Prescriptive Attitudes to English Usage
Summary and Keywords
Taking a sociolinguistic approach to prescriptivism in English usage, this article presents different methods by which highly frequent usage problems can be analyzed as to their current acceptability. These methods comprise different ways of studying a selected number of well-known items—try and/try to, the placement of only, the split infinitive and the dangling participle—focusing on their treatment in British and American usage guides from the beginning of the prescriptive tradition onward, combined with the application of special elicitation techniques to probe the views of informants. Such a multi-modal approach represents a distinct improvement from earlier attempts at presenting targeted groups of informants with attitude surveys only. By studying representative samples of British and American usage guides, the article shows that attitudes became more lenient across time (though not for all usage problems analyzed), with the sociolinguistic variable age playing an important role in the process, but also that instead of usage guides becoming more descriptive in the course of the history of the tradition, today in effect two trends can be distinguished in the type of usage advice given. While one trend indeed shows an increasingly descriptive approach to the items treated, a continuing proscriptive approach characterizes usage guides published down to the beginning of the 21st century.
1. An English Usage Problem Hits the News1
In January 2016, the American Dialect Society announced the winner of its annual word-of-the-year competition for 2015: singular they, as in Everyone has their off-days.2 Singular they received recognition “for its emerging use as a pronoun to refer to a known person, often as a conscious choice by a person rejecting the traditional gender binary of he and she.” Its election is remarkable; singular they, after all, is a grammatical feature and not a lexical item like the other candidates, words like ammosexual, microaggression, and yass. Furthermore, it has taken quite a while for singular they to gain official recognition; a plea for acceptance was already made 40 years ago by Bodine (1975), who argued that singular they has always been in existence. Mittins et al. (1970, p. 102)—the source of the above example—quoted well-known writers like Daniel Defoe, Henry Fielding, and Jane Austen to illustrate this. Singular they, despite the American Dialect Society’s claim, can therefore hardly be said to have been “emerging.”
Singular they, Bodine (1975) claimed, was first condemned by the 18th-century grammarian John Kirkby (1705–1754). In fact, the rule originated not with Kirkby but with the earliest English female (!) grammarian Ann Fisher (1719–1778), whose grammar Kirkby had plagiarized (Tieken-Boon van Ostade, 1992). The form was condemned because using a plural pronoun to refer to a singular antecedent was felt to offend against correct grammar. In an age of male-dominated grammar writing, breaking a rule that conflicted with gender weighed less heavily than when number was involved. The proscription received official sanction by a British Act of Parliament in 1850, which laid down the obligation to use he when the sex of the antecedent was unknown (Bodine, 1975, p. 134). England lacked an academy like those in France, Italy, or Spain that would normally impose correct language use upon speakers and writers from above, and it is remarkable to see how in this instance the government stepped in to do so. Usage of singular they has nevertheless continued, and the American Dialect Society noted a recent increase in usage. That this should happen despite formal guidelines against its use is of great interest.
Singular they is only one example of a “usage problem”—a feature of divided usage that conflicts with notions of correct grammar and that is often associated with a lack of education or social standing among particular groups of users (see Section 4.3). The English actor and writer Stephen Fry, well known for his language documentary Planet Word (2011) and the BBC Radio 4 series Fry’s English Delight (2008–2015), mentions quite a few of them in the online video Kinetic Typography—Language: split infinitives, disinterested/uninterested, imply/infer, preposition stranding, the greengrocer’s apostrophe, nouns being turned into verbs, less/fewer, none is/none are, would/should, will/shall, and that/which. There are in fact many more such debated features of language use: Don Chapman, for instance, calculated that there must be well over 10,000 prescriptive rules dealing with such items (Chapman, 2017, p. 246). Usage problems and prescriptive rules are the domain of usage guides: language advice manuals for people who are insecure about what is considered correct language. The short story “Mother tongue” by Ian McEwan (2001) explains how such linguistic insecurity may come about: describing his mother’s social mobility, with a first husband being a private soldier and the second, McEwan’s father, a sergeant, McEwan discusses “[h]er displacement within the intricacies of English class, and the uncertainty that went with it, [which] taught her to regard language as something that might go off in her face, like a letter bomb.” McEwan himself had to unlearn saying done for did (he done this for me) in school and, as a young writer, to avoid mistakes like using a dangling participle when complaining in a letter to the editor of the treatment he had received in a review (“Sir, Having destroyed my meaning with dishonestly juxtaposed quotation, I find myself perplexed by your reviewer’s sudden concession to probity when …”). The dangling participle is a much criticized grammatical error because of the lack of agreement between the underlying subject of the adjunct (your reviewer) and the overt subject of the main clause (I), and it is a staple item in usage guides (see Section 4.4).
Fry’s harangue, however, is not directed at the perpetrators of usage mistakes, as is usually the case in language criticism expressed in the media (see, e.g., Lukač, 2015), but at the pedants, the people who insist on correcting them. Fry thus positions himself on the other side of the so-called usage debate (see Kamm, 2015, pp. 78–104), as someone who considers it irrelevant to continue to harp on linguistic correctness. But in doing so, he uses quite strong language as well: “I hate that,” he says, it is “a silly approach to language,” even arguing that pedants are “a sorry bunch of semi-educated losers.” Strong language has characterized the debate about correct English since the earliest days of the prescriptive tradition 250 years ago. An inventory of the proscriptive metalanguage of 18th-century normative English grammars (Sundby, Bjørge, & Haugland, 1991, pp. 44–53) includes condemnations like absurd, ridiculous, barbarous, repugnant, and preposterous. Such terms are still found in usage guides today. Examples are “is no less absurd than …” in Eric Partridge’s Usage and Abusage (1942) on the wrong placement of adverbs, and “preposterous monstrosity” used by Bryan Garner in his Dictionary of American Usage (1998) when discussing preposition stranding. In Strictly English (2010), Simon Heffer calls sentence-initial hopefully a “tiresome usage [which] is now so ubiquitous that those who object to it are sometimes dismissed as pedants. It remains wrong, and only a barbarous writer … would try to pass it off as respectable prose.” Fry’s words show that strong language characterizes both sides of the usage debate: those who stand in a long tradition of criticizing perceived linguistic errors and those who believe that actual usage should dictate what is to be considered correct language use.
To throw further light on why this should be so, we will analyze four usage problems from a survey carried out by Mittins, Salu, Edminson, and Coyne (1970). We will study these features—try and vs. try to, the placement of only, the split infinitive, and the dangling participle—by taking into account current views on their acceptability, which we expect to have increased considerably since the time when the Mittins survey was first carried out, as well as the way they are treated in usage guides, drawing on a database that was created specifically for the purpose of being able to study these as well as many other usage problems. The broader context for the analyses presented here is the account of the standardization process discussed by Milroy and Milroy ( 2012), in which prescription forms the final—though never-ending—stage in the process, preceded by a stage called “codification” during which normative grammars and dictionaries were produced, largely during the 18th century, to lay down the rules of the English language. How all this happened despite the absence of a language academy in England and particularly how prescriptivism as such actually began is described in detail in Tieken-Boon van Ostade (forthcoming), while a very lucid account of the rise of so-called “proper” English may be found in Lynch (2009). The different beginnings of the usage guide tradition between England and the United States are focused on in Tieken-Boon van Ostade (2015), while Milroy (2001) presents an illuminating account of the language ideologies existing in these two countries.
What needs to be borne in mind when studying attitudes to usage is the basis on which informants make their judgments concerning the acceptability of a particular usage feature. While intrinsic and aesthetic differences between language varieties or usage features, according to which one variety or feature is considered to be “linguistically superior” or believed to “sound better” than another, have been dismissed by Edwards (2006, pp. 324–326) as a possible basis for making such judgments, it is social perceptions of speakers which most likely form the basis on which evaluations and linguistic judgments are made. Such evaluations tend to be evaluations not only of the speakers themselves, but also of the community they are thought to belong to. Edwards (2006, p. 326) points out how the language variety used by the most influential and powerful groups in society is frequently considered superior to other varieties, including nonstandard ones. Hence, the social perceptions of a speaker making use of nonstandard features can trigger evaluations and associations with a particular social group as a whole. Language use can thus function as a powerful instrument in the exclusion of certain speakers from a social group or restrict their access to certain domains in society.
