Stylish Academic Writing - Helen Sword 2012

Structural designs
The elements of stylishness

Essayist Annie Dillard describes writing as an architectural endeavor, a continuous cycle of design, demolition, and rebuilding. Sentences are the bricks; paragraphs are the walls and windows:

Some of the walls are bearing walls; they have to stay, or everything will fall down. Other walls can go with impunity.… Unfortunately, it is often a bearing wall that has to go. It cannot be helped. There is only one solution, which appalls you, but there it is. Knock it out. Duck.¹

Dillard’s metaphor strikes at the emotional heart of the writing process, which involves destruction as well as production, short-term losses as well as long-term gains. Stylish academic writers are craftspeople who regard their texts as intricate, labor-intensive structures that must be carefully planned and meticulously built, from the pouring of the foundation and the sourcing of the materials to the final polishing of the banisters—not to mention those rare but wrenching occasions when the wrecking ball must be called in.

A well-structured article or book, like a well-built house, requires careful thought and planning. Most academics enjoy a wider range of structural choices than they may realize, starting with the most basic decision of all: will their overall structure be conventional, unique, or something in between? As a general rule, disciplinary cultures that value creative expression (such as literary studies) encourage and reward creatively structured scholarship, whereas disciplinary cultures that privilege scientific rigor (such as biology) encourage and reward structural rigor. However, of the ten disciplines in my data sample, medicine was the only field in which 100 percent of the articles employed a conventional Introduction, Method, Results, and Discussion (IMRAD) structure or something very similar, with absolutely no variations. In every other discipline surveyed, I observed a range of structural approaches. Significant percentages of academics in computer science (90 percent), higher education (70 percent), psychology (56 percent), anthropology (50 percent), and even evolutionary biology (10 percent) adopted unique or hybrid rather than purely conventional structures. In the humanities, meanwhile, I noted a fairly even mix of articles with unique structures (that is, their section titles follow no recognizable pattern or convention), hybrid structures (whereby uniquely titled sections cohabit with conventionally titled sections), and sequential structures (sections that are numbered but not titled). More than one-third of the history and literature articles in my survey sample—36 percent and 38 percent, respectively—contained no section headings at all (see Figure 2.1 in Chapter 2).

For scientists and social scientists, the advantages of adhering to a conventional structure are many. Authors who follow the IMRAD model always open their articles with an introductory section that clearly states the purpose and scope of the current research, sums up previous work in the field, and probes gaps and flaws in the existing literature. Next, in sections with titles such as “Data,” “Methodology,” and “Results” (the exact labels vary from field to field), they describe the data collection and results. Finally, in the “Analysis,” “Discussion,” and/or “Conclusion” sections, they review their main findings, explore the wider implications of their work, and offer suggestions for further research. This paint-by-numbers approach prompts researchers to plan their research methodically, conduct it rigorously, and present it coherently, without leaving out any crucial information. Moreover, a conventional structure is relatively easy for new academics to learn; all they have to do is follow models established by others before them. Readers, meanwhile, know exactly where to look for key findings. They can skim the abstract, mine the literature review, scan the data, and grab the conclusions without wasting valuable time actually reading.

SPOTLIGHT ON STYLE

DONALD SHANKWEILER

At the beginning of our long collaboration, Isabelle Liberman and I were concerned with testing explanations of reading problems that were current at the end of the 1960s. At that time, ideas about causation regularly invoked neuropsychological concepts such as poorly established cerebral dominance. Reversals of letters and words were still considered to be the hallmark of dyslexia.… As for treatment, that was the heyday of motor patterning, balance beams, and eye exercises. Our early work was devoted more to showing what reading disability was not than to explaining what it was.

In an article that pays tribute to his recently deceased colleague Isabelle Liberman, linguist Donald Shankweiler explains how the concept of phonological awareness can help teachers help children with reading problems. Despite a plethora of be verbs and some sloppy locutions that demonstrate the pitfalls of abstraction (for example, “ideas about causation regularly invoked neuropsychological concepts”—can an idea invoke a concept?), Shankweiler’s writing style is for the most part lucid, personable, and example-driven. He structures his article as a series of seven numbered assertions about “the development of reading and its difficulties,” with each “assertion” constituting a section heading, for example:

1. Emergence of Phonological Awareness Follows a Developmental Pattern

2. Early Instruction Designed to Promote Phonological Awareness and Letter Knowledge Confers an Advantage in Reading and Spelling That Is Measurable Years Later

Readers come away from Shankweiler’s article with a clear understanding of his seven arguments and of the evidence he musters to support each one. Rather than wrapping up with a standard conclusion, he ends with a “promissory note” that describes new research advances. Such news, he notes, would have given great pleasure to the friend and colleague whose work his article memorializes.

