Corpus Linguistics and Technical Editing: How Corpora Can Help Copy Editors Adopt a Rhetorical View of Prescriptive Usage Rules

Abstract

Scholars have long argued that technical editing should be viewed as a rhetorical practice in which copy editors take “a situational approach to each individual task” (Buehler, 1980/2003, p. 458). Yet many editing pedagogies still treat some language-level editing tasks, like those that involve prescriptive usage rules, as mechanical rather than rhetorical. This article discusses how empirical data from corpora can help copy editors adopt a more rhetorical view of prescriptive usage rules and introduces corpus linguistics as a methodology that can contribute to technical editing pedagogy.

Keywords

technical editing prescriptivism usage corpus linguistics pedagogy

Modern technical editors do more than correct mistakes in other people's writing (Lang & Palmer, 2017), but in many standard technical editing pedagogies, teaching prescriptive language rules remains a core part of the curriculum. In such curricula, students learn rules that label the use of certain linguistic variants as “correct” and others as “incorrect.” Reference works that copy editors use, such as style manuals and usage guides, reinforce many of these rules. For example, both the Chicago Manual of Style (2017) and the Publication Manual of the American Psychological Association (2020) strongly recommend maintaining a distinction between “that” as a restrictive relative pronoun and “which” as a nonrestrictive relative pronoun, a distinction that written Standard English has been shown to increasingly favor over time (Hinrichs et al., 2015).

While an awareness of prescriptive usage rules is an important element of any technical copy editor's skill set, teachers of technical editing need to be careful not to teach these rules in a way that promotes uncritical adherence to them. More than 40 years ago, Miller (1979) argued that technical communication “should present mechanical rules and skills against a broader understanding of why and how to adjust or violate the rules” (p. 617). An increased focus on social justice and advocacy in technical communication scholarship (see Walton et al., 2019) suggests that, as a field, technical communication is moving toward clearly realizing technical communication as a “more humanistic and less coercive endeavor” (p. 611). Yet we sometimes still resort to teaching prescriptive usage rules as unrhetorical—despite calls to promote a more rhetorical approach to (technical) editing specifically (e.g., About – Conscious Style Guide, n.d.; Buehler, 1980/2003; Connatser, 2004).

Instead of unthinkingly adhering to a set of rules, technical communicators—including technical copy editors—should know the rules in order to make rhetorically informed choices about the kind of currency these rules carry with certain audiences (see Wolfe, 2009, p. 364). Such an understanding of different rules, I argue, requires more than simply looking them up in a style manual or usage guide. Instead, it requires empirically investigating the linguistic features associated with the rules using collections of actual language that represent the contexts (or registers) in which the language variety is used. The methodology of corpus linguistics is well suited to help copy editors with this task.

This article introduces corpus linguistics as a methodology that can contribute to technical editing pedagogy. Specifically, I describe how corpora can be used to help students make rhetorically sound decisions about whether and when to follow prescriptive usage rules. This approach of inviting copy editors to use corpora to inform their editorial decisions has been discussed in professional editing circles (e.g., Owen, 2016), but it has largely been neglected in technical editing classrooms. While corpora might not be as helpful for copy editors during developmental editing of a text, they can be quite helpful during copyediting, or language-level editing, the stage in which copy editors review a text for grammar, syntax, and usage issues (Van Buren & Buehler, 1980).

I begin by discussing the benefits of taking a rhetorical approach to technical copyediting, echoing calls from others—both scholars and practitioners—who have advocated for this position long before now. Then I describe the ways that empirical data, like those derived from a corpus, can help copy editors to view their work through a rhetorical lens. I supplement this discussion with two examples from my own experience working as a copywriter and editor for a dental manufacturing company. Finally, I conclude with three sample case studies that I developed for a graduate course in technical editing. Each case study defines a scenario and asks students to imagine themselves as a technical editor (or technical communication major) in three different settings. I present the prompt given to students for each scenario, and I describe what students find as they use corpora to investigate the questions. In each case, I note the particular value that corpora provide—a value that simple adherence to a style manual lacks. In this way, students can practice thinking critically about the ways they edit, and they can use data from corpora to inform the decisions they make.

Why Copy Editors Should Adopt a Rhetorical Approach to Their Work

Scholars, practitioners, and professional editing organizations have long acknowledged the importance of viewing editorial work as rhetorical. Van Buren and Buehler (1980) suggested that language-level editing “goes beyond the mere application of grammatical or syntactical rules” (p. 23). Buehler (1980/2003) elaborated this concept further by distinguishing between a “rhetorical” and a “programmatic” approach to editing, advocating that technical editors should prefer the rhetorical approach.

Rhetorical Approaches to Technical Editing

In Buehler's rhetorical model, the editor makes editorial decisions based on the rhetorical situation of the communicative event. In the programmatic model, the editor “simply applies a set of rules to all situations…without concern for all the varied elements of the situation itself” (p. 459). Similarly, Connatser (2004) contrasted prescriptive grammar with what he called “organic grammar,” or the grammar that is naturally acquired as children are exposed to linguistic input (p. 265). When organic grammar conflicts with prescriptive grammar, Connatser suggested breaking the prescriptive rules.