2. Attitudes Toward English Usage (1970)
While studying usage attitudes seems to have found a fruitful ground in the United States (see Leonard, 1932; Hairston, 1981; Albanyan & Preston, 1998; Gilsdorf & Leonard, 2001), for Great Britain only one usage attitude study focusing on Standard British English could be identified, Attitudes to English Usage (Mittins et al., 1970). In this study, the authors report on a survey carried out in the late 1960s in response to a call from the School Council in Great Britain which initiated several studies related to the field of English language teaching (Burgess, 1996, pp. 55–56). The aim of the survey was to identify current attitudes toward English usage in order to provide teachers with advice on what to teach in the light of growing uncertainties as to questions of linguistic correctness and about how to deal with the issue in an educational context. The survey consisted of an enquiry among teachers, examiners, lecturers, students, some managerial staff, salesmen, and administrators, and included 55 sentences of the type illustrated by singular they above (Mittins et al., 1970, pp. 5–6). These sentences represented issues of divided usage that were expected to meet with varying degrees of acceptability in different linguistic registers. Informants had to indicate acceptability across four different styles for 50 of them (informal speech, informal writing, formal speech, and formal writing) and for only two or three for the remaining five. The authors then calculated the general acceptability of the items tested, presenting them in the order of greatest to least acceptability. Least accepted turned out to be very unique (11%), imply for infer, and like for as if as in It looked like it would rain (12% each), while very amused, averse to (rather than from), and did not do as [for so] well as received general acceptability ratings of 76, 81, and 86%, respectively. Singular they gained 42% general acceptance.
As far as we know, the Mittins survey has not recently been repeated in toto to see whether acceptability of the tested features has increased over the years.3 That it would have done so is to be expected in view of a process called “colloquialization” which, according to Mair (2006, p. 183), shows an increase toward greater informality in the English language during the 20th century. Across the board, the 55 contested items of the Mittins survey show a greater acceptability in informal speech than in formal writing (1970), and we expect this trend to have continued over the years. There are, however, a few problematical aspects about the Mittins study. For one thing, the informants were predominantly highly educated, so that the acceptability ratings are not representative of (British) society in general. Nor was the informants’ social background sufficiently taken into consideration, as is nowadays routinely done within sociolinguistics. Age is given only scant treatment, while gender, which would have been relevant since women tend to behave differently when it comes to linguistic variants that carry prestige (see, e.g., Nevalainen & Raumolin-Brunberg, 2003, p. 110), is not considered at all. Another controversial point is the fact that the tested features were highlighted, which might have biased the informants.
For all that, the Mittins survey was repeated on the website of the Bridging the Unbridgeable project in the context of which part of the research for this article was carried out. It was kept as it was, with the usage problems being highlighted as in the original. Following Hedges (2011), however, we added the further option “netspeak” (Crystal, 2006, p. 402), which we described as “internet usage or chat language, texting,” since this has become an important new usage category that needs to be taken into account. Another option, “unacceptable under any circumstances,” was added in response to specific requests from informants. To keep the new survey similar to the original one and merely to identify changing acceptability trends for the items investigated by Mittins et al., we did not enquire into the informants’ sociolinguistic backgrounds either. Such variables do play a role in more detailed analyses of individual usage problems such as those listed in note 3. As we will show in Sections 4.4 and 5, specific attitude elicitation techniques—online questionnaires and face-to face interviews with informants living in England—play an important role in the attempt to gain insights into current attitudes to usage problems.
The results of the repeated Mittins survey are freely available online but merely serve to identify ongoing trends in usage. To compare, for example, the trend in acceptability of singular they today with the data from Mittins et al. (42%; item 49 in the survey), it appears that with an unacceptability rating of only 1.83% it is now almost fully acceptable. Though not actually an emerging new use as the American Dialect Society suggests, it is indeed common today and merits acceptance as such. Singular they would consequently no longer need to be considered a usage problem. Nevertheless, it is still treated in two recently published usage guides, Steven Pinker’s The Sense of Style (2014) and Oliver Kamm’s Accidence Will Happen (2015).
3. The HUGE Database
Usage problems come and go, and they all have a history of their own. To be able to study individual usage problems, we compiled a database of usage guides and usage problems in the Bridging the Unbridgeable project, called Hyper Usage Guides of English (HUGE). The HUGE database includes 77 selected usage guides, published between 1770 and 2010, and 123—mostly grammatical—usage problems (for the selection principles, see Straaijer, forthcoming). Except for the perceived problem of nouns being converted into verbs mentioned by Stephen Fry, all usage problems referred to above (and many others besides) are included in the database and can be analyzed accordingly.4 The database is freely accessible and admission is granted upon request. A User Manual is available from the Bridging the Unbridgeable website, where practice sessions are provided as well. Videos showing its construction process may be found on the project’s YouTube channel.
While the first usage guide ever published in England was Robert Baker’s Reflections of the English Language (1770), the earliest American usage guide (as far as we know) was Seth T. Hurd’s Grammatical Corrector (1847). The traditions began for different, if related, reasons: in England in response to the need for linguistic guidance upon the increased social mobility due to the Industrial Revolution, and in America upon the large numbers of foreign immigrants during the mid-19th century (Tieken-Boon van Ostade, 2015). The traditions, moreover, have different characteristics, due to the different language ideologies prevalent in these countries (Milroy, 2001). While the standard language ideology in the United Kingdom is strongly rooted in social class differences, as illustrated by Ian McEwan’s short story “Mother Tongue” (see Section 1), the usage debate in America is influenced by speakers’ regional and/or ethnic backgrounds. The HUGE database includes British and American usage guides so that both traditions can be studied separately as well as contrastively. An illustration of the kind of differences that may be encountered is that in American English have went (for have gone) is considered a usage problem, while in British English it is primarily viewed as a dialectal feature, even though it first occurred as a usage problem in Baker (1770) (Tieken-Boon van Ostade & Kostadinova, 2015). Another example is thusly, of which Burchfield (1996) claimed that “it has not been washed ashore in the UK yet.” Though first occurring as a usage problem—in the HUGE database—in an American usage guide, i.e. Krapp (1927), and being indeed treated mostly in American publications, thusly already made its appearance in the English usage guide by Greenbaum and Whitcut (1988), while it is also treated in the Cambridge Guide to English Usage (Peters, 2004) and the Oxford A–Z of English Usage (2007). Butterfield, the author of the latter text, notes that “[m]any people consider it superfluous, and it should probably be avoided in writing.” Who these “many people” are is unclear (Chapman, 2017), but Burchfield’s advice shows that thusly is highly stigmatized. For all that, it is more common in American than in British English, and despite strong proscription (“There is no such word in standard English” according to Trask, 2001), usage is on the increase, especially among young people (Lukač & Tieken-Boon van Ostade, forthcoming).
4. Usage Problems and the HUGE Database
The usage problems we decided to focus on are the following (Mittins et al., 1970):
Try and was selected for its different status in British English, where it is in general use (Hommerberg & Tottie, 2007; Tottie, 2012), and American English, where it isn’t. The placement of only was part of an attitudes survey posted on the Bridging the Unbridgeable website in May 2012, which invited informants to write short pieces of text explaining their views on the features in question (Tieken-Boon van Ostade, 2013). Though in general use today, the split infinitive has a special status among usage problems as a so-called “old chestnut” (Weiner, 1988, p. 173); it continues to feature in usage guides and is often cited as a prototypical usage problem. The dangling participle, finally, constitutes an interesting syntactical construction which is believed to cause ambiguity and confusion especially in writing (Ebner, 2014). As we will show, its acceptability over the years has not increased as much as would have been expected.
In analyzing these four usage problems, we adopted different methodologies, ranging from an earlier literature survey in the case of try and, an analysis of metalinguistic comments by informants for the placement of only, to a mixed-methods approach to elicit attitudes from informants combined with a sociolinguistic investigation of attitudes for the split infinitive and the dangling participle. In all cases we drew on information retrieved from the HUGE database. In interpreting these data we once again applied different methodologies, the one largely based on the use of pre- and proscriptive metalanguage (try and and the placement of only) and the other on a method developed for the classification of the treatment of preposition stranding in 18th-century normative grammar by Yáñez-Bouza (2015) (the split infinitive and the dangling participle). As we will demonstrate, these different approaches have their own merits depending on the purposes of the analyses.
