Robert Paul Wolff on writing a dissertation

Here. I endorse almost every part of this, and of course especially this:

In Philosophy, a dissertation is The Defense of a Thesis. [That is why a dissertation is referred to familiarly as a thesis.]

What is a thesis? It is a proposition, expressed in a declarative sentence. Here are some examples of theses:

Contrary to popular opinion, David Hume and Immanuel Kant have almost identical views on the role of the mind in empirical knowledge. [This is the thesis of my doctoral dissertation]

God is dead.

God is not dead; he has just been on vacation.

In all situations, I am morally obliged to choose the act that will produce the greatest happiness for the greatest number of people.

Here are some examples of things that are not theses:

Kant and Hume on the role of the mind in empirical knowledge

Nietzsche's view of religion

Act utilitarianism

Each of these is a topic, not a thesis. You cannot write a dissertation defending a topic.

I take exception to the boldfaced portion of this:

In order to write a dissertation, you must be prepared to defend a thesis. If you cannot state the thesis of your dissertation in a single declarative sentence, you are not ready to write. Do not make the mistake of thinking that if you begin writing, your thesis will become clear eventually. That way lies disaster. You ought to be able to begin your dissertation with the sentence, "In this dissertation, I shall defend the thesis that p." You should then be able to conclude your dissertation with this sentence: "Thus we see that p."

The beginning and end of that paragraph are certainly right. The middle is right only

"If you are like me, and work in your head,"

Some people can work through an article-length idea in their heads. It has become clear in reading Wolff's memoirs that he really can work through a book-length idea in his head. But most of us can't.

What is true is that until you know your thesis, you cannot write your introduction or conclusion. What is probably true is that if you do not know your thesis, much of what you are writing won't end up in your final dissertation. But the last thing graduate students need to think is "you are not ready to write," full stop. Write. Try an idea out. See where it goes. Maybe you have a thesis for one chapter that you're pretty sure about; write that. Or maybe you have an objection to a standard view in the literature that you're pretty sure about. Write that. (A literature review per se isn't writing, but it does get you in the habit of putting words onto paper, and it sometimes leaves you with a more useful resource than scattered notes.)

Wolff is absolutely right to emphasize the importance of slow-but-steady writing-- a page or two a day, every day, of new writing, not revising. But I doubt that any but a handful of students are well-served by telling them to : Start on Page 1, with the sentence, "In this dissertation, I shall defend the thesis that p." This is all of a piece with my concerns above. Almost every successful dissertation I've seen written was written from the inside out, not from the beginning to the end. If you're going to write from page 1, you have have to know your thesis cold before you start writing. If you wait until you know your thesis cold before you start writing, you'll wait far too long.

When you write from the inside out, you should make every effort to write a beginning and end that make it look as if you've written from beginning to end! And that will require some revision of the middle chapters. By the time you're writing your introduction, you know your story, and you'll want to adjust the middle portions to make them fit more cleanly together and more cleanly into the story.

Wolff's absolutel right on what a dissertation is, and in large part right about writing. But I fear that parts of his advice encourage too much delay. Start writing.