Is English usage changing more rapidly than it was before? Judging by the decreasing time between editions of ‘Fowlers’ it would seem so: thirty-nine years between Fowler's Dictionary of Modern English Usage (1926, from here on MEU1) and Gower's second edition (1965, from here on MEU2), and thirty-one years between the latter and Burchfield's The New Fowler's Modern English Usage (1996, from here on NFMEU). This led Manfred Görlach to write in his review of NFMEU that ‘[t]imes change, and we (and usage) change with them. A new edition of Fowler's classic book is thus in order every thirty to forty years’ (Görlach, Reference Görlach1997).
But now, just eighteen years later, there is the fourth edition: Fowler's Dictionary of Modern English Usage (2015, from here on FDMEU), edited by Jeremy Butterfield, a lexicographer at the OED and editor of the Oxford A–Z of English Usage (Butterfield, Reference Butterfield2007). Then again, we should not be too surprised considering that from the beginning of the genre in the late eighteenth century (Tieken-Boon van Ostade, Reference Tieken-Boon van Ostade2010), the rate at which new usage handbooks have been published has been increasing.
The genre has developed considerably over those two centuries, with possibly the biggest milestone occurring with the publication of MEU1 in 1926, whose ‘miscellany of items, the alphabetic format, and the didactic stance […] became the generic model for prescriptive usage books’ (Peters, Reference Peters, Aarts and McMahon2006: 763). Since the late twentieth century, roughly two types can be distinguished within the genre of usage guides: reference works such as Burchfield's The New Fowler's Modern English Usage, or The Cambridge Guide to English Usage (Peters, Reference Peters2004), and guides that are more of an entertaining nature, often meant to be read cover to cover, such as The Queen's English (Lamb, Reference Lamb2010) or What Not To Write (Sayce, Reference Sayce2006). The new FDMEU falls squarely into the former category. Butterfield identifies the book as doing so in the preface, writing that ‘the main use of a dictionary - which this is, albeit a dictionary of usage - is as an alphabetically ordered look-up resource.’ (Butterfield Reference Butterfield2015: viii). And as if to underscore this, the Oxford University Press put the word ‘dictionary’ back in the title after it had been taken out in NFMEU.
Because of my interest in the reception of Burchfield's third edition (Straaijer, Reference Straaijer, Percy and Tieken-Boon van Ostadeforthc.), my attention was immediately drawn to the fact that the publisher left the word New off the title of this new, fourth edition. When it appeared in 1996, NFMEU received much criticism from conservative reviewers who accused Burchfield of usurping Fowler's authority, as in the Buffalo Evening News:
I resent Burchfield trading on Fowler's name. Why didn't he call it ‘Burchfield's Modern English Grammar’ and see how many he'd sell? (Rooney, Reference Rooney1997)
Burchfield possibly put too much emphasis in his preface on how the third edition was different compared to the first and second. He begins his preface by stating that it is indeed a new edition of a classic handbook on English usage, but doesn't emphasise that it is different from what came before, stating its mission in positive terms instead:
‘This new, fourth edition of Fowler's Dictionary of Modern English Usage has been thoroughly revised and updated to reflect how English speakers the world over use the language now, in the twenty-first century.’ (Butterfield, Reference Butterfield2015: vii)
Those that published NFMEU underestimated how much of an institution (see also Busse & Schröder, Reference Busse and Schröder2010) ‘Fowler’ had become. Oxford University Press seems to know better now. Leaving the word New off the title is a way to reaffirm that this edition is simply another ‘Fowler’: not ‘New’, but just new.
Four editions of ‘Fowler’ have appeared over a period of nearly a century, and this provides us with an opportunity to compare usage advice in a single usage guide at various points in time. So let us look at some contested usage items and see how they are viewed and discussed in FDMEU, specifically compared to the previous three editions. This is not a systematic treatment of the usage items in FDMEU – which would take up too much space anyway – but merely a selection of those in which I expect some form of change in the way they are discussed, or those that reflect my own research interests. Most of them are in the HUGE database (Straaijer, Reference Straaijer2014), which contains usage advice about a selection of usage problems from usage guides published between 1770 and 2010. I have used this to find comments about them in the three previous editions of ‘Fowler’ as well as in other usage guides published in the past two hundred years.
ain't
The advice for ain't has essentially not changed since MEU1: ‘A(i)n’t is merely colloquial, & as used for isn’t is an uneducated blunder & serves no useful purpose. But it is a pity that a(i)n’t for am not, being a natural contraction & supplying a real want, should shock us as though tarred with the same brush.’ (Fowler, Reference Fowler1926: 44–45).