Other scholars have called for a similar emphasis on the rhetorical underpinnings of language use in general. Gorrell (1977) argued that “usage is not a question of grammar or linguistics or morality, but of rhetoric” (p. 20) and that “usage labels or other facts about language…contribute useful knowledge to any rhetorical choice, but do not automatically make the choice” (p. 23). For technical copy editors, then, considering the rhetorical situation allows them to confidently make choices about what is appropriate even when those choices do not align with prescriptive usage rules. In other words, the prescriptions outlined in usage guides and style manuals should not be taken as inflexible rules, never to be disobeyed—a point that even the Chicago Manual of Style (2010) and the Publication Manual of the American Psychological Association (2010) make in their introductory material. Writers and copy editors must consider the rhetorical situation within which the communicative event takes place so that they can then make the choices that they feel will best meet a given exigency for a given audience within given constraints (see Bitzer, 1968/2010).

Practicing editors and professional editing organizations have also recognized the importance of viewing editing as a rhetorical (rather than mechanical) task. Pope (2013), a Seattle-based technical editor and blogger, called editing a “pragmatic” (para. 3) exercise and recognized that rules should not be mechanically applied to all texts at all times. Another well-known editor and blogger, Owen (2020), often posts about the need to critically consider usage advice. For instance, he recently posted about his own wrestle with the changing meaning of “as such” from a prepositional phrase to a transitional adverb and thoughtfully discussed “what makes a usage right or wrong and how we as editors decide which rules to enforce and which ones to let slide” (para. 2). The actions of professional editing organizations support the views of Pope and Owen—and other like-minded practitioners. For example, the websites for both the Council of Science Editors and ACES: The Society for Editing provide resources for editors to learn about diversity, equity, and inclusion (DEI Scholarly Resources, n.d.; Diversity & Inclusion Resources, n.d.), which can help editors better understand the rhetorical situations in which they edit. Similarly, in its Professional Editing Standards, the Editors’ Association of Canada (2016) acknowledged that editorial decisions are not based on abstract rules of correctness but instead vary by audience and situation (see pp. 5, 13).

Those examples, I admit, are anecdotal. I have not carried out a systematic review of a representative sample of practicing editors and professional organizations. But the views summarized by these examples suggest a wider held belief of practicing editors and professional editing organizations: The task of copyediting is ultimately rhetorical.

Implications for Social Justice

Taking a rhetorical approach to understanding prescriptive language rules also helps copy editors to avoid advancing any social problems associated with the rules. Children do not naturally follow the prescriptive rules associated with programmatic (Buehler, 1980/2003) and prescriptive (Connatser, 2004) grammars when they acquire their first language. Instead, these rules must be explicitly taught. Bourdieu (1991) highlighted this fact when he described the concept of a prescriptive usage rule as a

correct, i.e. corrected, expression [that] owes the essential part of its social properties to the fact that it can be produced only by speakers possessing practical mastery of scholarly rules, explicitly constituted by a process of codification and expressly inculcated through pedagogic work. (p. 61)

Because prescriptive language rules are learned not through natural processes of language acquisition but through conscious pedagogical instruction, they covertly operate within a frame of classism, helping to sustain existing structures of social power by benefiting those who grew up in environments in which they were able to learn these rules from their parents or in school. Ebner (2017) emphasized the social nature of usage problems (a technical term referring to the sets of linguistic variants that prescriptive usage rules try to govern, see Tieken-Boon van Ostade, 2019) and the “divisive” (p. 7) role that they carry out in a society. Similarly, Chrisomalis (2015), summarizing the work of other researchers, stated that general prescriptivism “is used to index social status and moral propriety through appeals to tradition and authority within a national or even international linguistic context” (p. 68).

Many usage problems are long-standing shibboleths used by groups in power to suppress people in populations who acquired a language variety different from their own, and English manifests such shibboleths as well. Maintaining and perpetuating these shibboleths has social consequences, and technical editors need to be aware of these concepts so that they can consciously determine how they might work to change them. Of course, technical editors need not disregard all conventions of Standard English as a display of social protest, but they should be able to understand and critically consider the language rules they enforce, including (maybe even especially) the usage problems they often encounter in their work. Because “correctness,” when discussed in the context of usage problems, is ultimately an arbitrary construct (see Williams, 2005, chap. 2, for an excellent and practical discussion on correctness), I believe that editors should deviate from any usage rule if they have a reason for doing so.

Some usage problems are overtly linked to problematic and harmful ideologies. For example, the use of “he” as a gender-neutral singular pronoun is widely discussed within the framework of sexism and LGBTQ + rights because it clearly erases those who do not use this pronoun. To combat this issue, the use of “they” as an indefinite singular pronoun and a pronoun for people who do not use “he/him” or “she/her” pronouns is generally accepted across style guides now. Technical editors must understand and consider these issues 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. In fact, taking a rhetorical approach to editing can help editors work more successfully with authors. For example, Eaton et al. (2008) found that some of the authors they surveyed did not like working with editors “who were too stringent in following style guides” (p. 126).

