By Jordan Smith
https://doi.org/10.55177/tc377227
Abstract
Purpose: Prescriptivism—a concept concerned with “correctness in language use” (Tieken-Boon van Ostade, 2019, p. 8)—serves an important purpose when editors and other language professionals apply the findings from empirical linguistic studies to practical communication tasks (Oaks, 2021). Usage guides catalog usage rules, but they treat these rules with varying levels of prescriptivism. Therefore, advice varies across usage guides. This study empirically investigates levels of prescriptivism observed in usage guides.
Method: Using a scale from 1 (minimally prescriptive) to 4 (maximally prescriptive), two raters coded the level of prescriptivism observed in entries for eight well-known usage problems (e.g., who/whom and lay/lie) from 11 current usage guides relating to American English. Based on the codes assigned to these entries, an overall prescriptivism index was calculated for each usage problem and usage guide.
Results: A range in levels of prescriptivism was observed. Overall, the treatment of usage problems skewed high on the prescriptivism scale with six of the eight being treated as maximally prescriptive by at least two usage guides and six having mean indexes at or above the scale’s midpoint of 2.50. Similarly, seven of the 11 usage guides gave maximally prescriptive advice for at least one usage problem and eight had mean indexes at or above 2.50. While these findings indicate a bias toward prescriptive advice, a noteworthy amount of prescription-breaking advice was also observed.
Conclusion: The findings demonstrate that usage guides vary considerably in their levels of prescriptivism; therefore, writers and editors must critically consider which advice to follow.
Keywords: prescriptivism, technical editing, usage guides, usage problems, language ideology
Practitioner’s Takeaway:
- Overall, usage advice in the guides I analyzed skewed high on the prescriptivism scale, though a sizable number of usage-guide entries (36.05%) were classified as prescription-breaking.
- Even though usage advice analyzed for this study skewed high, it was still largely inconsistent, with some guides recommending writers and editors follow a rule and other guides recommending they could comfortably ignore the rule.
- Because of this variation in usage-guide advice, technical writers and editors—and the organizations they work for—should carefully consider which usage guides to adopt. To do this, they need some understanding of how prescriptive or not prescriptive usage guides are.
Introduction
Many English speakers have strong ideas about what constitutes correct and incorrect language use. An instance of imply when infer is intended, an apostrophe placed before a plural s, and the use of literally to mean figuratively are all examples of usage that, in the prescriptive tradition, would be considered “bad” or “incorrect.” This type of usage often incites passionate criticism from self-proclaimed sticklers and sometimes even elicits potent reactions from everyday people who otherwise do not give much thought to language use.
Prescriptivism—a concept concerned with “correctness in language use” (Tieken-Boon van Ostade, 2019, p. 8)—promotes the idea that when a language user can choose from multiple linguistic alternatives to express a single meaning, some of the alternatives are correct while the others are incorrect. Though not all groups of semantically equal alternatives in a language fit into the prescriptive tradition (e.g., speakers of English can express possession with either an ’s or an of-phrase, neither of which are contested in terms of correctness), the previously mentioned examples represent some well-known prescriptive rules: imply and infer have distinct meanings, plurals are not typically formed with apostrophes, and literally does not mean figuratively.
Technical communicators encounter prescriptive ideologies in their work, and an increasing body of research in technical communication journals has studied prescriptive rules or topics related to prescriptivism. These studies (reviewed in the next section) have focused on rules that people seem to care about, and the findings from them have informed recommendations for whether and when to follow certain rules. The current study adds to this growing body of research on prescriptivism by empirically analyzing the levels of prescriptivism found in popular, current usage guides that contain information about American English. The findings presented here can help writers and editors understand the extent to which the advice given in the guides varies. Understanding this variation can begin to help writers and editors analyze and evaluate the advice in the guides, rather than just accept it on faith and rotely follow it.
Literature Review
Prescriptivism has been a topic of interest in TPC research and is particularly relevant to copyediting, a task that many technical communicators—writers and editors alike—perform in their work. In this section, I review previous literature broadly related to prescriptivism in TPC research. I then describe prescriptivism’s role in copyediting and present the research questions this study seeks to answer.
Prescriptivism in TPC Research
Prescriptivism has been a topic of valuable research in technical and professional communication (TPC) for many years. Jordan (1999) argued persuasively that not all instances of unattached clauses (a typically proscribed feature) are equally bad, and we should therefore “adopt a more accepting attitude toward the less offensive forms of weak or unclear attachment” (p. 88). Similarly, in their study of the use of unattended this (another traditionally proscribed feature) in student technical writing, Boettger and Wulff (2014) found the advice to avoid unattended this may be less preferable in some circumstances, for example, when delivering bad news. These two studies directly critique prescriptive usage advice. In contrast, Malone and Roberson (2021) argued for following prescriptive rules regarding the mandative subjunctive, suggesting that doing so “serves the goal of stylistic clarity and consistency as well as semantic accuracy” (p. 72). Other studies in TPC scholarship have centered on issues related to prescriptivism such as error correction (e.g., Eaton, 2003; Quible, 2006), botheration levels (Boettger & Moore, 2018; Gubala et al., 2020), and features of linguistic style (e.g., Campbell et al., 2021; Conrad, 2018). These areas of research are related to prescriptivism because they analyze linguistic features of usage or style that some may view as correct (or preferred) while others may view as incorrect (or dispreferred). For example, Conrad (2018) provided a comprehensive profile of the use of passive voice in engineering writing in order to “inform the development of materials that will better prepare civil engineering students for writing in the industry” (p. 66). Her findings, she argued, can help students make “effective choice[s]” (p. 66) about the use of passive voice. Because passive voice is a common focus of prescriptive usage advice, Conrad’s study is a good example of how descriptive, objective research can be effectively used to apply scientific findings to practical communication tasks, which Oaks (2021) argued is a useful function of what he called an “informed and measured prescriptivism” (p. 5).
Prescriptivism and Copyediting
One technical communication-focused task in which prescriptive guidance can be usefully applied is copyediting. When writers and editors perform the task of copyediting, they “[act] as the author’s second pair of eyes, pointing out—and usually correcting—mechanical errors and inconsistencies; errors or infelicities of grammar, usage, and syntax; and errors or inconsistencies in content” (Einsohn, 2006, p. 3). Cunningham et al. (2019) suggested that because technical editors work with rules and guidelines, they must be prescriptivists. They further elaborated that because editors “may have some say in which rules [they] follow and expect others to follow,” they “should be…sensible and informed prescriptivist[s]” (p. 328). In this way, the task of copyediting inherently involves grappling with prescriptive ideas, but successfully performing the task does not always require copy editors to adopt and enforce those ideas.