Butterfield repeats Burchfield, writing that ‘Dictionaries of English tend to hold the word, as it were, in a pair of tweezers. The label ‘informal is applied to it. None admits it to the sacred unqualified ranks of standard English.’ (Butterfield, Reference Butterfield2015: 33). The positioning of the last two statements make it seem as though the latter is a result of the former, even though informal words are still part of the standard language.
Under the subheading ‘Current use’, Butterfield adds that ain't is used by users from a variety of social classes, including its being ‘espoused in intellectual circles as an affectation, which tends to confuse the issue’ (Butterfield, Reference Butterfield2015: 33). Actually, it doesn't confuse the issue of who uses ain't at all, since apparently everybody uses it. The real problem with ain't is revealed in the remark, added since the previous edition, that ‘many people on both sides of the Atlantic regard it as incorrect, or even non-existent’ (Butterfield, Reference Butterfield2015: 34). It is the status of ain't as a ‘non-word’ that is the main source of contention about its usage.
alright
Fowler's condemnation of the form alright was straightforward and without room for nuance. ‘The words should always be written separate; there are no such forms as all-right, allright, or alright. This rule holds … also where they may be regarded as forming a more or less fixed phrase’ (Fowler Reference Fowler1926: 16). Burchfield noted that the form alright ‘is preferred … by popular sources like … magazines’ (Burchfield, Reference Burchfield2000[1996]: 43)
Butterfield shows a major advancement from his immediate predecessor, whose rigidity he doesn't eschew to criticise on this point. He finds that ‘Burchfield (1996) snottily wrote “The … inability to see that there is anything wrong with alright reveals one's background, upbringing, education, etc.”’ (Butterfield, Reference Butterfield2015: 38). Butterfield's judgement is reasoned and in line with current usage, as promised in the preface, and based on linguistic arguments regarding alright's form and specific functions. He concludes that we should ‘feel free to use it in business or personal writing, and in dialogue in fiction’ while cautioning that ‘in any formal work, or any that is likely to be edited, its pariah status will ensure that it is dismembered’ (Butterfield, Reference Butterfield2015: 39).
disinterested
This is perhaps one of the more strongly disputed and divisive current usage items, and a genuine usage problem. Fowler did not mention it in MEU1, which could mean that either its usage was not problematic at the time, or that he simply missed it.
Butterfield first discusses its etymology: historical usage – ‘[t]he sense “impartial” can … be said to have had a longer period of continuous use’ (Butterfield, Reference Butterfield2015: 223); current usage – ‘[a] listing of examples underlines the current strong presence of both meanings’ (Butterfield, Reference Butterfield2015: 224); and corpus data – ‘people still use uninterested more often than disinterested’ (Butterfield, Reference Butterfield2015: 224). The first three come straight from Burchfield, the last is a new addition. His recommendation seems at first to follow actual use – ‘It is clear that a semantic change is well under way with this word. There is no reason whatsoever why it should be prevented from having two different meanings; most words do’ (Butterfield, Reference Butterfield2015: 224), but this is followed by a sizeable ‘however’ – ‘given the strength of feeling on this issue … it is advisable to restrict disinterested to its meaning of ‘impartial” in any writing that is not informal, or that is likely to be edited or reviewed’ (Butterfield, Reference Butterfield2015: 224).