4.1. Try and
From the repeated Mittins survey (Section 3), try and appears to be virtually acceptable today: the rating “unacceptable under any circumstances” received only about 4.4% of the votes. Acceptability of the different styles, however, ranged from 31.7% acceptability for informal speech to 7.1% for formal writing, so try and is still subject to a certain amount of usage criticism depending on the context in which it appears. Since the survey was published online, however, our informants may have been native or non-native speakers coming from anywhere. Despite its shortcomings from a sociolinguistic perspective, the informants of the Mittins survey at least appear to have been native speakers of British English. In that light, the low general acceptability of try and at the time, 27%, is striking, since Hommerberg and Tottie (2007, p. 48) found that usage is much more frequent in British than in American English; in written English, American and British, try to is most frequent. Perhaps informants, when it comes to expressing their attitudes in language surveys, which is after all a conscious process, are equally biased when assessing spoken usage as they would be when confronted with the choice between using try and or try to in writing (see Milroy & Milroy,  2012, pp. 35, 42).
Try and is discussed in 53 out the 77 usage guides in the HUGE database. In view of its greater acceptability by British speakers, it is striking that it was first condemned in a British work: The Bad English of Lindley Murray and Other Writers (Moon, 1868). Moon criticized his predecessor Henry Alford, the author The Queen’s English (1884), for the sentence please … to try and not think me so, calling it “the worst mistake [Alford] has made.” According to Ross (2013, pp. 120–121), the sentence had already attracted criticism in 1864 in Every Boy’s Magazine, so this was possibly the earliest time the form was criticized. Before Moon, try and had been discussed by Gould (1867), who merely noted that it “is a very common substitute for try to, in contemporaneous literature and in conversation.” Gould, therefore, allowed for both variants, and again it is of interest in the light of today’s differences between British and American English that an American usage guide took a descriptive rather than pre- or proscriptive approach to the issue: try and is common in literature and speech as an acceptable variant of try to. These different approaches do not, however, continue to characterize the subsequent tradition. The majority of the works in HUGE that deal with try and are American publications: 28 in all; 24 are British, and one, Trask (2001), might have been either since the author described himself as “an American who works in Britain” (Trask, 2001, p. 3).
Inspired by Gould’s descriptive and Moon’s proscriptive approaches, the usage guides were classified along these lines, with a prescriptive category being included as well. The following examples illustrate the three categories:5
Classification was not always straightforward. Batko, for instance, gives both proscriptive and prescriptive advice, while O’Conner’s treatment could be interpreted as prescriptive and descriptive at the same time:
Our classification of the usage guides into these categories may be found in Table 1.
Table 1. Usage Guides Classified According to Their Approach (Prescriptive, Proscriptive, Descriptive) to try and
TurckBaker1910(1938)/Am, Treble&Vallins1936/Br, Ebbitt&Ebbitt1939(1978)/Am, Bailie&Kitchin1979(1988)/Br, DeVries1991/Am, Howard1993/Br, Trask2001, Brians2003/Am
4 Am, 3 Br, Trask2001
Moon1868/Br, Ayres1881(1911)/Am, Vizetelly1906(1920)/Am, Payne1911/Am, Partridge1942(1947)/Br, WrittenWord1977/Am, Vermes1981/Am, Bryson1984/Am, Mager&Mager1993/Am, AmHerDict2005/Am, Sayce2006/Br, Fogarty2008/Am, Taggart2010/Br, Heffer2010/Br, Lamb2010/Br
9 Am, 6 Br
Gould1867/Am, Hall1917/Am, Fowler1926/Br, Krapp1927/Am, Vallins1951/Br, Nicholson1957/Am,7 Evans&Evans1957/Am, Wood1962(1970)Br, Fowler(=Gowers)1965/Br, Follett1966/Am, Swann1980/Br, OxfordEnglish1986(1990)/Br, Weiner1983(1994)/Br, Burchfield1984/Br, Greenbaum&Whitcut1988/Br, Randall1988/Am, Webster’s1989/Am, Wilson1993/Am, NYPubLib1994/Am, Burchfield1996(2000)/Br, Garner1998/Am, PocketFowler1999/Br, Peters2004(2006)/Br, OxfordA—Z2007/Br
11 Am, 13 Br
Morris&Morris1975/Am, Marriott&Farrell1992(1999)/Br, Ayto1995(2002)/Br, O’Conner1996(1998)/Am
2 Am, 2 Br
Total: 53 (28 Am, 24 Br, Trask2001)
The figures in the table show that the majority of the usage guides take a descriptive approach (24), followed by a proscriptive one (15). A prescriptive approach is found only in eight publications. The British and American publications do not differ very much in the approach taken, except for the proscriptive category.
Because we expect the treatment of usage problems to have become more descriptive over the years (see Peters, 2006), in line with the changing attitudes to linguistic correctness discussed in Section 2, we wanted to see whether this is reflected in the above classification. To this end, the accounts in Table 1 were arranged chronologically in Figure 1.
The graph confirms our expectations: though the overall numbers are small, the majority of the descriptive approaches identified dates from the second half of the 20th century, with a peak for the 1980s (7/10). But we also see a drop immediately after that. What is more striking is the continuing occurrence of usage guides taking a proscriptive approach to try and, with a peak for the 2000s (6/11): this goes against the increased acceptance of the form since the Mittins survey of the late 1960s. Moreover, four of the six proscriptive works from the 2000s are British publications: Sayce2006/Br, Taggart2010/Br, Heffer2010/Br, and Lamb2010/Br. It is therefore not the case that, across the board, usage guides become more descriptive over time. Instead, we see two trends here, an increasing descriptive approach on the one hand but another that continues to offer proscriptive usage advice. Remarkably, though, this trend is here found for the British tradition with a feature that shows greater acceptance in British than in American English: it is odd that British usage guides now turn against a pattern of usage that is considered quite acceptable. This confirms the unacceptability of try and in British English that was noted by Luscombe (2012, p. 150), who showed that the feature is among the “top four ‘pet hates’” of the BBC journalists she interviewed.
Two more points are worth making. Hall1917/Am, though published in America, presents an early descriptive account based on what seems a corpus-linguistic approach. His evidence, however, entirely consists of examples from British authors, so by saying that both variants are acceptable (“Try to is more strictly grammatical, but euphony [with respect to try and] has its rights”6) he appears to be upholding a British standard of correctness (see Kamm, 2015, p. 74). The other point concerns Fogarty2008/Am, who is “bothered” by try and while claiming that all the books she consulted state that “try and is an accepted informal idiom that means ‘try to’.” Considering this “obviously wrong,” she places try and firmly on her “list of pet peeves.” Unfounded and emotionally charged statements in popular usage guides like Fogarty’s (or her Grammar Girl website) will continue to keep usage patterns for try and/to in British and American English distinct, though it is unclear how this situation arose to begin with (Hommerberg & Tottie, 2007, p. 60).
4.2. The Placement of only
In the late 1960s, the placing of only outside the constituent within which it functions as a modifier (He only had vs. he had only [one chapter] to finish) was considered unacceptable by more than half the informants of the Mittins et al. survey (45% overall acceptability). Today, as our repeated survey indicates, acceptance seems virtually categorical (1.15% unacceptable). The placement of only is nevertheless still considered a usage problem by some writers: Lamb2010/Br calls only a “tricky word,” and Taggart2010/Br writes that “only should be positioned … as closely as possible to the word it qualifies.”
What kind of people, apart from these two authors, would still regard misplaced only unacceptable? To find out, we drew on the results of an attitudes survey carried out in 2012 for three contested sentences, including he only had one chapter to finish, about which informants were invited to write a short piece of text. The aim of the survey was to study the use of negative metalanguage (e.g., horrible, abomination, sloppy; see Section 1) that is common in discussions of usage problems and prescriptivism in general (Tieken-Boon van Ostade, 2013). For the only sentence, there were 171 informants, native and non-native speakers, men and women, aged between 16 and 91, and with a wide range of professions. For reasons of space, we will analyze the native speakers only (129); two-thirds of these are women, while 70% are British (91/129 informants; 28 Americans, 10 “other”). The largest number of informants are in their sixties (47), with 23 of them being in their fifties and 28 in their seventies. Only 10 informants are below thirty. This unequal distribution within the survey population is largely due to how the survey was distributed—on the Bridging the Unbridgeable website, though also through personal acquaintances, one of whom had the survey announced on the website of the (British) University of the Third Age. Another reason may be the factor age itself, since elderly people tend to be more concerned with the preservation of former standards of—linguistic—correctness (e.g., Luscombe, 2012, p. 170). Our findings will therefore probably not be representative of any speech community as such.
