Ian Cushing raises important issues, both specific and general, in his provocative article (Cushing Reference Cushing2019), so I am grateful to the editors for allowing me to offer a comment.Footnote 1 The main issue in his article is the role of language policy in language planning (and especially in language education); and in particular, who is responsible for prescriptivist bias against nonstandard varieties? This is an important question for sociolinguists who wish to campaign for change in education, and who therefore need to know who to persuade: those at the top (government ministers) or at the bottom (practising teachers)? Such questions seem to be as significant in contemporary sociolinguistics as they ever were (M. Lewis Reference Lewis2018). But even without these practical concerns this question would still matter for our attempts to understand how society treats language. Another issue for sociolinguistic research is methodology: how valid are the methods that Cushing advocates and uses? I discuss both of these issues, but my main focus is on the first.
My worry about Cushing's article is that it can easily be read as a very pessimistic analysis of societies such as ours in which the top-down power structures are so strong, and the motivation of government is so suspect, that nothing can change: prescriptivism is built into society. I find this at odds with what I know about Cushing's own intense activity to counter prescriptivism, which only makes sense when there is hope of change. He and I, together with a number of other colleagues (who I mention below), have invested many years of work for change in this area—not least in a recent co-authored book which argues strongly against prescriptivism (Aarts, Cushing, & Hudson Reference Aarts, Cushing and Hudson2018:10), and I believe that collectively we have already had a positive impact on schools. The aim of this comment is therefore to present a more optimistic interpretation of the data that Cushing reports on.
Turning to matters of methodology, Cushing presents his theoretical assumptions in the following passage:
this article assumes that language prescription is part of language policy, adopting a critical discursive approach… in order to trace metalinguistic discourse across policy ‘levels’, from government-produced curricula through to classroom practice. Here then, critical involves unpacking ‘[linguistic] strategies for maintaining social power’…, and discursive involves examining the interplay between policy levels and the degrees of agency that policy arbiters such as teachers, management, the media, and government have… (Cushing Reference Cushing2019:425)
As we all know, research is based on evidence that comes from the world rather than from the initial assumptions; but Cushing's project seems to reject this principle by assuming from the start precisely what it claims to prove as a matter of fact. It assumes that prescription is part of policy, and that this is a strategy for maintaining social power. I would argue that both of these claims should be seen as matters of fact subject to verification against the available research evidence; and researchers should beware of being guided by such assumptions to find confirmatory evidence and to ignore all else. Personally, I find both assumptions hard to accept, so I need evidence. I am persuaded by Cushing that prescriptivism is a serious issue in our schools; but I am more inclined to believe that this is the result of cultural inertia: incompetence in our leaders and lack of training in our teachers, rather than a power-grabbing conspiracy.
Cushing discusses three kinds of data:
• Government policy as revealed in official documents: the 2014 National Curriculum for England, and the annual grammar tests for eleven-year-olds prescribed by the government for all English primary schools
• School policy as revealed in media coverage of prescriptivism in schools
• Teachers’ policy as revealed in sixteen interviews
I do have access to the same sources of data as Cushing regarding government and school policy, but not to his interview data; consequently, most of my more specific comments focus on the former, with some rather general comments about the survey of teachers.
Cushing's general conclusion is summarised below.
Broadly, then, I argue that instances of language policing in schools arise as a result of contact points between macro-level curriculum policy and micro-level aspects of teacher knowledge, and that language policing in schools is endorsed and legitimised by the media, who engage with such policies in uncritical ways. (Reference Cushing2019:427)
My counter-argument is that when a school or teacher decides to ‘police’ nonstandard speech, this is in spite of the official government policy, rather than being ‘endorsed and legitimised’ either by government policy or by the media.
Government policy
The National Curriculum has gone through several revisions since it was first introduced by act of parliament in 1990 (UK Government 1990), but in every version there have been explicit rejections of prescriptivism. (The early versions of the curriculum are hard to find online, so interested readers are referred to the list at ‘government documents’ in the website of the LAGB Education Committee.) The first curriculum was published in 1990, and revisions followed in 1995, 1999, 2007, and 2014.