It is important to acknowledge that the task of copyediting is increasingly done not only by people with the title of “copy editor” but by almost anyone who writes. This is especially true with the emergence of AI writing tools like ChatGPT, which have been found to reduce the amount of time writers spend drafting while increasing the amount of time they spend editing (Noy & Zhang, 2023, p. 7). As more and more AI tools are adopted into professional writing workflows, writers will likely continue to spend more time editing the output—both on a substantive and copy-edit level—so that it meets the needs of their audience. Indeed, “the necessity of knowing how to edit in the workplace is greater than ever” (Cunningham et al., 2020, p. 2). Therefore, I use the term “copy editor” to refer to anyone who engages, at least on some level, in the task of copyediting.
Previous work in technical communication scholarship has suggested that writers and editors should not follow prescriptive rules uncritically (see Buehler, 1980/2003; Connatser, 2004; Smith, 2023). But why not? What characteristics of prescriptive rules should writers and editors be critical about? In many cases, following a prescriptive rule can be beneficial. For instance, following prescriptive rules can help writers and editors add to a document’s clarity and semantic accuracy (see Malone & Roberson, 2021). In addition, a controlled language such as Simplified Technical English has many “writing rules” (i.e., prescriptions) that make documents easier for machines to translate and humans to understand. Furthermore, the prescriptive rules that govern Aviation English allow air traffic controllers and pilots to communicate clearly to ensure safe air travel. And the prescriptive rules that the law mandates in warning labels and other critical communications help to ensure public safety. The rules used in each of these scenarios are arguably beneficial, and a writer, editor, or speaker working in any of these domains must follow the rules.
Prescriptive guidance can also be used to promote social justice. Many organizations have adopted usage rules that prohibit the use of racist, sexist, and ableist language. For instance, the Google Developer Documentation Style Guide (n.d.) plainly advises that writers should “avoid ableist language” like “Before launch, give everything a final sanity-check” or “There are some crazy outliers in the data” (underlining added for emphasis). Writers and editors who follow these prescriptive rules help to promote a culture of fairness and equality.
In other cases, following a prescriptive rule may be neither beneficial nor harmful. That is, following a rule may neither substantially improve the clarity of a text nor would it drastically harm the text. In other words, following these kinds of rules is neutral. An example of neutral prescriptive rule-following may be a case where an editor changes all instances of towards to toward, which Owen (2020) found to be a common change editors make, in order to conform to standard usage advice that dictates the use of toward in American English. Such a change may follow prescriptive guidance, but it ultimately does nothing to improve or to harm the text in meaningful ways.
While there are, undoubtedly, instances in which prescriptive rules can be beneficial or neutral, there are also cases where enforcing prescriptive rules may have negative effects. For example, copy editors who mechanically enforce language rules simply because they are supposed to may ultimately hinder a reader’s experience. Connatser (2004) identified several prescriptive rules that, when followed, can inhibit a reader’s experience. For example, he argued that the oft-prescribed rule to use singular units of measure for any value between 1 and –1 (inclusive) goes against what he terms a reader’s “organic grammar” and therefore hinders the reading process. For example, the use of singular gram in the sentence “Add 0.5 gram to the mixture” follows a usage rule, but plural grams may be a more natural-sounding alternative for many writers. Writers and editors who engage in this type of mechanical or uncritical following of prescriptive rules can be doing so at the expense of the experience they create for their readers.
More seriously, some prescriptive advice can go beyond inhibiting a reader’s experience to promoting “problematic and harmful ideologies” (Smith, 2023, p. 199). Smith pointed to the example of using he as a gender-neutral singular pronoun and how doing so “clearly erases those who do not use this pronoun” (p. 199). Ultimately, Smith argued that
Technical editors must understand and consider [issues related to the harmful effects of prescriptivism] in their work, and technical editing teachers must help their students understand that technical communication does not reveal absolute reality (Miller, 1979, p. 616) but instead functions rhetorically, requiring editors to think critically, be aware of their audience, and make conscious decisions about whether to follow or flout prescriptive usage advice. (p. 199)
In order to consider whether language rules enforced by a copy editor or an organization are harmful, writers, editors, and the organizations they work for need to critically consider the usage advice they adopt. This means examining the usage guides they use as well as the rules they include in their in-house style guides for any potentially harmful prejudices they may promote.
Many novice technical communicators may not be aware of just how much variation exists in usage-guide advice for all three types of prescriptive rules described above (beneficial, neutral, and negative). As I show below, one guide can give advice that runs directly counter to the advice found in another guide. Because of this, writers and editors should know something about the ideologies expressed in the usage guides they follow.
The rules in reference materials that copy editors use, such as style manuals and usage guides, offer a starting point from which writers and editors can determine what is considered appropriate usage in different contexts. These reference materials constitute similar, though not identical, genres. A usage guide “lists the meanings of words, but is not a dictionary[; it] discusses grammatical structures, but is not a grammar” (Straaijer, 2018, pp. 11–12); and it offers readers “a short cut to the acquisition of habits” that are “acquired, not automatically—through growing up among speakers of the language—but through a conscious educational process” (Weiner, 1988, p. 172). Usage guides are different from style guides like the Microsoft Style Guide and style manuals like the Chicago Manual of Style because style guides and manuals are “usually designed for in-house use by the organizations that produce them . . . as well as for external writers producing texts for such organizations” (Straaijer, 2018, p. 15) while usage guides, on the other hand, are written for a general audience. Usage guides are different from writing handbooks because unlike writing handbooks, usage guides do not “aim to instruct the reader in various aspects of writing, including the composition and structure of texts” (Straaijer, 2018, p. 15). Prescriptivism in writing handbooks has been studied in previous TPC scholarship (Mackiewicz, 1999). Mackiewicz’s work served as an inspiration for this article, but her study and the current one differ in the object of analysis (writing handbooks in Mackiewicz’s study and usage guides in this one) as well as the publication dates of the reference works under investigation. All of the guides in the current study—except for one—were published in 2004 or later, at least six years after the most current guide included in Mackiewicz’s study, which was published in 1998. Because usage advice can change over time, new studies analyzing levels of prescriptivism in all kinds of reference materials need to be conducted, especially as new reference materials (and new editions of existing reference materials) are released.