To address the social harms that can be perpetuated when copy editors do not apply a rhetorical lens to their work, practicing editors have recognized the need for contextual approaches to copyediting. For instance, Saller (2017) noted that “checking even the simplest document for grammar, spelling, and style requires knowledge, restraint, and editorial judgment,” adding that “in copyediting, the right answer is often a surprising ‘That depends’” (p. 108). Scholars have also posed potential solutions. Popham (2019) advocated an approach to editing informed by feminist theory. This approach “include[s] an awareness of the role that emotion and empathy can play in editing, that is, a distinct turn away from a dispassionate, impartial façade” (p. 102). By acknowledging the emotions some prescriptive usage rules invoke—and not seeing the rules as simply dispassionate and logical—teachers of technical editing can encourage students to apply critical lenses in their own editorial work. Teachers can apply Popham's calls for feminist methods in technical editing courses when they address usage problems in the classroom. Specifically, they can ask students to think critically about the power structures that specific usage problems uphold in societies and to acknowledge the ways that uncritically following arbitrary rules of Standard English could harm and belittle marginalized groups. In sum, by applying socially just theories to editing pedagogies, teachers can help students understand that simply following a set of rules outlined in a style guide is not a neutral exercise.

Viewing technical editing as a rhetorical practice can help editors recognize implications for social justice in their work, but how might they go about adopting this viewpoint? Flanagan (2019) notes that a rhetorical approach to editing is “suitable for editing students who have developed editorial judgment and confidence in their editorial decisions” (p. 29). Similarly, Cunningham et al. (2019) advised editors to become what they called “informed prescriptivists”:

A technical editor works with rules (or at least guidelines) and therefore must be a prescriptivist, but you may have some say in which rules you follow and expect others to follow. You should be a sensible and informed prescriptivist. (p. 328)

For technical editors to develop their editorial judgment and become “informed prescriptivists,” they must have information about the exigencies, audiences, and constraints of a text's situation in order to accurately gauge what will be considered appropriate. These elements all relate to the rhetorical situation as Bitzer (1968/2010) defined it. I argue that one additional element that can help technical editors become informed prescriptivists is empirical data. Empirical research on prescriptive usage problems encourages a more critical examination of usage rules that technical editors encounter regularly in their work, and it can help technical editors understand that not all prescriptive language rules contained in style manuals need to be observed in all cases. Empirical research can help pinpoint such cases.

Empirical Data From Corpora Can Support a Rhetorical Approach to Technical Editing

Thus far, I have outlined some benefits of technical editors’ taking a rhetorical approach to their work and suggested that empirical data can complement a rhetorical approach to technical editing. Here, I describe how copy editors can use data from corpora to make empirically informed decisions in their work.

Technical editing teachers and students—as well as practitioners—can learn to use general corpora of the English language to study usage questions that might arise. Corpora are machine-readable samples of language collected to represent a linguistic register (or registers). Many studies in technical communication have used corpora (e.g., Boettger & Wulff, 2014; Conrad, 2018; Lambrecht, 2021; Zhang et al., 2020). But studies that have used corpora in technical editing are rare. One exception is Boettger (2016), who asked his graduate-level technical editing students to build their own corpora to learn about writing variation. He found that students generally enjoyed using corpora in his technical editing class and “somewhat agreed” (p. 9) that being exposed to corpora in a technical editing course would help them in the professional editing work they conducted.

Corpus linguistics is a branch of applied linguistics that uses corpora to study language structure and use (see Biber et al., 1998). The corpora that corpus linguists use to study language come in many different sizes, from small, specialized corpora of less than 100,000 words to corpora containing billions of words. Some corpora are freely available and easy to search. One of the most widely used general corpora of American English is Davies’s (2008) Corpus of Contemporary American English (COCA). COCA is composed of 1 billion words balanced among eight different registers or genres: newspapers, academic texts, fiction, spoken, magazines, subtitles (from TV and movies), blogs, and other web texts. Each register is further divided into more specific subregisters (e.g., academic texts from specific disciplines and news texts from specific sections of a newspaper). Users can search the entire corpus at once, or they can refine their searches to specific registers or subregisters within the corpus. Each word in COCA is tagged for its part of speech, allowing users to conduct powerful searches not just for certain words but also for more general grammatical constructions. In addition to COCA, other corpora are available on the english-corpora.org website that represent other registers. Users can find corpora composed of web registers, online news, British English, movie scripts, and more. Table 1 highlights some of the corpora available for free on english-corpora.org that teachers of technical editing might find useful in the classroom.

Table 1.

Selected Corpora on english-corpora.org, Size and Makeup of Each Corpus, and Respective Applications for Technical Editing.

