Skip to main content

As you begin to write your dissertation, you begin to understand that not all of your thoughts are nearly as profound as you imagined. Dr. Michael Munger confirms that writing, editing, and responding to feedback is usually far from glamorous. As Dr. Munger says in the interview below:

“I think if you’re not a little bit confused, embittered, and angry, you’re not working hard enough on your writing.”

But he is quick to point out that dedication to consistency and hard work pay off. Here are a few of his thoughts on keeping the perfect from becoming the enemy of the good as you write, finishing your dissertation sooner rather than later, finishing graduate school (on time), and getting a job afterward. The podcast below contains the full interview with Dr. Munger.

Transcript: 10 Truths to Help You Write Your Dissertation

Jeanne Hoffman

Welcome to this Kosmos online podcast. I’m Jeanne Hoffman. Today our topic is dissertation writing and research agendas, and my guest is Mike Munger. Dr. Munger is a professor at Duke University in the political science and economics department and the School of Public Policy, as well as the director of the joint UNC Duke Philosophy Politics and Economics Program. Welcome, Dr. Munger.

Mike Munger

It’s a pleasure to be on.

Jeanne Hoffman

It’s a pleasure to have you. Our interview today is based on your essay, “Writing your Dissertation and Creating your Research Agenda,” which was originally written for IHS’s guide for graduate students, Scaling the Ivory Tower. Where do you place the importance of these activities in relation to other graduate school responsibilities?

Mike Munger

it seems to me that people, when they’re in graduate school, have to recognize that their chief job is to redirect their energies from where they were as an undergraduate to where they will be as a professional.

A graduate student is somewhere between an undergraduate and a professional. The people who succeed are those who make that transition faster. What’s disturbing about it is that faculty often don’t really tell you that. This is something that you have to pick up on your own.

The problem is faculty would like you, and I’m no different, faculty would like for you to validate their pathetic lives by taking their classes very, very seriously. You’re not going to get a job taking classes. You’re going to try to get a job as an independent researcher who has their own ideas. Who’s able to make his or her own ideas clear to someone in writing.

The sooner you start working on that, the sooner you’re going to be able to make that transition. And the easier it will be to write your dissertation.

I’ve noticed that in the second or third year of graduate school there’s something that’s approaching an inversion where the people who were the real stars in the first and second year, who managed to make the faculty smile and pat them on the head and say, “Good student. Here’s a biscuit.” These are the ones who, after their second or third year, are thinking, “You know, I’m going to take more classes.”

Whereas the other people, and frankly I was one of the other people, were thinking, “You know, classes are sort of boring. I’m going to try to write some stuff on my own.”

Those are the ones who end up succeeding.

The stars in the first and second year, a lot of them don’t finish a thesis and they don’t get a job. Be of good cheer, if you feel like an odd ball, if you’re someone who’s working on research and taking the classes. Getting through the classes is fine. But you’re not going to get a job taking classes. You’re going to get a job doing your research.

Start working on writing. Start working on your dissertation as soon as you can.

Jeanne Hoffman

In your essay, you also listed 10 truths about writing. I just want it go through them all briefly because I think each of them is a great piece of advice for students to keep in mind when they’re writing. The first is writing is an exercise. What do you mean by that?

Mike Munger:

Suppose you were going to run a marathon six months from now. You wouldn’t wait until the night before the marathon and train all night. What you would do is run a little bit at a time. Now, some days you’d go slow. You wouldn’t feel like it, but you would recognize that your performance in this marathon is going to be based on having practiced in situations where you developed the lung capacity, the ability of your muscles in your legs to perform.

Writing is the same way.

If you write every day, some days it’s not going to be very good. You’re going to throw a lot of it away. But when it comes time to do the marathon, you’ll be ready to write the dissertation that you have to finish in order to leave graduate school.

And as I often tell graduate students, “How will be ever miss you if you won’t go away?”

The key to this is to treat it as an exercise. Writing is something that you do every day, and like training for a marathon you will get better. The fact that some of what you do is a waste isn’t the point.

The point is to develop the skills so when the dissertation comes up, you’ll be ready.

Jeanne Hoffman:

You also mentioned “set goals.” What type of goals should graduate students be setting?

So many people have this fetish about input based metrics. I worked for three hours today. Yeah, well you didn’t do a thing. You need output based metrics. I wrote three pages today. That’s a goal. Not I went to the office. Think of all the times that you as a graduate student said, “Well, I worked for six hours today. I read a book.”