Of the 129 native-speaker informants, 35 considered the sentence He only had one chapter to finish acceptable, calling it merely “Fine!” for instance. Forty-two informants added a specification as to the context in which the sentence would be acceptable or not, as in: “This sounds perfectly OK to me, and I would use it. ‘He had only one chapter …’ sounds a bit more formal, so I might write it that way round—but I'd certainly say it the first way.” Fifty-five informants were aware of the existence of a rule prescribing the correct placement of only in a sentence:
We learn at school that we should place the “only” before the relevant phrase or clause (“he had only one chapter … ,” but I find that many people, including myself, feel the compulsion to insert the “only” earlier, typically before the verb, and so I think that I and many colleagues would note the solecism, but not worry unduly.
Thirty-one informants rejected the sentence, 14 of whom were in their seventies, 9 in their sixties, and 5 in their fifties; the remaining three were 17, 42, and 91 years old. If age plays a role in the rejection of the Mittins sentence—since the 14 informants who rejected it make up 50% of the seventies group, with only 19 and 23% for the sixties and fifties groups, respectively—it is striking that the youngest informant (female, British) also belongs to this category: “I wouldn’t use it myself, but grammatically it is correct I think. There could be mad[e] use of it.” The eldest informant condemned the sentence as follows: “Only in this case refers to ‘he’ Should be: ‘He had only one chapter to finish’, ‘only’ referring to chapter. Too many people would use it! I think and hope I would not.” Not unexpectedly, the high condemnation rate among the eldest informants (50%) agrees with that in the Mittins survey (55%). Our figures show considerably greater tolerance in the age groups immediately below them, which suggests an ongoing process of increasing acceptability over the years. In this light the youngest informant stands out; had the results had been based on oral interviews, we could have explored this further. As for the informants’ nationality, about a quarter of the British ones rejected the sentence (24/91) compared to only about 18% of Americans (5/28). Though the figures are small, the placement of only is possibly considered more of a usage problem by—older—British than American speakers.
To determine whether similar differences can be detected in the usage guides published over the years, their treatment was classified along the same lines as that of try and (see Section 4.1). Thus, the entries on only in the usage guides in the HUGE database were classified into prescriptive, proscriptive, or descriptive treatments and, since the approach taken was not always straightforward here either, into combined pre- and proscriptive and descriptive/prescriptive treatments. Examples of these categories are the following:
The classification of the 62 usage guides in the HUGE database that deal with the placement of only is presented in Table 2.
Table 2. Usage Guides Classified According to Their Approach (Prescriptive, Proscriptive, Descriptive) to the Placement of only
Live&Learn1856?/Am, Vizetelly1906(1920)/Am, TurckBaker1910(1938)/Am, Krapp1927/Am, Ebbitt&Ebbitt1939(1978)/Am, Gowers1948/Br, Evans&Evans1957/Am, Follett1966/Br, WrittenWord1977/Am, Bryson1984/Am, Greenbaum&Whitcut1988/Br, Carter&Skates1988(1990)/Am, DeVries1991/Am, Marriott&Farrell1992(1999)/Br, Mager&Mager1993/Am, Blamires(1994)/Br, Ayto1995(2002)/Br, O’Conner1996(1998)/Am, Garner1998/Am, Burt2000(2002)/Am, Trask2001, Brians2003/Am, Sayce2006/Br, Fogarty2008/Am, Taggart2010/Br, Heffer2010/Br
16 Am, 9 Br, Trask2001
Baker1770/Br, 500Mistakes1856/Am, Gould1867/Am, Ayres1881(1911)/Am, Fitzgerald1901/Am, Vallins1953(1960)/Br
4 Am, 2 Br
Moon1868/Br, Hall1917/Am, Vallins1951/Br, Morris&Morris1975/Am, Crystal1984(2000)/Br, Randall1988/Am, Webster’s1989/Am, Wilson1993/Am, Howard1993/Br, Amis1997(1998)/Br, Peters2004(2006)/Br, OxfordA—Z2007/Br
5 Am, 7 Br
Baker1779/Br, Payne1911/Am, Burchfield1981/Br
1 Am, 2 Br
Alford1864/Br, Fowler1926/Br, Treble1936/Br, Nicholson1957/Am, Wood1962(1970)/Br, Fowler1965/Br, Bailey&Kitchin1979(1988)/Br, Swan1980/Br, Weiner1983(1994)/Br, Burchfield1984/Br, OxfordEnglish1986(1990)/Br, Burchfield1996(2000)/Br, PocketFowler1999/Br, AmHerGuide2005/Am, Lamb2010/Br
2 Am, 13 Br
Total: 62 (28 Am, 33 Br, Trask2001)
Table 2 shows that the placement of only has been a usage problem throughout the usage guide tradition: it is found from the earliest publications in the 1770s for British English and from the 1850s onwards for American English, down to the most recent British and American usage guides in the HUGE database. It is clearly among the oldest of old chestnuts in the canon of usage problems. Overall, slightly more British than American usage guides deal with the problem, and it is striking that, over the years, American publications are predominantly prescriptive, while combining a prescriptive with a descriptive approach is more typical of British usage guides. Hall1917/Am again stands out for his descriptive approach, citing “104 authors misplacing only in over 400 passages.” As with his treatment of try and ( see Section 4.1), Hall’s list largely comprises British English writers, though this time it also includes Americans like Franklin, Jefferson, and Washington. His aim here is to show that there is no real ground for condemning the allegedly incorrect construction since among his “offenders” we find many “pretty high authorities.”
Ordering the different categories of usage guides chronologically in Figure 2, we found that the misplacement of only is no longer proscribed after the 1950s. Burchfield’s combined pre- and proscriptive approach in 1981 is exceptional; in his subsequent publications (1984, 1996) he is more lenient, which again may reflect the changing attitudes toward the placement of only.
It is, finally, noticeable that during the last two decades of the period covered by the HUGE database prescriptive treatments predominate, both for American publications (7) and British ones (6); few usage guides during this period are classified as descriptive.
Both the survey results and the analysis of the usage guides show that attitudes to the placement of only are changing. Not only do we no longer find a proscriptive approach to the usage problem, informants have also become more tolerant of misplaced only over the years. The remarkable consistency between the over-70 informants and attitudes from the late 1960s confirms the importance of the factor age in studying changing attitudes to usage problems over the years. That so many usage guides today take a prescriptive approach to the feature, as exemplified by Heffer2010/Br—“Only should be positioned as close as possible to the word it qualifies”—suggests that these usage guides are fighting a losing battle here.
4.3. The Split Infinitive
Splitting the infinitive by inserting an adverb between to and the infinitive, as in the infamous Star Trek trailer to boldly go where no man has gone before, was first considered problematical in the early 19th century. The anonymous author’s comment in The New-England Magazine in 1834 is probably not the first written objection to the split infinitive, but it does include the first rule against splitting infinitives (Bailey, 1996, p. 248):
The particle, to, which comes before the verb in the infinitive mode, must not be separated from it by the intervention of an adverb or any other word or phrase; but the adverb should immediately precede the particle, or immediately follow the verb. (P. 1834:469)
The rule is based on Latin grammar and reflects the single-word status of infinitives in Latin as opposed to English. The author also provides an insight into how usage of the split infinitive was perceived at the time by stating that splitting infinitives was a practice frequently found among “uneducated persons” and even occasionally in newspapers whose editors had not received “a good education” (P. 1834:469). That the use of split infinitives should be attributed to uneducated speakers is no surprise given the Latinate origin of the stricture. These findings contradict the long-held belief that the rule against split infinitives originated with the 18th-century grammarian Robert Lowth (1710–1787) (Tieken-Boon van Ostade, 2011, pp. 117–118). In the United Kingdom the first critical comment against the use of split infinitives was made by Richard Taylor, an editor and the co-founder of the publishing company Taylor & Francis, who stated in 1840: “Some writers of the present day have a disagreeable affectation of putting an adverb between to and the infinitive” (as quoted by Visser, 1966, pp. 1036–1037). Mittins et al. (1970, p. 70) note that the first instances of the split infinitive go back as far as the early 14th century, but a considerable increase in usage is said to have caused 19th-century grammarians to proscribe the construction (Evans & Evans, 1957, p. 469; Mugglestone, 2012, p. 354). The split infinitive has subsequently become one of the most widely discussed usage problems and is now frequently identified as an old chestnut.