Name	Corpus Size and Makeup	Applications for Technical Editing
Corpus of Contemporary American English (COCA)	1 billion words of modern American English (1990–2019) from eight genres: spoken, fiction, magazines, newspapers, academic texts, TV and movie subtitles, blogs, and other web texts	Users can quickly get a sense of general, edited English when searching all genres except spoken, subtitles, blogs, and web texts. Users can also search in specific texts (e.g., technically oriented magazines or academic texts) to get targeted results.
iWeb: The Intelligent Web-Based Corpus	14 billion words of English-language text taken from 22 million web pages	Technical editors working with internet language will find the iWeb corpus useful because its large size makes searching for rare linguistic features possible.
Wikipedia Corpus	1.9 billion words from 4.4 million English-language Wikipedia articles	Users can create virtual corpora (i.e., smaller, customized corpora drawn from certain texts in the full corpus) based on technical topics to easily study usage in narrower topical domains. Creating virtual corpora is possible for any of the corpora in this table, but for technical editors using the Wikipedia Corpus, it can be especially useful.
News on the Web (NOW)	16 + billion words from English-language news sources across the world with new content added continually	News writing is typically not seen as technical communication, but the size of the NOW corpus makes it useful to technical editors who want to study rare lexical items. It is also one of the only corpora on the English-corpora site to be continually updated, ensuring the corpus always reflects current language use.
Global Web-Based English (GloWbE)	1.9 billion words of English-language text on the web from 20 different countries	Because the large GloWbE corpus includes language from many different countries, it can be useful for technical editors working with varieties of English other than American English.
Corpus of Online Registers of English (CORE)	50 million words of English language text on the web	While CORE is much smaller than any other corpus in this table, its texts have been carefully categorized into genres, making it especially useful for genre-specific research.

While COCA and other publicly available corpora are quite user friendly, they do require some training to use properly. Of course, users should review the documentation provided by the corpus administrator to learn about how and why the corpus was built, what register(s) it is intended to represent, and how to use it. Once users are familiar with a corpus, they might wonder how specifically to use it to help them with editorial work.

Examples of Using Corpora in the Workplace

To help illustrate how copy editors can use corpora to answer usage questions that arise, I share the following examples from my own experience working as a copywriter and editor for a dental manufacturing company. In both examples, I describe an editing-related question that arose and discuss how corpora can be used to answer it. The language highlighted in both examples appeared in various marketing and technical genres—promotional mailers, catalogs, technical brochures, and informational web copy—intended for dentists, hygienists, and other dental professionals.

Example 1: “Aesthetic” Versus “Esthetic” in Dentistry-Related Texts

The first example involves the variant spellings “aesthetic” and “esthetic” in dentistry-related texts. Dictionaries list both spellings, indicating that both are generally accepted, but usage guides (e.g., Garner, 2016) suggest that writers prefer the spelling “aesthetic.” In addition, corpus data from COCA suggest that the variant “aesthetic” is more than 20 times more common than “esthetic.” But the dental company I worked for adopted “esthetic” as the preferred variant because, I was told, it was more common in dentistry-related texts. Rather than uncritically adopting this less common spelling, copy editors in a similar position to mine could use corpora to study this domain-specific question (i.e., whether “aesthetic” or “esthetic” is the preferred spelling in dentistry-related contexts). Indeed, one major advantage that corpora provide editing students, teachers, and practitioners is this ability to study domain-specific usage questions like this one. Copy editors could then use the results from the corpus to determine which spelling would be more rhetorically appropriate.

To ensure that the results returned by the corpus came from other dentistry-related texts, I included the word “dentistry” in my searches (i.e., “aesthetic dentistry” and “esthetic dentistry”). Neither phrase appeared in COCA, but both appeared in the larger, nearly 14-billion-word iWeb corpus (Davies, 2018). Having access to extensive corpora such as iWeb makes researching rare phrases such as “aesthetic dentistry” and “esthetic dentistry” possible because the large size of the corpus increases the likelihood that these words will appear at frequencies high enough to make data-driven decisions. I found that the phrase “aesthetic dentistry” appeared 136 times and the phrase “esthetic dentistry” appeared 179 times (as of June 20, 2022). Users of the corpus can view the words in context to see their sources. As Figure 1 illustrates, many of the sources appear to come from websites within the dentistry discourse domain. Some have the word “dental” or “dentist” directly in the URL (e.g., 1800dentist and dentalproductsreport.com). And several of the results come from agd.org, the URL for the American Academy of General Dentistry website.

Figure 1.

Sample concordance lines from the iWeb corpus (Davies, 2018) when “esthetic dentistry” is searched. The second column in the output shows the source for each line. Many of these sources are dentistry related, suggesting that the spelling “esthetic” is not uncommon in dental texts.

While the number of instances for both terms is relatively small, and the difference between the two frequencies is not dramatic, the results of this quick corpus search indicate that the spelling “esthetic” does appear in dentistry-related websites and at a frequency just slightly higher than the otherwise far more common “aesthetic.” Equipped with this information, then, copy editors working on dentistry-related documents can feel confident that adopting the spelling “esthetic” is not uncommon within that domain.

Example 2: “Light Cure” Versus “Light Cured” as a Compound Modifier in Dentistry-Related Texts

The second example involves the compound modifier “light cure” or “light cured.” This modifier is common in writing about dental products because many of the materials that dentists use to treat oral disease need to be cured with a specialty light in order to polymerize and become semipermanent. Copy editors working in the dentistry field may question whether “light cure” or “light cured” is the preferred variant because both are used. This question arose early on in my experience as a dental copywriter and editor, and I found that there were passionate preferences for either variant. Unfortunately, no general language usage guide is likely to have an entry that answers this question. Style manuals may offer general guidance on whether to hyphenate compounds composed of a noun + base-form verb (i.e., “light cure”) or a noun + past participle (i.e., “light cured”), but they would not say which combination is preferred in dentistry-related texts.