However, conventional structures also have some significant drawbacks. Generic section titles such as “Method” and “Conclusion” provide very little real information about an article’s content, a handicap for skimmers as well as for readers. In the following outline excerpted from a higher education journal, only the title tells us anything specific about the topic being addressed:

Another disadvantage of identically structured articles is that they all end up looking and sounding more or less alike, thus offering the subliminal impression that they all say more or less the same thing. Even more worryingly, academics who always plan, research, and write to a template risk thinking to a template as well.

Hybrid structures offer an alternative for scientists and social scientists who want to add some unique architectural features to work that is otherwise safely grounded in disciplinary norms. In a research article with a hybrid structure, sections with conventional titles such as “Introduction,” “Method,” or “Conclusion” sit side by side with uniquely titled sections such as “Gender and Developmental Issues Relative to Interest Structure” (psychology), “Pre-Classic Settlement, Ceramics, and Social Conflict in the Rio Grande del Rancho Drainage” (anthropology), or “Legalism in East Asian Regional Economic Integration” (law). The following outline of an article from a computer science journal offers a fairly typical example of a hybrid structure:

Note the parallel sequencing of the two main sections (Definitions, Expressivity, Analysis) and the numbered outline indicating structural hierarchies (a mandatory feature in many science and engineering journals). The authors of this article are not trying to impress anyone with their inventive structure and clever section titles. Nor, however, have they followed a predetermined template dictating how they must present their research.

Stylish academic writers often adapt conventional and hybrid structures to suit their own needs, as when psychologist Bob Altemeyer offers two brief Method-Results-Discussion studies within a single article, or when management researchers David Guest and Neil Conway set up a study based on five “Hypotheses,” each of which is explained in the opening section, reported on in the “Results” section, and further analyzed in the “Discussion” section.² Some authors use unique subsection headings to enliven and individualize conventionally titled main sections (a common ploy in evolutionary biology, among other fields). At the opposite end of the stylishness spectrum are articles so carelessly assembled that their structure exposes cracks and fissures in their authors’ thinking. One of the higher education articles in my data sample, for instance, contains a section promisingly titled “Findings and Interpretations,” which opens with the following sentence: “Four dominant discourses shaping images of leadership emerged from our analysis: autonomy, relatedness, masculinity, and professionalism” [my italics]. The reader therefore anticipates that the section will consist of four subsections arranged in the following sequence:

• Autonomy

• Relatedness

• Masculinity

• Professionalism

Instead, however, when we skim through the section, we discover that the authors have broken it into five subsections:

• Autonomy

• Gender and Masculinity

• Professionalism

• Masculinity

• Relatedness

Not only do the subsections occur in a different order than the opening sentence has led us to expect, but the “Masculinity” subsection has suddenly spawned a semi-redundant offshoot, “Gender and Masculinity.” This lack of attention to structural detail—indeed, to structural fundamentals—leaves readers feeling rather as though we followed signs marked “Auditorium” and found ourselves in a broom closet. Worse, the structural inconsistencies make us doubt the validity of the authors’ analysis; how could such shoddy construction techniques possibly result in a watertight building?

SPOTLIGHT ON STYLE

ROBERT J. CONNORS AND ANDREA LUNSFORD

As we worked on this error research together,… we started somewhere along the line to feel less and less like the white-coated Researchers of our dreams and more and more like characters we called Ma and Pa Kettle—good-hearted bumblers striving to understand a world whose complexity was more than a little daunting. Being fans of classical rhetoric, prosopopoeia, letteraturizzazione, and the like, as well as enthusiasts for intertextuality, plaisir de texte, différence, etc., we offer this account of our travails.