In Mittins et al. (1970), the split infinitive in the sentence He refused to even think of it received an average acceptability rating of 40%, and was ranked 24th of the 50 usage problems studied (see Section 2). The acceptability rates ranged from 66% in informal speech to 19% in formal writing (Mittins et al., 1970, p. 72). For the present article an almost identical stimulus sentence was used, He refused to even think about it, and a survey among British informants showed that attitudes toward the split infinitive have become more lenient. Instead of having one category for netspeak (see Section 2), however, a distinction was made between degrees of online formality (formal online/mobile vs. informal online/mobile language use) following Crystal (2006, p. 52), who argues that online language use is not uniform but is situated between written and spoken usage. The results of the survey show that the average acceptability rating of the split infinitive has increased to 66.2%, with the highest acceptability rating of 85.3% for informal speech and the lowest, at 46.6%, for formal writing. Compared to the Mittins findings, this increase in acceptability is quite drastic. To see whether this change correlates with social variables like age, gender, education, and social class, Mann-Whitney U-tests were performed on the data, which showed that greater intolerance of the split infinitive could be attributed to age (U = 257.5, p = .018): the older the informants, the more likely they were to find the split infinitive unacceptable.
While a quantitative—statistical—analysis of usage problems provides interesting insights into usage attitudes, a qualitative approach based on interviews and comments of informants is necessary to obtain a fuller picture of attitudes toward usage problems (see Section 4.2). To this end, first an online questionnaire was conducted, which allowed participants to comment on usage problems like the split infinitive. The majority of the comments relating to this linguistic feature showed the participants’ awareness of the issue at hand. Examples of comments that readily identified the construction may be found in (15) through (17):
Despite the growing acceptability of split infinitives, their problematical status is still acknowledged by some informants, who comment on the construction’s common occurrence, as in (18), and express their awareness of other people’s reactions toward this construction—see (19) and (20).
As the comment in (20) shows, prescriptive attitudes toward splitting infinitives held by others may influence one’s own practice, even if the stricture is considered “pedantic and pointless.” The mixed-methods approach adopted here (see Giles, 1970; Soukup, 2015) helped to gain a better understanding of usage attitudes. While acceptability of the split infinitive was shown to correlate with age, the comments confirm the near-mythological status of the construction. Despite the increased acceptability rating from Mittins et al.’s 40% to 66.2% today, the split infinitive remains problematical in that its current negative status may make speakers refrain from using it in order to avoid being judged negatively by others.
As for the treatment of the split infinitive by usage guides, we found it in 62 of the 77 works in the HUGE database. The entries collected from HUGE were analyzed according to the approach taken by Yáñez-Bouza (2015) in her study of preposition stranding—another old chestnut in the usage guide tradition. Yáñez-Bouza devised a threefold categorization depending on the 18th-century normative grammarians’ treatment of preposition stranding, as critics or advocates or as reflecting a neutral position. This method is useful as its categories are exclusive, thus enabling a good overview of the way in which the treatment of the split infinitive developed over the years (Yáñez-Bouza, 2015, p. 30). If a usage guide author disapproves of splitting infinitives, the entry is categorized as “criticized”; approval of split infinitives results in the categorization “advocated,” while if an entry either contains no judgment on the acceptability of the construction or suggests a contextual preference for its use, it is considered “neutral.” Examples of each category are presented in (21) through (23).
Since attitudes toward the split infinitive were only analyzed for British English speakers here, we will mainly focus on its treatment in the British publications included in HUGE, 35 of them altogether, including Trask2001 (see Section 4.1). As shown in Table 3, the majority of the usage guides published in the United Kingdom do not straightforwardly approve or disapprove of the split infinitive. Only four works criticize the use of split infinitives, while six publications advocate its use.
Table 3. British Usage Guides Classified According to Their Treatment of the Split Infinitive (Criticised, Neutral, Advocated)
Alford1864/Br, Fowler&Fowler1906(1922)/Br, Treble&Vallins1936/Br, Heffer2010/Br
Hall1917/Br, Fowler1926/Br, Partridge1942(1947)/Br, Gowers1948/Br, Vallins1951/Br, Vallins1953(1960)/Br, Fowler(=Gowers)1926(1965)/Br, Wood1962(1970)/Br, Swan1980/Br, Burchfield1981/Br, Burchfield1984/Br,Weiner&Hawkins1984/Br, Bailie&Kitchin1979(1988)/Br, Greenbaum&Whitcut1988/Br, Dear1986(1990)/Br, Weiner1983(1994)/Br, Crystal1984(2000)/Br, Howard1993/Br, Amis1997(1998)/Br, Marriott&Farrell1992(1999)/Br, PocketFowler1999/Br, Burchfield1996(2000)/Br, Peters2004(2006)/Br, Sayce2006/Br, Lamb2010/Br
OxfordDictionary1981(1984)/Br, Cutts1995/Br, Ayto1995(2002)/Br, Trask2001, Burt2000(2002)/Br, OxfordA—Z2007/Br Taggart2010/Br
Total: 35 (34 Br + Trask2001)
If we present the usage guides in Table 3 chronologically (Figure 3), it appears that, though the split infinitive was first dealt with in the mid-19th century, it really is a 20th-century usage problem.
Figure 3 shows that, with one exception (Heffer2010/Br), criticism of the split infinitive is not found after the 1940s. The first time, moreover, usage is actually advocated is in The Oxford Dictionary for Writers and Editors (1981). There we find a brief statement directing the reader to Fowler’s Modern English Usage (1926) for further guidance: “split infinitive, the separation of ‘to’ from its verb by other words, as ‘he used to often say’. Objected to by many, but the cure is often worse than the disease. See Modern English Usage, pp 579–82” (OxfordDictionary1981(1984)/Br). Figure 3 also illustrates how the split infinitive has increasingly been advocated over the last three decades in the British usage guide tradition. The treatment in Heffer2010/Br deserves special comment, since it is an example of the mistaken attribution of the origin of the split infinitive to Lowth (see above):
This began with Latinists, notably Lowth, arguing that since the infinitive was intact in that language, it had better be as intact as possible in our own too. There is no reason in that sense why this should apply in English. However, the division of to from its verb was seized on by the Fowlers, correctly in my view, as inelegant.… for the sake of logic and clarity to and the verb whose infinitive it forms are always best placed next to each other rather than interrupted by an adverb. In nearly 30 years as a professional writer I have yet to find a context in which the splitting of an infinitive is necessary in order to avoid ambiguity or some other obstruction to proper sense. (Heffer2010/Br)
Not only is Heffer mistaken in attributing the split infinitive’s origin to Lowth, his discussion of the construction’s unacceptability suggests a peculiar divergence from the overall growing acceptability of the construction.
Interestingly, the American usage guide tradition shows an earlier shift toward a more lenient treatment of the split infinitive than the British one, with Follett1966/Am describing and advocating the construction as having “its place in good composition” in the 1960s. An overview of all entries, however, reveals that usage guides in both traditions become less critical toward the split infinitive from the 1940s onward. The contextual acceptability of the split infinitive is acknowledged by various usage guide writers in this period, as exemplified by Crystal1984(2000)/Br:
Split infinitives are extremely common in everyday conversation and in informal writing (letters and postcards, for instance), which tends to follow the rhythms of speech. You'll hear them a lot on radio and television, especially in plays, sitcoms, and all forms of unscripted discussion.
You won’t hear them a great deal in formal broadcast speech, however—announcers, newscasters, programme presenter and others are generally anxious to avoid such things, because they are all too aware of the complaints which will come if they let one slip into their speech.…
Despite the growing acceptability of the split infinitive both among the survey informants and within the usage guide tradition, proscriptive attitudes toward the split infinitive can still be identified in both groups. The survey data discussed here show that the myth of the split infinitive still exerts power over linguistically insecure language users.
