Stylish Academic Writing - Helen Sword 2012

On being disciplined
Style and substance

discipline (n.)

• A branch of instruction or education; a department of learning or knowledge; a science or art in its educational aspect.

• The order maintained and observed among pupils, or other persons under control or command, such as soldiers, sailors, the inmates of a religious house, a prison, etc.

• Correction; chastisement; punishment inflicted by way of correction and training; in religious use, the mortification of the flesh by penance; also, in a more general sense, a beating or other infliction (humorously) assumed to be salutary to the recipient.¹

To enter an academic discipline is to become disciplined: trained to habits of order through corrections and chastisements that are “assumed to be salutary” by one’s teachers. Scholarly commentators have variously alluded to the academic disciplines as “silos,” “barricades,” “ghettos,” and “black boxes,” using metaphors of containment that implicitly critique the intellectual constraints imposed by disciplinary structures.² Yet disciplinarity remains a robust and even sacred concept. University of California chancellor Clark Kerr is said to have described the mid-twentieth-century research university as “a series of individual faculty entrepreneurs held together by a common grievance over parking,” and his censure still rings true six decades later: academics often seem more intent on fencing off and tending their own patches of disciplinary turf than on seeking common ground.³ Even within disciplines that appear relatively homogeneous to an outsider, scholars may belong to warring subdisciplinary clans that have established and entrenched separate identities marked by distinctive ideologies and idiolects. Sociologist Andrew Abbott compares the “fractal distinctions” between subdisciplines to segmental kinship systems: “A lineage starts, then splits, then splits again. Such systems have a number of important characteristics. For one thing, people know only their near kin well.”⁴

Recently, a colleague from my own university’s medical school told me that she had decided not to enroll in an interdisciplinary faculty development course because it would be “a waste of time” for her to learn about academic writing from anyone outside the medical profession. Her comment reminded me of a news story that I came across a few years ago involving an unlikely but productive collaboration between medical and nonmedical experts. In 2006, surgeons from the Great Ormond Street Children’s Hospital invited a team of Ferrari Formula One pit stop mechanics to observe them at work. The mechanics noted a number of inefficiencies in the surgeons’ procedures and recommended some key changes, particularly in the areas of synchronization, communication, and patient relocation. The doctors consequently developed new surgical protocols, forged new lines of communication with nurses and technicians, and even designed a new operating gurney to smooth their young patients’ transition between the operating room and intensive care. According to one of the participating surgeons, the surgical unit has been transformed into “a centre of silent precision” where “the complications of operations have been substantially reduced.”⁵ Academic writing is not brain surgery, of course. However, like surgeons and Formula One mechanics, academics do engage daily in a number of complex and highly specialized operations, and our ability to write effectively about our work requires not only training, commitment, and skill but also a willingness to change, grow, and learn from others.

In an article on “signature pedagogies,” education researcher Lee Shulman urges university faculty to look beyond the conventional teaching styles of their own disciplines—the demonstration lab (science), the discussion seminar (humanities), the Socratic dialogue (law), the studio session (fine arts), the clinical round (medicine)—and to borrow ideas from elsewhere: for example, an English professor might encourage students to undertake a “live critique” of each other’s work (the fine arts studio model) or a mathematics professor might engage students in a structured discussion of key conceptual issues (the humanities seminar model).⁶ Similarly, academic writers can make a conscious effort to question, vary, and augment the signature research styles of their own disciplines—which often embody deeply entrenched but unexamined ways of thinking—by appropriating ideas and techniques from elsewhere. Looking around my university, I can’t help noting how many of my most eminent colleagues have earned their academic reputations through interdisciplinary endeavors of one kind or another: the evolutionary psychologist who imports into the domain of comparative linguistics classification methods that he learned from studying zoology; the professor of education whose training as a statistician underpins his meta-analysis of educational research from around the world; the anthropology professor who deliberately weaves together historiographic and anthropological methodologies; the literature professor whose groundbreaking work on the origin of stories draws on extensive readings in the fields of evolutionary biology and psychology.⁷ All of these distinguished academics have been well schooled in the norms and expectations of their own disciplines, yet none of them toes a predictable party line.

