How to Start Your Thesis

Jerry Bannister

Starting a graduate thesis is hard. Anyone who tells you otherwise is lying, delusional, or one of those bizarre people who find it easy. December in Canada brings awful holiday specials on TV, complaints about freezing rain and, for those of us in universities, worries over what’s left undone from the Fall term. This is that liminal time when classes have ended but the final grades are not entered. It’s the anxious time when both professors and graduate students are supposed to have accomplished nearly half their work for the academic year. While the end of classes stirs warm illusions of productivity, the spectre of the holidays brings cold memories of what invariably happens to all those carefully laid plans. If, as I said last June, early summer is when stress becomes as unavoidable as black flies, by December it’s stuck to us like ice to a car windshield.

So, instead of finishing my own grading today, I want to give some unsolicited advice to graduate students and their supervisors. My advice is meant to be pondered now but acted upon in January, because, if you’re like 94.7% of the academic world, you will get precious little work done once the holidays are upon us. You can fight it and make yourself unproductive and miserable, or give into seasonal reality and be unproductive yet happy. Like finishing a thesis, starting one depends as much on emotional momentum as it does on ability or work ethic. If you feel crappy about your thesis, reading twenty books each week is not going to help much. That crappy feeling is typically caused by a sense of not feeling in control. In this emotional state, the thesis is everywhere yet nowhere, something always in your thoughts but rarely on your screen. Gaining a sense of control is, of course, an exercise in creating fiction, because so much of the thesis process lies, like so much of life, beyond your actual control. You can control what you write but not how others will respond to it.

Convincing yourself that everything will work out may be an exercise in illusion, but it’s a useful starting point. We need our rationalizations and fictions every bit as much as we need coffee and wine. (If you need neither, please stop reading this). Supervisors know well that much of what new graduate students plan in September will either never happen or turn out very differently after January. Almost no one ends up doing precisely what was in their mind’s eye when they applied to graduate school. Which is why supervisors need to cut students some slack at this time of year. They need to listen patiently as students work through half-baked ideas, test drive alternate plans, and bounce ideas around the seminar room. They need to think carefully about when to let a student travel down a particular research path and when to caution against going down a rabbit hole. Making that decision is often a difficult judgement call, but it’s made best when the supervisor and student talk openly about expectations and stress. The road through graduate school is, in a sense, the distance between the ideal thesis that a student wants to write and the real one that they finish.

Like most roads, this one comes with signposts. Let’s call them the five laws for starting your thesis. Number one, which I often call the first law of history, is the law of selection. The virtue and vice of history is its open-endedness. You have to decide what you research. You have to choose the chronology, geography, as well as the person, people, or thing that you want to research. Once you choose your basic parameters, you have to follow the law of selection like peeling an onion. You must select the scholarly literature, the primary sources, the examples, the theories, and the methodology on which your thesis will be based. Once you find your primary sources, you then have to choose which ones you will focus on and how you will use them. You will do that, of course, with plenty of advice from your supervisor, other professors, friends, and Google. Everyone will have lots to say, but the choice is ultimately yours and yours alone. How good your thesis is will depend, in large measure, on how well you make your selections. You will want to make the best choices, of course, but you need also to make practical ones, because the perfect thesis is the one that’s never written.

The law of selection contains within it important rules. The first is the prohibition against gapstritis, an unfortunate condition that afflicts many thesis proposals and SSHRC applications. It’s a condition in which the sufferer justifies the selection of their topic solely on the basis of a perceived gap in the scholarly literature. This justification is based on the delusion that there is an entire unfarmed pasture simply awaiting their historical plough. The problem is that a topic may have been ignored for excellent reasons (not least of which is its unimportance), or, more likely, there are plenty of relevant studies lurking either just underneath the topsoil or in adjacent fields. Saying that your thesis will boldly go where no one has gone before will make you feel good but give your supervisor a headache. All good theses are, in a way, conversations between yourself, your evidence, and other scholars working in your field.   Rather than thinking in negative terms of gaps, think in positive terms of connections. You get to decide which conversations you have, but the onus is on you to persuade the reader that you’ve made the best extant choices.

