博士生涯第二年，終於要進入正式寫作的階段。原先以為博士論文應該先處理「知識論（epistemology）」與「本體論（ontology）」，釐清我所認識的世界觀，到底是屬於主觀主義（subjectivism）或客觀主義（objectivism）的脈絡，我所謂的knowledge來自何處，然後再正式進入我要研究的主題。另一方面因為我不是本科系出身，社會學理論唸得不多，經常覺得超心虛，不知道怎樣才能完成一本社會學的博士論文。

最近一次跟老師meeting，卻被老師五雷轟頂--「最有趣的東西是你的研究，不要雜七雜八寫一堆文獻回顧」、「文獻回顧是用來點出你所要強調的重點與論文脈絡，最差的論文是文獻跟實證資料連不起來」、「口試委員想看的是你的發現、你的貢獻、你的想法，才不想看你對某某大師作品的註解或解讀」。
我把老師講的幾個重點整理在這裡跟好朋友分享，A、E代表我的兩位老師。

A: Traditional ideas of literature reviews are not very helpful often. You should orientate the literatures that you review to the problems, you want to engage with, you want to show why you are dealing with that literatures.

Q: Is it 'ethnomethodology' that I am doing? Or is there other epistemology? Should I explain first about what I think the knowledge is in my thesis?

A: One of the way you can defence this empirical orientation is that most of the theories are related to the west anyway, so we must be cautious about jumping in, applying it. But in anyway, one of the dangerous, there is a common bad model of thesis where you get huge discussions of theories at the beginning and then you have empirical study of Lancaster or what, and then there is no joint between them. And it is much better to have, as with methodology, an introduction to the simplest point about it at the beginning and say that you are going to be developing these and they will become more complex. In the context to the interpretation of the actual data, you sort of work it out—you move from the simple to the complex, you take the reader through it; but you don't want to give them full your newest, complexest interpretation right from the beginning. So I agree that you don't have to put all you want to say about methodology at the front, but you have to put some in the front which orients the reader. You can elaborate it at some proper point later on. Pretty much the same as theory I think.

Q: when I read other phd theses, they always say that they follow a Marxist approach or Foucaultian approach and explain how they view the world.

A: but generally they can't stick with one view of world through the way anyway. That's a bit rigid I think. And that is not necessary a good model.

E: You certainly don't have to do that.

Q: Does a phd thesis need to clarify the points that we choose, for example between objectivism and subjectivism, things like that, in philosophy?

A: No not in philosophy, but you are looking into both objective circumstances, like one find themselves in the yard, and what they think, and what we do. You are dealing with both, the relationship between them. Bourdieu writes a lot about that. I think that could be put very simply.

E: The interesting thing is your work. The least interesting theses are the ones that don't realize that or don't make that enough of the new and interesting work that you have done; and, in a way, pretend it grounded, or important or, walk around loads of stuff. Anyway you are not good at doing that either.

A: One of the common problems in many students is that they want to write an extended essay on all the theories which they read, a few little comments, and it is hard to get them put their own data in their own interpretation, their research. External or internal examiners will want to know what you think, what they can learn from you, what you think, what you find out, rather than your footnotes to Bourdieu or what ever.