However, there are also those that are more liberal, and it seems to partly depend on which corpus one looks at. Peters, for instance, finds a different pattern from Butterfield: ‘[t]he primary and most frequent meaning of disinterest is ‘lack of interest,’ and in [British National Corpus] data this is its use about 90% of the time.’ (Peters, Reference Peters2004: 158). This is not to say that Butterfield is overly conservative. There are current usage guides that unequivocally state that ‘“[d]isinterested” means “impartial”, “neutral” or “unbiased”. “Uninterested” means “not interested”’ (Sayce, Reference Sayce2006: 69). Others also reject the sense of ‘uninterested’, but for a different reason: ‘In four surveys spanning thirty-five years … the Panel has overwhelmingly disapproved of it. … This usage thus is likely to be viewed as a sign of ignorance of the traditional use of the word.’ (Pickett, Kleinedler and Spitz, Reference Pickett, Kleinedler and Spitz2005: 145).
gay
Fowler's only concern was with the various inflections and modifications of the word - ‘gay makes gayer, gayest, gaily, gaiety’ (Fowler, Reference Fowler1926: 338) is the entire entry in MEU1 - and Gowers has no entry on ‘gay’ at all. But then again, it was a little bit before their time, as the first usage guide to have such an entry seems to have been the Longman Guide to English Usage (Greenbaum & Whitcut, Reference Greenbaum and Whitcut1988). It is Burchfield who first introduces an entry on gay with the meaning of homosexual in a Fowler. His is descriptive and fairy neutral, noting that ‘[s]ince the 1950 … widespread resentment has been expressed about the “loss” of a treasured word and about the omnipresence of the sexual uses of gay’ (Burchfield, Reference Burchfield1996: 324) and that ‘word has it, homosexual men in America are beginning to opt for the word queer instead’ (Burchfield, Reference Burchfield1996: 325), and predicting that ‘further developments can be expected’ (Burchfield, Reference Burchfield1996: 325). Butterfield shows that Burchfield was right, adding a new derogatory meaning for the word:
‘While the word gay was being enlisted in the fight for equality, a new, slang meaning of “foolish, stupid” or “insipid, unimpressive” was developing, thereby putting old prejudice in a new bottle.’ (Butterfield, Reference Butterfield2015: 339).
The entry for gay also shows Butterfield to go further than merely paying attention to political correctness. In his discussion he states that ‘[r]arely has it been so obvious that ostensibly linguistic anxieties – in this case, over a dated meaning of a not particularly frequent word – are a cloak for much darker sociological concerns.’ (Butterfield, Reference Butterfield2015: 338). This shows that he inserts social judgements into his comments in addition to linguistic ones, and not just about the usage items themselves but also about those using or arguing against or in favour of certain usages. This is in itself not new, but it is refreshing to see it done in such an open and measured way.
hopefully
Today, hopefully is perhaps the single most highly contested usage problems. Can hopefully be used to mean ‘it is to be hoped that' or can it only mean ‘in a hopeful manner’. It seems almost unbelievable now that Fowler and Gowers did not concern themselves with hopefully at all. But because ‘[s]uddenly, round about the end of the 1960s, and with unprecedented venom, a dunce's cap was placed on the head of anyone who used just one of them – hopefully – as a sentence adverb’ (Burchfield, Reference Burchfield1996: 702, cited in Butterfield, Reference Butterfield2015: 383), NFMEU spent three and a half columns on sentence adverbs, and most of those on hopefully alone, in addition to a short separate lemma on hopefully itself.
Hopefully seems to have first appeared as a usage problem in Modern American Usage: A Guide (Follett, Reference Follett1966). It is a rather sad testament to the enduring controversy about ‘hopefully’ that, despite being accepted in the sense of ‘it is to be hoped that’ by style guides such as those from the Associated Press and the Guardian, Butterfield still has to spend four full columns discussing its actual and recommended uses in FDMEU, and concludes that they are still not the same.
literally
Like hopefully, literally is one of the most highly contested items of English usage at present. And like hopefully, it is used as a shibboleth to distinguish ‘correct’ speakers from ‘careless’ ones. And it seems to have become more contested over time, requiring an equally more nuanced treatment in handbooks of usage. The first one seems to have been William Strunk, in the privately published, pre-E.B. White Elements of Style (1918).