This process relies on what I call the law of evidence. According to this law, the burden of proof is entirely yours. Theses must be based on positive proof about what actually happened in the past, not what didn’t happen or could have happened. We make all sorts of counter-factual calculations as we cope with history, but we make them in our minds, not in our writing. Starting your thesis entails boiling the historical process down to its essence: identifying evidentiary patterns   I often find myself at this time of year advising students to write towards the evidence. What I mean by this is that they need to free themselves from presumptions based on what they expected to find, or what they read somewhere, in order to follow the evidentiary trail like miners following a seam of coal. Setting up a proverbial straw man to burn down is always fun and often useful in the early planning stages of a thesis; however, when you sit down to write, you need to think about what you are arguing for, not just against. Negative proof is precisely that; you should remember that everyone else does not have to be wrong for you to be right. The evidence that you use to prove your argument must be sufficient in both quantity and quality: it’s up to you to demonstrate that you’re deploying the best data. Make sure to avoid the danger of overreach, because a qualified argument that makes reasonable concessions is more persuasive than an argument in which analytical reach exceeds evidentiary grasp.

Mulling over your evidence raises the law of entry. In my discussions with students about their writing, we often spend a lot of time talking about what I call the entry-point. Deciding when and where to plunge into the writing is arguably the single most important decision in determining the success of a thesis.   In making that decision, students should collapse the distance between themselves and their topic, in order to avoid wasting time on extended prefatory discussions related to background contexts or historiography. Such discussions, if they turn out to be necessary, can always be added after the first draft. In my experience as a supervisor, theses invariably expand outward, not inward: what a student chooses to write about first will typically become what they write about the longest. First chapters often end up subdividing, like cells following mitosis, as students realize that the 30 pages they planned became the 70 pages they wrote. I advise students to frame their entry-point as sharply and as narrowly as possible, because they can always spread out in different directions, depending on how their writing evolves. Starting with a concise question and working methodically through primary sources not only facilitates analytical clarity and organizational cohesion, but also mitigates the risk of a thesis spinning out of control and becoming morbidly bloated.  It offers a type of insurance plan, in the likely event that the student will run out of either time or space. Theses rarely offer neat closure, in the sense of completing everything students originally set out to accomplish; rather, students typically walk away when it’s time and their thesis meets the standard of the graduate program.

Adopting a writing plan that allows for some components to be left out if necessary is an important step that should be taken deliberately and early in the process. This is often done by organizing chapters chronologically so that a later period can be dropped. Students should avoid organizational structures that depend on best-case scenarios in which all of the projected parts of a thesis must be fully completed for it to succeed. When advising students at the start of their writing, I encourage them to think about questions rather than theories or arguments.   I talk with them about the evidence that they’ve found, and we work to create and to calibrate questions best suited to their data. I find that when students think about their theses as answers to questions, rather than abstract points of analysis, their writing is much clearer and more persuasive.

Related to the goal of clarity is the law of organization. This law states that successful writing requires copious and careful record-keeping. For most graduate students, their thesis will be the first thing they write that cannot be read in one sitting. It’s the first project they’ve undertaken that requires extensive record-keeping painstakingly maintained over many months and across markedly varied contexts. One of the first things I advise all new graduate students is to keep a journal in which they dump their daily notes and ideas about their thesis. I encourage them to jot things down as they work, to keep a running bibliography of the primary and secondary sources that they consult, and to start drafting their table of contents as soon as possible. Maintaining a private journal for their own use gives students a place where they can freely express their ideas and experiment with concepts. As I often tell students, one of the ideas that they jot down will eventually become the seed for their thesis. Graduate students are not only researchers and writers but also their own personal archivist. They need to build and maintain a successful system for understanding their primary sources and connecting them to the scholarly literature. Effective record-keeping mitigates the risk of not only researching the same source twice (a mistake that most of us make, sooner or later) but also seeing archival sources as isolated islands of data rather than parts of a larger historical ecosystem. We tend to think and talk about graduate theses too much in terms of discrete phases of research and writing; keeping a journal helps students to make and maintain a creative bridge between these two interrelated processes. It helps to ensure that their outline reflects their ideas and arguments, in order to avoid the danger of staring at a blank laptop screen without a clear sense of what they want to say.