No, what you have to do is set a goal of writing. Have an output based metric. Focus on that, and once you’ve done that, yes, then go do something else. But make sure you have an output that you produce every day. Not input.

Nobody cares about the labor pains. They just want to see the baby.

Jeanne Hoffman:

Now, this other one sounds really profound, but could you explain it to me? It says “write for the ages.” What does that mean?

Mike Munger

This comes from an experience that I had that was pretty darned embarrassing.

I was interviewing in 1984 at George Mason University, and one of the people I was interviewing with was James Buchanan. Now, this was two years before he won the Nobel prize in economics. He won it in 1986. This was 1984, but he was pretty scary even without a Nobel prize. The first question he asked me in our interview … (Note that this was a job I really wanted. I so wanted to be at George Mason. This meant the world to me.) His first question was, “What are you working on? What are you writing that somebody might read 10 or 100 years from now?”

I had nothing. His point, and he says this pretty often, is you ought to be working on something that people are going to want to read years from now. Because if you know when you’re working on it that’s really of no consequence, why are you spending your time on it? How are you going to stay excited enough about it to be able to produce a decent quality piece of work if you already know that it’s trivial?

Now, a lot of the things that you think are interesting at first turn out to be not as important as you had hoped, but you have to have some aspiration to write on the kinds of questions that people care about and that might conceivably they would want to read in 10 or even 100 years.

Jeanne Hoffman

Next you have, “give yourself time.” Do you mean time to write?

Mike Munger

I certainly do mean time to write. We’re used to being patted on our heads for our prolixity–and the fact that even the night before a paper was due we could produce a decent quality paper.

Well, if you try to carry that over into graduate school or into your professional career, you’re going to fail. Look at something that Adam Smith wrote or that Ayn Rand wrote (or whatever writing that you care about), they didn’t sit down the night before it was due and write the whole thing. They wrote it then they went for a walk, had dinner, talked to someone. Then they wrote it again.

They worked on it over and over again.

In a way, this is the same thing as write for the ages. This is take your work seriously enough to treat it as something that’s worthy of your full attention over an extended period.

Not I work really well under pressure. The fact is nobody works really well under pressure. You’re just smart enough that you’ve been able to get away with it up until now.


Jeanne Hoffman

Speaking of people leaving things to the last minute, you have as your fifth truth, “edit your work.” Which I know a lot of undergraduates don’t get to because they do their paper overnight. But what does this mean for graduate students who have an extended amount of time to work on their papers?

Mike Munger

A lot of people don’t like the idea of editing their work, and I think there’s two reasons. (Maybe they’re related, but students always give two different reasons.)

  • One is it’s boring to edit (especially as you write your dissertation), and it’s more interesting and fun to work on something new.
  • The other is, you know, they hate to waste anything that they’ve written.

Once you get used to editing, it’s really quite liberating. Try to find out if you can shorten everything into something that’s better.

When you’re writing, often less is more.

The first paper that I published came out in the journal public choice in 1984. Started out as 22 pages of calculus and proofs. When it came out in the journal it was two-and-a-half pages, and it had two short equations. It bore no resemblance to the paper I started out with, but it was much better.

That first paper was unpublishable. If I had sent it to the journal it would have just been turned out, and then I would have thought, “Oh I was born under a dark star.” No, I was too lazy to edit the thing.

If you don’t like editing your own papers, find another graduate student and switch as you.

It’s often much easier to edit and find the mistakes or infelicities that other people have made. One of the things that we’re all good at is criticizing, so find a writing buddy and switch papers and edit each other’s stuff. There’s one other thing about editing that Deirdre McCloskey always says. Deirdre McCloskey’s claim is “Let editors edit.”

Which means that if someone, an editor or a friend who you’ve gotten to edit your paper, volunteers that there’s a problem with that sentence, there is. Nobody cares what you think. The fact is that when someone else looks at that and says, “You know, I don’t understand this.” Or, “I think you should reword it.” You should. Don’t get defensive. Just do it.

Let the editor edit.

Jeanne Hoffman:

Okay, now here you have, “pick a puzzle.” That is a puzzle to me. What do you mean by that?

Mike Munger

There are a lot of ways of making a paper that you’re working on seem, and in fact be, more interesting.

One of the key ways to do that when you’re setting it up is to choose one of the classic kind of puzzle formulations. Some examples that I could give are, well, we start with the problem of, “there’s a lot of people that have noticed empirically X happens, but the theory says that Y should happen. Why is it that our theory implies something different from what we actually observe?”