Copy editors, undoubtedly, refer to usage guides for advice about contested usage, but it is important to acknowledge that usage guides do not completely dictate the decisions editors make—even when an organization has designated a particular usage guide to follow. Editorial practices vary widely, even among editors with similar training who do very similar tasks (Owen, 2020), and editors do not always follow prescriptive advice, even when it would seem likely that they would (Lukač & Stenton, 2023, p. 281). Pillière (2020) found that a number of different factors influence the choices copy editors make, including the perceived level of formality of a text, whether or not passages appeared in dialogue, and sometimes simply the editors’ personal preferences. In the same vein, Mackiewicz and Durazzi (2023), in their study of editors’ judgments of they used as a singular pronoun, found that textual cues influenced editors’ decision-making. For example, when they appeared with a nongendered antecedent (e.g., “child”), 69% of editors found the use acceptable. But when they appeared with a gendered proper name as an antecedent (e.g., “Maria”), 58% of editors said they would query the author to confirm whether they was the preferred pronoun, and 13% said they would change the pronoun from they to her.
Other constraints, such as deciding when to overrule an author’s original preference, determining when to maintain consistency within a document or document set, and even the sometimes-arbitrary preferences of the editor or their supervisor, can influence how copy editors do their work. Sometimes an author may want an editor to enforce any and all prescriptive rules because they feel it will enhance their image. In many cases, copy editors are expected—and even required—to follow the rules outlined in their house style guide and the reference materials designated as authoritative by the organization or client they work for. In these cases, copy editors might feel compelled to enforce rules that are contrary to their own preferences or even their own standards because of organizational policies, author preferences, or supervisor direction. Even though they may not have the freedom to deviate from the rules they find objectionable, copy editors should feel empowered to “raise for discussion” (Graves & Graves, 1998, p. 412) rules or guidelines that they may feel are harmful, problematic, or otherwise unhelpful to the reader. In ideal work environments, supervisors will participate in these discussions and carefully consider endorsing the recommended changes. If, however, the copy editors’ concerns are not persuasive to those with decision-making power, the copy editors may still have to follow rules they disagree with, but they can take assurance in knowing they did what they could to resolve the issue.
Certainly, many reasons contribute to a copy editor’s decision (and sometimes ability) to follow or not follow one or more prescriptive rules. This study focuses on one of these potential reasons: the fact that usage guides exhibit a lack of uniformity in the advice they offer. Some copy editors might assume that the advice in one usage guide would be the same as—or at least similar to—the advice in any other. However, “usage guides do not necessarily speak with one voice” (Pillière, 2020, p. 258). Algeo (1991) noted that most usage guides are “based mainly upon the author’s knowledge and [express] the author’s taste and judgment” (p. 6). Indeed, some usage guides are even written by multiple authors under a single editor or several editors. Tieken-Boon van Ostade (2023) observed that some entries in usage guides may not be prescriptive or proscriptive at all but can instead be simply descriptive in which the author provides information about usage patterns but does not offer explicit usage advice. As such, usage guides can sometimes be idiosyncratic collections of the author’s (or authors’) opinions and can therefore express varying levels of prescriptivism in the advice they offer—if in fact they offer any advice at all.
For some prescriptive rules, agreement among usage guides seems to be widespread (i.e., the degree to which usage guides treat the rule prescriptively is the same). For others, however, usage guides present diverse judgments. When usage guides treat usage rules with differing levels of prescriptivism, copy editors can find themselves dealing with the difficult task of having to choose one guide’s advice over another’s. Once a decision has been made, copy editors should record their decisions in an in-house style guide or on a style sheet. However, the problem may persist as new usage questions arise—questions that are not already addressed in the in-house style guide. Understanding how prescriptively a guide treats a given usage rule, or how generally prescriptive (or not prescriptive) a usage guide’s advice is, can help copy editors make more informed decisions.
In this article, I empirically analyze levels of prescriptivism by answering these research questions:
RQ1: To what extent do 11 current usage guides relating to American English recommend following the traditional rules for eight well-known usage problems?
RQ2: To what extent does the overall level of prescriptivism observed in the usage guides vary when the advice they give for the same eight usage problems is considered collectively?
To answer RQ1, I describe a prescriptivism profile for the usage problems, noting which usage problems are treated most and least prescriptively by the usage guides. To answer RQ2, I describe a prescriptivism profile for the usage guides, based on a four-item scale, noting which guides gave the most and least prescriptive advice for the usage problems I studied.
Three terms used in the research questions above require definition. Usage refers to “actual [language] use in edited and printed American English” (Wachal, 2000, p. 199). Usage problem is a technical term (see Ilson, 1985, for an early use and discussion) used to refer to prescriptive rules and the linguistic variants associated with them. The linguistic variants associated with a usage problem comprise “a set of two or more ways of saying the same thing” (Grieve, 2016, p. 38) in which the correctness of one of the variants is contested. Traditional rules refers to guidelines that prescribe how the linguistic variants associated with a usage problem should be used. Cameron (2012, 1995) called these guidelines “received ideas” that seem to have no originating source, and Chapman (2017) noted that “prescriptive rules constitute a tradition which valorises rules independent of other benefits claimed for the rules, such as their usefulness for enhancing clarity or elegance in language” (p. 238).
In the following sections of this article, I describe the criteria I used to select the usage problems and the usage guides, and I describe the methods I used to analyze them. Then, I present the results of the analysis and discuss implications.
Selecting and Analyzing Usage Problems and Usage Guides
To determine which usage problems to analyze, I used the Hyper Usage Guide of English (HUGE) (Straaijer, 2014) to identify the 10 usage problems that appear in the highest number of usage guides that contain information about American English. I reasoned that 10 usage problems would provide an amount of data that would be manageable yet still robust enough to allow me to conduct a useful analysis. The HUGE database contains 77 usage guides (44 of which are categorized as relating primarily or in part to American English; the others relate primarily to British English) and catalogs a total of 123 usage problems. Thus, the HUGE database allows users to easily compare which usage problems are contained in different usage guides and, where copyright permissions have been obtained, to read the entries for these usage problems from the usage guides.
Because the analysis reported here was part of a larger study that used corpus methods to study adherence to prescriptive usage rules in formal and informal written English, I eliminated two of the 10 usage problems (SHALL/WILL and PLACEMENT OF ONLY) because they could not be feasibly analyzed given the constraints of the larger study—either because authentic examples of the linguistic variants associated with the usage problems were infrequently observed (as was the case with SHALL/WILL), or because determining adherence to a given rule would have required interviewing or surveying writers to assess their intended meaning (as was the case with PLACEMENT OF ONLY). Table 1 shows the remaining eight usage problems included in this analysis.