In a now classic 1988 article titled “Frequency of Formal Errors in Current College Writing, or Ma and Pa Kettle Do Research,” professors of composition Robert J. Connors and Andrea Lunsford report on a large-scale study of how composition instructors mark formal errors in student writing. Humorously describing their own awkward attempts to negotiate a research paradigm to which they brought plenty of naïve enthusiasm but no real disciplinary training or experience, they write with a stylistic audacity that matches their interdisciplinary chutzpah. Each section title pairs a weighty label drawn from classical rhetoric with a wry summary of “Ma and Pa Kettle’s” shambolic progress:

• Proem: In Which the Characters Are Introduced

• Exordium: The Kettles Smell a Problem

• Narratio: Ma and Pa Visit the Library

• Confirmatio I: The Kettles Get Cracking

• Confutatio: Ma and Pa Suck Eggs

• Confirmatio II: Ma and Pa Hit the Road

• Amplificatio: Ma and Pa Hunker Down

• Peroratio: The Kettles Say, “Aw, Shucks”

Working collaboratively in a field where single authorship is a disciplinary norm, Connors and Lunsford push against every stylistic and structural boundary they can think of, playfully reflecting on both the processes and the products of their own research.

Figure 11.1. Virginia Woolf’s sketch of the structure of To the Lighthouse.

A seemingly unstructured but in fact well-crafted article provides a more satisfying reading experience—and certainly a more persuasive demonstration of authorial skill—than a conventionally structured one with weak supporting walls and confusing signposting. Virginia Woolf famously described her experimental 1927 novel To the Lighthouse as “two blocks joined by a corridor,” sketching two large rectangles to represent the bulk of the novel, in which time moves very slowly, connected by a narrow band depicting the “Time Passes” section, in which years fly by in the blink of an eye (see Figure 11.1).³

As Woolf’s example reminds us, structure becomes more rather than less important when an author deviates from generic norms and expectations. Unique and experimental structures can open up new ways of approaching familiar issues, a form of intellectual displacement that parallels the physical displacement we feel when we traverse an unfamiliar landscape or enter a room where the walls sit at unusual angles. Only if the route is well signposted and the rooms are well lit will readers be able to take such displacement in their stride.

In a conventionally structured academic article, section headings function like centrally positioned, neatly labeled doorways that lead us from one well-proportioned room to the next. In a uniquely structured article, by contrast, we never quite know where we are going or why, unless the author makes a special effort to keep us on track. In some humanities articles, the section headings feel more like partitions randomly inserted to break up a cavernous space than like the coherent components of an architectural plan:

The author of this article from a prominent cultural studies journal, for example, has missed a golden opportunity to use his punning title as a structuring device: “Godard Counts” suggests not only that the filmmaker Jean-Luc Godard is important (“counts”) but also that his aesthetics is bound up with tropes of numbering, ordering, and counting. The article’s first three section titles (“Ordering Evidence,” “Dirty Hands,” “Counting on Your Fingers”) echo the counting pun; but then the author drops the ball. The remaining six section titles have nothing much to do either with the main title or with one another, and the numbers impose a sequential flow that is not reflected in the titles. For a reader skimming the article in search of information and direction, the cryptic section titles prove more mystifying than helpful.

To be sure, not all humanities scholars want to provide their readers with clearly marked entry and exit points. Some compose playful section titles that mimic the brightly painted doors of a fun house, deliberately enticing us into halls of mirrors or other surprising spaces. Others eschew section titles altogether, calling on more subtle structuring techniques—the gradually unfolding argument, the controlling metaphor—to direct their readers’ attention, as when literary scholar Linda Brodkey stitches together her childhood memories of reading, writing, and sewing in an article whose title, “Writing on the Bias,” puns on the relationship between textile and text.⁴ Peter Elbow’s influential book Writing with Power contains many intriguing suggestions for variations on conventional structures (for example, the collage essay, the dialogic essay, or the critical-creative essay).⁵ Elsewhere, in an essay titled “The Music of Form,” Elbow notes that, while section headings help readers get a quick visual overview of an article, there is still something to be said for the linear, time-bound experience of moving through a piece of writing one word at a time:

I’m not arguing against the usefulness of traditional organizational techniques like signposting, mapping, and thesis statements—which can powerfully compensate for how texts are trapped in the glue of time. But … the traditional techniques are not the only way to give readers a sense that an essay hangs together and is well organized.⁶