While the iWeb corpus is useful for answering the question regarding whether to use “aesthetic” or “esthetic,” it is less helpful in this case. The phrase “light cure” followed by any noun appeared only 58 times; “light cured” followed by a noun appeared only 39 times. And while some nouns following either phrase appeared to be related to dentistry, others were not (e.g., “nails,” “glass,” and “acrylics”), so ensuring that the results come from dentistry-related texts is difficult.

In cases like this—in which questions related to the use of a specific linguistic feature in a specific context arise—a small, register-specific corpus can be valuable. Building a small, ad hoc corpus of just a few texts is relatively easy and can be useful in classroom settings where time constraints favor fast, efficient projects. But building a more comprehensive corpus is not such a simple task. Care needs to be taken to ensure that the texts included in a corpus are systematically selected and representative of some register or variety of language. Once the texts are selected and gathered, they must be converted to plain-text format and cleaned to fix any character or formatting issues that may have resulted from the conversion. They might also be tagged for parts of speech to allow for the more powerful searches that corpora like COCA allow. Some taggers, such as TagAnt (Anthony, 2021) and LancsBox (Brezina et al., 2020), are convenient and free to use. Other taggers, like CLAWS (Garside, 1987), can be purchased for a fee.

Given the amount of work needed to construct a comprehensive corpus, it would not be practical for technical editing teachers to require their students to build one during a single semester. Similarly, it may not be practical for freelance editors working on their first dentistry-related article to build a full corpus of dentistry-related material. It would, however, be practical for a dental manufacturing company with an in-house communications team to build such a corpus that could be referenced when domain-specific language issues arise. This corpus could comprise academic articles from dental journals, web texts from a variety of leading dental companies or clinics, and sections of popular dental textbooks. Copywriters, technical writers, and technical editors could look to the corpus for any issues that are not already covered in the company's chosen style guide. Then when decisions are made based on corpus data, the communications team could amend the style guide accordingly.

Corpora—both those that are freely available, such as COCA and iWeb, and those that are built for register-specific reasons—are valuable tools to help copy editors answer usage questions that arise in their work. They provide data to support the decisions that copy editors make, which is particularly useful if those decisions are challenged by others within the organization.

These two examples are particularly relevant for workplace settings in which copy editors must use or build a corpus to get the empirical data they need to answer language questions. Technical editing teachers can use these examples as pedagogical springboards to teach students the basics of searching corpora and how to use the results of those searches to inform editorial decisions. I will now provide three case studies that I designed specifically for technical editing teachers to use in their classrooms in order to teach students how to use corpora to answer their usage-related questions.

Pedagogical Case Studies

The three case studies I describe here ask students to imagine themselves as copy editors and to edit within specific contexts. Emphasizing the importance of asking students to edit within specific rhetorical situations, Bridgeford (2019) recommended that technical editing instructors “create a context within which students will be working and completing assignments, as well as using to begin imagining themselves as editors.” These contexts, Bridgeford continued, should also be able to “be read quickly and understood easily” (p. 84). In this way, case studies can be useful for introducing technical editing students to corpora as editing tools.

The following case studies describe different situations that technical editors might encounter in the workplace. In each case, the use of a corpus can help answer the questions that arise. Instructors can invite students to work with a partner to complete the task associated with each case study. For each case, I first note the corpus features that students practiced using, then I present the editing scenario, the specific task that students are asked to complete, and a brief discussion of the expected outcomes.

Case Study 1: “E-Mail” Versus “Email”

In Case Study 1, students practiced the corpus features of filtering results by register and viewing usage over time.

Scenario. You’re a technical editor working at Acme Investments, a financial firm founded in 1953. The firm has a reputation of being slow to accept change and feels a strong need to maintain a highly polished, professional tone in their communication and documentation. You run across the following sentence in some document you’re editing: “You will receive an e-mail when the password has been reset.” You check your company's style guide to see if “e-mail” should be written with the hyphen or as a closed compound (i.e., “email”). Sure enough, the style guide states that “e-mail” is preferred. But you also notice that the last revision of the style guide was done in 2006. You want to propose that the company adopt “email” as the preferred form because you have a sense that you’ve seen it written that way in a lot of other documents you have read, and you feel that continuing to use the hyphenated form will make the firm seem old-fashioned and stodgy.

Task. Use COCA to research the use of “e-mail” versus “email.” Use the findings from your research to build an argument to upper management (who you know will be hard to persuade without convincing evidence) that the company should adopt “email.” Write 300–400 words and include screenshots of COCA in your write-up.

Discussion. Students are given information about the context of the scenario, including the industry and the attitudes of the audience they need to convince, so that, equipped with this information, they can do more than just a simple frequency search for “email” versus “e-mail.” Using COCA's filtering tools, they can refine their search to look only at language in subregisters whose texts have similar situational characteristics (e.g., financial magazines, money-related news, and business-related academic articles) in order to give them a sense of how this word is used in the specific language domain in which they are writing and editing.