To select the usage guides to include in the study, I identified the 10 most current usage guides coded as relating to American English in the HUGE database (some of the guides I selected, for example Peters [2004], were coded as relating to both American and British English). Again, I selected 10 because it was a manageable number that still provided enough data to conduct a useful analysis. If a guide in that list had a more recent edition than the one in the HUGE database at the time data was collected in late 2018–early 2019, I used the entries from the more current version. Two of the guides that appeared in the list, The New Fowler’s Modern English Usage (Fowler, 1996, 2000) and Pocket Fowler’s Modern English Usage (Allen, 1999), had new editions both edited by the same person, so I included only the more comprehensive guide—Fowler’s Dictionary of Modern English Usage (Butterfield, 2015)—in this study. I also included current editions of Merriam-Webster’s Dictionary of English Usage (1994) and The Elements of Style (Strunk & White, 2009) because of their popularity and influence1. Many editions of The Elements of Style have been published or reprinted, likely because the book’s contents are now in the public domain. The edition I included in this study is the 50th anniversary edition, which is a reprint of the fourth edition. Finally, new editions of Woe Is I (O’Conner, 2019) and Garner’s Modern English Usage (Garner, 2022) were published while I was collecting and analyzing the data presented in this study (in the case of O’Conner [2019]) or after data collection and analysis were complete (in the case of Garner [2022]). As a result, the editions included in this study for these two works are no longer the most current. I briefly discuss one change from the third to the fourth edition of Woe Is I below. Table 2 shows the complete list of the 11 usage guides included in this study.
I acknowledge that some of the usage guides used in this study (e.g., Strunk & White [2009] and Butterfield [2015]) have long publishing histories—in the case of Strunk and White, dating back more than 100 years. The ideas expressed in previous versions of these works may have some influence on the levels of prescriptivism observed in the editions I analyzed, but determining the extent to which this is true falls outside the scope of this study.
After selecting the usage guides, I compiled the relevant entries from the guides into a single Word document to use as a rating sheet. Because usage guides sometimes addressed a single usage problem in more than one entry (e.g., Garner [2016] addressed the I FOR ME usage problem under entries titled “B. Between you and me; *between you and I,” “CLASS DISTINCTIONS,” “HYPERCORRECTION,” and “PRONOUNS”), I used the HUGE database as a starting point for determining which entries to include in the rating sheet and added other entries that dealt substantively with the usage problems under investigation. To rate the level of prescriptivism of each usage problem, I developed a scale to calculate what I call a prescriptivism index for each usage problem in each usage guide. The scale I developed is reproduced in Figure 1. Ratings on the scale range from 1 (the guide rejects the rule and approves of breaking the rule in any context) to 4 (the guide suggests upholding the rule in all contexts). Levels 2 and 3 on the scale could be applied to advice that acknowledges the role context plays in a reader’s choice to follow the rule or not. Note that a rating of 1 or 2 is considered “prescription-breaking” while a rating of 3 or 4 is considered “prescriptive.”
Two people (the author and a PhD candidate in Applied Linguistics) used the prescriptivism scale to code the data. Neuendorf (2017) pointed out that “often, a principal investigator of [a] study serves as a coder” (p. 157), which was true in this study’s case. While not ideal, “it is a logistic fact of content analysis life” (Neuendorf, 2017, p. 157). The second rater was recruited as part of a “reciprocal arrangement,” which Geisler and Swarts (2019, p. 173) suggested can be a good way to recruit second raters. Each rater assigned a value from the prescriptivism scale to the content related to each usage problem from each usage guide in the rating sheet. The second rater was trained before completing the task. Training included explaining the prescriptivism index, reviewing exemplars for each point on the rating scale, and coding a practice set of data. We did not code each individual entry from the usage guides; instead, we considered all relevant entries for a single usage problem from the guides together when coding the level of prescriptivism for each of the eight usage problems. The inter-rater reliability coefficient (K) of the coding was 0.608, which can be considered “good,” according to Fleiss, Levin, and Paik (2003, p. 604). The weighted coefficient (KW), which can be used for ordinal data and accounts for nearness of agreement, was 0.706. The percent agreement was 70.93%; the percent adjacent agreement was 94.19%. The total prescriptivism index for each usage problem was calculated by averaging the prescriptivism index from both coders. The total prescriptivism index for each usage guide was calculated by averaging the total prescriptivism index of each usage problem from each guide.
Exploring Prescriptivism in Usage Guides
In this section, I present a prescriptivism profile for the usage problems and the usage guides based on the results of the coding. I highlight the usage problems and guides with the highest and lowest prescriptivism indexes as well as those with the most variation in their ratings, and I provide illustrative examples from the guides throughout the discussion.
Prescriptivism Profile of Usage Problems
The results of the coding are shown in Figure 2. Descriptive statistics for each usage problem are shown as box and whisker plots with data points overlaid. The upper boundary of each box represents the third quartile, or “the middle of the upper half of the data set” (Box Plots, n.d.); the lower boundary of each box represents the first quartile, or “the middle of the lower half of the data set” (Box Plots, n.d.). The whiskers, where present, represent the maximum and minimum values, and the line dissecting the box, where present, represents the median. The absence of an upper whisker indicates that the third quartile is the same as the maximum value, and the absence of a lower whisker indicates that the first quartile is the same as the minimum value. In boxes with no visible line dissecting the box, the median is the same as either the first or third quartile. The “+” inside each box represents the mean (average) prescriptivism rating for each usage problem. Numerical summaries of the minimum, maximum, and mean scores for each usage problem are shown in Table 3.
Taken on average, the usage guides took the most prescriptive stance for the I FOR ME (3.50), LAY/LIE (3.41), and LESS/FEWER (3.32) usage problems with each receiving an average prescriptivism index above 3.00. The NONE IN PLURAL CONTEXT usage problem was treated the least prescriptively by the usage guides with an average prescriptivism index of 1.45. I elaborate further on these findings below.
The usage problems with the highest prescriptivism indexes
The I FOR MEusage problem had the highest average prescriptivism index at 3.50, and six of the 11 usage guides (more than any other usage problem) treated it with maximal prescription: Batko (2004), Brians (2013), Fogarty (2008), Garner (2016), O’Conner (2009), and Strunk and White (2009). Four of these same authors (Batko, Fogarty, Garner, and O’Conner), as well as Trask (2006), took a maximally prescriptive stance toward the LAY/LIE usage problem as well.
