Gosh, if anyone knows, tell me! Help a guy out!

Seriously, my experience has been one of years of steady development. Aside from the required courses and languages (which will differ from field to field), I got an idea in a Latin reading seminar, which while I was working towards my MA I explored in several research seminars. When I became a candidate I had a lot of further research to do, but those seminar papers are the backbone of several chapters of my dissertation.

It wasn't a matter of coming up with an idea from scratch; it's something I've been working on at one level or another for years. Doesn't make the writing easier, and there's always more reading to do, but it does feel like a natural extension of of an ongoing process.

If it's not done by the end of the year, I'll be in some real trouble.

But I know other people have very different experiences with dissertations.

I can use any sources primary or secondary I want. I'm mostly relying on Roman historians and then modern scholars interpreting them in various ways; I'm using one series of coins; I could use a lot more archaeology if I knew more about it, or linguistic evidence. I just don't have the expertise for those.

And yes, it's difficult to be original. As I'm conducting my research I keep finding published scholarship that ALMOST says what I'm trying to say, and that's a little scary. So far I've worked around it all, so it's cool; it's also nice knowing that I'm not completely in left field.

In my university, a dissertation is expected to be complete in three years after reaching candidacy.

I'm two years into my research, and am planning to take 1.5 more to finish. That would mean graduating with 7 years of grad school (MA + PhD), which is a little faster than many are able to manage in the US. We'll see what actually happens, though.

So, what sources? A dissertation's purpose is to prove that you can interpret a body of primary sources within the context of current historiography (secondary sources). So you have a topic, questions you want to answer about that topic, and a set of primary sources (documents, archaeological artifacts, old government records, historical or literary texts, etc) that you will use to answer your questions. Some dissertations focus on a specific group of texts (for example, one medieval preacher's sermons, or the records of a specific city that are preserved in their archives) and analyze this discrete body of material in terms of the research questions (for example, to see what that medieval preacher's views were on clerical marriage). Other projects don't have a single body of primary documents and have to reach more broadly to adequately answer the chosen questions about the topic (my own project incorporates archaeological reports, law codes, hagiography, and a bit of art). Some projects will choose a few discrete sets of documents and compare them to each other.

Then the analysis of the primary sources is worked into the conversation other historians are having about the topic (so, how do that medieval preacher's views compare with those of his contemporaries?). It's not a simple two-step process of reading primary sources and then secondary, though - most of us go back and forth between the two types of sources, finding new questions and possible answers in both primary and secondary material that change how we read the other type of source. But at the end of the day, the goal is to contribute something new from the primary sources that isn't found in other (secondary) scholarship, so it always comes back to the primary sources.

Finding something new isn't always hard, depending on the field you're working in. If you were going to read Augustine's Confessions or Beowulf and try to get new historical insights from them, you'd have a long slog to find your own voice amidst all the other thousands of books and articles discussing them. But for most topics, there are often still large archives that haven't been studied (and have a ton of exciting information hidden inside them), or there are documents that have only been read with certain questions in mind that don't overlap much with the questions you're asking. So unless the person studying the topic before you had the same interests and questions as you, and used the same sources, you'll probably find something that's not only new, but which also broadens the picture. Historical theory and methods are constantly evolving, too, and that means each new generation can find something different and valuable in the same materials. And there is still a lot of material that has simply not been considered interesting enough to study, yet (my study draws on 20 years of archaeological publication that hasn't ever really been systematically studied with any of the questions I'm asking - it's easy to find lots to say, and the challenge is staying focused on the important points, not finding something new).

But this is my perspective right now, after I've read most of my primary sources and found my own voice and understand the unique contributions my work is making. For the first three or four years of grad school I wasn't able to easily find new things to say, for two main reasons. The first was that I hadn't yet learned the right questions to ask - I spent all my time in my seminars reading other historians' questions, so whenever I looked at a primary source I tended to interpret it the same way others already had. And I didn't always look at the most original sources, because I hadn't learned how to find them yet. I had a few really original ideas (one of them became my Ma thesis, another evolved into my dissertation topic), but that was it. Then I started my dissertation research and spent a year reading hundreds of archaeological reports, and the written (textual) primary sources from my period, and I started to find my own voice as I learned new things that weren't structured by the secondary literature. And that was reason number two that had held me back - it had been hard to be original at first because I didn't have enough familiarity with the primary source material and enough freedom to focus just on understanding it until I had waded deeply into my own research. I think that's probably a pretty normal experience (do others agree?). But now I have lots of ideas every time I sit down and work with my material. Most of them aren't very good, but some of them are shaping up into side projects (articles, and maybe a second project to pursue once my dissertation is done). And lots of them help my dissertation's originality.

All that said, people also remind me that a dissertation's primary purpose isn't to do something so original that it changes the field. It's a journeyman project designed to show you're competent at analyzing primary sources and showing how your analysis fits in with the conversation that established scholars are having. It's your proof that you have the skills of a doctor of history, and it authorizes you to do more and better work in the future. This is all that a lot of dissertations do: they look at something new, or something old in a new way, and show how it makes our knowledge a little deeper. Of course, we all hope our dissertations will also be exciting enough that turning them into an interesting book after we have our degrees won't be too difficult. You're always trying to think three steps ahead, while still making steady progress toward the immediate goals. It's like a really long, crazily high-stakes game of axis and allies.

What sources (if any) can you or can't you use?

You can use primary and secondary sources.

Was it difficult to be original?

Not for me but it depends on your topic, if it's well-studied or not. You spend the first couple years at least mastering the vast historiography of your field anyways, so it becomes pretty clear what's new and what's not.

How long did it take you?

Years and years!