumentative in genre. The contributors, representing eleven European nationalities, were all advanced learners of English studying in tertiary education. The ICLE is not the only public learner corpus but it is recognised as the most well established (see the survey of Pravec, Reference Pravec2002). Obviously, it is not possible to do an automatic search for language in the corpus which is innovative. The following examples were identified during a manual parsing of the corpus as part of a separate project (Rimmer, Reference Rimmer, Taylor and Weir2009). The approach then is admittedly ad hoc. However, the examples selected are for illustrative purposes only. It is not claimed that their specific linguistic features represent the corpus, or learner language, as a whole. They do provide a basis for the argument that learners are involved in the creative process which underpins the mechanism for language change.
Vocabulary is more susceptible to change than grammar (Mair & Leech, Reference Mair, Leech, Aarts and MacMahon2006) and lexical variation is certainly at play in the ICLE. To illustrate, there is no phrasal verb to hermit away in the dictionary so the following line, labelled with its code in the ICLE, is striking.
… people get hermitted away through computers and electronic devices. ICLE-BG-SUN-0040.1
The meaning is clear: technology is isolating people. The metaphor of a hermit gives a powerful image of miserable loneliness. The writer uses two grammatical devices to make hermitted away seem a legitimate coinage. First, the clause is marked by being in the passive voice. This markedness creates a certain tolerance for novelty. Second, the copula is get rather than be, although, from data in the British National Corpus (BNC) (Biber et al., Reference Biber, Johansson, Leech, Conrad and Finegan1999: 481), get is (1) much less common than be in the passive and (2) virtually restricted to speaking. Get is chosen because it is associated with accidents and unfortunate events (Fleisher, Reference Fleisher2006). This writer expertly manipulates the grammar to achieve a memorable metaphor.
In this instance hermitted away is convincing because it is motivated by awareness of norms and sensitivity to context. Error or incongruity would spoil the effect. Nevertheless, this citation does not constitute evidence that a new phrasal verb has been added to the lexicon. Isolated usage, however linguistically-informed or effective, cannot be incorporated into the universal record. The writer has played with the language, not changed it. However, in the spawning of a new word combination, the writer shows an appreciation that language can be manipulated and taken beyond its boundaries. This awareness is critical to change. Without confidence and competence in moulding existing language rules to new environments, change is a non-starter. Thus, the occurrence of hermitted away itself while striking is relatively insignificant. Of much greater import is the recognition that learner language can display a considered and deliberate interplay of form and meaning which may override linguistic convention if communicatively expedient.
There is something amusing about hermitted away that belies the serious social point being made. Humour seems bound up with language play, as evident in the next example.
Mum and dad are looking at each other like “What did we do to deserve all this?” ICLE-DB-KVH-0058.2
There is a comical contrast of the touching and ridiculous in the hopelessness of the parents. Grammatically, this sentence includes one of the most discussed recent developments in English, the quotative be like structure as in I'm like, ‘What does he know about it?’. Carter & McCarthy (Reference Carter and McCarthy2006) attribute its usage to situations when ‘the report involves a dramatic representation of someone's response or reaction’ (p. 102). Using a spoken corpus of Toronto English, the sociolinguistic study of Tagliamonte & D'Arcy found that since its emergence in the 1980s, quotative be like is now the most common way of representing direct speech amongst under thirty year olds. They surmise that be like ‘gained prestige as a trendy and socially desirable way to voice a speaker's inner experience’ (p. 212). Tagliamonte & D'Arcy arrived at this conclusion through analysis of the many variables which impact grammatical change, including age and gender of the speaker, context and pragmatic force. It is noteworthy that the language status of speakers was not considered although in a culturally diverse city such as Toronto it is conceivable that a proportion of the informants did not speak English as a mother tongue. The assumption was that innovation is powered exclusively by native speakers.
However, it is not certain that change is always unidirectional from first to second-language speakers. Advanced users could initiate change themselves or at least be part of the change mechanism. Consider again the example from the ICLE.
Mum and dad are looking at each other like “What did we do to deserve all this?”
Unlike any of the examples in the Toronto corpus cited by Tagliamonte & D'Arcy, be like is discontinuous: arelooking at each other like. This is a syntactic refinement on the basic be like construction. It allows more flexibility in sentence construction as the predication of the subject can be fuller. Compare the three sentences below.
Mum and dad are like ‘What did we do to deserve all this?’
Mum and dad are looking like ‘What did we do to deserve all this?’
Mum and dad are looking at each other like ‘What did we do to deserve all this?’