Garner’s (2016) view demonstrated the highly prescriptive tone that some usage guides took with the I FOR ME usage problem:
Here is the characteristic view of the modern descriptive linguists: “The meaning is clear; ‘I’ is no less, or more, euphonious than ‘me’; if the usage offends, it does so because the hearer (occasionally) or the reader (more frequently) is in the habit of expecting ‘me.’ Why is such a habit worth fighting about?” Ellsworth Barnard, English for Everybody 25 (1979). This view ignores the reality and the importance of the thousands of settled views of English usage. I, as an object of a preposition or a verb, has long been stigmatized. Using it in the objective case simply relates doubts about the speaker’s ability to handle the language. (p. 111)
Trask’s (2006) comments on the LAY/LIE usage problem were also highly prescriptive, saying that “standard English absolutely requires a distinction between intransitive lie and transitive lay” (p. 167). And, similar to Garner’s suggestion that others will look down on users who fail to follow the I FOR ME rule, Trask also suggested that following the LAY/LIE rule is required of readers who want to be respected: “Native speakers of vernacular English will often find this distinction unnatural and difficult, but mastery of it is essential if you want to be regarded as literate” (p. 167).
In contrast to looking at the average prescriptivism index, another way to determine the most prescriptive usage problems is to look at which ones had the highest minimal score among the usage guides. When viewed this way, the I FOR ME usage problem was still the most prescriptive usage problem, but it was joined by the LESS/FEWER usage problem, as both had minimum prescriptivism indexes of 2.5. The two guides that treated I FOR ME with minimal prescription were Merriam-Webster’s Dictionary of English Usage (1994) and the American Heritage Guide to Contemporary Usage and Style (2005). The only guide that treated LESS/FEWER with minimal prescription was Peters (2004). A prescriptivism index of 2.5 falls squarely in the middle of the scale (see Figure 1) and does not clearly indicate whether the usage-guide author recommends erring on the side of following the rule or breaking it. So, for these two usage problems, no guides suggested that writers and editors could comfortably err on the side of breaking rule.
This lack of a clear recommendation is evident in excerpts from the usage guides. Peters (2004) said the decision to use less or fewer is a “stylistic choice” (p. 205) based largely on the formality of the context. Similarly, after offering a catalog of historical commentary on the I FOR ME issue, Merriam-Webster’s Dictionary of English Usage offers the following advice that underlines the contextual considerations its author(s) feel writers should keep in mind when determining whether to follow the rule or not:
Conclusion: you are probably safe in retaining between you and I in your casual speech, if it exists there naturally, and you would be true to life in placing it in the mouths of fictional characters. But you had better avoid it in essays and other works of a discursive nature. It seems to have no place in modern edited prose. (p. 183)
The I FOR ME, LAY/LIE, and LESS/FEWER usage problems are associated with rules that usage-guide writers seem to care about a lot. Writers and editors may consider treating these usage problems prescriptively if they wish to align with prescriptive guidelines. On the other hand, writers and editors who may wish to ignore these prescriptive rules should do so with the understanding that many usage guides—and likely many readers as well—will disagree with their decision. Smith (2023) argued that copy editors should feel free to break any prescriptive rule if they have a reason for doing so. Therefore, writers and editors who choose to ignore these rules are advised to first have a clear, identifiable reason for doing so.
The usage problem with the lowest prescriptivism index
The NONE IN PLURAL CONTEXT usage problem had the lowest average prescriptivism index (1.45) and by far the lowest maximum score (2). Five of the 11 usage guides—the most of any usage problem—treated it with the lowest possible amount of prescriptivism: Brians (2013), Garner (2016), Merriam-Webster’s Dictionary of English Usage (1994), Peters (2004), and Trask (2006). Three other usage problems (DIFFERENT TO/THAN/FROM, SINGULAR THEY, and SPLIT INFINITIVE) had at least one guide that was assigned a prescriptivism index of 1, but the NONE IN PLURAL CONTEXT usage problem was the only one to receive the lowest prescriptivism score by five different guides.
NONE IN PLURAL CONTEXT was also the usage problem that caused the most disagreement between the two raters during coding. Of the 10 guides that commented on none in plural context, the two raters disagreed on half of them, and of those five disagreements, four ratings were more than one value apart (i.e., not adjacent agreement). This high level of disagreement may have resulted at least in part from the way the traditional prescriptive rule was defined for this study. The study adopted the most extreme definition, stating that none is always singular and therefore should be used only with singular verbs. However, as Kim (2018) found, “most dictionaries and grammar/usage books . . . dismiss the myth that none can only be singular” (p. 52). Most of the entries in the guides I studied also suggested that none can be used as both a singular and a plural pronoun; however, some entries seemed to clearly reject the traditional rule while at the same time suggest that the rule should be followed in certain circumstances. This may be one reason that achieving agreement for the entries on this usage problem was so difficult.
Butterfield’s (2015) entry is indicative of this apparent dual nature. The entry begins with a clear rejection of the rule, saying, “It is a mistake to suppose that the pronoun is singular only and must at all costs be followed by singular verbs or pronouns” (p. 551). But it ends with advice that acknowledges the contextual constraints writers and editors must consider when choosing whether or not to follow the rule: “Verdict: use a singular verb where appropriate but if the notion of plurality is present a plural verb has been optional since the OE period and in some circumstances is desirable” (p. 552).
The American Heritage Guide to Contemporary Usage and Style (2005), Batko (2004), Butterfield (2015), and Strunk and White (2009) all had the highest prescriptivism index for the NONE IN PLURAL CONTEXT usage problem, each receiving a 2. No other usage problem had such a low maximum prescriptivism index. The nearest one is SPLIT INFINITIVE, with an index of 3. With a maximum prescriptivism index of 2, writers and editors can comfortably err on the side of ignoring the traditional rule for the NONE IN PLURAL CONTEXT usage problem.
The usage problems with the most variation in their ratings
Two of the usage problems received the full range of ratings from the prescriptivism index scale: DIFFERENT TO/THAN/FROM and SINGULAR THEY. The average prescriptivism indexes for both usage problems were similar; however, the level of prescriptivism with which usage guides treated SINGULAR THEY was slightly lower (2.5 vs. 2.64 for DIFFERENT TO/THAN/FROM). Excerpts from usage guides on the DIFFERENT TO/THAN/FROM usage problem demonstrated the very different views usage guides express about whether or not to follow this rule. Fogarty’s (2008) entry (prescriptivism index: 4) stated the rule concretely and then offered a mnemonic to help readers remember it so they can follow it in their writing: “Different from is preferred to different than. I remember this by remembering that different has two f’s and only one t, so the best choice between than and from is the one that starts with an f” (p. 22).