When carrying out this search in these three subregisters, students find that “email” appears 582 times while “e-mail” appears 1,752 times. These results seem to favor keeping the current style recommendation of the hyphenated “e-mail.” But a general search for both terms in COCA using the Chart feature shows a marked increase in the use of “email” between 1990 and 2019 and a marked decrease in the use of “e-mail” from 2000 to 2019. So while the use of “e-mail” may be more common in financial texts, students can still make the argument—supported with data—that this use is declining in general English and should be jettisoned in favor of “email.”

Case Study 2: “A Historic” Versus “an Historic”

In Case Study 2, students practiced the corpus features of filtering results by register and exploring variation across academic disciplines.

Scenario. You’re a technical communication major with an interest in technical editing. Your friend is thinking of declaring a history major and is taking her first history class. Given your expertise in technical editing, she asks you to look over her first paper before she turns it in, and you agree. While editing the paper, you come across the following sentence: “Washington's crossing of the Delaware was a historic event and is commemorated in Emanuel Leutze's famous 1851 painting.” You pause after reading the sentence because you wonder if “a historic” should be changed to “an historic.” You have heard people say “an historic” before, but you’re not sure if this usage is accepted (or even expected) and, if so, in what contexts. You would like to know which usage is common in history writing so that you can help your friend fit in with her new academic group.

Task. Use COCA to research the use of “a historic” versus “an historic” in academic writing. Determine which usage is more common in which disciplines. Write 300–400 words describing your research methods and findings. Use screenshots in your write-up.

Discussion. This scenario asks students to compare the use of two phrases in a specific register (academic texts) and across subregisters (individual disciplines within academic texts). Again, using COCA's filtering options, students can search for the two phrases in different disciplines and see frequencies to determine which variant is more common in which discipline. Because students were free to explore a variety of disciplines, I have not provided specific results here.

Case Study 3: “Said” Used as an Adjective Modifying a Noun

In Case Study 3, students practiced the corpus features of searching using part-of-speech tags and reviewing results for false positives.

Scenario. You’re a technical editor working for an up-and-coming tech start-up in the Silicon Hills area of Austin. The atmosphere at the office is very casual, and the type of marketing the company produces is creative and fun—almost goofy. On the whole, the company doesn't appear to take itself too seriously. You’re editing some documentation for a new piece of hardware the company is launching, and you see these sentences: “The labels can be reapplied with the adhesive included in the starter pack. Apply said adhesive to the outer casing of the box, and let it stand for 20 minutes to fully cure.” The word “said” catches your eye. You’ve seen it used like this—as a noun modifier—before, but it strikes you as being particularly formal—like something you’d see in an academic paper or a legal brief. You’re questioning whether it fits the casual, goofy voice and tone of the company you work for, so you decide to do some investigating.

Task. Use COCA to research the use of “said” as a noun modifier. Be careful to restrict your searches to its use as an adjective and not its use as a verb. What do you learn about this use of “said?” Were your intuitions right that it's mostly used in formal writing? Would you leave the sentence as is, or would you change it to something else? Write 300–400 words describing your methods and findings. Use screenshots in your write-up.

Discussion. In this scenario, students practice searching COCA using patterns rather than just words or phrases. Because “said” is most often used as a verb, students can alter their search to return only those instances of “said” that have been tagged as an adjective by the part-of-speech tagger, followed by any noun. (Typing “said_jj* NOUN” into COCA's search bar returns instances of “said” tagged as an adjective followed by any noun.) Then students can analyze the registers in which this use of “said” is common, the kinds of nouns that it typically accompanies, and the frequencies of use across registers in order to make a decision about whether this use of the word in the context described in the scenario is appropriate.

Students will also see that some of the results returned by corpus searches are false positives. For example, one of the results when using the search syntax “said_jj* NOUN” is “said state.” Reviewing the concordance lines for this result shows that in some of the results, “said” is used as a reporting verb (e.g., “…, said state Rep. Steve Reick”) and should therefore be ignored. No corpus is perfect, so instructors should use this exercise as an opportunity to teach their students to scrutinize their search results to make sure they get accurate information.

These three scenarios are relatively benign in that they do not ask students to engage with usage problems that are overtly demeaning or discriminatory. But they still invite students to think critically about the language use they encounter, and they show them how corpora can help answer questions that arise about specific linguistic features in specific linguistic domains.

Conclusion

I have argued here that adopting a linguistically informed approach to technical editing—specifically using corpus-linguistic methods—can help technical copy editors view their work rhetorically. In conclusion, I respond to three obstacles that might discourage some instructors and editors from implementing corpus-linguistic methods in their work.

The first obstacle concerns research about how valuable technical editors find linguistic instruction. Kreth and Bowen (2017) reported that the technical editors they surveyed considered linguistics courses less helpful in their career preparation than courses in technical writing, journalism, and even nonfiction writing. But linguistics is a broad field, and some courses (e.g., grammar or corpus-linguistic methods) are likely more relevant to a career in editing than are others (e.g., phonology). Kreth and Bowen do not report the kinds of linguistic courses their respondents took, so we do not know the reason that they felt linguistics courses were less helpful than others. In addition, the mean score for linguistics courses in Kreth and Bowen's study suggested that respondents still saw them as (somewhat) helpful. As a result, I believe that corpus-linguistic instruction can still be a highly promising pedagogical tool in technical editing classrooms.