4.4. The Dangling Participle
The mismatch of subjects in sentences containing a dangling participle (see Section 1 for a definition) is considered problematical because it is said to cause ambiguity as well as unintentional hilarity, as the example from O’Conner1996(1998)/Am illustrates: “Born at the age of forty-three, the baby was a great comfort to Mrs. Wooster.” The proscription against dangling participles is often considered a grammatical error rather than a stylistic one (Aarts, 2014). It is an interesting issue, since ambiguity is usually only caused when no context is available. Constructions with participles like assuming, speaking, considering, and provided have acquired idiomatic status and are no longer considered dangling participles but rather absolute constructions (DeBakey & DeBakey, 1983, pp. 233–234). The ellipsis of the subject in such instances is rarely problematical in context.
There are various studies on the dangling participle that date from the early 20th century, such as Tressler’s (1917) investigation of errors made by (American) high school students in their written and spoken compositions, which showed that 2% of all grammatical errors recorded were dangling participles (Tressler, 1917, p. 649). According to Bartlett (1953), dangling participles are also common in English literature and with good authors. She further explains that “[t]he ubiquitous dangler which offends against sense and style is the fault of half-educated writers, trying seriously and awkwardly to sound like a book” (Bartlett, 1953, p. 354)—an attribution which reminds us of the origin of the debate on the split infinitive (see Section 4.3).
In the Mittins survey, the dangling participle in the sentence Pulling the trigger, the gun went off unexpectedly received an average acceptability rating of 17%. Ranked 46th among the 50 items studied, it was one of the least acceptable usage problems in the study. More than four decades later, the feature achieved only a slightly higher acceptability rating from the (British) informants consulted: 22.7%. Despite this slight increase, the construction is still overwhelmingly considered unacceptable. With an acceptability rating of 12.9%, formal speech is the context in which the dangler is least acceptable, while the highest acceptability rating, 36.2%, was found for informal speech. Falling in between these two contexts, the online/mobile contexts show acceptability ratings of 14.7% (informal online/mobile) and 30.2% (formal online/mobile). Correlating relative acceptability with the social variables age, gender, education, and social class did not produce any statistically significant results, which possibly confirms the construction’s widespread controversial status.
A qualitative analysis of comments made by the questionnaire respondents provides an insight into speakers’ attitudes to the construction. Categorizing the comments brings a large variety of themes to light. While many respondents were able to identify the problem, ambiguity and insecurity about the sentence’s correctness were frequently mentioned. As the comment in (24) shows, the mismatch of the two subjects is identified by a female student as the source of the confusion, while the one in (25) provides us not only with a correction— “He pulled the trigger, and the gun went off”— but the retired teacher also hints at the importance of contextual information.
Comments (26) through (28) highlight the confusion which a sentence like Pulling the trigger, the gun went off may cause without any context. The retired teacher in (27) states that the context would clarify who pulled the trigger.
Comparison with the Mittins study indicates that dangling participles are still widely considered unacceptable. However, the comments made by the questionnaire respondents highlight the importance of the context. Reading a sentence like the Mittins one in isolation, one will certainly wonder who pulled the trigger, while contextual information would have made this clear (Ebner, 2014, pp. 3–4).
The construction is discussed in 49 of the 77 usage guides in the HUGE database. Applying Yáñez-Bouza’s (2015) threefold categorization—criticized, neutral, advocated—to the dangling participle entries brings to light clear tendencies in its treatment. Examples of the different treatments can be found in (29) and (30):
None of the 27 British publications (including Trask2001), however, advocates the use of the dangling participle without any restrictions. In Table 4 we see that, in contrast to the split infinitive (Table 3), the categorization of the entries on the dangling participle shows that there are far fewer neutral discussions of the feature.
Table 4. British Usage Guides Classified According to Their Treatment of the Dangling Participle (Criticised and Neutral Only)
Fowler&Fowler1906(1922)/Br, Fowler1926/Br, Treble&Vallins1936/Br, Partridge1942(1947)/Br, Gowers1948/Br, Vallins1951/Br, Fowler(=Gowers)1926(1965)/Br, Burchfield1981/Br, Burchfield,Weiner&Hawkins1984/Br, Bailie&Kitchin1979(1988)/Br, Swan1980/Br, Greenbaum&Whitcut1988/Br, Dear1986(1990)/Br, Weiner1983(1994)/Br Blamires1994/Br, Amis1997(1998)/Br, Burchfield1996(2000)/Br, PocketFowler1999/Br, Trask2001, Burt2000(2002)/Br, Ayto1995(2002)/Br, OxfordA—Z2007/Br, Taggart2010/Br, Heffer2010/Br, Lamb2010/Br
Hall1917/Br, Vallins1953(1960)/Br, Peters2004(2006)/Br
Total: 27 (26 Br + Trask2001)
Not only did the HUGE survey show that the dangling participle is still widely considered problematical, the feature was also rated less acceptable by the respondents of the questionnaire.
Figure 4 illustrates the historical treatment of the dangling participle in the British usage guide tradition, showing that the dangler is likewise a typically 20th-century usage problem (see Section 4.3), which was only first discussed in Fowler and Fowler’s The King’s English (1906).
Comparing the treatment of the dangling participle in British publications to the American tradition shows no great deviance from the pattern discussed here. The majority of American publications likewise condemns the use of dangling participles. Of all the usage guides studied, the entry in Peters2004(2006)/Br reflects the most lenient approach toward the usage. That she mentions the importance of context is of particular interest:
Castigation of “dangling” constructions almost always focuses on sentences taken out of context. In their proper context of discourse, there may be no problem.… The third example [Now damaged in the stern, the captain ordered the ship back to port.] would sound natural enough in the context of narrative:
The bows of the vessel had been scarred by pack ice.
Now damaged in the stern, the captain ordered the ship back to port …
The narrative keeps the ship in the spotlight—in the topic position in both sentences.… If we rewrite the sentences to eliminate the dangling participles we lose the topicalizing effect they have. Any sentence in which they create a bizarre distraction should of course be recast. But if the phrase works in the context of discourse and draws no attention to itself, there’s no reason to treat it like a cancer in need of excision. (Peters2004(2006)/Br)
Having discussed both the attitudes questionnaire and the HUGE survey, it appears that attitudes toward the dangling participle differ not only diachronically but also synchronically, though there does appear to be agreement on its problematical status, among both usage guide writers and members of the speech community.
5. Different Trends in Usage Guides
Our analysis of try and, the placement of only, the split infinitive, and the dangling participle has confirmed our expectations that all four usage problems would have increased in acceptability since the Mittins survey was conducted in the late 1960s. Remarkably, however, acceptance of the dangling participle increased least of all—from 17% overall acceptance in 1970 to only 22.7% today. This low acceptability stands in somewhat inverse proportion to the number of usage guides that discuss the usage problem: only 49 of the 77 publications in the HUGE database, compared to 53 for try and and 62 for the placement of only and the split infinitive. This has to do with the fact that the dangling participle is the youngest of the four usage problems analysed here: it was only first discussed in The King’s English by Fowler and Fowler (1906). With one exception, Alford1964/Br, the split infinitive, too, gained considerable currency as a usage problem after it made its appearance in The King’s English. Both features are therefore typically 20th-century usage problems, though their current status differs considerably. While the split infinitive has undergone an enormous increase in usage, both in British and American English, and has indeed been advocated rather than proscribed by usage guide writers since the 1980s, the dangling participle is never advocated and rarely even treated neutrally by usage guides. For all that, attitudes may be about to change in view of the recognition that ambiguity, the main problem critics have with the construction, is hardly ever a real issue since context usually helps to solve the problem. The split infinitive nevertheless remains problematical in that even today, British informants fear they might be judged negatively when using the construction. It may therefore be concluded that “the split infinitive syndrome,” as Burchfield called it in The English Language (1985), is still alive today. The extent to which this is the case may appear from several instances in the survey discussed in Section 4.2 in which informants identified the misplacement of only as an instance of a split infinitive (“He only had to finish one chapter would be better and would eliminate the split infinitive,” British, retired, over 60, female).
