Article Design

Many years ago, as a kind of philosophical exercise, I tried to imagine the perfect object, the ideal thing. I quickly decided that it would be one that you'd immediately know what to do with when you see it. It would need no scientific investigation to understand. No owner's manual to operate. It would simply be obvious what it was for, and once you put it to that use you'd discover that it was perfectly suited to the task.

Scholars demonstrate that they know something by writing articles; and knowledge, I usually say, is the ability to write a coherent prose paragraph in 27 minutes. An article consists of about 40 paragraphs, so the problem of writing an article—or at least the first draft of an article—is solved during 20 hours of work.

The problem of article design is the problem of planning those twenty hours of work. It is the problem of mapping out the forty parts that will go into the larger whole. As a rough estimate, I normally say that an article is 3 parts introduction, 2 parts conclusion, 5 parts background, 5 parts theory, 5 parts method, 15 parts analysis, and 5 parts implications. If you go at it with the right attitude, these parts can be written in any order, but it's a good idea to write the introduction and conclusion first. Also, once they are written, you may find that my "ballpark" sense of their distribution is a bit off, i.e., that your paper will have a somewhat different form. That's fine. The individual parts still need to be made, and made "to specification".

The more articles you write, the better you will understand what a paragraph in each section must accomplish. While every discourse has its own requirements, its own particular style, the general rule is that a paper must engage your peers in conversation. In the coming weeks, I'll try to say something about how that can be done. But when designing our articles we should always keep Virginia Woolf's simple dictum in mind: "To know whom to write for is to know how to write."

In the background section you are trying to be informative. Who are you trying to inform? In the theory section you are evoking expectations of the object. Whose expectations? In the methods section you are trying to gain the reader's trust. Whose trust? Etc. The introduction is for the reader who has not yet read but would very much like to read your paper. The conclusion is for the reader who has just finished reading it. Once you know who that reader is, you know how to write. If you don't feel you know how to write, it may well be because you don't have a clear image of your reader.

Your article design, your image of your paper, should always be developed alongside an image of your reader. So, as I write about this in the weeks to come, I will always be asking you to think of your reader. Since I don't know you or your reader, my advice will necessarily be somewhat schematic. You fill it in, not just with your knowledge of what you want to say, but your knowledge of who you want to say it to.

Intended to Accomplish Goals

(I'll follow up on my post about academic cheating later today. I want to stay on schedule with my series on article design.)

The object that is specified by a designer must serve some specific set of ends. To understand a design is, in part, to understand its purpose; if we cannot see the purpose of an object, if we do not know what it's for, then we do not understand its design. This is captured by the third part of Ralph and Wand's definition: a design is "intended to accomplish goals".

When designing your article, therefore, it is useful, and quite necessary, to have some clear goals in mind. You are obviously designing it to be published somewhere. This goal can be defined by making a list of potential journals that indicate the public space in which the conversation you want to participate in goes on.

But you also have decide what effect you want to have on the conversation. Do you want simply to inform others about your results? Do you want to change their minds? Do you want to correct a misconception? Do you want to re-orient the field and take it in another direction? (I'm going to leave aside secondary goals like impressing future employers or earning tenure. As goals, these don't have a very specific effect on the design on the article.)

You can take this goal-orientation down to the level of the paragraph (and even the sentence). Ask yourself what you want your reader to do with it: believe it, or agree with it, or understand it, for example. These all set up slightly different tasks, slightly different rhetorical problems. You may be saying something that you know the reader will find difficult to believe; your job here will be to overcome their doubts. Or you may be engaging in an argument that has clearly defined sides where you want them to come over to yours. Or you may be saying something that is difficult to understand and your job is explain it clearly and effectively.

In any case, thinking about your article as an object to be designed demands that you make yourself aware of your goals. It will be useful to have one overarching goal and a 40 smaller ones. One for each paragraph.

Manifested by an Agent

The word "design" has recently been given new importance by its association with creationism. This is the belief that life on earth did not (just) evolve, i.e., develop through a series of random mutations that were then "selected" by the environment. Rather, argue the "intelligent design" theorists, someone or something made us. Something or someone wanted us to exist, intended us to be (more or less) as we are. Our bodies and their capacities are the expression of a plan, not just, as evolutionists believe, the fortuitous result of a long, natural process. Whether the designer is God or an advanced alien life form, the important thing is that it possesses agency, it is able to act with the aim of bringing something about.