Good writing depends on good habits and routine. Students need to find a place, time, and pace that works for them. They need to figure out, for example, whether they are early-morning writers, night owls, or something in between. Supervisors should give students ample space to make their own decisions and to caution them against feeling that they must follow whatever their peer-group is doing. When I was a graduate student at the University of Toronto, there was a group of serious students who met every day in the library cafeteria for lunch. They had important talks about important things but, try as I might, I could never bring myself to join them. The last thing I wanted to do over my lunchbreak was not to have a break. I also couldn’t get much writing done in the early mornings, regardless of how much coffee I guzzled or how much the serious students bragged about their pre-lunch productivity. I was, until I became a parent, a night owl, and I benefitted enormously from the fact that my supervisor never badgered me once about my habit of staying up all night and sleeping in late. Thirty years later, I couldn’t work all night even if I tried, but I try not to forget that graduate students in their twenties need to devise their own habits that work best for them.

Like all actions, this law has a reaction. It is the law of disorganization. This refers to the creative oxygen that all good writing needs to survive. Too much habitual organization and structural rigidity can be as deadly as writer’s block to a thesis. What is often most needed when a student is writing is a judicious mix of messiness to keep the creative juices flowing. This often entails switching back-and-forth between analyzing patterns in the primary sources and reading the scholarly literature. I often encourage students, when they have completed their basic archival research, to return to their early notes and to re-think articles and books related to specific problems. I advise them that the literature can be used in three different ways: to help with establishing the larger historical context; to assist with understanding the prevailing scholarly interpretations; and as a source for further primary texts. Jumping between tasks helps to avoid the complacency that can easily set in when a student spends too much time alone with the same set of sources. It can help to ensure that their thesis is not source-driven but rather flows from their own questions, analyses, and arguments.

Each of these laws brings with it choices about what to do and how long to do it. As I work with students making these decisions, we typically build a list of pros and cons based on worst- and best-case scenarios. We talk about likely outcomes for each option. I try to explain that all that anyone can expect is that they work hard, think carefully, and give it their best shot. If I’m lucky, I’ll remember to tell them the story of Milo, the main character from one of my favourite books, The Phantom Tollbooth. As Milo traveled through different lands, he discovered that the way to get to the island of conclusions was to jump there. Regardless of how hard a student works, or how organized they are, as they start writing they will confront the tyranny of a computer screen perfect in its blankness. Each sentence they type will require further explanation and elaboration, raising questions as fast as answers, and erasing hope for perfection.

From a legal perspective, historians do not work in the realm of the criminal law, where the burden of proof is beyond a reasonable doubt. On the contrary, doubt can be the historian’s best friend: you can tell if you are talking with a poor historian if they never admit to having any doubts about what they’re doing. And if they start talking about the truth, chances are you’re speaking with a philosopher, not a historian.  If you listen carefully to what historians say when they get together, you will hear them discuss whether they buy a particular argument. They are not referring to whether they believe that it reflects some ultimate truth, but rather whether it persuaded them. In legal terms, this means that historians think more like lawyers in civil litigation, where the person bringing the action needs merely to prove the case on the balance of probabilities. Historians try to be as careful, complete, and certain as possible, but they know in their bones that good history requires plenty of imagination and occasional guess-work. The trick is to be honest with yourself and your readers about your choices. The trick for good writing is, in other words, knowing how to jump.

Jerry Bannister is an associate professor of history at Dalhousie University.