Writing, and its alter ego, reading, are the backbone of academic work. The practices that make scholarship what it is.

In the PhD there are multiple places and purposes for writing.

We often focus on the final text, the thesis, the writing that communicates what we claim to know, that explains the research we have done. We may of course also write other texts that deal with our research – conference papers, journal articles, and perhaps articles for professional publications, blog posts and tweets. We may write about our research in audit reports to the university or a funder, or send regular updates and small textual chunks to supervisors.

This kind of writing is product and action oriented. We assemble information, plan what we will say, attend to the order and structure of the text – we craft the writing so that it has the best chance of succeeding. We write so that the thesis is awarded, the conference abstract is accepted, the conference paper is engaging and encourages others to make connections with us, the journal article makes it through review, funding is continued, the supervisor and institution have confidence in us and our work.

We don’t entirely know ahead of time what this kind of writing-for-another reader will be – it is often the case that as we start to write, we see more and know more as we go along. We consolidate, as well as find new emphases and possibilities, through the writing.

However there’s other writing that scholars do, and that’s writing that is central to learning. This type of writing helps us to figure things out. During the PhD – and in any scholarly work – we write all the time. Not to publish. Not to send the text anywhere. We write to help us make sense of things. We make notes continually. We summarise and synthesise, compare and contrast what other people have said and written. We shape and craft our interpretations of texts into something that speaks with our particular research project. We make tables, codes, images, doodles, graphs, maps and diagrammes. We have piles of files of ideas, tentative explorations of data, emerging analysis, experimentation with theory. These writings help us to sort out what we might mean, can mean, can say and not say.

We start an idea, consider a text, speculate, attempt to put things together. Ideas coalesce and gel in ways we may not have expected. Connections and contradictions are clarified. Putting an idea into words, finding the words to consolidate thinking and talking, creating something more orderly out of scattered fragments – this is writing to learn. We also learn more about the writing itself. We experiment and play with ways to describe, to categorise, to explain. We learn how to craft a sentence, a paragraph, an argument. Our writing goal is simply to find out more.

There is a lot of this kind of learning-writing at the start of the PhD. Early writing is process oriented, we are less concerned with a well turned phrase, we are more interested in a clearer line of thinking. We are the primary reader of these learning texts. We are writing for ourselves. Of course, learning-writing can lead to writing intended for other readers, and indeed some of this learning-writing may be shared with supervisors or writing groups.

Some very famous scholars write in exactly this way – they start with an idea and then write it to fruition. Michel Foucault for instance described his writing just this way.

I don’t write because I have something in mind, I don’t write to show what I have already demonstrated and analysed for myself. Writing consists essentially of doing something that allows me to discover something that I hadn’t seen initially. When I begin to write an essay or a book, or anything, I don’t really know where it’s going to lead or where it’ll end up or what I’m going to show. I only discover what I have to show in the actual movement of writing, as if writing specifically meant diagnosing what I had wanted to say at the moment I begin to write. (Foucault, 2013 p 46.)

Of course, this doesn’t mean that Foucault started off his writing with nothing. I imagine him surrounded with books and notes but not yet a clear idea of how these would come together. Nor do I imagine that his written work didn’t go through many iterations and drafts in order to reach the point where it was fit to go to other readers. He may have even made a plan as his first step. He doesn’t say. But many scholars do begin their writing process with some kind of written plan or abstract, a small piece of writing that begins the process of pulling materials together. Other scholars do something more free form, and work through successive texts, shifting and refining as they go along.

And it is writing to learn that is crucial at the very start of the PhD. It is important to develop the habit of writing summaries of texts, jotting down ideas about your research design, keeping a reading journal, perhaps also a journal of reflections on your own learning, worrying away at the wording of your research question or hypothesis, writing regular updates of where you are for yourself and your supervisor. All of these kinds of writings will, even if they feel scattered and messy at the outset, help you to think through your research problem and your field.

And of course you need to make writing to learn a systematic, regular and frequent practice. Make sure that you find a way to organise all of this early writing so that you can find things again easily. Use bibliographic software as well as a system of naming, dating and grouping texts so that you can find things again. Work out when it is possible and good for you to do different kinds of writing – noting, reflecting, generating ideas – and set up regular times in your diary to do them. That’s because a key to the PhD is to see writing as your close companion. A friend to help you make sense of what and how you are learning. A friend to cultivate now, rather than later.