• 1990: Pupils should be encouraged to respect their own language(s) or dialect(s) and those of others. (Programme of Study 1:11 in Hudson Reference Hudson1992:269)
• 1990: Pupils should be able to show in discussion an awareness of grammatical differences between spoken Standard English and a non-standard variety; e.g. take note of different ways in which tense and person are marked in the verb ‘to be’ after listening to recordings or participating in classroom improvisations. (Attainment Target 1:6 in Hudson Reference Hudson1992:273)
• 1995: All pupils are therefore entitled to the full range of opportunities necessary to enable them to develop competence in standard English. The richness of dialects and other languages can make an important contribution to pupils’ knowledge and understanding of standard English. (Department for Education 1995:2)
• 1995: Pupils should be introduced with appropriate sensitivity to the importance of standard English. Pupils should be given opportunities to consider their own speech and how they communicate with others, particularly in more formal situations or with unfamiliar adults. … They should be introduced to some of the features that distinguish standard English, including subject-verb agreement and the use of the verb ‘to be’ in past and present tenses. Pupils may speak in different accents, but they should be taught to speak with clear diction and appropriate intonation. (Department for Education 1995:5)
• 1999: Pupils should be taught: (a) how written standard English varies in degrees of formality (b) some of the differences between standard and non-standard English usage, including subject-verb agreements and use of prepositions. (Department for Education and Employment 1999:57)
• 2007: When teaching standard English, it is helpful to bear in mind the most common non-standard usages in the UK for subject-verb agreement (they was), formation of past tense (have fell, I done), formation of negatives (I ain't), formation of adverbs (come quick), use of demonstrative pronouns (them books), use of pronouns (me and him went) and use of prepositions (out the door). (Qualifications and Curriculum Authority 2007:64)
• 2007: The study of English should include… variations in written standard English and how it differs from standard and non-standard spoken language. (Qualifications and Curriculum Authority 2007:72)
• 2014: At this stage, pupils should start to learn about some of the differences between Standard English and non-Standard English and begin to apply what they have learnt. (Department for Education 2014:41)
• 2014: [Pupils should use] Standard English forms for verb inflections instead of local spoken forms [for example, we were instead of we was, or I did instead of I done]. (Department for Education 2014:78)
• 2014: knowing and understanding the differences between spoken and written language, including differences associated with formal and informal registers, and between Standard English and other varieties of English (Department for Education 2014:84)
• 2014: Standard English can be recognised by the use of a very small range of forms such as those books, I did it and I wasn't doing anything (rather than their non-Standard equivalents); it is not limited to any particular accent. It is the variety of English which is used, with only minor variation, as a major world language. Some people use Standard English all the time, in all situations from the most casual to the most formal, so it covers most registers. The aim of the national curriculum is that everyone should be able to use Standard English as needed in writing and in relatively formal speaking. (Department for Education 2014:103)
• 2014: Examples in the glossary:
• I did it because they were not willing to undertake any more work on those houses. [formal Standard English]
• I did it cos they wouldn't do any more work on those houses. [casual Standard English]
• I done it cos they wouldn't do no more work on them houses. [casual non-Standard English] (Department for Education 2014:103)
These passages reflect a very different policy from the ‘prescriptive language policies and agendas’ that Cushing claims to find in government documents (Cushing Reference Cushing2019:430). They consistently contrast Standard English with non-Standard English—not with ‘incorrect’ or ‘bad’ English. Many of them include examples classified as non-Standard English, and not as ‘errors’. And many of them reflect the influence of linguists (and especially of sociolinguists), because they were commented on or even created by linguists; for instance, the Committee for Linguistics in Education submitted comments on the draft for the 1995 and 2014 revisions (see ‘responses’ on the Committee's website), and I myself supplied the examples and drafted the definition of Standard English. Indeed, linguistics has been influential since before the first version of the National Curriculum was drafted thanks to a number of distinguished colleagues who sat on the two committees that laid the ground for the new curriculum in English: Gillian Brown and Henry Widdowson on the Kingman Committee and Katharine Perera and Michael Stubbs on the Cox Committee (Kingman Reference Kingman1988; Cox Reference Cox1989). The linguistics community should celebrate the impact that it has had, via government documents, on government policy.