Authors of scholarly books—the mansions of academe—have the luxury of constructing architectural features that would not easily fit within the confined footprint of an individual research article: staircases and turrets, fountains and follies. Some build whole volumes around a unifying theme or metaphor, as when literary scholar Robert Pogue Harrison, in a book about forests in the Western literary imagination, invokes different forms and uses of forests in chapters with titles such as “Shadows of Law,” “Forests of Nostalgia,” “Dwelling,” and “The Ecology of Finitude.”⁷ Some authors focus on the reader’s journey, as when classicist David Ulansey structures his book on Mithraic religious rituals “as a gradual unveiling of a mystery … allowing the story to unfold step by step, slowly adding separate pieces to a puzzle whose final image does not become clear until the end.”⁸ Some even set their readers loose in conceptual mazes deliberately designed to disorient and amuse, as when cognitive scientist and jack-of-all-disciplines Douglas Hofstadter, in books with titles like I Am a Strange Loop and The Mind’s I, foregrounds the self-referential intricacies of his own writing.⁹ Bold structural choices such as these are available not only to book authors but to dissertation writers as well, provided they have the necessary personal confidence and institutional support. As with any other aspect of academic writing, the key to producing a well-structured book, article, or thesis is neither slavish imitation nor willful anarchy but carefully considered craftsmanship.

SPOTLIGHT ON STYLE

VICTORIA ROSNER

“Yes? No?” No. The opening lines of Good Morning, Midnight capture what seemed so wrong with the forms of private life in the first part of the twentieth century. Rootless and solitary, protagonist Sasha Jensen passes her time in a fruitless search for rooms. Rooms speak to her, tell her in suggestive tones what they’re about.… Sasha warns the reader later of the latent power in the rooms she inspects: “Never tell the truth about this business of rooms, because it would bust the roof off everything and undermine the whole social system.”

In a citation announcing the Modernist Studies Association’s 2006 Book Prize for a “significant contribution to modernist studies,” the selection committee drily noted, “It is a rare thing to be seduced by a table of contents.” Victoria Rosner’s multidisciplinary book Modernism and the Architecture of Private Life explores the domestic interiors of early twentieth-century art, literature, and thought by inviting us to wander through a series of beautifully composed chapters appropriately titled “Kitchen Table Modernism,” “Frames,” “Thresholds,” “Studies,” and “Interiors.” Architecture provides Rosner not only with the structural design for her book but also with a treasure trove of evocative metaphors, from the literal “impasse” where novelist Jean Rhys’s protagonist Sasha Jensen conducts her fruitless search for a room of her own—“a narrow alley that arcs and cuts off in a dead end”—to the complex web of relationships that shaped modernist culture:

This book proposes that the spaces of private life are a generative site for literary modernism. These spaces compose a kind of grid of social relations that shifts and slips, often upending the individuals who traverse it.

Rosner’s highly spatial vocabulary—space, site, grid, shift, slip, traverse—illuminates abstract ideas about society and selfhood. Her book takes us from the laundry room to the library, from the closet to the study, and to many other places in between.

Things to try

• If you are a scientist or social scientist, decide in advance whether you want your journal article to have a conventional, hybrid, or unique structure. Pros and cons: a conventional IMRAD structure (Intro, Method, Results, and Discussion) encourages scientific rigor but discourages independent thinking; a unique structure promotes creativity but risks disorienting readers; a hybrid structure offers flexibility but is neither fish nor fowl.

• Consider using a metaphor, theme, or series of sequential steps as a structuring device.

• If you have never before strayed from IMRAD and its cousins, consider developing a hybrid structure or, at the very least, introducing some unique subsection titles. Look in journals from both within and beyond your discipline for examples.

• Make an outline of your article or book based only on its chapter titles or section headings. How well does that outline, on its own, communicate what your work is about? Are you using section headings to inform, engage, and direct your readers, or merely to carve up space?

• To fine-tune your structure, make a paragraph outline. First, identify the topic sentence of each paragraph (that is, the sentence that most clearly states its overall argument); next, arrange those sentences in a numbered sequence. This process can help you identify structural weaknesses both within and between paragraphs: for example, a paragraph that has no clearly stated argument or one that does not logically build on the one before.