I use the iceberg metaphor quite a bit when talking about academic writing. “The dignity of the movement of an iceberg,” said Hemingway, “lies in only one eighth of it being above water.” This is generally taken to suggest that we should leave some things unsaid in our writing, but Hemingway was quite clear that it also means we should have a great deal under the surface. The key isn’t just to write less than you know; it is to know more than you write.

For Hemingway, the essential thing was to have the relevant experiences. If you were going to write about war, bullfighting, or love, you should have some experience with war, bullfighting, or love. The dignity of your writing depends on those experiences. But in academic writing, experience is not required in the same way, or, perhaps better, experience is not enough. The essential thing in an academic context is to be knowledgeable about your subject, not merely experienced. That means that the iceberg beneath your writing is somewhat more complex.

Experience does play an important role but in a different way than Hemingway proposed. The part of a research paper or dissertation that is most directly informed by actual or “lived” experience is your methods section, which tells the story of what you did to collect your data and why you did it that way. Writing it is mostly a matter of being honest about how you converted what Thomas Kuhn (following William James) called the “blooming, buzzing confusion” of experience into “data” — how you turned the things that are just lying around casually, if you will, into serious, empirical objects ready to be analyzed. This sort of writing isn’t easy, of course, but it’s a place where you can feel your authority (your status as an “author”) most naturally. It’s a good place to find your voice. Just try really explaining how you collected your data and what problems you encountered in doing so. Why do you trust these data? Why should your reader? Tell the story in that spirit — always honestly, of course — and you will give yourself an occasion to improve your writing quite directly. There’s no knowledge problem here. You know what you did.

But you don’t make methodological decisions in a vacuum. You don’t just do whatever you feel like doing. You try to do things in ways your reader will respect, and you get a sense of what your reader demands by reading other writers. This gets us to another portion of the iceberg under the surface of your writing: the scientific literature. This literature doesn’t inform your methods section directly but it does shape the experiences (and decisions) you write about in your methods section. So you will of course cite it as needed even there. The bulk of your references to “the literature”, however, will be found in your theory section (and sometimes an actual literature review) which is where you provide a framework within which the reader is to make sense of your analysis (and, indeed, your methodological decisions). You are setting up your readers’ expectations.

Hopefully, you can sense the balance that I’m trying to suggest here. The mass of your iceberg is centered, under the surface, on the experiences you had collecting your data, which are shaped by your reading of the scientific literature. These three components (literature, experience, data) in turn support, above the surface, your theory, methods, and analysis sections (respectively). Much of the “scientific” or “academic” content of your paper or dissertation is determined by how you do this work and how you write this prose. But that is not all.

You will often have to give your reader background information about the company, country, region, product, industry, or practice that you are studying. Though your reader is a “peer”, and therefore familiar with your theories and methods, there is much about your empirical object that will be new to them. You must tell them what you know and you must do so on the basis of publicly available documents (newspapers and magazines, company reports and press releases, government reports and official statistics) that confirm what you are saying. You want to put your reader in the same confident position to assert these facts that you enjoy.

But scholarly writing, finally, isn’t just based on reading and experience. You have to do a great deal of thinking, reasoning. So does your reader. And this will become apparent especially in the discussion section of your paper, where you make the implications of your results explicit. Here you are not (usually) introducing new sources or new information (and certainly not new data points) to your reader, you are merely noting the (logical) consequences of granting the correctness of what you’ve been arguing so far. These consequences may be practical or theoretical — they may propose changes to the way we do things or the way we see them — but they should always be reasonable. Ideally, the reader should have been able to think of them themselves. You’re just saving them the trouble.

“A writer’s problem does not change,” said Hemingway. “He himself changes, but his problem remains the same. It is always how to write truly and, having found what is true, to project it in such a way that it becomes a part of the experience of the person who reads it.” Like I say, in academic or scholarly writing the problem is slightly different, but the idea of “projection” is an apt one. The reader of a short story has to square the events with their experience of living in the world. The reader of a research paper has to square the results with their knowledge of the facts as they are. The paper has to make the result (a “truth” of sorts) part of the reader’s knowledge by the end of the reading. In your writing, you can only ever show your readers the surface of what you know; it’s what they imagine you have beneath the surface that counts.