When I first embarked on the research that underpins this book, I harbored a fantasy that I could map a coherent landscape of disciplinary styles, zooming in on specific regions and making informed pronouncements about their inhabitants: “Anthropologists write like this; computer scientists write like that.” By the time I had assembled my initial data set, however—one thousand peer-reviewed articles from sixty-six different journals in ten disciplines across the arts, sciences, and social sciences—I realized that a panoptic overview of signature writing styles across the disciplines would be an impossible task. In the 2003 edition of their book Academic Tribes and Territories, Tony Becher and Paul Trowler note that “there are now over 1000 maths journals covering 62 major topic areas with 4500 subtopics,” and a similarly daunting set of statistics could be generated for most other major academic fields.⁸ Casting my porous nets into various disciplinary waters, I felt less like a mapmaker or surveyor than like a lone fisherman at the edge of a vast and seething ocean.

My choice of disciplines for the study was prompted by a mixture of curiosity, expertise, ignorance, and serendipity. In the sciences, I chose medicine because I wondered whether leading medical journals allow for any variation in writing style, evolutionary biology because the field has produced some dazzlingly engaging popular science writers, and computer science because a colleague in that discipline had pointed me to some examples of intriguingly playful peer-reviewed articles. In the social sciences, I included higher education because I was already familiar with research journals in the field, psychology because of its diversity, and anthropology because of the discipline’s long tradition of self-reflective writing about writing. In the humanities, I picked philosophy for the distinctiveness of its style, history because colleagues often claim that “historians are good writers,” and literary studies, my own home field. To round the number of disciplines up to ten, I tossed in law, which sits somewhere between the social sciences and humanities and has many unique stylistic features of its own.

In most of the disciplines surveyed, I selected five representative journals—another researcher might well have chosen differently—and downloaded the twenty most recent articles from each journal. After the entire data set had been cataloged by a diligent research assistant, I undertook a detailed analysis of five hundred articles (fifty from each discipline). For the most part, I posed quantitative questions designed to yield unambiguously objective answers, for example: How many authors does each article have? What is the average page length per discipline? How many of the articles use first-person pronouns? What percentage of certain types of words can be found in each article? At times, however, I also ventured into more subjective terrain, as when, working from a detailed rubric, my research assistant and I rated the title and opening sentence of each article as “engaging,” “informative,” or both. (For more details on my sources, selection criteria, and methodology, see the appendix.)

Predictably, as soon as I started presenting the results of my analysis to colleagues from the ten disciplines surveyed, they noted that if I had chosen articles from this anthropology journal or that computer science journal, my findings would look very different. I also heard grumbles from academics in fields ranging from nursing, fine arts, and engineering to management studies and tourism, whose disciplinary journals had not been part of my survey sample. Both groups of colleagues—those whose disciplines were represented and those whose disciplines were not—felt that I had somehow neglected them, whether by failing to grasp the nuances of their particular field or subfield or by ignoring their discipline altogether. Such responses, of course, miss the point of the exercise. The purpose of this book is not to hold a mirror up to academics and show them what they already know about themselves. Instead, I want to encourage readers to look beyond their disciplinary barricades and find out what colleagues in other fields are up to. Like surgeons who believe they have nothing to learn from pit stop mechanics, academics who think they have nothing to learn from researchers outside their own discipline risk missing out on one of the greatest pleasures of scholarly life: the opportunity to engage in stimulating conversations, forge intellectual alliances, and share ideas with people whose knowledge will nurture and stimulate our own.

My data analysis confirmed some disciplinary stereotypes and upended others (see Figure 2.1). For example, I had anticipated that the science journals in my sample would all be highly prescriptive, tolerating very little variance in structure, titling, or other points of style. This expectation proved true for medicine, a field in which researchers tend to work in large teams and to publish their findings using a standardized template. In evolutionary biology and computer science, however, I found considerably more expressive diversity. Ten percent of the evolutionary biologists in my sample opted for a unique or hybrid structure in a field where the standard Introduction, Method, Results, and Discussion (IMRAD) structure predominates; 8 percent of the computer scientists use the IMRAD structure in a field where hybrid structures predominate; and 11 percent of the evolutionary biologists and 8 percent of the computer scientists include at least one “engaging” element in their titles, such as a quote, a pun, or a question. These results were fairly evenly spread across journals in both disciplines; that is, roughly 10 percent of the articles across the board diverged from any given disciplinary trend.

Figure 2.1. Percentage of articles with various stylistic attributes in ten academic disciplines (n = five hundred; fifty articles per discipline). For more details, see the appendix.