In the light of the increasing acceptance of the four usage problems analyzed, the sociolinguistic variable age proved an important factor, particularly with the placement of only and the split infinitive (see the increased use of thusly; Section 3). For the first of these two items, the oldest informants proved least accepting of “misplaced” only, in which they did not differ greatly from the Mittins informants 45 years ago. For the split infinitive, too, it was found that acceptance correlated in inverse proportion with age. We expect this to be true for other usage problems as well, particularly those commonly called old chestnuts, though possibly also newer ones. Further research will have to bear this out. Usage problems tend to be associated with particular groups of users (Weiner, 1988), the “uneducated,” for instance, in the case of the split infinitive. It is interesting to see that in some cases such criticism originated outside the usage guide tradition. The split infinitive is a good example of this, but so is try and, which was likewise first criticized in a journal intended for the general public. In this case, however, the attack, by Moon (1868), was directed at a fellow grammarian, hardly an uneducated person. Once again, we see usage criticism originating not in a usage guide or grammar but in a public journal, as was discovered by Percy (2008, 2009) in the context of the 18th-century normative tradition. It remains for us to find similar early attributions for the placement of only and the dangling participle—not something, however, that can easily be undertaken through systematic analysis.
Perhaps our most important finding is that usage guides do not as a rule become more descriptive in the course of time, advocating constructions that were proscribed or criticized in earlier days. Our findings consequently go against the account presented in Peters (2006) in this respect. Proscriptive advice continues to be offered, as our discussion of try and and the split infinitive has shown. Two trends may therefore be distinguished that exist alongside each other: one in which usage guides adopt a more descriptive approach over the years, and another, in direct continuation of the origins of the tradition, that offers advice that in effect keeps the old stricture alive. Heffer2010/Br is a good example of this. Though from a linguistic perspective it must be said that such usage guides are fighting a losing battle—“misplaced” only and the split infinitive, for instance, as well as singular they (Section 1) are in such widespread use that they are not really perceived as usage problems any more—this may simply be the kind of advice insecure readers hope to find when consulting a usage guide. It is this that may explain the popularity of usage guides like Heffer2010/Br and Taggart2010/Br. What we consider more seriously problematical, however, is that unfounded emotional accounts continue to be part of usage advice—as in Fogarty2008/Am—and that misattributions are still made to 18th-century normative grammarians like Lowth, as in Heffer2010/Br. Performing a full-text search for “Lowth” in the HUGE database shows that Heffer does not stand alone here. It is by creating HUGE as a tool for (socio)linguists and future usage guide writers alike that we hope to have contributed to further work on prescriptivism in English.
Ager, D. (2003). Ideology and image. Britain and language. Clevedon, U.K.: Multilingual Matters.Find this resource:
Bex, T., & R. J. Watts (Eds.). (1999). Standard English: The widening debate. London: Routledge.Find this resource:
Cameron, D. (1995). Verbal hygiene. Oxford: Oxford University Press.Find this resource:
Curzan, A. (2014). Fixing English: Prescriptivism and language history. Cambridge, U.K.: Cambridge University Press.Find this resource:
Ebner, C. (Forthcoming). Proper English usage: A sociolinguistic investigation of usage attitudes in British English (PhD thesis). University of Leiden, Leiden, The Netherlands.Find this resource:
Ebner, C. (Forthcoming). Attitudes to British usage. In I. Tieken-Boon van Ostade (Ed.), English usage guides: History, advice, and attitudes. Oxford: Oxford University Press.Find this resource:
Finegan, E. (1998). English grammar and usage. In S. Romaine (Ed.), The Cambridge grammar of the English language (Vol. 4, pp. 536–588). Cambridge, U.K.: Cambridge University Press.Find this resource:
Fry, S. (2011). Planet word. BBC documentary. DVD.Find this resource:
Hitchings, H. (2011). The language wars: A history of proper English. London: John Murray.Find this resource:
Kaunisto, A. (2017). Which items need to be standardised? Variation in the choice of entries in usage guides. In I. Tieken-Boon van Ostade & C. Percy (Eds.), Prescription and tradition in language: Establishing standards across time and space (pp. 202–220). Bristol: Multilingual Matters.Find this resource:
Kostadinova, V. (Forthcoming). Attitudes to usage in American English (PhD thesis). University of Leiden, Leiden, The Netherlands.Find this resource:
Lynch, J. (2009), The lexicographer’s dilemma. The evolution of “Proper” English, from Shakespeare to South Park. New York: Walker & Company.Find this resource:
Smits, M. (2017). “Garnering” respect? The emergence of authority in the American usage tradition. In I. Tieken-Boon van Ostade & C. Percy (Eds.), Prescription and tradition in language: Establishing standards across time and space (pp. 221–237). Bristol: Multilingual Matters.Find this resource:
Straaijer, R. (2017). A perspective on prescriptivism: Language in reviews of The New Fowler’s Modern English Usage. In I. Tieken-Boon van Ostade & C. Percy (Eds.), Prescription and tradition in language: Establishing standards across time and space (pp. 185–202). Bristol: Multilingual Matters.Find this resource:
Tieken-Boon van Ostade, I. (2011). The Bishop’s grammar: Robert Lowth and the rise of prescriptivism. Oxford: Oxford University Press.Find this resource:
Tieken-Boon van Ostade, I. (2012a). The codification of English in England. In R. Hickey (Ed.), Standards of English. Codified varieties around the world (pp. 34–54). Cambridge, U.K.: Cambridge University Press.Find this resource:
Tieken-Boon van Ostade, I. (2012b). Codifying the English language. In A. Schröder, U. Busse, & R. Schneider (Eds.), Codifications, canons, and curricula. Description and prescription in language and literature (pp. 61–77). Bielefeld, Germany: Aisthesis Verlag.Find this resource:
Tieken-Boon van Ostade, I. (Forthcoming). Towards prescriptivism: A single norm for standard English? In L. Brinton (Ed.), Approaches to English historical linguistics. Cambridge: Cambridge University Press.Find this resource:
Tieken-Boon van Ostade, I. (Ed.). (Forthcoming). English usage guides: History, advice, attitudes. Oxford: Oxford University Press.Find this resource:
Tieken-Boon van Ostade, I., & Percy, C. (Eds.). (2017). Prescription and tradition in language: Establishing standards across time and space. Bristol, U.K.: Multilingual Matters.Find this resource:
Note: For bibliographical details of the usage guides discussed in this article, see the information provided in the HUGE database.