In contrast, the entry in Merriam-Webster’s Dictionary of English Usage (prescriptivism index: 1) reviewed the history of the DIFFERENT TO/THAN/FROM usage problem and concluded with the proclamation that all three variants are standard and therefore unproblematic:
In summary we can say that there need have been no problem here at all, since all three expressions [i.e., different to, different than, and different from] have been in standard use since the 16th and 17thcenturies and all three continue to be in standard use (p. 343)
Unlike DIFFERENT TO/THAN/FROM, SINGULAR THEY is a usage problem that carries with it a heavy social and cultural load. The use of they as a singular pronoun has received increasing attention in recent years due to its growing use as the preferred pronoun for people who do not identify with traditionally masculine or feminine pronouns. Indeed, major style guides now accept this use of they as a singular pronoun, and Merriam-Webster declared they its word of the year in 2019 (Harmon, 2019).
Given the changing attitudes about the use of they as a singular pronoun, it is not surprising to see a wide range of opinions on its preferred use in the usage guides. Peters’s (2004) entry (prescriptivism index: 1) showed an uncompromising acceptance of this use: “Yet that kind of response to singular they/them/their [i.e., avoiding them as singular pronouns] is no longer shared by the English-speaking population at large. Writers who use singular they/them/their are not at fault” (p. 538, emphasis in original). O’Conner’s (2009) entry (prescriptivism index: 4), on the other hand, took the opposite view. She called the use of singular they “careless” and “a mistake,” and she advised readers who do not want to use generic masculine pronouns or other workarounds to simply reword the sentence rather than using a singular they (pp. 13–15). Some parts of the entry can even sound jarring to today’s progressive readers: “Strictly speaking, one person can’t be a they. Yes, it’s tempting to use they and them when you don’t know whether the somebody is a he or a she. But resist the temptation” (p. 13). As noted above, the edition of O’Conner’s book included in this study is the third, published in 2009. The fourth edition, published in February 2019, was released while data was being collected and analyzed for this study and is therefore not included in the formal analysis. However, reviewing the fourth edition’s commentary on singular they reveals a marked shift in the advice O’Conner offers on this issue, as the new edition now approves of the use of singular they with indefinite antecedents. This change reflects O’Conner’s response to the rapidly shifting views on the cultural norms associated with this usage problem, and it underscores the need for studies like this one to repeatedly analyze the extent to which usage guides treat usage rules prescriptively—especially as the guides are updated and revised.
Prescriptivism Profile of Usage Guides
The previous section has reviewed the prescriptivism indexes for the usage problems I studied. In this section, I discuss the prescriptivism indexes for the usage guides. Figure 3 shows the descriptive statistics for each of the usage guides included in this study. Table 4 shows the minimum, maximum, and mean scores for each usage guide.
As the figure and the data in the table make clear, some usage guides tended to be much more prescriptive in general than others, and vice-versa. Strunk and White (2009) had the highest average prescriptivism index (3.56), indicating that, for most of the usage problems analyzed in this study, Strunk and White suggested their readers should uphold the rule. In fact, LAY/LIE is the only usage problem for which Strunk and White were not among those who showed the highest level of prescriptivism. Batko (2004) had the second highest average prescriptivism index (3.43). She did not comment on the SPLIT INFINITIVE usage problem, but, for the remaining seven, she took a maximally prescriptive stance on four (I FOR ME, LAY/LIE, LESS/FEWER, and WHO/WHOM) and the most prescriptive stance among the other guides for one more (NONE IN PLURAL CONTEXT). Neither Batko nor Strunk and White took a minimally prescriptive stance on any usage problem.
In contrast to these two highly prescriptive usage guides, Merriam-Webster’s Dictionary of English Usage (1994) had the lowest average prescriptivism index (1.81), suggesting that in most cases, the authors of this usage guide allowed their readers to make their own decisions on whether to follow the traditional rules or not. Peters (2004) had the second lowest average prescriptivism index (2.00). These two guides both rely heavily on data (historical commentary and usage in the case of Merriam-Webster’s Dictionary of English Usage and corpus data in the case of Peters) to make their recommendations. Taking this data-driven approach likely minimizes the personal bias of the authors and results in a less prescriptive approach overall.
Brians (2013), Fogarty (2008), Garner (2016), O’Conner (2009), and Trask (2006) showed the greatest variation in their assessments as the only five guides that had scores spanning the whole range of the prescriptivism index. In contrast, the American Heritage Guide to Contemporary Usage and Style (2005) and Butterfield (2015) were the most moderate guides, both with the smallest range between their minimum and maximum scores and the only two guides that did not rate any of the usage problems with scores at either extreme end of the prescriptivism index. Writers and editors looking for advice that is neither minimally or maximally prescriptive might consider adding the American Heritage Guide to Contemporary Usage and Style or Butterfield to their collections.
The results presented in this section have demonstrated the degree to which current usage guides treat commonly discussed usage problems differently, and they highlighted the overall level of prescriptivism observed in each usage guide. As I noted earlier, someone unfamiliar with usage guides might assume that all usage guides offer the same (or similar) guidance on whether to follow a prescriptive usage rule or not. However, the findings discussed here suggest otherwise. Variation can be observed in the advice usage guides offer, which can result in some usage guides being more (or less) prescriptive than others.
Conclusion
One of my goals in carrying out this study has been to demonstrate that usage advice does indeed vary—and not just from one usage problem to another, but also in terms of how prescriptively individual usage guides treat the same usage problem. As this study shows, advice for the same usage problem can vary widely across different usage guides. The study presented here has showed the extent to which this is the case—at least for the small sample of usage guides I analyzed.
Overall, the findings from this analysis show that the results skew high on the prescriptive scale for the data I analyzed. Six of the eight usage problems and eight of the 11 usage guides had mean indexes at or above the midpoint of the prescriptivism scale. In addition, six of the eight usage problems were treated with maximal prescriptivism by at least two usage guides, and seven of the 11 guides gave maximally prescriptive advice for at least one usage problem. Given that we generally think of usage guides as prescriptive works, the fact that the overall results skew high is unsurprising.
However, there are still important indicators of a lack of prescriptivism observed in the data. One usage problem, NONE IN PLURAL CONTEXT, received a maximum prescriptivism index of 2, which is on the “prescription-breaking” half of the scale shown in Figure 1, and all but four guides gave at least some minimally prescriptive advice. In addition, of the 86 total ratings made by the coders2, more than one-third (31/86 = 36.05%) were considered prescription-breaking (i.e., received a rating between a 1 and 2, inclusive, on the prescriptivism scale). So, while the majority of the entries rated for this study were at least somewhat prescriptive, a sizeable amount was not, suggesting that usage guides regularly give advice that allows writers and editors to comfortably break traditional prescriptive rules—at least in some contexts.
