I was reading a final draft of a thesis written by one of the doctoral researchers I was working with. I’d just started and the text was going along very nicely indeed until I reached the end of the first chapter. There was a paragraph on the final page which just didn’t gel. It read as if it was written by someone else entirely – no flow, more tentative. When I asked the researcher what this paragraph was about, he laughed and told me that it was something he had cut and pasted from work he’d written in his first year….

Now this is an illustration of what, writ large, can happen if you mix together writing from different years of research learning. I want here to focus on why this is the case – it’s about text work/identity work (see Kamler and Thomson, 2006, Thomson and Kamler, 2012). In essence, writing the thesis – and by that I mean the final text you hand in for examination – from year one is a risky practice because the researcher who writes in year three after the research is completed is not the same scholar as the researcher who began the study.

We recognize this reality easily with children. We would NEVER assume that a Year 3 child’s knowledge, and their writing about what they know, would be the same in Year 5 as it was two years before. If it was, we’d worry and go running for the special needs teacher. We just assume that a child will have learned over time how to write more fluently, effectively and in more complex ways, and about more difficult topics. But it seems as if we assume that adults stop learning how to learn and to write. We particularly seem to assume this in research education, which is THE most protracted and focused supervised learning experience on offer in the education system.

Let’s think about the doctoral researcher in year three. After the field work/library work is completed, they know an ‘answer’ to the research question they initially proposed. In my experience, doctoral researchers can usually talk about their research ‘findings’ in great detail. Doing the research over two years has also changed them in ways they may not be able to fully articulate, but they are not the same learner that they were when they began the study. They know MUCH more than they did before. They are actually an expert in their chosen field.

And after the initial literature review in year one, the doctoral researcher did get on top of a lot of the ‘stuff’ out there. But they will have kept reading throughout the study, and added new perspectives. Indeed, they may have had to read quite a bit more in order to make sense of what they ‘found’ in the research. It’s not at all uncommon for people to have to get to grips with a new bit of theory in order to explain their ‘findings’ in ways that are worthy of the title of doctor of philosophy. All of this reading – and writing about that reading – adds to the expertise that the doctoral researcher possesses at the end of the two and a bit years’ worth of work.

Acquiring authority and expertise of course doesn’t end when the field work is done. The research must be communicated, because that is how we contribute to scholarly knowledge and conversation. This requirement is accomplished in and through the thesis writing. And very importantly, the writing not only communicates the research – but is also integral to the process of forming the scholar.

We write ourselves into being as scholars, as particular kinds of scholars who stand for particular things, who belong to particular traditions of research, who are interested in and ‘profess’ for specific agendas. We are known by and for our writing and we can consciously shape how we represent ourselves via our writing practices.

Now, the problem for doctoral researchers is that they may not feel like experts, even though they are. The struggle in the thesis is often as much about how to write with the level of authority that their research potentially allows them to have, as is it about finding a plausible way to wrestle the unwieldy stuff of ‘findings’ into an argument.

However, importing text written by the novice researcher-self into the thesis will certainly not only prevent the text having a flowing authoritative ‘voice’ but will also actually impede the job of scholarly formation. Parts of the text will literally be written by another and former version of the self – a less knowledgeable, more tentative and less assured researcher.

As my opening anecdote suggests, I can very often pick early writing that is cut and paste into a thesis. It reads differently. It has a different author and author-ity. If it is a literature chapter – or the methods chapter – and these are the pieces that people most often seem to think that they can just move holus bolus from year one to the final text – they are invariably the worst and most tedious read and the most disconnected from the argument the researcher is trying to make. The text really is not written in the same way in the same voice as the remainder of the thesis. It is a novice text, not that one of an expert. And what do examiners judge that half and half doctoral researcher to be? Ready to be doctored – or incomplete?

And I won’t be the only person who can read early and later researcher selves in a thesis text. Not by a long shot. If YOU are tempted to write the thesis from day one remember what it might show. I know that some people DO start writing chapters right from the start and it apparently works for them, but I want to suggest that it is a very risky strategy. Cutting and pasting may not cause you to fail – although it could mean some rewriting – but it will certainly mean that the process of scholarly formation is less than it might be.This is because learning to talk critically about other people’s texts (the literatures), and to economically and clearly discuss knowledge, truth and methodology, are two of the hardest things to do. It’s actually easier to talk about your own research project. So the researcher who dodges doing these two things and substitutes some earlier work is really dodging the harder parts of ongoing doctoral education. For this reason, the researcher of the half-and-half text may not be nearly so well positioned to do post PhD publications as the one who has had the experience of successfully authoring a cogent and coherent new text from go to whoa.

AND it will mean that the thesis is not nearly as good as it could be if it is a new text, written after the field work/library work, purposefully structured to present the argument for the contribution to knowledge in the most lucid and persuasive writing that can be managed.

Post script.
This post is another response to the recent assertion by author of The Three Month Thesis suggesting that writing from day one is nuts. This line worried a lot of people and there have been a number of responses on The Thesis Whisperer blog, as well as posts by Explorations of Style – Links: can you write too early ? and clamorous voice – Why writing from day one isn’t nuts. I agree with their arguments but also with one of the points James Hayton was, I think, making about the importance of not writing the actual thesis from the start.