Pentimento is the term used to describe the traces of an earlier work glimpsed through layers of paint on a canvas. Marks from the previous composition bleed through the newer surface, a reminder of what went before, a sign of the artist’s corrections and/or new thoughts. The presence of brushstrokes, images and/or forms intended to be hidden is a reminder of the artistic processes that led to the final work.

Paintings are not the only type of text where previous versions re-emerge, bleed through, haunt the present. Pentimento happens in writing too. A version of writings past appears in a new text – this often happens when we are revising. When we cut and paste from one text into another, or rewrite and restructure using bits of what went before, we can carry forward vestiges of the previous version(s). And this can be a problem.

Take the case of radically revising a paper. You are writing a first draft. But while you are writing you discover that your initial argument isn’t working. By the time you finish the draft you’ve worked out what the paper is really about. You know that you have to go back to restructure it. There are a couple of places in the results section where you really like what you have written so you want to keep those bits. You also really can’t bear to go back and do the literatures again. So you don’t rewrite the literature section, nor the methods which have, of course, been written specifically for version one of the paper. So you rewrite the introduction to fit your first draft conclusion and play about a bit with the order of the results and the headings. This second draft, the revised paper consists of a new section, as well as sections that are rewritten, tweaked and/or transferred over.

Now, this second draft runs a real risk of bringing over too many phantoms from your first text. And your readers may not see the traces of the previous paper as clearly as an X Ray shows previous versions of an art work, but they can usually point to something in a text where there is is a mis-step, a something out of place, a jolt rather than flow in the text, a shift or turn where none was needed. Pentimento.

The presence of previous text, pentimenti, is more of a problem in the thesis. A thesis takes a long time to write, and it is almost inevitable that some chapters are written way ahead of others. Writing through in one go is an important part of revising. One of the goals of revision is to smooth out and eliminate the places where the narrative arc falters, where the continuity of the argument drops away, or goes down a side tunnel. And some of the real danger spots in the thesis are when entire slabs of text are taken from versions written not months, but years earlier. This tends to happens most with text devoted to literature work.

There are two reasons why taking text about literature from a much earlier time – most likely the proposal – might be problematic:

1. The literatures work you did for the proposal is not the same as the literatures work you need for the thesis, although there is overlap. At the outset of the doctorate, you were looking for literatures that would help you see where your research might make a contribution as well as literatures that would help you to conceptualise and design your research project. But at the end of the project you know your results and you now know exactly what literatures they refer to, build on, speak back to. So you may need ot jettison a lot or some of what you started out with. You also know which literatures helped you to make sense of your analysis. You have also kept up with the literatures and there has been more published since your initial extensive work. The thesis text you have to write to help the examiner understand the chapters to come will therefore not be the same as the text you wrote for your proposal. It may need ot be radically different. Simply inserting a few new bits will not be enough to make the new argument you write when you know your research results. What’s more…
2. The scholar who first wrote about the literatures is not the same as the scholar who is completing their thesis. Over the course of your PhD programme you have become more expert, more evaluative, more authoritative. Writing the thesis is still forming you as a scholar, but you are starting from a very different place you were in some years before. Your new thesis text is both making and representing your doctorateness, not your readiness to embark on a doctorate. Cutting and pasting from your initial proposal brings the spectre of your former doctoral self into the dissertation. And the eery presence of an earlier writer is one which is often very obvious to examiner readers (it always is to me) – the early PhDer is generally much more tentative, writes assignment style rather than scholarly argument, and is much less present in the text than the completing nearly PhD. You don’t want your former self to be examined, it’s the current nearly and ready to be Dr that you want the examiners to encounter.

Potentially adverse risks of the carry-over traces – or writing pentimento – apply equally to the task of publishing from the PhD. Cutting and pasting from the PhD thesis into papers, or more commonly, to the book of the PhD is potentially a problem. The PhD is written for a particular audience, the text must meet examiner expectations. Academic book readers have different expectations of a book or paper; this is why the advice about converting theses into books and papers is to think about them as new text, as rewrites for a different audience. Writing from the PhD is not just a matter of making a few cuts and a lot of cut and paste. Ask any publisher, and they will tell you that they really dont want to see the PhD in the book that comes from it. Although I can imagine a book which is a deliberate conversation between the early PhDer and the completed PhD, the general rule of academic writing thumb is – Try not to let your present and unwary readers be spooked by the apparition of your PhDer past.

Of course I’m not saying don’t cut and paste. I’m not saying you can’t write into an pd over an existing text. We all do. It’s just about being aware of the potential for hauntings. Do keep pentimento in mind – it is a helpful concept to guide revising.