Ralph and Wand's (2009) definition of design, completely unrelated to creationism of course, also stipulates an agent. Design must be "manifested by an agent", they tell us. A designed object is always "artificial", man-made. And, though I suppose this is open to stylistic variation and shifting tastes, a designed object is generally made to look artificial. The attempt to make an object look like something nature made often results in kitsch. In an important sense, the object must not just be the product of design; it must manifest the will of the designer.

In any case, scholarly articles, too, must manifest agency. They must appear to be created by an intelligent being, who wanted the text to be as it is, who had a "plan" for it, and exercised his or her (or its) own capacity for action to realize that plan in the "specified object" (the article). A journal article must not look like it came about through a series of fortunate accidents, random mutations that just happened to survive the dangers in some hostile environment (the peer-review process). A journal article is a paper that is manifestly trying to say something and there should be a sense, in reading it, that there is some agency behind it, some one who is trying to say it.

The agency that a journal article manifests is known simply as "the author". In the case of the co-authored paper, this "one-ness" of the agent is important, I should add. A paper should not look like it is the result of struggle for dominance between two brutes. It should manifest a "meeting of minds" that is, for all intents and purposes, a single intelligence. This authorial persona is of course a construction—it is in many ways part of the design. It certainly should be. Indeed, the sense we get of the agency behind the text is part of the meaning of the text as a whole.

A Specification of an Object

"Mais Dégas, on n'écrit pas des poèmes avec des idées, on écrit des poèmes avec les mots."

I'm going to work through each element of Ralph and Wand's (2009) definition of design over the next two weeks, trying to show that article writing involves a component of "design".

One thing to keep in mind is that I'm here talking about the work of planning the article, not actually writing it. The reason for this will become clear as we look at the first element of Rand and Wand's definition: design is "a specification of an object". It is not, that is, the construction or assembly or production of that object. There is a difference between designing a coffee pot and mass-producing it. In designing it, you are only specifying its properties.

But it is important to keep in mind that you are specifying the properties of an essentially physical object; you are not imagining some ideal "intellectual" thing-in-the-mind or spectral entity. An article is, ultimately, an arrangement of words across 20 or 30 pages of an issue of a journal. You are deciding how those words will be arranged. To do this, you will, have to group those words into sections and paragraphs. You will have to decide (at least roughly) how many words the whole paper will arrange. Then how many there will be in each section.

When specifying an object, be specific. What will each section say and how much of the paper will be devoted to saying it? Consider making a list of the core concepts you'll be using in each part of the paper along with a short statement of the section's overall purpose.

Like a designer, you are imagining an object without yet having to construct even a model of it. You might make some sketches on a piece of paper of course.

Article Design

Every now and then I get a great surge in traffic here at RSL. Lately, these surges have been owed to a kind tweet by Oliver Reichenstein, whose firm, Information Architects, has given us the iA Writer, which looks worth trying. When I get around to it, I'll write a post about it too. For now, however, in a pretty bald attempt to pander to the design community, I'm going to write a few posts this week about article writing as a "design" problem.

The Wikipedia article provides us with the following definition of design:

a specification of an object, manifested by an agent, intended to accomplish goals, in a particular environment, using a set of primitive components, satisfying a set of requirements, subject to constraints. (Ralph and Wand 2009)

It suits me nicely because, on this definition, I've been talking about article "design" forever. I have long been encouraging writers to think of their problem as that of constructing an object with particular goals in view. Also, my focus on the paragraph is very much an attempt to identify "primitive components", while the "environment" of an article is, of course, the discourse or conversation that it is attempting to engage with. Finally, there is no question that publication in this environment is dependent on satisfying certain requirements and respecting particular constraints. In this sense, then, your problem as a writer of a journal article is a design problem.

Many years ago, as a kind of philosophical exercise, I tried to imagine the perfect object, the ideal thing. I quickly decided that it would be one that you'd immediately know what to do with when you see it. It would need no scientific investigation to understand. No owner's manual to operate. It would simply be obvious what it was for, and once you put it to that use you'd discover that it was perfectly suited to the task.

Clearly, we're striving for the same kind of perfection in our article writing. We want the reader to be able see at a glance what the article is for and, then, while reading it, to feel that the article is perfectly suited to accomplish that goal. Just as designers must constantly keep the user in mind, writers must be ever mindful of their readers. They must imagine what the reader will do with the object they're constructing.