On Writing an MSc Dissertation

Bryan W. Roberts

LSE Philosophy, Logic & Scientific Method

Enjoy the Ride

Philosophy is infectious, fun, and deep. Indulge in sumptuous amounts of time for thinking and taking notes. Explore big ideas. Think outrageous thoughts. Draw on the skills you already have, and build up some new ones along the way. Your MSc is an opportunity to do something great.

Don't be shy about discussing your ideas and plans with your friends, peers and teachers. Book an office hour with your degree coordinator when you're ready to chat about some possible topics. For most people, philosophy is done best as an interactive discipline. At LSE you'll see that this is very common and natural.

Here are some thoughts to guide you along the way.

The "sweetheart" research phase

Serenade your questions until you find your match

You're on your way to becoming immersed in interesting ideas, questions and problems in the philosophy of science. Your aim is to turn that flow of fascinating topics into a problem that you can state and solve in a 10,000 word dissertation. How do you do that?

I don't know of any algorithm that will do it. If you find one, please tell me! But I do know a few strategies that are very reliable.

The most important thing is to focus on topics that you find really interesting. If you fall in love with your topic, writing your dissertation will be an absolute pleasure. On the other hand, if you choose a topic that it pains you to work on, then your dissertation writing will be like a three month trip to the dentist.

That said, you should adopt an important moderating principle: only pursue those interests that lie within your abilities. If you're interested in the AdS-CFT duality in string theory but don't know the first thing about physics, then you may want to switch to an easier topic.

The final thing to remember is that this stuff takes practice. Isolating a worthy topic in philosophy is a skill akin to playing the violin: very few people just nail it the first time. Explore lots of ideas. Attempt to solve lots of problems. Take your formative essays seriously and write lots of papers. The most reliable roads to success, I fear, are paved with blood sweat and tears.

As you start your exploration, don't wait to start thinking about what it means to write a dissertation. Have a look a previous dissertations. Check out the MSc handbook. And have a look at the wealth of online resources out there. This dissection of a philosophy paper, for example, illustrates many important and conventions.

The analysis phase

Get to know your question deeply

Eventually, you'll settle on a topic and a question that you would like to see answered. That's when the serious philosophical analysis begins.

In the analysis phase of your research, you'll start to understand your problem more deeply, and eventually produce a solution. This may sound ineffable, but don't worry! You can do it. But you'll most likely find that it takes regular hard work.

State the question precisely, and think about it. Then do it again.

Your question may be one that has come up in your courses, or some variation, or a question that you've stumbled upon from another source. However you found it, make it the centre of your attention. Write it down on a napkin at lunch. Explain the question to a friend. Don't let the question stray far from your thoughts.

Like a date, you can't just go through the same motions over and over again or you'll get nowhere. Be creative. Look for slight variations or equivalent ways to ask your question. What do all the terms in the question mean exactly? What would it take to clarify them? What are the possible answers that one might give to your question, and how might they be justified?

Read (and re-read) relevant texts with your question in mind

Your course readings may be relevant to the question, so think about your question while reading and re-reading. Seek out as much as you can about what others have said about the question, and write down any further questions this may lead to. Summarise the arguments of other authors in your notes and study them. Take note of their strengths and weaknesses, and where you agree and disagree. It may help to go over your lecture notes. Although you will be constructing your very own argument in your dissertation, it may help your case to discuss the argument of another author as well.

You may search for other sources, but almost without exception they should be 'scholarly'. Google Scholar is an indispensable resource. Another crucial strategy is to check out the bibliography section of readings you think are important to your question, and follow up on the titles that seem relevant.

By the end of this process, you should have formulated your view about how to answer the question.

Write down or revise your 1-2 sentence thesis statement.

When you've started to come to grips with how to answer your question, it's time to start sketching the thesis of your dissertation. This may change down the line, but it's important to keep at least a preliminary thesis in mind in order to organise your thoughts.

Your thesis, and only your thesis, is what your dissertation will be arguing for. By keeping it in mind you'll make sure that your research is not getting off track on a tangent. Your thesis should be precise and clear, and demand an argument in order to establish it.

The thesis of a dissertation should be expressible in no more than two sentences, preferably one. Your thesis might have two parts, such as a negative claim and a positive claim. But it always be short and sweet. Here are some sample thesis statements from previous dissertations.

This dissertation proposes a new way to justify why patients with Huntington's disease ought not to naturally conceive, and that a light nudge policy would be a reasonable way to discourage this behaviour.
I will argue that the supposition that a physical supertask can be used to compute a non-Turing computable function leads to serious conceptual confusion, if not outright inconsistency.
I argue that that the reason prediction markets fail comes down to how they undermine the core requirements for reliable collective wisdom - specifically, independence - such that prediction markets fail doing the best job they can at eliciting and aggregating the information that is avialable to aid in prediction.

Organize the support for your thesis in argument form

This is the most important part of the process — give yourself plenty of time! Your thesis must be accompanied by an air-tight argument. If you can't provide one, go back and modify your thesis.