The third sentence are + participle + prepositional complement + like is the most expressive. The discontinuous structure is recursive so, syntactically, there is no limit on the size and number of intervening elements.
Mum and dad are looking at each other in a blank daze of horror like ‘What did we do to deserve all this?’
The structure now is are + participle + prepositional complement + adverbial + prepositional complement + like. The writer does not select such a long sequence but grammatically it is available. In expanding the construction yet preserving its grammaticality, the writer shows its potential development. Here, learner language highlights the direction of change. This does not mean quotative be like will develop on these lines – indeed, it may fall out of fashion completely – but an argument can be made which meets the syntactic and pragmatic dimensions of the emerging construction.
The last point is critical for there needs to be a convincing linguistic argument for claiming that non-canonical learner language is well motivated. Otherwise, any novel usage could be defended as creative, resourceful, etc. There has to be a boundary between error, ignorance of the rules, and language play, recognition and exploitation of the rules. Obviously, only language play is relevant to language change if the latter is regarded as competence based. A study (Cross & Papp, Reference Cross, Papp, Gilquin and Papp2008) of verb + noun combinations in the ICLE agrees that novel usage is only legitimate if it can be proved that learner language is informed. This is not an objective science and in practice there are many borderline cases between error and well-motivated usage. The credibility of the final judgment depends on the adequacy of the argument made. Consider the comparative clause below.
Has our schoolsystem gone? this mad that, already in nursery school, they have to prepare the children for their future jobs. ICLE-DB-KVH-0016.2
One would expect so in place of this. There are citations of so mad + finite clause in the BNC.
He says he was so mad that he got a knife to scare them. K23 4010
There are no examples of this adjective + finite clause in the BNC. This is basically either a determiner or a pronoun. Functions of this as an adverb are semantically very restricted and confined to speaking. For example, from the Longman Corpus of 300 million words (Summers, Reference Summers2005: 1727), this is used to indicate size when the speaker uses corresponding hand signals, as in the boastful fisherman's line ‘It was this big!’ While the meaning of this adjective + finite clause is perfectly clear on analogy with the standard comparative construction, its usage here is idiosyncratic and not part of the shared language community.
Still, it is problematic to reject the construction on purely linguistic grounds. That is well-attested as a modifier, even appearing in the ICLE.
…this soil is not that fertile anymore. ICLE-BG-SUN-0101.1
There is also a pattern that adjective + comparative clause, labelled in the Longman Corpus (ibid: 1716) as British, spoken, informal.
I was that embarrassed I didn't know what to say.
Why should this not occupy the same functional ground as that? The deictic distinction between this / that could be maintained with so as the unmarked variant.
The original sentence can be restructured to represent a system of linguistic choices.
Has our school system gone so mad that, already in
that
this
nursery school, they have to prepare the children for their future jobs.
Grammatically, there is a gradient from standard to colloquial to marginal. Pragmatically, there may also be variation. So + adjective + that is unmarked. That + adjective + that emphasises the undesirability of the situation. The Longman dictionary entry (ibid) points out that this structure comments on ‘bad’ states of affairs. More generally, that has the force of problematising situations (Carter & McCarthy, Reference Carter and McCarthy2006: 180). The repetition of that, once as a strong and then as a weak form, is also emphatic. This + adjective + that internalises the situation more deeply. The focus is not so much on blame and anger as personal hurt and disappointment. The meaning of the sentence encourages this interpretation. The use of the plural personal determiner in our schoolsystem identifies the speaker with a whole community and culture. The pronoun they is intriguing for it has no obvious anaphor – the school system is singular and non-human. [T]hey is plausibly the teachers or all the mandarins in the educational machine. In any case, they is portrayed as an external faceless entity that is part of the negative system.
In short, an argument can be made for this + adjective + that based on the deictic system and the meaning expressed by the sentence. If one of these elements is missing, the argument falls down and the construction is an error. For example, my + adjective + that is indefensible as there is no my / that deictic contrast and my is never a modifier. Out of context, this + adjective + that is illegitimate – one would surely not teach it to second-language learners – but it can be condoned, rather than penalised, if it can be demonstrated that the writer is making a systematic choice based on an awareness of the grammatical options available. It is undeniable that the structure is marginal, but it is less clear that it results from a lack of grammatical competence.