The Craft of Research Series starts this week, as many students are beginning to think about their final projects, due this summer. Some are doing their first-year projects; some are doing their master’s theses. Some are already well on their way, and some still need to get properly started. I thought I’d say a few words about that this morning.

It is common to suggest you begin with a question. But another way, which amounts to the same thing, is to begin with what you know and what you don’t know. A question always presumes a great deal of prior knowledge and some limit to it, or gap in it. You can’t wonder what caused the failure of Lehman Brothers without knowing that it failed and, usually, a little something about what kind of failure it was. The meaning of the question simply depends on the knowledge you already have about the firm and, no doubt, the global financial crisis it was embroiled in. You should begin with this meaning, this understanding that suggests an inquiry. Maybe a better word for this is curiosity. That’s as good a place to begin as any.

You should also take stock of your resources. Begin with how much time you have. Don’t think in terms of months or even days or weeks. Try to get a realistic sense of how many hours you’re going to put into this project. Include the theories and methods you’ve already learned among your resources, along with some sense of your own intellectual abilities. How much can you expect to learn in the hours you’ve given yourself? Will you need to develop new (qualitative or quantitative) analytical skills? How much reading will you have to do? How long will it take to collect the data you need? Go back to your question and make sure that you will be able to answer it in time.

Also, and I can’t stress this enough. Find an opportunity for pleasure in doing your project; once you have found it, don’t be ashamed of it. Much of this has to do with being genuinely curious about the answer, not just vaguely anxious about your grade. But it also has to do with appreciating the aesthetic dimension of research. Maybe you like reading. Make sure there is time for that. Maybe you like writing. Definitely, make sure there is time for that. But you might also enjoy conducting interviews or observerving people or drawing graphs or doing calculations or running simulations or writing code. Make sure you approach your question in terms that afford you opportunities for such pleasure.

About a year ago I had an epiphany that, I think, can bring all this together. Our methodologies are always rooted in personal, lived experience. At the end of the day, the methods section of our paper simply describes, as honestly as we can, what we did and why we did it. In an important sense, it is a story told from experience. So begin where you are, with the questions and skills and resources that you have, and then embark on investigation — an adventure, if you will — that will become the story you will tell in your paper or thesis. Try to imagine the journey and set yourself some goals along the way — some experiences you want to have. Like any journey, it’s not going to turn out exactly as you imagine. But you have to begin somewhere and your own imagination isn’t a bad place to start.

To me, this is the original meaning of “master class”. I discovered it over a decade ago, and it has subtly shaped my writing instruction ever since. I don’t, of course, presume to be a “master” at anything like Segovia’s level. The range of my humor and the depth of my wisdom about scholarly writing is nowhere near his on the matter of playing the guitar. I imagine that the writers who come to see me are correspondingly less ambitious, though I sometimes have to remind myself and my students that I am not really an accomplished scholar at all. Still, this form of instruction seems to work when I do it. Maybe what I lack in humor and wisdom, I make up for in a vaguely Socratic irony. I know, at least, what I don’t know.

“As far as practice and suffering are concerned,” said Roland Barthes, “any writer can be compared to the greatest.” (He was talking about whether one could compare Philippe Sollers to Marcel Proust.) The master class gives me access, not just to a writer’s text, which may then be “corrected”, but to their practice and suffering, which may be disciplined. As writers, we cannot avoid suffering, but I like to think we can make it more precise. This is best done by direct engagement.

But unlike the playing of Segovia’s students, another’s writing can’t happen right in front of us in real time. You can’t tell how good a writer is, nor where their writing can be improved, by watching them work for a few minutes. That simply isn’t how writing works. On the other hand, you can’t simply point out errors of grammar and punctuation in a draft that they’ve spend the past week writing either. (That approach has, thankfully, been pretty widely rejected by writing teachers long ago.) Our instruction needs to bring together what Robert Graves called “the huge impossibility of language” with what I might call the “appreciable finitude of writing”.