You might begin by brainstorming all the claims that you think justify your thesis. Then try to write down an argument in correct premise-conclusion form. To take a famous example, here is the premise-conclusion form of an argument suggested by Socrates, for the thesis that holiness has nothing to do with god.

  1. Holiness has nothing to do with god unless either (a) things are holy because god loves them, or (b) god loves things because they are holy.
  2. Premise 2. It is false that things are holy because god loves them.
  3. Premise 3. It is false that god loves things because they are holy.
  4. Conclusion. Therefore, holiness has nothing to do with god.

If you have formulated your argument correctly, the premises will necessarily imply the conclusion. Your remaining task is then to provide as much support as possible for each premise. Make sure you jot down the best support you can think of for each of your premises. Now you're prepared to start writing.

The writing phase

Turn blank pages into a dissertation

The aim of a philosophy dissertation should be to present an argument for an interesting philosophical thesis in a clear and compelling way. Nothing more, nothing less.

You can forget most things you learned in a creative writing class. A philosophy dissertation will be more like a business proposal. Your aim is to persuade, using the clearest and most convincing arguments you can.

Sketch an outline of you dissertation

This will include an introduction, a development of relevant background material, and the main body in which you present your thesis and your argument. There may also be a section in which you discuss objections. The last section should be the conclusion.

The introduction to a philosophy paper is often written last. That's because it's hard to introduce something that hasn't been written. You should include a brief discussion of the issue and the question, a statement of your thesis, and a few comments about how you will argue for that thesis. It is also recommended that you commit a paragraph to outlining your paper before you begin.

The background material and main body should introduce any background material needed to understand your question, discuss the context of this question with respect to existing literature, and present your thesis and argument. It is advisable to divide it up into subsections to keep things nice and organised.

The conclusion should contain a restatement of your thesis, and (optionally) some brief comments on loose ends, such as what kind of future work remains to be done on this topic. It is often nice to have a brief section on open problems that remain to be solved as well.

As you construct your outline, you may include relevant quotes from your notes and from the relevant literature. And make sure you spend plenty of time thinking it through. A careful and detailed outline will vastly decrease suffering and increase efficiency when it comes to writing the dissertation itself.

Write the body of your paper, then the introduction, then the conclusion

Begin with the main body, because this is the most relevant and difficult part. You should make very clear what the premises of your conclusion are, and how they imply your thesis. Spend plenty of time explaining your support for each premise. Then, when you turn to write your introduction and conclusion, you will be able to more accurately summarise your argument. Follow your dissertation outline and your argument outline during this process. And don't forget to include a References section and cite every piece of information that is not utterly trivial common knowledge!

Some day, the moment will come when you have produced the fabled first draft of your dissertation. It's very important that this moment is not the night before the deadline.

Set the paper aside for 2-5 days, then revise

This can be the difference between good papers and great dissertations. If you give yourself time to digest what you wrote down and the read it with fresh eyes, you will almost always find ways to improve your work.

Avoiding pitfalls

Skirt some common errors

1. Get started early

Sometimes you can wait until the last minute and still do a decent job. This is rarely the case in philosophy. Philosophy requires time to reflect.

2. Take notes as you read

Write down what you think about the relevant scholarly material, jotting down your ideas as they come to you. Have these notes in front of you when you write your outline.

3. Produce a clear answer to a clear question.

Everything that you write should be aimed at addressing the question. Your thesis is your answer to that question. Your argument supports that thesis. Almost everything else is irrelevant!

4. Assume your audience is rather dull

You might think that your teachers and professors, having read a great deal of philosophy, will immediately pick up on your hints and suggestions. We aren't that smart. Write your paper as if you are explaining your point to an utter buffoon. Explain every point you make as completely as possible.

5. Don't try to cover too much

A great philosophy dissertation will state something interesting but humble, and then argue well for it. Don’t try to solve the biggest problems in philosophy in a 10,000 word dissertation.

6. Say what you are doing often

Keep the reader informed as to what part of your argument you are in. The purpose of every paragraph should be insultingly obvious to the reader. For example, you can say things like:

7. Use overly simple and concise language

There is no need to use language that is long or overly refined unless you absolutely have to. Slang is not appropriate; however, your sentences should be short and easy to read. You can refer to yourself using "I" freely in a philosophy paper, especially to explain where you are in your argument.

8. Don't say anything you can't support

Every sentence you write that isn’t absurdly obvious should be supported, and not by just saying, "I believe that X." You must persuade your audience, by giving an argument, considering alternatives, giving examples, citing sources, etc. Avoid sentences like "Since the dawn of time, mankind has...." It is rather unlikely that you will be able to adequately support a sentence like this.

9. Focus on scholarly references

Terms in a philosophy class may have precise or technical meanings that are not the same as the ones you will find in other fields or in a dictionary. Stick to scholarly sources, and ask one of your teachers if you're not sure what those are. Wikipedia is a fine place to start learning about something, but it is not an adequate reference for a philosophy paper.