Unfortunately, as we all know, successes tend to be incomplete: the 2014 National Curriculum does contain two passages which confuse standardness with correctness.
• [Pupils] should be taught the correct use of grammar. (Department for Education 2014:11)
• [Pupils should ensure] correct subject and verb agreement when using singular and plural, distinguishing between the language of speech and writing and choosing the appropriate register. (Department for Education 2014:48)
These passages were presumably drafted by a different civil servant who was unaware of the stance taken elsewhere, and they may not have been part of the draft that was open to public comment. Moreover, Cushing may be right in viewing the policing of language by government as ‘an increasingly pervasive phenomenon’ (Cushing Reference Cushing2019:426). It is striking how the liberal tone of the first National Curriculum gradually hardens through later revisions, even though it has not been replaced by downright proscription of nonstandard varieties. In 2006 it was possible to write that the National Curriculum ‘has totally rejected the prescriptive mentality. Standard English continues to be seen as a major educational goal, but it is viewed in an inclusive way, with all varieties… spoken and written, formal and informal, professional and everyday, taken into account’ (Crystal Reference Crystal2006:206). By and large this continues to be true in 2019.
Another element of government policy which Cushing discusses is the national system of tests for primary children in which ‘students are required to identify grammatical constructions in decontextualized sentences, and ‘correct’ any examples of nonstandardised language’ (Reference Cushing2019:429).Footnote 2 It is true that the 2017 test did include a question which required a nonstandard form to be ‘corrected’, but this question is the exception, not the general pattern. Every other test also includes a question about standardness, but framed differently, and much more acceptably: ‘Which sentence is in Standard English?’, followed by a list of four sentences, three of which contain nonstandard grammar. Once again, we can celebrate the fact that the linguists who advise on these tests (Bas Aarts and myself) have had so much impact, even if the odd error slipped past us. It could of course be objected that any attempt to teach or test Standard English is objectionable for social and political reasons, but this is a minority view. It is unusual nowadays to meet disagreement with the principle expressed by the Kingman Committee who write:
one of the schools’ duties is to enable children to acquire Standard English, which is their right. This is not a matter of controversy: no item of evidence received by the Committee contained disagreement with this point. (Kingman Reference Kingman1988:14)
To summarise the discussion so far, I have contested Cushing's claim that government policy is prescriptive. On balance, every version of the National Curriculum has taken a rather positive approach to non-Standard English, though the most recent version contains a couple of lapses. Nor do the grammar tests support Cushing's claim, because only one question (out of fifty questions) addresses standardness, and this is generally (with just one lapse) phrased without prescription. In contrast with Cushing, I have noted, and celebrated, the influence that linguists have had on the government's official documents.
School policy and the media
Cushing's study of school policy on non-Standard English is based on twenty-eight newspaper reports uncovered on the internet by googling for the terms school, language, dialect, and ban. The articles concerned reported school-wide bans on non-Standard English, enforced via printed lists of banned words.
Unfortunately when I tried to replicate this search it only produced thirteen relevant hits (among a large number of irrelevant newspaper articles about other countries or academic publications);Footnote 3 maybe this discrepancy was due to the way in which Google tailors its search to the user's personal history, but it does raise serious questions about whether internet searches can ever produce solid evidence for a scientific study.
More to the point, my thirteen articles presented a very different picture from Cushing's, which he summarised as follows: ‘An overall picture of the data reveals a dominant prescriptive discourse’ (Cushing Reference Cushing2019:434). Only one article (from The Express) clearly supported the reported ban on nonstandard words, while one (the 2014 one from the Daily Mail) reported positive exam results following the ban. All of the other reports gave far more prominence to criticism of the ban from local parents, who felt affronted by this attack on their dialect; so, contrary to Cushing's analysis, the articles could be read as an attack on prescriptivism. Of my thirteen reports, four (SecEd, TES, Guardian, The Day) even quoted the views of a sociolinguist, and one (Daily Mail 2013) included a detailed and objective discussion of the local dialect. Moreover, six of the schools, when interviewed, qualified the ban by saying that they respected the local dialect and that the ban only applied to its use within the school; and one report even denied that there was any kind of ban.