This example nicely shows the tension between error and creativity that makes language play a rather risky undertaking for learners. If it works, the impression is of high proficiency and ingenuity; if not, the conclusion is a flawed competence. Many learners, particularly in critical situations such as high-stakes examinations, may prefer to play safe, so to speak, and avoid the risk of failure. Language change thus has a precarious existence in its initial stages because it depends upon a weight of evidence which has a dubious status as individual structures, particularly when learners are supplying the data. Language play may be pervasive but that does not mean it is always appreciated either on a local level, as an attempt to maximise meaning, or a global level, as a principle behind language change. Learners are likely to be more sensitive to the former rather than the latter for they will be judged as individuals producing individual utterances. This, as demonstrated, does not make learners less involved in the process of language change but it probably makes them more conscious of the costs as opposed to benefits of challenging norms.
Conclusions
For learners, there are heavy restrictions on the permissible degree of deviance from standard English. Change results from variation on established patterns rather than the creation of unattested forms. The coining of a morphological or syntactical unit, e.g. a new subordinator, would certainly not be admissible. Language play then demands considerable competence and ingenuity in manipulating the rules while adhering to the basic lexical and grammatical framework. Why do learners do this? Put more bluntly, what is in it for them? Given that the risk of condemnation, even misunderstanding, is high, there must be some internal stimulus for language play. It is unlikely that learners see themselves as instruments of language change so their choice of language must come down to the specific context of production. Obviously, in a corpus methodology the author's original intention is irrecoverable so the following analysis of factors favouring language play is strictly speculative.
As mentioned earlier, users have a natural propensity to play with language. This might be prompted by linguistic curiosity, creativity, non-conformity, a search for self-identity, or just a sense of fun. These factors are intertwined and difficult for users to isolate, let alone articulate. It is impossible to exclude learners from such a natural process. Indeed, as asserted by Cook (Reference Cook2000), language play is fundamental to the learning experience. Developmentally, it allows learners to experiment with language and take it in new directions. Psychologically, it provides a sense of ownership of the second language which is critical in forging a linguistic identity. Language play is as natural as making mistakes and both behaviours are equally formative. The difference is that language play is largely a positive outcome that is incremental with, not detrimental to, language level.
In addition, language play has a super-communicative function, that is to say it transcends considerations simply of communicating content. Certainly, the basic message can be transferred in conventional language constructions. But there are occasions when learners want to go beyond getting their meaning across. For example, one goal may be style, the considered selection and arrangement of language to enhance rather than deliver the message. Style has literary associations but it is a device open to all users in diverse contexts. The quotative structure discussed above, Mum and dad are looking at each other like “What did we do to deserve all this?”, was an example of a spoken register being transferred to the written form and hence engineering a colloquial style which appeals to the reader. Playing with language allows finer distinctions of not just meaning but tonality, building a deeper relationship between participants by adding semantically-loaded layers to the language exchange. An aesthetic motivation for language play may also be a factor.
There are pedagogic implications of these remarks. Teachers should have an interest in encouraging language play. The ability to extend the language resource to its limits results in greater linguistic sensitivity and more meaningful communicative events. To be in this position, learners need to be given opportunities to apply their linguistic competence in tasks which demand a degree of creativity and self-expression. Routine and formalised situations need not exclude language play but they will struggle to inspire learners in a classroom scenario. Personalised and open tasks on the other hand allow the degree of engagement and free response which is essential for opening up learners to the latent possibilities within the language system. To reiterate, knowledge and rules are still important so no credible second-language pedagogy could exclude a focus on form. However, the next, and trickier, step is to foster, not inculcate, within learners the skill of fashioning meaning through the deliberate and artful manipulation of the language available to them.
Change comes as a result of many factors and language play is but one ingredient in the mix. It is difficult to appreciate its significance without the availability of more learner data. A major drawback to appreciating the role of learners in language change is the paucity of learner corpora. For example, the ICLE used in this study is the largest learner corpus but it contains only written English of a loosely academic genre. A greater variety of larger and more representative learner corpora will allow second language production to be valued on a level which matches its sociolinguistic importance. Second-language speakers now outnumber native speakers, Crystal (Reference Crystal2008) suggests, by a ratio of three or four to one, but it is more than a numbers game. Native speakers will remain an oligarchy with a disproportionate influence over English unless learner language is appreciated on a fuller range of dimensions. Understanding the relationship between language play and language change is one way of ensuring a more level playing field.
WAYNE RIMMER has taught English to second-language learners in Moldova, Russia, Thailand, Germany and the UK. He is co-author of Cambridge Active Grammar. a pedagogical grammar for teenagers to be published by CUP in spring 2011. His PhD research topic investigates grammatical complexity in the International Corpus of Learner English. Email: wrimmer@hotmail.com