While the findings from this study are particularly relevant to copy editors and those who engage in the task of copyediting, they are useful for other people and other reasons as well. For example, an awareness of varying levels of prescriptivism can inform writers’ choices of style for different audiences and purposes. Understanding that some usage guides are more or less prescriptive can help writers tailor their style to suit the needs of specific audiences and contexts because it gives them the freedom to do so. In addition, the findings presented in this analysis have implications for the teaching of technical and professional writing. Instructors could use the findings to guide discussions of prescriptivism and usage with their students. Students learning to become technical writers would benefit from understanding levels of prescriptivism in common usage guides. Finally, the findings of this study can have broader implications for communication across disciplines and cultures. Prescriptive rules vary across fields and cultures, so an awareness of differing levels of prescriptivism can help technical communicators navigate multiple discourse communities.
The analysis for this study has been robust in many ways, but it is not free from limitations. For example, I collected entries from only a small number of current usage guides to assess current recommendations for a small number of usage problems. While the data analyzed was sizable (amounting to roughly 60,000 words of usage-guide advice) and enough to point to potential patterns, it is still not enough to make conclusive pronouncements about the state of usage guides and usage problems as a whole. Future studies might analyze the usage advice for additional usage problems from a larger number of usage guides (including not only those intended for an American audience) to provide a more comprehensive view of the level of prescriptivism present among usage guides.
In addition, in the current study, raters assigned only one prescriptivism index for each usage problem per usage guide. Future studies might break the entries down into smaller codable units or comments (see Yáñez-Bouza, 2015), assign a rating to each comment in which a usage problem is discussed, and then average these ratings together to create a potentially more accurate prescriptivism index for each usage guide.
Another limitation was the fact that the author was one of the raters of the data. This can introduce biases into the coding (Neuendorf, 2017), but as Neuendorf also explained, “in unfunded academic research,” having a principal investigator act as a coder “is a logistic fact of content analysis life” (p. 157). Having a second rater code the data and averaging the ratings each coder assigns should help to mitigate any biases introduced into the analysis.
Despite the limitations, the findings presented here have important implications for technical communicators as they consider what usage advice to follow. Knowing which usage problems are treated more or less prescriptively, and knowing approximately how prescriptive or lenient a usage guide is, can help writers and editors to more critically consider the advice given in the guides, rather than simply accepting the advice that one usage guide gives. When usage-guide advice conflicts, and writers and editors are unsure of which course to take, they should carefully consider the rhetorical situation of the text they are editing (Buehler, 1980, 2003). They might also follow Smith’s (2023) advice to use corpora to help them make their decisions.
Acknowledgment
I would like to thank Joe Geluso for his help rating the usage-guide entries.
References
Algeo, J. (1991). Sweet are the usages of diversity. Word, 42(1), 1–17. https://doi.org/10.1080/00437956.1991.11435829
Allen, R. E. (1999). Pocket Fowler’s modern English usage (1st ed.). Oxford University Press.
The American Heritage guide to contemporary usage and style. (2005). Houghton Mifflin.
Batko, A. (2004). When bad grammar happens to good people: How to avoid common errors in English. The Career Press.
Boettger, R. K., & Moore, L. E. (2018). Analyzing error perception and recognition among professional communication practitioners and academics. Business and Professional Communication Quarterly, 81(4), 462–484. https://doi.org/10.1177/2329490618803740
Boettger, R. K., & Wulff, S. (2014). The naked truth about the naked this: Investigating grammatical prescriptivism in technical communication. Technical Communication Quarterly, 23(2), 115–140. https://doi.org/10.1080/10572252.2013.803919
Box Plots. (n.d.). MathBitsNotebook.Com. https://mathbitsnotebook.com/Algebra1/StatisticsData/STboxplot.html
Brians, P. (2013). Common errors in English usage (3rd ed.). William, James & Co.
Buehler, M. F. (2003). Situational editing: A rhetorical approach for the technical editor. Technical Communication, 50(4), 458–464. (Original work published in 1980).
Butterfield, J. (Ed.). (2015). Fowler’s dictionary of modern English usage (4th ed.). Oxford University Press.
Cameron, D. (2012). Verbal hygiene. Routledge. (Original work published in 1995).
Campbell, K. S., Naidoo, J. S., & Smith, J. (2023). When your boss says, “you need to sound more professional”: Writing style and writer attributions. International Journal of Business Communication, 60(4), 1071–1094. https://doi.org/10.1177/23294884211025735
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). Multilingual Matters.
Connatser, B. R. (2004). Reconsidering some prescriptive rules of grammar and composition. Technical Communication, 51(2), 264–275.
Conrad, S. (2018). The use of passives and impersonal style in civil engineering writing. Journal of Business and Technical Communication, 31(1), 38–76. https://doi.org/10.1177/1050651917729864
Cunningham, D. H., Malone, E. A., & Rothschild, J. M. (2020). Technical editing. Oxford University Press.
Eaton, A. (2003). The effectiveness of two methods of correcting formal error. Business Communication Quarterly, 66(2), 79–83. https://doi.org/10.1177/108056990306600209
Einsohn, A. (2006). The copyeditor’s handbook: A guide for book publishing and corporate communications (2nd ed.). University of California Press.
Fleiss, J. L., Levin, B., & Paik, M. C. (2003). Statistical methods for rates and proportions (3rd ed.). Wiley.
Fogarty, M. (2008). Grammar Girl’s quick and dirty tips for better writing. Holt Paperbacks.
Fowler, H. W. (2000). The new Fowler’s modern English usage (R. W. Burchfield, Ed.; 3rd ed.). Oxford University Press. (First impression published in 1996).
Garner, B. A. (2016). Garner’s modern English usage (4th ed.). Oxford University Press.
Garner, B. A. (2022). Garner’s modern English usage (5th ed.). Oxford University Press.
Geisler, C., & Swarts, J. (2019). Coding streams of language: Techniques for the systematic coding of text, talk, and other verbal data. The WAC Clearinghouse; University Press of Colorado. https://doi.org/10.37514/PRA-B.2019.0230
Google Developer Documentation Style Guide. (n.d.). “Write inclusive documentation.” https://developers.google.com/style/inclusive-documentation
Graves, H. B., & Graves, R. (1998). Masters, slaves, and infant mortality: Language challenges for technical editing. Technical Communication Quarterly, 7(4), 389–414. https://doi.org/10.1080/10572259809364639
Grieve, J. (2016). Regional variation in written American English. Cambridge University Press. https://doi.org/10.1017/CBO9781139506137
Gubala, C., Larson, K., & Melonçon, L. (2020). Do writing errors bother professionals? An analysis of the most bothersome errors and how the writer’s ethos is affected. Journal of Business and Technical Communication, 34(3), 250–286. https://doi.org/10.1177/1050651920910205
Harmon, A. (2019, December 10). ‘They’ is the word of the year, Merriam-Webster says, noting its singular rise. The New York Times. https://www.nytimes.com/2019/12/10/us/merriam-webster-they-word-year.html
Ilson, R. F. (1985). Usage problems in British and American English. In S. Greenbaum (Ed.), The English language today (pp. 166–182). Pergamon Press.