Another surprising finding was the predominance of first-person pronouns in the sciences. The high percentages in medicine, evolutionary biology, and computer science (92, 100, and 82 percent, respectively) confound the commonly held assumption that scientists shun the pronouns I and we in their research writing. By contrast, only 54 percent of the higher education researchers in my data sample and only 40 percent of the historians use first-person pronouns, a finding I discuss in further detail in Chapter 4. Overall, I could identify no particularly strong correlation between pronoun usage and the number of authors per article; that is, single-authored articles are neither more nor less likely than multiple-authored articles to contain first-person pronouns. Nor did I find a single discipline in which first-person pronouns are either universally required or universally banned. Even in literary studies, where first-person pronouns predominate, I counted two I-less articles among the fifty surveyed.

Higher education researchers topped the table in their enthusiasm for nominalizations, those multisyllabic abstract nouns formed from verbs or adjectives—obfuscation, viscosity, fortuitousness—so beloved by academic writers. In 78 percent of the higher education articles, at least seven words out of every one hundred, and often many more, ended with one of seven common nominalizing suffixes (-ion, -ism, -ty, -ment, -ness, -ance, -ence). By comparison, only 16 percent of the history articles contained a comparatively high density of nominalizations. Surprisingly, the philosophers in my sample—academics who specialize in abstraction—employ fewer nominalizations on average than their colleagues in evolutionary biology, computer science, higher education, psychology, or law. Philosophers do, however, turn to two other clusters of words associated with dense, passive prose—is, are, was, were, be, been and it, this, that, there—more than twice as often as academics in any of the other disciplines surveyed.

Psychology and anthropology proved the most challenging disciplines to characterize in terms of a “typical” style. Both are vast and varied social sciences with one foot each in the sciences and the humanities; the range and complexity of their subdisciplines cannot possibly be captured in a single snapshot. The five anthropology journals in my sample, for example, span a wide range of research activities—from the carbon dating of ancient jawbones to the development of new algorithms for explaining how social networks function—and differ starkly in their methodology, content, and style:

Because the orientation of the femur could impact this measurement, the inferior curvature of the femoral necks of the specimens measured in this study were aligned with a photograph of a gorilla femur to standardize the superior-notch-depth measurement. [Journal of Human Evolution]

It was shown in Dorogovtsev and Mendes (2000) that if the ageing function is a power law then the degree distribution has a phase transition from a power-law distribution, when the exponent of the ageing function is less than one, to an exponential distribution, when the exponent is greater than one. [Social Networks]

It wasn’t that I set out to test drive a sports car. Rather, on my way to work, I noticed rows of BMWs underneath a huge sign saying come and drive one, raise money for breast cancer. [Cultural Anthropology]

A similarly broad range of styles can be found in psychology, a discipline that ranges across all four quadrants of the “hard/soft,” “applied/pure” typology first defined by Anthony Biglan.⁹ Such disparities are, however, flattened in Figure 2.1, which represents average results across journals from ten different subdisciplines: applied psychology, biological psychology, clinical psychology, developmental psychology, educational psychology, experimental psychology, mathematical psychology, multidisciplinary psychology, psychoanalysis, and social psychology.

Figure 2.2 shows the average authorship, page length, and citation statistics for the ten disciplines surveyed. Most academics are aware that researchers in some disciplines publish short, multiauthored research reports while those in other fields favor long, single-authored articles. Nevertheless, the statistics for medicine (9.6 authors and 29 citations per 9 pages) versus law (1.4 authors and 152 citations per 43 pages) provide a striking visual contrast. For anyone who has ever sat on a multidisciplinary grant committee or promotion panel, Figure 2.2 offers a useful reminder that academics should never judge their colleagues’ productivity or citational practices based solely on their own disciplinary norms.

Overall, my stylistic analysis confirms that most academic writers—except in highly prescriptive disciplines such as medicine—are shaped rather than ruled by convention. For nearly every disciplinary trend I identified, I noted stylistic exceptions: philosophers who opt not to employ first-person pronouns (8 percent); higher education researchers who opt not to begin every article with a bland, abstract sentence defining the significance of the research topic (“Academic writing is increasingly acknowledged as an important area of inquiry for higher education research”) but instead capture their readers’ attention with an opening anecdote, quotation, or question (10 percent). These statistics will, I hope, give courage to academics who want to write more engagingly but fear the consequences of violating disciplinary norms. A convention is not a compulsion; a trend is not a law. The signature research styles of our disciplines influence and define us, but they need not crush and confine us.

Figure 2.2. Average number of authors, page numbers, and citations or footnotes in articles from ten academic disciplines (n = five hundred; fifty articles per discipline). For more details, see the appendix.