The second obstacle to introducing corpus-linguistic methods into technical editing work concerns efficiency. Technical editors must be efficient in their work, and asking technical editors to add any extra steps to their workflow might cause some hesitation. Although it may take some extra time, the corpus-informed approach I have described can be efficiently implemented into professional technical editing workflows for the following reasons:

Many corpus searches—especially those using already-available corpora such as those described in Table 1—take little time. Once users are familiar with the user interface for a given corpus, they can conduct searches in minutes.

Corpus searches do not need to be conducted for every usage question. Using their rhetorical judgment, editors can choose which usage questions merit such extra attention.

For any given usage problem, corpus research is infrequently needed. When editors do use corpus research to answer usage questions, they should record their decisions in a company style guide or style sheet so that they do not need to do the research again until circumstances inside or outside the organization change. In fact, editors (and whole technical communication teams) should plan to carry out such searches when an organization's style guide is up for review.

Therefore, although the task of using corpora to research individual usage problems may seem relevant for classroom instruction but not for professional settings, corpus-linguistic methods can be efficiently incorporated into the work that practicing technical editors do. In fact, professional editing organizations are teaching practicing editors how to use corpora to study usage. For instance, several recent national meetings of ACES: The Society for Editing have included a panel on corpus linguistics and editing led by professional editor Jonathon Owen (e.g., Owen, 2016).

The third obstacle to implementing corpus-linguistic methods in technical editing work involves editors’ perceptions of their own power. While some technical editors recognize the need to deviate from style advice,¹ not all technical editors may feel that they have the authority to violate usage rules for rhetorical purposes. But as Graves and Graves (1998) argued, technical communicators play an important ethical role in ensuring that the language used in technical documents does not promote offensive or harmful social problems, such as racism, classism, or sexism—even when such language falls in line with approved guidelines:

Editors, writers, and researchers of technical communication are in a unique position to use their expertise not only to produce top quality documents, but also to examine and raise for discussion those linguistic constructions and conventions that portray reality in questionable ways. (p. 412)

My primary pedagogical goal here has been to encourage technical editing instructors to teach students how to use empirical data from corpora to answer usage questions. Doing so, I have argued, is crucial toward helping editing teachers, students, and practitioners adopt a more rhetorical view of editorial work, as other scholars have called for (Buehler, 1980/2003; Connatser, 2004). Teaching students how to use corpora, such as COCA, iWeb, and others that are freely available, gives them the skills to be able to research usage questions within specific linguistic domains. That ability to account for specific rhetorical situations in this way is in turn important for copy editors in order to more fully adopt a rhetorical view of editing.

Footnotes

Acknowledgments

I would like to thank Chris Lam for his comments on a draft of this article. I would also like to thank Jonathon Owen, whose presentation materials on corpus linguistics and copyediting at the meetings of the American Copy Editors Society were influential. Much of this article was developed as part of my dissertation work. I would like to thank Jo Mackiewicz, Bethany Gray, and the other members of my committee for their insights and thoughtful comments.

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Jordan Smith

Notes

Author Biography

Jordan Smith is an assistant professor in the Department of Technical Communication at the University of North Texas. His research interests lie at the intersection of technical communication and applied linguistics.

References

About—Conscious Style Guide. (n.d.). https://consciousstyleguide.com/about/

Anthony

(2021). TagAnt (Version 1.3.0) [Computer software]. Waseda University. https://www.laurenceanthony.net/software

Biber

Conrad

Reppen

(1998). Corpus linguistics: Investigating language structure and use. Cambridge University Press.

Bitzer

L. F.

(2010). The rhetorical situation. In Burgchardt

C. R.

(Ed.), Readings in rhetorical criticism (4th ed., pp. 46–55). Strata. (Original work published 1968).

Boettger

R. K.

(2016). Using corpus-based instruction to explore writing variation across the disciplines: A case history in a graduate-level technical editing course. Across the Disciplines, 13(1), 1–21. https://doi.org/10.37514/ATD-J.2016.13.1.03

Boettger

R. K.

Wulff

(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

Bourdieu

(1991). Language and symbolic power. Polity Press.

Brezina

Weill-Tessier

McEnery

(2020). #LancsBox (Version 6.0.0) [Computer software]. Lancaster University. http://corpora.lancs.ac.uk/lancsbox

Bridgeford

(2019). Imagination as agency: Communities of practice and editing pedagogy. In Flanagan

Albers

M. J.

(Eds.), Editing in the modern classroom (pp. 66–90). Routledge.

10.

Buehler

M. F.

(2003). Situational editing: A rhetorical approach for the technical editor. Technical Communication, 50(4), 458–464. (Original work published 1980).

11.

Chicago manual of style . (2010). (16th ed.). University of Chicago Press.

12.

Chicago manual of style . (2017). (17th ed.). University of Chicago Press.

13.

Chrisomalis

(2015). What’s so improper about fractions? Prescriptivism and language socialization at Math Corps. Language in Society, 44, 63–85. https://doi.org/https://doi.org/10.1017/S0047404514000748

14.

Connatser

B. R.

(2004). Reconsidering some prescriptive rules of grammar and composition. Technical Communication, 51(2), 264–275.

15.

Conrad

(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

16.