Another famous and common kind of puzzle that’s quite useful is, “There’s this theory about phenomenon X, and there’s another theory about apparently very different phenomenon Y. It turns out that the same underlying explanation accounts for both of these apparently very different things.”

Well, there again, if you start with that, you have the readers interest. It’s a way of organizing your discussion. It’s a way of getting started. A lot of people have trouble writing the first page of their paper, so we sometimes jokingly say, “Okay start with the second page!” It’s hard to set the thing up.

Using the puzzle (even if it seems rather mechanical at first) is a good way to get past that first hurdle of presenting your work. Tt has the bonus of grabbing the reader’s attention.

Jeanne Hoffman

Now, your seventh truth seems to tie into your fourth truth, your “give yourself time” one. Your seventh truth is “schedule time for writing.”

Mike Munger

The reason that it’s different is the “give yourself time” means that you start long before it’s due.

Scheduling means that you have to think when is my most productive time. Am I a morning person? Am I a night person? Then make sure that you schedule your writing during your most productive time. For instance, take classes or teach them (as the case may be) at a time that doesn’t conflict with your most productive writing time.

Now, it’s perfectly true that when I teach I get enough of an adrenaline rush. It happens that I’m a morning person, so I always schedule my writing in the morning and then I teach in the late afternoon. The late afternoon, a lot of times, I might need a little nap. I’m a little tired after lunch, and I’m kind of nodding off, but when it comes to teaching I get a rush of adrenaline. It’s like having two different peak times because you spend your writing time when you’re naturally most productive. Then you do your teaching at a time that otherwise would be a down time. It makes you enormously more productive.

The problem that I see is that people say, “Well, I’m so busy, so I’ll write when I get a chance.” It’s a residual category. It’s what I do after I’ve done everything else.

You have to turn that on it’s head. Schedule your writing first, and make everything else fit. First and foremost, if you’re going to succeed, you need to be a writer.

Jeanne Hoffman

Now, I want to tie your eighth and ninth together because I think one flows from the other one. Your eighth truth is “not all of your thoughts are profound,” and your ninth one is “your most profound thoughts are often wrong.” Why aren’t all my thoughts profound? I think they are.

Mike Munger

You know, they are as long as they’re thoughts. Actually this is going to tie into the next one also, so let me tie all three together.

Jeanne Hoffman:


Mike Munger

When I think of an idea, or when I talk about an idea in a bar to my friend, we’re having cigarettes and beers, and it’s 1:30 at night. You know, you think that’s really clever. That’s something.

The problem is I sit down to write it, it turns out to be much more complicated than I thought it was.

When I was in graduate school, and I was a beginning assistant professor, would keep a list of what I thought were interesting ideas. Half of them by the time I had worked on them for a day turned out to be not that important. A lot of the things that seem interesting and important are not as important once you start to write them.

That means that you need to start to write them as soon as possible. You learn about your own thoughts by trying to write them. Not by just repeating them and having other people say, “Oh, that’s a clever nugget.”

Writing is how you learn whether your ideas are profound.

The ninth one that you mentioned was your most profound thoughts are often wrong.

I have a friend who made this suggestion that you should kill your children. What he means by “kill your children” is after you finished a paper, and you think it’s almost ready to send to the journal, go through it and underline the three most clever and profound sentences, and delete them because you’re bored with this paper. You have read this paper so often. You thought about it so much. The three sentences that you think are the best are almost all non sequiturs. They’re usually something that they have nothing to do with the paper, or they’re an ad hominem attack on someone you should leave alone. You think, “Oh I really got that guy.”

Go through and take all of those “profound” thoughts out. They’re the things that are going to make referees angry or that are going to side track the reader.

You’re no longer competent to judge whether or not these thoughts are profound.

This leads to the tenth truth, which is that everyone’s unwritten work is brilliant.

In my essay, I try to conjure a figure that we’ve all met. Usually, it’s a like seventh year graduate student, and it could be a fourth year assistant professor who hasn’t published anything, but he’s extremely glib. He generally has a cigarette and probably a black turtle neck sweater and an imported beer. Probably a Hefeweizen from Germany. A person that excludes good taste. He holds court in this bar or in this coffee shop with his cigarette and tells you a three-or-four-hundred-word summary of what he’s going to write his dissertation or his next book on in the case of a professor.

You think, “Wow that’s so interesting.”

Then he asks you, “What are you working on?”