To do this, I require participants in my master classes to bring a well-formed paragraph that has been written during a deliberate writing moment. I imagine Segovia expected that his students had practiced something — were “working on something” as it were — in the days leading up to their meeting. The session begins with them playing it as well as they can. Likewise, I expect my students to have written several paragraphs in the days and weeks before they attend the master class. The key sentence and its rhetorical posture had been decided in advance, and the writer has spent 27 minutes composing at least six sentences and at most 200 words to overcome the difficulty that they expect the reader to experience during one minute of their deliberate attention. Having written several such paragraphs (having “practiced”) they select one to bring the master class (where they will “suffer”).

The important thing is that I can now assume that I have before me the product of one deliberate attempt to say one intended thing. I can engage with it directly. First, I read it out loud, straight off the page, having never seen it before, to show the writer how hard or easy that is to do (for a academically literate reader and native speaker of English). I then try to identify the key sentence (the writer knows which sentence they’re hoping I pick) and determine whether it is being supported, elaborated, or defended (the writer, again, was presumably trying to do one of these things, with some exceptions). All the while I’m thinking out loud and moving words around to show what I mean. I even draw on the other participants to get multiple points of view on the same text, multiple readings. Sometimes I make some cryptic remark about “nuance” or “style”. The writer is supposed to sit quietly and take it all in. After 13 minutes (usually) I stop. We then move on to the next participant, another paragraph.

I don’t do a lot of language editing in these master classes, except where things are easy to fix and make the rest of the work go more smoothly. And I don’t ever suggest that I am correcting or even improving the text. That is for the writer to decide on their own. All I can do is reveal the contingencies in the composition, the fact that the paragraph could have been written differently. That, I’m told, is also the value of the experience to those who have tried it. To steal a line from the Segovia, I try to impart “a delicate lack of respect” for the grammar and focus on the possibilities of meaning instead. What participants get is not a judgment of their work but a very explicit experience of being read. “Writing as freedom,” said Barthes, “is perhaps the most explicit in history.” A master class provides an occasion to become, perhaps not “great”, but at least better, through a moment’s suffering, followed by many moments of subsequent practice.

____________

If you’re curious, you can experience my master classes as part the four-week Writing Process Re-engineering self-study course.

I approach university students as budding scholars. Like scholars, I tell them to write for their peers. They should imagine their reader as a fellow student in their class, someone who is doing the same assignment they’re doing. Since every paragraph poses and overcomes a difficulty for the reader, the writer needs to have a good understanding of the reader’s knowledge of the subject. The context of a course provides this understanding because the reader is someone who enrolled in the course with the same prerequisites and is now following the same syllabus. To know the reader, the student merely has to be themselves and keep up with the reading.

Some students forget this and leave most of the reading until the end of the semester when they’re doing their final exams. These students will often leave their writing til the end too. Not only does this cause a lot of unnecessary stress, it robs them of a community of learning, one that can give them the important experience of forming their ideas gradually as part of an ongoing conversation. This sense of a place in a conversation is what the classroom is all about. By participating actively in a course — attending lectures, engaging in discussions, doing the reading, and writing along the way — students are finding out what it means to know things for “academic purposes”. This is not at all a trivial state of mind; it suggests an ideal that universities exist to promote.

Scholars should remember this ideal student. After all, a working scholar is really just a very successful student, someone who has earned the privilege of remaining a “student” (of their subject) “for a living”. But just as their students have to keep up with the readings in their courses — not just, like I say, in order to understand the lectures, but also in order to understand their implied reader — scholars have to keep up with “the literature” — not just to stay current with developments in their field, but in order understand the needs of their readers and their means to make sense of what they’re saying. A good academic writer has to be a good reader of the literature to which they are contributing. Their readers, after all, are reading that same literature.

This is one of the reasons that it’s important to choose your discipline carefully. If you are going to get any pleasure out of your research, you will have to find a literature that you enjoy reading. This, too, is like the student who must choose their program carefully. If you don’t like any of your fellow students — if you’re not at least a little “like” them in your beliefs and desires, your curiosity and your ambition — you will not enjoy imagining them as the readers of your papers. If you can’t respect their interest in the papers that are published in your discipline, year in and year out, then you will not find it useful to imagine the difficulty they have understanding your key sentences.