The entries for hopefully are the same in MEU1 and MEU2. Fowler seems to have been conflicted about this word, on the one hand recognising its function as an intensifier (or ‘emphasizer’ as he calls it), but on the other saying that it is a use ‘we ought to be at pains to repudiate’ (Fowler, Reference Burchfield1926: 328). Interestingly, it is not Burchfield but Allen (Reference Allen1999) who brings more explicit nuance to the discussion. Whereas Burchfield merely notes the use of literally in a non-literal sense, it is Allen who also defends the legitimacy of this use, though still concluding: ‘avoid using literally when the effect might be distracting or comic; but it can be used to good effect in cases where it reinforces a strong verbal image’ (Allen, Reference Allen1999: 380).
Butterfield spends more than three columns on it, and comes to a nuanced – and consequently somewhat unspecific – recommendation: ‘[u]se your judgement’ and ‘reread what you have written so as to avoid hilarity or ridiculousness you did not intend’ (Butterfield, Reference Butterfield2015: 486).
sneak
The description of snuck shows a nice diachronic development as well. There is no entry for sneak in either MEU1 or MEU2; it first occurs in NFMEU, with the following comment regarding its status:
in a little more than a century, a new pa. t. form, snuck, has crept and then rushed out of dialectal use in America, first into the areas of use that lexicographers label as jocular or uneducated, and, more recently, has reached the point where it is a virtual rival of sneaked in many parts of the English-speaking world. But not in Britain, where it is unmistakably taken to be a jocular or non-standard form. (Burchfield, Reference Burchfield2000[1996]: 720)
Butterfield's entry on sneak illustrates the progress that this former non-standard form has made since the publication of NFMEU:
in a little more than a century, a new past tense and past participle form, snuck, has crept and then rushed out of dialectal use in America, first into the areas of use that lexicographers label as jocular or uneducated, and, more recently, has reached the point where it is a virtual rival of sneaked in many parts of the English-speaking world, and is recognized as standard in AmE. There is evidence of snuck sneaking into BrE in a big way as well, though probably here it would be regarded as informal. (Butterfield, Reference Butterfield2015: 756)
This confirms the prediction that ‘data on frequency of use suggests that the tide is turning and that broad acceptance of snuck is inevitable’ (Pickett, Kleinedler & Spitz, Reference Pickett, Kleinedler and Spitz2005: 435).
singular they
Discussed under the lemma they, Butterfield mentions Fowler's disapproval of the use of plural pronouns with a singular verb form in MEU1 and Gowers' noted in MEU2 that in ‘colloquial usage the inconvenience of having no common-sex personal pronoun in the singular has proved stronger than respect for the grammarians’, but that ‘few good modern writers would flout them’ (Gowers, Reference Gowers1965:635). Burchfield wrote that ‘such constructions are hardly noticed any more or are not widely felt to lie in a prohibited zone’. Butterfield notes that the use of he as a gender neutral pronoun is ‘now politically verboten’ (Butterfield, Reference Butterfield2015: 814), adding a condemnation of the formerly prescribed use to the acceptance of the formerly proscribed one.
We see here clearly illustrated that, following actual usage, a pattern of usage advice completely reversed over a ninety-year period. In addition, perhaps incidentally, Butterfield has taken more notice of political correctness, a subject that Burchfield was criticised for neglecting: ‘Burchfield is generally insensitive to the use of “politically correct” language, a subject of great importance in American discourse’ (Sheidlower, Reference Sheidlower1996).
unique and other absolute adjectives
The entry on ‘Absolute’ adjectives is partly what one would expect: adjectives such as perfect and unique cannot be modified by adverbs like largely, more or very. Exceptions are said to usually be literary – exemplified with Orwell's famous Animal Farm ‘more equally than others’ quote. It is noted that they can be modified by adverbs denoting an extreme, such as ‘absolutely’, ‘since that is consistent with their meaning’ (20), but there is no mention of a more contentious use in which unique is modified by almost. This use is still considered unacceptable in some 21st century usage guides, see for example Heffer (Reference Heffer2010: 131) and Sayce (Reference Sayce2006: 93), even though Fowler accepted it in MEU1 (1926: 679–680), and indeed even earlier in The King's English (Fowler & Fowler, Reference Fowler and Fowler1902: 58–59).