In short, my little attempt at replicating Cushing's research project reveals a reassuringly liberal approach to nonstandard varieties not only among the individuals quoted in the reports, but also among the newspapers and (presumably) their target readership. Moreover, the fact that the newspapers saw these school bans as newsworthy should in itself give us pause for thought. As Cushing himself points out, the typical is not newsworthy, so by reporting these bans, journalists are telling us that they are exceptional. Even if we take Cushing's figures, newspapers only found seven schools with dialect bans during the period from 2008 to 2019—a tiny number among the 24,000 schools in England. By implication, the vast majority of schools have no blanket ban on nonstandard varieties, but teach Standard English as an additional variety for use in educational settings. Once more, linguists may be able to take some credit for this state of affairs, thanks to the writings and campaigning in schools of our colleagues such as David Crystal and Deborah Cameron (Crystal Reference Crystal1984, Reference Crystal2003; Cameron Reference Cameron1995), a tradition continued more recently by a number of linguists including Rob Drummond, Dan Clayton, Sue Fox, and Ian Cushing himself, who continue to spread the message of linguistics not only through the traditional medium of books (Clayton & Drummond Reference Clayton and Drummond2018) but also through blogs and other more modern media.
Teachers’ policy
Finally we come to the classroom teachers and their alleged prescriptivism. Cushing's account is based on interviews with sixteen primary teachers, who certainly provided clear examples of prescriptive attitudes; it is unfortunate that we are not told how many of the sixteen shared these attitudes. Since Cushing does not distinguish prescriptive and nonprescriptive teachers, but draws gloomy general conclusions, maybe we must assume that most, or even all, of them were prescriptive. But were they typical of the teachers in our primary schools? After all, they were not selected randomly but selected themselves by offering to work with Cushing, so it is quite possible that they saw the research as an opportunity to promote their prescriptivism.
A good starting point for the discussion is the Bullock Report of 1975, which reported as follows.
A substantial number [of teachers] considered that the express teaching of prescriptive language forms had been discredited, but that nothing had been put in its place. They could no longer subscribe to the weekly period of exercises, but they felt uneasy because they were not giving language any regular attention. It seems to us that this uncertainty is fairly widespread, and that what many teachers now require is a readiness to develop fresh approaches to the teaching of language. (Bullock Reference Bullock1975:172)
This passage suggests that by 1975 ‘a substantial number’ of teachers had already abandoned prescriptivism, at least in their classroom practice. The same seems to have been true as recently as 2010.
In other European countries argot and slang are not allowed into the classroom; children know exactly what is ‘correct’ usage in their main language, and what is not. In this country, by contrast, primary school teachers—dedicated as many of them are to ‘childled’ education—don't feel that it's their role to interfere with self expression in any shape or form. (Gross Reference Gross2010:28)
The teachers described in these two extracts are polar opposites of those in Cushing's cohort. Admittedly neither extract reports a systematic survey, but neither does Cushing's report, so the only reasonable conclusion is that we simply do not know how many primary (or secondary) teachers hold prescriptive views and apply them in teaching. This is clearly an important gap in our knowledge which calls for research, and Cushing has done well to put it on our collective agenda.
Discussion
I have argued that government policy, as expressed in the various versions of the National Curriculum, is relatively benign on the issue of prescriptivism. It only applies terms such as correct in discussion of spelling and punctuation, where they are more or less uncontentious; and in particular it almost always uses the term non-Standard English rather than prescriptive terms such as error or bad English. I have also argued that the media reports on prescriptivism in schools actually suggest that this is a rare phenomenon, and that when it happens it is out of step with parts of the local culture; so although the reports alert us to the continuing attractions of prescriptivism, they give little idea of how many schools adopt it as school policy. And finally I have suggested that, although some teachers hold prescriptive views, these teachers contrast with an unknown number who do not. In short, the following claim made in 2005, while clearly overoptimistic, was not far from the truth:
prescription is dead. Non-standard varieties are tolerated, as are informal registers; variety is accepted, but different varieties are suited to different occasions so the focus is now on the matching of variety to context. (Hudson & Walmsley Reference Hudson and Walmsley2005:615)
Leaving aside the unknowns, are there any general lessons for sociolinguists to learn from Ian Cushing's article and my commentary? It seems to me that the main issue is whether sociolinguists and other linguists can have any influence on the world. Sociolinguistics inevitably brings us face to face with social concerns, and there is a particularly strong natural link between sociolinguistics and education, so the question is how we academics can influence education, preferably for the better.