Writing is a social act because it depends on the existence of readers for its meaning. You are always writing for someone who recognizes your theories and your methods, not just from their own work, but from the work of the peers you have in common. You are writing for someone who can not only make use of your results if they are true but identify your errors if they are false. I tell students to imagine the most intelligent, most knowledgeable, student in their cohort as their reader. Scholars should imagine a well-versed peer in their discipline. You train this imagination by reading.

(Part 1)

Crispin Sartwell (1991, 1992) thinks that knowledge is merely true belief. Most philosophers think it takes more than true belief to know something (e.g., Alper Gürkan). And many say it even takes more than justified true belief. Who is right? Do they even believe these things? Who knows what knowledge is? (Cf. Lycan 1994) I’m not going be able to answer those questions, of course. But here’s what I think.

Knowledge is true belief held with good reason. It’s not “merely” true belief, nor is it merely belief held with “reasons” (as the kids say). To know something, you must believe it and you must have your reasons, but the belief must also be true and your reasons must be good ones. When writing, then, you should know what you’re talking about and when you’re trying to decide whether or not to write something you should ask yourself (1) whether you believe it, (2) what your reasons are, (3) whether those reasons are good, and (4) whether it’s actually true. Maybe that’s also the best order to do it in. Let me try to think about each in turn.

In academic writing, the most important thing is to believe what you say. That’s because the purpose of academic writing is to expose your ideas to criticism, to give your peers a chance to correct you if you are wrong. Students unfortunately sometimes get the impression that the important thing is to say what their teachers believe. They are often rewarded for this with a good grade but this leaves their own beliefs on the subject untouched. From a strictly educational point of view, it’s better to get a C on your own ideas than an A on someone else’s. After all, you’re still going to believe your own ideas; you just don’t know that they deserve a C.

The same goes for working scholars, who may be tempted to write papers about ideas they think are publishable rather than their own. Here, again, they are missing the opportunity to be corrected on what they actually think by people who are qualified to tell them that they are wrong. At best, they are testing their hypothesis about what people want to hear, or, rather, about what their editors think people want to hear. They are not exposing their own thinking to the scrutiny of other knowledgeable people.

Beliefs and reasons are “subjective” in the sense that they belong to individuals. They are “states of mind”. It’s probably best to think of “reasons” as just more beliefs, but, if so, they are also beliefs about beliefs or at least beliefs that are held in relation to other beliefs. I may believe that my house is on fire and I may believe that smoke is rising from it. I take the second belief as a reason for the first. Importantly, if I discover that the smoke is coming from my neighbor’s house rather than my own, I no longer have a reason to believe that my house is the one that is on fire. Having a reason, rather than a mere belief, works that way. It’s still a belief, but it is implicated in other beliefs.

Now, those implications may be more or less, let’s say, tenuous — more or less of “a stretch”. Many people think of their own house for a brief second when they see a fire truck — sirens blaring, lights flashing — driving by. They don’t believe (not even for a second) that their house is on fire. It’s just that the thought occurs to them and they do a mental check to see if it’s possible that they left their iron on. Then they put it out of their minds. They believe that the fire truck is going somewhere and there’s probably a fire of some kind (or a false alarm, of course) but they don’t think that it’s their house. That’s just too unlikely.

Seeing smoke rising from your roof is, let’s say, a good reason to believe that your house is on fire, while seeing a fire truck drive by isn’t a good reason. You can make up your mind accordingly. But, at the end of the day, you may believe that your house is on fire though it isn’t. And we’re not going to say that you know something if it isn’t true. If you believe something and you have very good reasons to believe it you are being perfectly rational but you may still be entirely wrong. The question, then, is whether all those reasons you had were really so “good” after all. And what was the point of having them?

My answer is that the important thing about “knowing” something isn’t just that you believe something that is the case but that you believe a number of things in a ways that hold implications for each other. If you’re a knowledgeable person, then revising one of your beliefs should affect some of the other beliefs you hold. The better your reasons are, the more easily those revisions can be made, because the implications of your beliefs for other beliefs are simply clearer. Being knowledgeable isn’t just a matter of having your mind made up about it and happening to be right. It’s about being in a position to change your mind in an orderly fashion.

This isn’t over yet. More later.