Regarding the gradability of the word unique in general, Burchfield came to the following advice: ‘It must, I think, be conceded that unique is losing its quality of being “not gradable” (or absolute), but copy editors are still advised to query such uses while the controversy about its acceptability continues.’ (Burchfield, Reference Burchfield2000[1996]: 808–809). Butterfield does not mention the loss of unique's absolute quality, but amends Burchfield's advice for copy editors of ‘serious or formal prose’ (Butterfield, Reference Butterfield2015: 847). What is relevant here is that the advice is meant for use of a specific register and for a very specific group of users.
For all the useful entries, as a reference work, FDMEU also suffers from what many of such usage handbooks show: bloating, or the inclusion of as many items as possible in an attempt to be complete. The result is that certain entries make you wonder to what extent the items discussed are in fact problematic. The entry for caoutchouc is a case in point. Here it is in full:
This strange-looking word, adopted in the 18c. from a Quechua word via Spanish and French, is pronounced /kaʊtʃhʊk/. (Butterfield, Reference Butterfield2015: 131).
Conspicuous by its absence is the meaning of the word. Without the meaning, why have the entry in the first place? It seems to me that if the meaning is known, than so will its pronunciation or spelling, the last of which will be equally problematic (in fact the word's pronunciation is not at all difficult considering the spelling). An entry such as this seems merely to inflate the size of the guide in an attempt to be comprehensive, or to be perceived as such. And its inclusion in the book is questionable: it is not an example of genuinely problematic usage, and a regular dictionary would give more complete information.
In addition to the emergence of the two subgenres mentioned above, comparing the four editions of Fowler with each other illustrates another development of the genre: the increasing influence of linguistics on usage guidance. This process, which seems to be ongoing and which will hopefully lead to a revalidation of linguists as usage experts, is also reflected in the latest Fowler.
In February 1997, when Robert Burchfield's The New Fowler's Modern English Usage had been out for three months, The Economist wondered whether, just as the first and second editions of the Dictionary of Modern English Usage came to be known as ‘Fowler’, the next edition would be known as ‘Burchfield':
The 1926 Fowler is already a period piece, though no one has ever gone wrong by taking its advice. The same, no doubt, will in time be true of ‘Burchfield’ – as, perhaps, by 2097 the publisher of its latest revision will dare to call it. (The Economist, 1 February 1997)
It turns out that The Economist didn't have to wait for 100 years to get an answer, and the answer is a resounding ‘No’. And perhaps this is why the latest edition again has the name Fowler in the title: it is still the name when it comes to guides to modern English usage.
The use of the Oxford English Corpus, a 2.5 billion-word monitor corpus of 21st-century English, is a great improvement in the way real non-literary language data is used to trace patterns of current usage, a method that was still comparatively simple in the previous edition, which used Burchfield's personal database of examples of (mostly literary) usage. The inclusion of English usage data from the Oxford English Corpus is a useful and welcome feature of this new Fowler, and makes it a good analogue resource for up-to-date and well-argued usage advice for English usage in the twenty-first century.
Those that approved of the influence of linguistics in Burchfield's third edition will certainly be pleased with this latest, fourth edition as well. In many ways Butterfield's fourth edition seems to be as much an updating of Burchfield's third, as Gowers' second was of Fowler's first – though perhaps more critical and even more linguistically oriented. As for those that found this influence an unwelcome intrusion, well, other guides, as well as a multitude of self-proclaimed authorities on the web offer plenty of ipse dixit usage advice.
ROBIN STRAAIJER received an MA in English Language & Literature from the University of Amsterdam and obtained a PhD from the University of Leiden with his dissertation Joseph Priestley, Grammarian. He built the Hyper Usage Guide of English (H.U.G.E.) database of English usage guides and usage problems as a member of the project ‘Bridging the Unbridgeable: linguists, prescriptivists and the general public’ in Leiden. He is currently working as an independent writer and reviewer, and planning a popular work about standard English in America. Email: r.straaijer@gmail.com