You’re a little confused because the chapter that you’re currently working on, you’re not quite sure it’s going right. The direction isn’t at good. You finished another chapter, but you’re not sure it hangs together, and so you sort of stutter. From this guy, you get a smug smile, “It’s hard, isn’t it?”

Well, the truth is that guy is a poser.

He’s not actually working on anything. The reason that his glib superficial description of his work is so impressive is that he’s been saying the same thing for five years.

You are the winner here. You’re the one who’s actually working on something. You’ve written several pages today. You wrote several pages yesterday. You finished a chapter in the past month.

It’s hard to know because you’re in the middle of a project. Beware of the people that have this description of their work that is practiced and sounds like it’s good. People who are writing are often confused, embittered, and angry.

In fact, I think if you’re not a little bit confused, embittered, and angry, you’re not working hard enough on your writing.

This guy’s work is unwritten. That’s why it sounds brilliant. Everyone’s unwritten work is brilliant. You have to encounter how hard your idea is by writing it.

Jeanne Hoffman:

If you could give grad student just one piece of advice over all about their dissertation what would that be?

Mike Munger:

I’m going to give three. I’m a professor. I have a hard time giving just one.

Jeanne Hoffman:

Okay. I’m a lawyer, so I can negotiate, so that’s fine.

Mike Munger:

I’ll try to be brief about it, though.

The first is that graduate students elevate the dissertation in their mind to the status of something that’s enormous.

In a way, that absolves them of responsibility for not finishing it very quickly. You should think of the dissertation as being a sort of glorified class requirement. It doesn’t have to be publishable. It doesn’t have to be close to publishable. In fact, when you start to think of your dissertation what you’re doing is saying, “I have to write something that four or five people who may not like each other very much have to all sign off on.”

Which brings me to the second point. How good does your dissertation have to be?

Well, one of my professors told me a good dissertation is a done dissertation, and a done dissertation is good.

What that means is that as soon as you get it finished then you’ve already accomplished the first piece of advice was don’t elevate this to having the status of some gigantic important book. You recognize it’s just a glorified class requirement. Now you look and there’s four different people on your committee. Maybe they don’t like each other very much. Maybe they don’t even talk to each other, and so they communicate only through you. Where you bring a draft to one, and they give you corrections. You make the changes, and then another person says, “No, no. Change all that back.”

Well, what you need to do is have them talk to each other. But what you really need to do is recognize that a done dissertation is good. Just finish what they say. Don’t let them use you as a pawn in a personal war that for them goes back ten years. Just get the work finished, and once you’re in a position to have it done you can work on a book that then you won’t have to please four masters who are making different demands on your time.

The way to do that is the third piece of advice which is don’t read. Write.

I ask my students, the students who work with me on their dissertation, to put up a three by five card in their work cubicle or their library carrel or wherever they’re working, that says, “Don’t read. Write.” Because writing is an output-based metric. Reading is an input-based metric. You should always avoid input-based metrics.

“I read two books today” means you did nothing. If you wrote something then you had an accomplishment. What are you supposed to read though? The answer is let other people, including your faculty advisors, be your research assistants. You can hire them as research assistants, and you don’t have to pay them. You give them a draft, and they say, “Oh, here’s four things you should cite.” Well, go look up precisely those four things. Cite those four things, and add them to your references because then you’re not reading to decide whether or not it’s relevant. You already know that it is. You have used your faculty advisors as unpaid, really smart, highly-trained research assistants.

Make the system work for you.

Jeanne Hoffman:

That’s brilliant. You mentioned that students make the mistake of elevating their dissertation much higher than it should be. What are some other common mistakes that you see students making as they work on their dissertation?

Mike Munger:

I guess an elaboration of that first one is the one that’s most important which is after I’ve been working on it for six months, and I haven’t written anything, the explanation has to be that it’s really hard and really profound. After 18 months or two years, I knew a guy who worked for seven years on his dissertation. He had 600 pages written, but since he’d been working on it for six years it had to be something enormous. Now, he could have turned in what he had, and it would have been fine, but it wouldn’t have been fantastic. It wasn’t good enough because he couldn’t explain to himself why he’d done it for six years. He ended up not finishing, although he had 600 pages written.

Well, believe me, by that time the faculty really just wanted him to go away. Write an introduction, write a conclusion, hand it in. Say, “That’s enough. I’m done.”

Jeanne Hoffman:

What else separates a student who gets their dissertation done from a student who doesn’t complete their dissertation on time?