Albanyan, A., & Preston, D. R. (1998). What is Standard American English? Studia Anglica Posnaniensia, 33, 29–46.Find this resource:
Bailey, R. (1996). Nineteenth-century English. Ann Arbor: University of Michigan Press.Find this resource:
Baker, R. (1770). Reflections on the English Language. London: J. Bell.Find this resource:
Bartlett, A. C. (1953). Dangling participles. College English, 14(6), 353–354.Find this resource:
Bodine, A. (1975). Androcentrism in prescriptive grammar: Singular “they”, sex-indefinite “he”, and “he or she.” Language in Society, 4(2), 129–146.Find this resource:
Burchfield, R.W. (1985). The English language. Oxford: Oxford University Press.Find this resource:
Burchfield, R.W. (1996). Fowler’s modern English usage (3d ed.). Oxford: Oxford University Press.Find this resource:
Burgess, T. (1996). English. In P. Gordon (Ed.), A guide to educational research (pp. 53–81). Abingdon, Oxon, U.K.: Routledge.Find this resource:
Butterfield, J. (2007). Oxford A–Z of English usage. Oxford: Oxford University Press.Find this resource:
Chapman, D. (2017). Stalwarts, Snoots and some readers: How “traditional rules” are traditional. In I. Tieken-Boon van Ostade & C. Percy (Eds.), Prescription and tradition in language. Establishing standards across time and space (pp. 238–252). Bristol, U.K.: Multilingual Matters.Find this resource:
Crystal, D. (2006). Language and the Internet. Cambridge, U.K.: Cambridge University Press.Find this resource:
DeBakey, L., & DeBakey, S. (1983). Syntactic orphans and adoptees: Unattached participles. I. Mischievous intruders: first of two parts. International Journal of Cardiology, 3(1), 67–70.Find this resource:
Ebner, C. (2014). The dangling participle—a language myth? English Today, 30(4), 3–4.Find this resource:
Edwards, J. (2006). Language attitudes. In K. Brown (Ed.), Encyclopedia of language and linguistics (pp. 324–331). Boston: Elsevier.Find this resource:
Evans, B., & Evans, C. (1957). A dictionary of contemporary American usage. New York: Random House.Find this resource:
Fowler, H.W. (1926). A Dictionary of Modern English Usage. Oxford: Clarendon Press.Find this resource:
Giles, H. (1970). Evaluative reactions to accents. Educational Review, 22(3), 211–227.Find this resource:
Gilsdorf, J., & Leonard, D. (2001). Big stuff, little stuff: A decennial measurement of executives’ and academics’ reactions to questionable usage elements. The Journal of Business Communication, 38(4), 439–471.Find this resource:
Greenbaum, S. & Whitcut, J. (1988). Longman Guide to English Usage. Harlow: Longman.Find this resource:
Hairston, M. (1981). Not all errors are created equal: Nonacademic readers in the professions respond to lapses in usage. College English, 43(8), 794–806.Find this resource:
Hedges, M. (2011). An assessment of attitudes to language change based on the use of like (Master’s thesis). University of Leiden, Leiden, The Netherlands.Find this resource:
Hommerberg, C., & Tottie, G. (2007). Try to or try and? Verb complementation in British and American English. ICAME Journal, 31, 45–64.Find this resource:
Kamm, O. (2015). Accidence will happen: The non-pedantic guide to English usage. London: Weidenfeld & Nicolson.Find this resource:
Krapp, G.P. (1927). A comprehensive guide to good English. New York: Rand McNally.Find this resource:
Leonard, S.A. (1932). Current English usage. Chicago: The Inland Press.Find this resource:
Lukač, M. (2015). Linguistic prescriptivism in letters to the editor. Journal of Multilingual & Multicultural Development, 37(3/4), 321–333.Find this resource:
Lukač, M., & Tieken-Boon van Ostade, I. (Forthcoming). Attitudes to flat adverbs and English usage advice.Find this resource:
Luscombe, A. (2012). Sending the right message. Forty years of BBC radio news (PhD thesis). University of Utrecht, Utrecht, The Netherlands.Find this resource:
Mair, C. (2006). Twentieth-century English: History, variation, and standardization. Cambridge, U.K.: Cambridge University Press.Find this resource:
Milroy, J., & Milroy L. ( 2012). Authority in language: Investigating standard English (4th ed.). London: Routledge.Find this resource:
Milroy, L. (2001). The social categories of race and class: Language ideologies and sociolinguistics. In N. Coupland, S. Sarangi, & C. N. Candlin (Eds.). Sociolinguistics and social theory (pp. 235–260). London: Pearson Education.Find this resource:
Mittins, W. H., Salu, M., Edminson, M., & Coyne, S. (1970). Attitudes to English usage: An enquiry by the University of Newcastle upon Tyne Institute of Education English Research Group. London: Oxford University Press.Find this resource:
Mugglestone, L. (2012). English in the nineteenth century. In L. Mugglestone (Ed.), The Oxford history of English (pp. 340–378). Oxford: Oxford University Press.Find this resource:
Nevalainen, T., & Raumolin-Brunberg, H. (2003). Historical sociolinguistics: Language change in Tudor and Stuart England. London: Longman.Find this resource:
Nicholson, M. (1957). A Dictionary of American-English Usage: Based on Fowler’s Modern English Usage. New York: Oxford University Press.Find this resource:
Percy, C. (2008). Mid-century grammars and their reception in the Monthly Review and the Critical Review. In I. Tieken-Boon van Ostade (Ed.), Grammars, grammarians and grammar writing in eighteenth-century England (pp. 125–142). Berlin: Mouton de Gruyter.Find this resource:
Percy, C. (2009). Periodical reviews and the rise of prescriptivism: The Monthly (1749–1844) and Critical Review (1756–1817) in the eighteenth century. In I. Tieken-Boon van Ostade & W. van der Wurff (Eds.), Current issues and late modern English (pp. 117–150). Bern, Switzerland: Peter Lang.Find this resource:
Peters, P. (2004). The Cambridge Guide to English Usage. Cambridge: Cambridge University.Find this resource:
Peters, P. (2006) English usage: Prescription and description. In B. Aarts & A. McMahon (Eds.), The handbook of English linguistics (pp. 759–780). Oxford: Blackwell.Find this resource:
Pinker, S. (2014). The sense of style: The thinking person’s guide to writing in the 21st century. New York: Viking.Find this resource:
Ross, D. (2013). Dialectal variation and diachronic development of try-complementation. Studies in the Linguistic Sciences: Illinois Working Papers, 108–147.Find this resource:
Soukup, B. (2015). Mixed methods in the study of language attitudes. In A. Prikhodkine & D. Preston (Eds.), Responses to language varieties: Variability, processes and outcomes (pp. 55–84). Amsterdam: John Benjamins.Find this resource:
Straaijer, R. (Forthcoming). The usage guide: Evolution of a genre. In I. Tieken-Boon van Ostade (Ed.), English usage guides: History, advice, attitudes. Oxford: Oxford University Press.Find this resource:
Sundby, B., Bjørge, A. K., & Haugland, K. E. (1991). A dictionary of English normative grammar 1700–1800. Amsterdam: John Benjamins.Find this resource:
Tieken-Boon van Ostade, I. (1992). John Kirkby and the practice of speaking and writing English: Identification of a manuscript. Leeds Studies in English, 23, 157–179.Find this resource:
Tieken-Boon van Ostade, I. (2011). The Bishop’s grammar. Robert Lowth and the rise of prescriptivism. Oxford: Oxford University Press.Find this resource:
Tieken-Boon van Ostade, I. (2013). Studying attitudes to English usage. English Today, 29(4), 3–12.Find this resource:
Tieken-Boon van Ostade, I. (2015). Five hundred mistakes corrected: An early American English usage guide. In M. Dossena (Ed.), Transatlantic perspectives on late modern English (pp. 55–71). Amsterdam: John Benjamins.Find this resource:
Tieken-Boon van Ostade, I., & Kostadinova, V. (2015). Have went—an American usage problem. English Language & Linguistics, 19(2), 293–312.Find this resource:
Tottie, G. (2012). On the history of try with verbal complements. In S. Chevalier & T. Honegger (Eds.), Words, words, words: Philology and beyond. Festschrift for Andreas Fischer on the occasion of his 65th birthday (pp. 199–214). Tübingen: Narr Francke Attempto.Find this resource:
Trask, R.L. (2001). Mind the Gaffe: The Penguin Guide to Common Errors in English. London: Penguin Books Ltd.Find this resource:
Tressler, J. C. (1917). High-school grammar. II. An investigation. The English Journal, 6(10), 645–657.Find this resource:
Visser, F. Th. (1966). An historical syntax of the English language. Part two. Syntactical units with one verb. Leiden, The Netherlands: Brill.Find this resource:
Weiner, E. (1988). On editing a usage guide. In E. G. Stanley & T. F. Hoad (Eds.), Words: For Robert Burchfield’s sixty-fifth birthday (pp. 171–182). Cambridge, U.K.: Brewer.Find this resource:
Yáñez-Bouza, N. (2015). Grammar, rhetoric and usage in English. Cambridge, U.K.: Cambridge University Press.Find this resource:
(1.) This article was written in the context of the project Bridging the Unbridgeable: Linguists, Prescriptivists and the General Public, directed by Ingrid Tieken-Boon van Ostade and financed by the Netherlands Organisation for Scientific Research.
(2.) Throughout this article, bold is used for emphasis.
(3.) Individual items from the Mittins survey as well as others have meanwhile been studied, such as have went, flat adverbs (go slow/slowly), and the variation between -ic/ical in adjectives like historic/historical (Tieken-Boon van Ostade & Kostadinova, 2015; Lukač & Tieken-Boon van Ostade, forthcoming, respectively). These studies allow us to place our own findings into a wider perspective.
(4.) For copyright reasons, not all material in the HUGE database is, however, freely available.
(5.) Full bibliographical references to the usage guides, including page numbers of the quotations, may be found in the HUGE database. Here, for the sake of brevity, references only show American (Am) or British (Br) origin of the works—based on their place of publication—and the dates of their first publication and of the actual editions used for HUGE (in brackets).