It is tempting to think, as I myself have in the past, that the way to do this is to target those who hold the levers of power: the government, its agencies, and its official documents such as the National Curriculum for England. This is certainly important, and I hope to have shown that academics have had some influence at this level. But I think the real lesson is that governments actually have much less control than they would like over what happens in classrooms, so convincing government is only half the battle. After all, not only Cushing's teachers with their prescriptive views but also the schools with prescriptive policies seem to be operating as free agents beyond the control of the government.
A particularly clear example of this disconnect between official policy and actual practice is the status of grammar teaching in our primary schools. This has been mandatory in primary schools since the first revision of the National Curriculum in 1995, and each revision spelled out in some detail what should be taught; but it has only become controversial since the government also introduced mandatory tests of grammatical knowledge in 2013. Many teachers have complained bitterly since then about the intellectual load of these tests and the difficulty of teaching grammar, but without noting that such teaching had been part of the National Curriculum for two decades. Presumably this omission can be taken as evidence that teachers had simply ignored the grammar requirements until the tests made that impossible. The point of this digression is that government policy does not necessarily turn into classroom practice.
The key to any educational reform is the classroom teacher. Teachers live and work in a complex culture which influences their attitudes, so if prescriptivism is part of that culture, they may well have prescriptive views and teach in prescriptive ways. This world is very different from the neat world of government policy, where there is a clear command structure: it is diffuse and multi-faceted, and the agents are individual human beings with the strengths and weaknesses that we know so well. They can be persuaded by a good argument, as witness the teachers quoted in the previously mentioned Bullock report of 1975 (‘A substantial number considered that the express teaching of prescriptive language forms had been discredited…’). But humans also tend, for very good evolutionary reasons, to be conservative, so if their own education was prescriptive their teaching may also be prescriptive unless they heed the familiar arguments (as of course they should). No doubt they would avoid prescription if government policy was clearly against it and a government inspector was sitting in their classroom; but given half a chance, they would do what they honestly (but mistakenly) think is the right thing.
How, then, can we influence classroom teachers? One answer is to go on as we have done in the past fifty years: writing for teachers and the general public, visiting schools, talking to educationalists, including as much linguistics as possible in initial-training courses for future teachers and CPD for practising teachers. This has clearly produced some worthwhile results, though we have very little idea how many people we have influenced, or in what way. But we should also do more of what we do best: research. What we need is strong research evidence that prescriptivism is counterproductive: that, for at least some children, it makes Standard English harder to learn by setting it in opposition to the local vernacular, the language of peers and that of home. We should team up with educationalists to explore the effects of explicitly comparing standard and nonstandard varieties on children's success in mastering Standard English. If (as we predict) the effect is positive, this would be a superb argument for persuading both classroom teachers and headteachers.
But another lesson is that culture change is slow, and should be measured in generations rather than decades. Eliminating prescriptivism will take a long time, and we must expect to meet resistance and unconverted prescriptivists along the way.
And finally, there is a lesson for government: changing attitudes requires firm and explicit leadership. It is not enough simply to avoid prescriptivism in curriculum documents; what is needed is an open and clearly argued condemnation of prescriptivism, combined with sincere recognition of non-standard varieties as full equals outside education and other ‘high’ domains. The case for persuasion is well made in the following quotation (G. Lewis Reference Lewis1981:262, quoted in Kircher & Fox Reference Kircher and Fox2019:851):
Any policy for language… has to take account of the attitudes of those that are likely to be affected. In the long run, no policy will succeed which does not do one of three things: conform to the expressed attitudes of those involved; persuade those who express negative attitudes about the rightness of the policy; or seek to remove the cause of the disagreement.
In short, whereas Cushing argues that the government has too much negative influence on language attitudes in schools, I would argue that it has too little positive influence.
