Adrift in drafts

Less than six months into my first year of my PhD and I am on draft number 9 of my preliminary literature review (in Cambridge, we write a first year research proposal of 12,000-20,000 words and have a viva on it before we get such privileges as a locker in the PhD room). As a relative newbie, perhaps I am being naive but the process of redrafting is actually pretty satisfying.

The Merriam-Webster definition of a draft (incidentally, a word that originates in the Middle English word draght, related to the word for drawing) is this:

a version of something (such as a document) that you make before you make the final version. (Merriam-Webster, 2014)

However, I feel that writing drafts is not as linear as implied by this definition. As Elbow points out,  “writing is a way to end up thinking something you couldn’t have started out thinking” (1998. p.15). For Elbow, each draft marks a development of thought that may be returned to, refocused and revised. It is an iterative, circular and systematic process of clarification and discovery. This approach is reminiscent of the Vygotskian view of the intertwined nature of thought and language.

My supervisors and my incredibly wise father both gave me a similar piece of advice – keep former drafts. I may have cut a paragraph from my current draft due to word count constraints, but will want it in the extended edition. On a contrary note, I may inadvertently regress my research questions or focus and not remember quite why I altered it in the first place. So, having previous drafts to refer to is an incredibly useful idea. So, instead of just having one document (imaginatively) called ‘Literature Review’, I currently have Literature Review drafts 1.1-1.4 from before my supervisor gave me initial feedback and Literature Review drafts 2.1-2.5 responding to points my supervisor highlighted. There will undoubtedly be drafts 3.1-3.x and possibly even 4.1-4.x. Most of these will be filtered out and archived by the time I submit my interim report, but even now it is rather heartening to go back and see how my thinking and research has developed and refined.

The distinction between drafts is not that I rename it every time I made a slight edit, nor do I label it a redraft for the same reasons every time. Sometimes, I change drafts because I have added or taken away a significant amount of text (usually at least a page). Sometimes it is that I have refocused my ideas. For example, following a discussion with my supervisor, I re-drafted to make my focus much more process-orientated, rather than concentrating exclusively on outcomes. While I may not have added many more references, my approach has undergone a change, so a new draft is justified.

Between each draft, I tend to take a couple of days off to get my mind clear and give some distance. Then, I often print out the document and go through it, highlighting the main points of each paragraph, figure and section. Are they clear? What is their purpose? How do they relate to the surrounding information? Are my references useful and correct? Finally, I ask my long suffering family to read over it to check that it is clear and purposeful for an independent reader.

My tips for drafting

  • Allow more time than you think you need to redraft, always. If you don’t use it, enjoy a brief and guiltless break. Far better to have extra time at the end than to procrastinate and then panic.
  • Due to the healthy fear of losing what I am working on, I save all my drafts directly into Google Drive and, just to make doubly sure, email myself everything at the end of each day. Don’t let your work be lost through carelessness!
  • Are you saying what you want to say? Is it worth saying? Will the intended reader be able to understand it? What are alternative interpretations or arguments? These are simple questions but worth bearing in mind.
  • One of the most memorable pieces of advice I have come across about re-drafting is the piratical ARRR approach. This suggests; Adding, Rearranging, Removing and Replacing. I think that the order in which this is proposed is a useful one, but it assumes that your general focus or approach has remained unchanged. It is, I suppose, a bit granular for a first re-drafting, but useful in the final stages.

Drafting is an incredibly useful process in developing an argument and, even more fundamentally, in identifying exactly what your argument is, why you are proposing it and what evidence you can provide. I try not to approach it as a box ticking exercise or as a self-satisfied grammar check, but as a thoughtful and engaging activity in its own right. Hopefully, I will continue in this happy attitude in years to come.


Elbow, P. Writing Without Teachers. 2nd ed. New York: Oxford UP, 1973, 1998. p.15

Merriam-Webster (2014) ‘Draft; definition. Available at: [Accessed: 2 Feb 2014]

Using referencing software

I recently discussed with other PhD candidates the use of bibliographic software such as EndNote or Zotero. They were all of the opinion that a well-organised thesis should not need software to keep track of the referencing. They also felt that learning how to use it would take time better spent doing other things. I disagree with both opinions.

People no longer mention word-processing skills or digital literacy on CV’s because it is, to a large extent, assumed. It is vital to remain au fait with current technologies in your career field, whether or not it is an explicitly technological field. In fact, I would argue that it is a professional responsibility to have a working knowledge of the most common tools. So, in any field of research, it is important to learn at least one referencing software and preferably more, in the same way as it is important to learn methodological tools or understand different epistemologies within your field. Therefore, in pursuing a career path in research, I would argue that it is short-sighted not to learn the functions and possibilities of referencing software.

Professional researchers are reading an increasing number of research articles per year (Tenopir and King, 2001), possibly due to the facilitation of online journals and databases. Conversely, while the overall time spent reading articles has increased, the time allocation per article is decreasing (Renear and Palmer, 2009). It is therefore ever more important that researchers (even amoebic ones like myself) work smart – reference management software is easy, quick and gives instant returns in time saved.

Instead of copying and pasting bit by bit every salient detail of the authors, journal, publication date, doi etc., you click a button and that information is automatically stored for you. Many of the reference managers will also download a pdf of the source and connect it with the citation information also by that one click. Then, when writing your thesis, you just type in the first few letters of the author into the add-ins box (if you are using Word) and like magic the appropriate citation appears in the format of your choice.

Many of the software packages store the library remotely, so you can access your library from any computer or tablet. This is incredibly convenient if, like me, you have family in other countries and don’t want to lug a laptop through security, if you want to get some work done on the train, or if you just have several different devices. You can also choose to share your library with someone else – for example your supervisor or a reading group.

For me, however, one of the most appealing aspects of using reference management software is the ability to add your own tags and notes connected to the citation information. These can then be used to call up all the articles relevant to, for example, ‘argumentation’ or ‘psycholinguistics’ and so on. I have also used this capacity as a useful way of ranking articles. I add a tag that says ‘rank 1’ if I feel it is incredibly useful, persuasive and relevant to my research. Conversely, a ‘rank 5’ will be something I consider as mostly un-useful, unconvincing or irrelevant. Again, this means that when writing my literature review I can recall the most relevant and important research according to my own classification.

In summary; the benefits of using reference management software are:

  1. speed – it is much quicker to create a citation using software than manually
  2. convenience – rather than transferring or emailing a document with all your citations, the software will store it for you
  3. organisation and annotation – tagging and making notes easy and connected to citations
  4. professional development – if everyone around you is using it, you most certainly need to be too (and many universities offer training in their favoured software – MIT has a good comparison of the major software packages ).

The only drawback to reference management software, as far as I can see, is the one proposed by my colleagues – learning it in the first place. If you are a student, you are there to learn, so this is part and parcel of that process. If you are a researcher, you are still invested in the learning process, so the motivation to learn and develop should still be present.

Go forth and love your referencing software!


Renear, A. H., & Palmer, C. L. (2009). Strategic reading, ontologies, and the future of scientific publishing. Science, 325(5942), 828–832. doi:10.1126/science.1157784

Tenopir, C., & King, D. (2001). The use and value of scientific journals: past, present and future. Serials: The Journal for the Serials Community, 14(2), 113–120.