Jordan, M. P. (1999). “Unattached” clauses in technical writing. Journal of Technical Writing and Communication, 29(1), 65–93. https://doi.org/10.2190/41PB-WPVV-0VXY-JM1Q
Kim, S. (2018). ‘Two rules are at play when it comes to none ’: A corpus-based analysis of singular versus plural none. English Today, 34(3), 50–56. https://doi.org/10.1017/S0266078417000554
Lukač, M., & Stenton, A. (2023). Copy editors, (not) all alike. In J. C. Beal, M. Lukač, & R. Straaijer (Eds.), The Routledge handbook of linguistic prescriptivism (pp. 264–282). Routledge.
Mackiewicz, J. (1999). Gauging prescriptivism in writing handbooks. IEEE Transactions on Professional Communication, 42(1), 55–58. https://doi.org/10.1109/47.749369
Mackiewicz, J., & Durazzi, A. (2023). Editors’ perceptions of singular they. Technical Communication Quarterly, 1–18. https://doi.org/10.1080/10572252.2023.2184499
Malone, E. A., & Roberson, E. M. (2021). The mandative subjunctive in technical writing, or the gap between subconscious and conscious grammatical knowledge. Technical Communication, 68(2), 61–86.
Merriam-Webster’s dictionary of English usage. (1994). Merriam-Webster.
Miller, C. R. (1979). A humanistic rationale for technical writing. College English, 40(6), 610–617.
Neuendorf, K. A. (2017). The content analysis guidebook (2nd ed.). Sage.
Noy, S., & Zhang, W. (2023). Experimental evidence on the productivity effects of generative artificial intelligence. Science, 381(6654), 187–192. https://doi.org/10.2139/ssrn.4375283
Oaks, D. D. (2021). Linguistic encounters in real world prescriptivism: Acknowledging its place and role. Lingua, 264, 1–12. https://doi.org/10.1016/j.lingua.2021.103159
O’Conner, P. T. (2009). Woe is I: The grammarphobe’s guide to better English in plain English (3rd ed.). Riverhead Books.
O’Conner, P. T. (2019). Woe is I: The grammarphobe’s guide to better English in plain English (4th ed.). Riverhead Books.
Owen, J. (2020). Practicing prescriptivism: How copy editors treat prescriptive rules. In D. Chapman & J. D. Rawlins (Eds.), Language prescription: Values, ideologies and identity (pp. 280–294). Multilingual Matters.
Peters, P. (2004). The Cambridge guide to English usage. Cambridge University Press.
Pillière, L. (2020). US copy editors, style guides and usage guides and their impact on British novels. In D. Chapman & J. D. Rawlins (Eds.), Language prescription: Values, ideologies and identity (pp. 253–279). Multilingual Matters.
Pullum, G. K. (2010). The land of the free and The elements of style. English Today, 26(2), 34–44. https://doi.org/10.1017/S0266078410000076
Quible, Z. K. (2006). Impact of error labeling on error elimination in business writing. Business Communication Quarterly, 69(1), 8–24. https://doi.org/10.1177/1080569905285597
Smith, J. (2023). Corpus linguistics and technical editing: How corpora can help copy editors adopt a rhetorical view of prescriptive usage rules. Journal of Business and Technical Communication, 37(2), 194–216. https://doi.org/10.1177/10506519221143125
Straaijer, R. (2014). Hyper usage guide of English (HUGE) [dataset]. Leiden University Centre for Linguistics. http://huge.ullet.net
Straaijer, R. (2018). The usage guide: Evolution of a genre. In I. Tieken-Boon Van Ostade (Ed.), English usage guides: History, advice, attitudes (pp. 11–29). Oxford University Press.
Strunk, W., & White, E. B. (2009). The elements of style (4th ed.). Longman.
Tieken-Boon van Ostade, I. (2019). Describing prescriptivism: Usage guides and usage problems in British and American English. Routledge.
Tieken-Boon Van Ostade, I. (2023). Usage guides as a text type. In J. C. Beal, M. Lukač, & R. Straaijer (Eds.), The Routledge handbook of linguistic prescriptivism (pp. 159–174). Routledge.
Trask, R. L. (2006). Mind the gaffe: A troubleshooter’s guide to English style and usage. HarperCollins.
Wachal, R. S. (2000). Review of two handbooks on style and usage. American Speech, 75(2), 199–207.
Weiner, E. (1988). On editing a usage guide. In E. G. Stanley & T. F. Hoad (Eds.), Words: For Robert Burchfield’s sixty-fifth birthday. D. S. Brewer.
Yáñez-Bouza, N. (2015). Grammar, rhetoric and usage in English: Preposition placement 1500–1900. Cambridge University Press.
About the Author
Dr. Jordan Smith is an assistant professor in the department of Technical Communication at the University of North Texas. He earned his PhD in 2019 from Iowa State University, co-majoring in Rhetoric and Professional Communication as well as Applied Linguistics and Technology. Smith also holds a master’s in professional communication from Weber State University and a bachelor’s in English language from Brigham Young University. His research interests include corpus linguistics, register variation, English grammar and usage, and technical editing, focusing on linguistic variation in different writing styles. His work has appeared in Technical Communication Quarterly, IEEE Transactions on Professional Communication, International Journal of Business Communication, and Journal of Business and Technical Communication. He can be reached at jordan.smith2@unt.edu.
1 In his critique of The Elements of Style, Pullum (2010) acknowledged its popularity and influence by pointing to its high sales numbers and the fact that “many college-educated Americans revere [it], swear by it, carry it around with them” (p. 34). It continues to sell well with the 24-year-old fourth edition occupying the number eight slot on amazon.com’s best-seller list for grammar reference books as of September 9, 2023. While Merriam-Webster’s Dictionary of English Usage has likely sold fewer copies than The Elements of Style, its influence is still beyond doubt. Wachal (2000) called it “the gold standard” “for copy editors and others interested in usage as defined by what regularly gets into print” (p. 207).
2 Coding eight usage problems from 11 usage guides would yield 88 total ratings. However, Batko (2004) did not address the SPLIT INFINITIVE, and Fogarty (2008) did not address the NONE IN PLURAL CONTEXT usage problem. Therefore, there are 86 total ratings.