Mike Munger:

I would say that completing your dissertation on time largely just requires you sitting down and writing out a timeline that starts with finishing. Then give yourself reasonable amounts of time to do all the things you need to do in the middle, and work backwards to now. Now, I finished my dissertation when I was 25. I wrote the whole thing in six months. Was it good? No. It was terrible, but a good dissertation is a done dissertation. After I finished it I was able, on my own without the infighting and bickering of faculty members looking over my shoulder, to be able to fix the problems, and send it to journals, and I got it published.

Think of it as a job. Think of it as something that has a schedule that you can produce. Rather than “I’m working on this, and my lack of productivity is a sign of my profundity,” which is a trap many of us fall into.

Having a deadline is an important part of becoming a professional because you have to learn to generate internal deadlines. Journal editors do not have deadlines. They would prefer that you don’t send your paper to them.

The editor of the philosophy journal, economics journal, political science journal, they’re not going to call you and say, “Hey, are you going to send that paper in?” They hope you don’t. You have to generate your own internal deadlines. You might as well start doing it now. Think of it as a job. Have deadlines, and meet the deadlines. If it isn’t perfect that’s fine. The faculty will tell you what you have to add.

Jeanne Hoffman:

In this timeline that you’re talking about (other than sending your work to journals) what other steps should people consider in making it?

Mike Munger:

I guess I would suggest you send papers to journals before a lot of people actually send their papers to journals. Let me see if I can explain that. For anyone who does computer programming, you know something called “machine intensive debugging.” Machine intensive debugging means that I don’t stare at the program and try to figure out what errors of programming or logic are in it. I send it to the computer, and it comes back with error messages.

Now, there are no obvious mistakes. I don’t send it to the computer so that it bombs in the first line. I do the best that I can to make sure that it doesn’t immediately bomb. Well, you can think of journals the same way. Since I, myself, was the editor of the Journal Public Choice for five years, let me say immediately that doesn’t mean send in half-finished papers. It means get the paper to a certain point of being good and then finish it in the sense that you edit it. You correct the references. You make sure the title page is right. There are no typos. There’s no hanging widow titles. It looks like a professional paper.

Then the referees are going to tell you what you should work on.

You need to have a portfolio of papers. People are surprised when they first come out. You know, they’re teaching, they’re spending time trying to get their dissertation into shape, and two or three years pass by. Start sending papers to journals right away, and that sort of generates it’s own momentum, its own logic. You get the paper back, and you’ve got comments. Basically these are like error messages from the computer. Fix those. Send it back again. You’ll be publishing papers before you know it.

Send the paper out before you think it’s perfect because the editors aren’t going to think it’s perfect anyway. Even the paper you worked on for three years is going to come back with error messages.

Jeanne Hoffman:

This is my final question. Do you have any advice on how to generate really interesting ideas that spark quality papers?

Mike Munger:

I do, and the answer is write a lot of different papers, and recognize that not all of your ideas are as profound as you thought they were. As we’ve already discussed. When the paper comes back from the journal … This has happened to me several times. My most cited paper is the 1986 American Political Science review paper with Arthur Denzau. We had sent it to four other journals before it was finally accepted. When the paper started out, it wasn’t very good, but we got comments from referees that said, “You know, this is stupid, but if you were really going to do this here’s what you could go and do.”

Well, we took those seriously. By the time I had worked on it for two and a half years, it had been turned down at four places. We had gotten comments from twelve different referees. The paper was excellent, and it was partly because of the important ideas that we had gotten from machine intensive debugging.

From having those very smart anonymous referees make suggestions.

Again, you can use smart people as your unpaid research assistants as long as you keep at it, and take comments seriously. Let the editors edit. Let the referees tell you. If they think there’s something wrong with a passage, there is. Don’t be defensive about it. Fix it.

Jeanne Hoffman

Well, thank you very much for joining us, Dr. Munger and for your great advice.

Mike Munger

It was a pleasure.


Here is the timeline for our application process:

  1. Apply for a position 
  2. An HR team member will review your application submission  
  3. If selected for consideration, you will speak with a recruiter 
  4. If your experience and skills match the role, you will interview with the hiring manager
  5. If you are a potential fit for the position, you will interview with additional staff members
  6. If you are the candidate chosen, we will extend a job offer


All candidates will be notified regarding the status of their application within two to three weeks of submission. As new positions often become available, we encourage you to visit our site frequently for additional opportunities that align with your interests and skills.