Cunningham

D. H.

Malone

E. A.

Rothschild

J. M.

(2019). Technical editing. Oxford University Press.

17.

Davies

(2008). The Corpus of Contemporary American English (COCA). https://www.english-corpora.org/coca/

18.

Davies

(2018). The iWeb Corpus. https://www.english-corpora.org/iWeb/

19.

DEI scholarly resources. (n.d.). https://www.councilscienceeditors.org/resource-library/diversity-equity-and-inclusion-resources/

20.

Diversity & inclusion resources . (n.d.). https://aceseditors.org/resources/diversity-inclusion-resources

21.

Eaton

Brewer

P. E.

Portewig

T. C.

Davidson

C. R.

(2008). Examining editing in the workplace from the author’s point of view: Results of an online survey. Technical Communication, 55(2), 111–139.

22.

Ebner

(2017). Proper English usage: A sociolinguistic investigation of attitudes towards usage problems in British English. Leiden University.

23.

Editors’ Association of Canada (2016). Professional editing standards. https://www.editors.ca/sites/default/files/pes-2016-formatted-for-web_0.pdf

24.

Flanagan

(2019). The current state of technical editing research and open questions. In Flanagan

Albers

M. J.

(Eds.), Editing in the modern classroom (pp. 15–46). Routledge.

25.

Garner

B. A.

(2016). Garner’s modern English usage (4th ed.). Oxford University Press.

26.

Garside

(1987). The CLAWS word-tagging system. In Garside

Leech

Sampson

(Eds.), The computational analysis of English: A corpus-based approach (pp. 30–41). Longman. http://ucrel.lancs.ac.uk/papers/ClawsWordTaggingSystem RG87.pdf

27.

Gorrell

R. M.

(1977). Usage as rhetoric. College Composition and Communication, 28(1), 20–25. https://doi.org/10.2307/356888

28.

Graves

H. B.

Graves

(1998). Masters, slaves, and infant mortality: Language challenges for technical editing. Technical Communication Quarterly, 7(4), 389–414. https://doi.org/10.1080/10572259809364639

29.

Hinrichs

Szmrecsanyi

Bohmann

(2015). Which-hunting and the Standard English relative clause. Language, 91(4), 806–836. https://doi.org/10.1353/lan.2015.0062

30.

Kreth

M. L.

Bowen

(2017). A descriptive survey of technical editors. IEEE Transactions on Professional Communication, 60(3), 238–255. https://doi.org/10.1109/TPC.2017.2702039

31.

Lambrecht

(2021). Tracking the differentiation of risk: The impact of subject framing in CDC communication regarding COVID-19. Journal of Business and Technical Communication, 35(1), 94–100. https://doi.org/10.1177/1050651920958394

32.

Lang

Palmer

(2017). Reconceiving technical editing competencies for the 21st century: Reconciling employer needs with curricular mandates. Technical Communication, 64(4), 297–309.

33.

Miller

C. R.

(1979). A humanistic rationale for technical writing. College English, 40(6), 610–617. https://doi.org/10.2307/375964

34.

Owen

(2016, April 2). Copyediting and corpus linguistics [Presentation]. ACES 20th Annual National Conference, Portland, OR. https://aceseditors.org/assets/files/pdfs/2016/Copyediting-and-Corpus-Linguistics.pdf

35.

Owen

(2020, August 27). Right, wrong, and relative [Blog]. Arrant pedantry. https://www.arrantpedantry.com/2020/08/27/right-wrong-and-relative/

36.

Pope

(2013, April 2). Copyediting principles [Blog]. Mike’s web log. https://www.mikepope.com/blog/AddComment.aspx?blogid = 2401

37.

Popham

S. L.

(2019). Teaching editing through a feminist theoretical lens. In Flanagan

Albers

M. J.

(Eds.), Editing in the modern classroom (pp. 91–108). Routledge.

38.

Publication manual of the American Psychological Association (6th ed.). (2010). American Psychological Association.

39.

Publication manual of the American Psychological Association (7th ed.). (2020). American Psychological Association.

40.

Saller

C. F.

(2017). Toward accuracy, clarity, and consistency: What copyeditors do. In Ginna

(Ed.), What editors do: The art, craft, and business of book editing (pp. 106–115). University of Chicago Press.

41.

Tieken-Boon van Ostade

(2019). Describing prescriptivism: Usage guides and usage problems in British and American English. Routledge.

42.

Van Buren

Buehler

M. F.

(1980). The levels of edit (2nd ed.). Jet Propulsion Laboratory.

43.

Walton

Moore

K. R.

Jones

N. N.

(2019). Technical communication after the social justice turn: Building coalitions for action. Routledge.

44.

Williams

J. M.

(2005). Style: Ten lessons in clarity and grace (8th ed.). Pearson.

45.

Wolfe

(2009). How technical communication textbooks fail engineering students. Technical Communication Quarterly, 18(4), 351–375. https://doi.org/10.10.1080/10572250903149662

46.

Zhang

Gosselt

J. F.

de Jong

M. D. T.

(2020). How large information technology companies use Twitter: Arrangement of corporate accounts and characteristics of tweets. Journal of Business and Technical Communication, 34(4), 364–392. https://doi.org/10.1177/1050651920932191