10. Always cite your sources, but paraphrase, and quote sparingly

A few quotes in a five page paper is fine, but beyond that, stick to paraphrasing. But don’t forget, you must cite every word or idea in your paper that is not your own. It doesn't matter which style guide you choose (Chicago, MLA, etc.). Just pick one and stick with it.

On fads in philosophy

Avoid the higher-order analysis of Chmess

Identifying and avoiding fads

It is healthy for everyone interested in philosophical research to have a look at Daniel Dennett's comment on Higher-order truths about chmess. Have a look. Then come back for a few comments.

Part of what makes philosophy so endlessly interesting is that you can find philosophical puzzles in just about anything. But sure enough, this also means that you can find philosophical puzzles in a baloney sandwich. As a result, many philosophers have a tendency to find themselves literally or figuratively talking baloney.

What do we mean by that? As an example, take the game of chess.

chess

Now, instead of adopting the rule that the King piece can move one square at a time, let's adopt a rule in which the King can move two squares at a time. We'll call that game Chmess.

Chmess is not a real game. Dennett just made it up. But it is clear that one could imagine a great number of puzzles that might arise. Under what circumstances is checkmate possible? When is it impossible? An entire theory of Chmess could be constructed, in principle. And it would all be perfectly pointless.

It is a well-known sociological phenomenon that, even when one is basically producing a lot of baloney, this may continue with great vigour when enough participants are engaged. One could imagine this happening with Chmess, with whole groups of enthusiasts studying its subtle and complex properties. This phenomenon can happen from time to time with philosophical fads as well.

This is not a problem merely for philosophers, but really for all academics. As Dennett points out, even the experimentalist is often faced with a real problem when an expensive apparatus is no longer really anymore. A community might well choose to continue accumulating lists of data that doesn't matter at all, before finally retiring the expensive machine once and for all.

But we have an advantage as philosophers: we can decide that something is a fad and just switch topics, without losing any money on an experimental device. What is important is that we sometimes stop and think about this question.

Dennett proposes two strategies to test for philosophical fads.

  1. Try explaining it to someone outside of philosophy.
  2. Try teaching it to a classroom of smart undergraduates.

If after considerable effort both groups just don't get it, then you may be dealing with a fad. Sometimes it takes an unbiased outside observer to determine that a topic is pointless.

This should not at all discourage you from trying to understand complex things. On the contrary, you are bound to run into complexity: you should explore the best work within your philosophical interests, and good philosophy is sometimes very complex.

However, it may help, as you go forward, to remember that some pretty silly fads are out there lurking. They'll quickly come under elaborate scrutiny by many brilliant teachers and students, and then forgotten just as quickly.

Elements of a Good Philosophy of Science

At LSE, we find that two further criteria can provide excellent indicators of the success and relevance of a problem in philosophy:

If the answer is "Yes" to either, then you may be on track to a promising topic.

Here are some further positive features that some of the best philosophy of science has often done. It is not an exhaustive list, but rather illustrates some of the most common traits of a great work of philosophy of science. See if you can think of examples of each among your philosophy of science readings.

  1. Correction of a standard story. A standard scientific episode tend to be described as a parable that is more fiction than truth. Philosophy of science helps inform both the history and the philosophical analysis of an episode by correcting such a story.
  2. Explication of a central concept in science. Sometimes philosophy of science points out how to make a vague concept in science more precise in an interesting or illuminating way.
  3. Clarification of traditional philosophy (e.g. epistemology, metaphysics, ethics) using results from science. For example, old philosophical views about the passage of time, the composition of matter, or the nature of the mind might be sharpened or corrected when properly informed by modern science.
  4. Characterisation of a theory's foundation. Sometimes the structure or basic elements of a theory are not clear, or may be characterised or interpreted in multiple different ways. In such situations, philosophy of science may help by giving a clear characterisation of a theory's foundation.
  5. Analysis of a paradox. Sometimes an example in philosophy or science may appear paradoxical, contradictory, or otherwise intractable. In such cases, philosophy of science can help by showing how to overcome the paradox.
  6. Synthesis of philosophical views and scientific results. Sometimes philosophy and science are actually developing similar ideas in tandem without knowing it. Philosophy of science can sometimes bring them into fruitful contact.
  7. Characterisation of epistemic or metaphysical limits. There are some things that we cannot do or cannot know, given an appopriate analysis of our best scientific theories. Philosophy of science can often clarify those limitations precisely.

Thus, there are paths to choose and obstacles to avoid on the road to your MSc dissertation! As I said in the beginning, philosophical research is an art akin to playing the violin. A good deal of regular practice will be required. But when you finally arrive, and with that long-awaited dissertation in hand, I hope you'll find that it was a very rewarding journey indeed.


Bryan W. Roberts, LSE Philosophy, Logic & Scientific Method