Authoring a PhD

The Emperor, in 
The Return of the Jedi
3
Writing up a chapter plan into the very first joined-up version
of your text will produce literally dozens of changes in what
you expected to do. All of them will be disappointing. What
seemed feasible, concise, coherent or original on your plan will
turn out weaker, lengthier, less accessible or more familiar in
practice. Howard Becker notes that many PhD students adhere
to the illusion that there is some ‘one best way’ of authoring
any given piece, sitting out there in a landscape of potentiali-
ties, just waiting to be discovered.
4
The writer’s task then is to
hunt high and low for this optimal path. Taking this view, you
D E V E L O P I N G Y O U R T E X T
◆
1 3 5

---

may easily get the feeling part-way through writing that you
have been thoroughly mistaken about where this best way lies,
and now have lost track of it entirely. Countering these setback
feelings entails taking a longer view of, first, the whole set of
stages involved in developing a professional text; and second,
the process of exposing it to consideration and debate by others.
Stages in the writing process
Don’t get it right, get it written.
James Thurber
5
Outlines can help, but not if you begin with them.
If you begin, instead, by writing down everything,
by spewing out your ideas as fast as you can type,
you will discover … the fragments you have to
work with.
Howard Becker
6
The major myth of the authoring process is the critical character
of breaking fresh ground, filling a blank screen or a blank page
de novo
. An essential antidote is to recognize that this is only 
a first stage in authoring, and not necessarily the key one for
the development of your argument. Authoring is a multi-stage
process and, as the quote from James Thurber above makes
clear, there are divergent rationales to go with these different
stages. The logic of a first draft is to make text where there was
none, to get something written, to get the elements you have
in play more or less defined, even if only in a preliminary way
and often in the wrong order. As your text grows you will also
necessarily lose some control over it. By the time a chapter is 30
or 40 pages long you cannot possibly hold its entire argument
in your head at one time. Nor can you even fully understand
what you have written or why the argument turned out as it
did. Rebuilding this mastery is a key element in the second
stage of the writing process, where you can follow through 
a logic of organizing text in a coherent fashion. Building an
extended text will necessarily change your thinking. It will
make clear aspects of your own views that you could not have
1 3 6
◆
A U T H O R I N G A P H D

---

known in advance, and allow you to weigh, test and sift the
varying levels of commitment you have to different proposi-
tions. Someone quoted the maxim, ‘Never begin a sentence
until you know how to end it’ to the novelist E. M. Forster. He
replied: ‘How can I know what I think till I see what I say?’
7
A second essential philosophical change of view with this
approach is to recognize that there is no ‘one best way’ of 
saying something. There is no Platonic perfect form sitting out
there waiting to be searchlit by a peculiarly perceptive advance
plan, or, once identified, capable of being written up intact by
a more self-consistent or more talented author. Instead all that
you can say is constructed, created, not found or discovered
ready-made. Difficulties arise because very often we confront
authoring dilemmas, choice points in the creative process where
two or more options lead further on but you can only maximize
one of your valued goals or purposes at the expense of another.
There may be no ‘right’ choice in such dilemma situations.
There often is no common currency in which to measure the
different kinds of costs attaching to each of the options leading
forward. So you can only make conditional choices to follow
one route rather than another and to see what happens. But
later on it will be helpful to recall those prior decision points in
re-evaluating what you have done. Perhaps an alternative
choice might be better after all.
Going from a poor version of your ideas to a radically
improved and viable text takes time, distance, alternative 
perspectives and a concerted effort at remodelling. Writing is an
act of commitment. So no one can constructively renounce text
that they have just produced – that is, see what is wrong with
it or what might be changed to remedy defects. With a 
newborn text you can only renew and reiterate your commit-
ment (perhaps tinkering around with perfectionist embellish-
ments) or reject it non-constructively (‘It’s all rubbish – I’m
wasting my time!’). You need at least some days to pass, other
things and other thoughts to intervene, and other people to
read what you have written in order to begin to see things 
differently. And when you start to revise and replan it can be
helpful to have built that stage into your thinking and your
timetable in advance, and to have some appropriate expecta-
tions about it.
D E V E L O P I N G Y O U R T E X T
◆
1 3 7

---

You will almost always need to carry out five operations on
any piece of text: print, edit, revise, upgrade and remodel:
◆
Print
your material to achieve a shift of perspective from
writing on your PC. If you only edit text on-screen 
your changes will be too confined to small corrections and
changes at a verbal level. Working on paper will help you
see how more thorough-going alterations are feasible, such
as moving large chunks of text around over several pages.
◆
Edit
means a word-level edit of your raw text to remove 
mis-spellings, grammar mistakes, tiresome repetitions of the
same word or phrase, and other infelicities. Do not leave your
text untouched with these problems still around. So long as
they remain uncorrected, their presence will tend to obscure
other defects from you. Getting to a clean text lets you see
beyond the clutter, to any deeper intellectual problems.
◆
Revise
covers a paragraph-level reconsideration of how one
idea chains to the next. It focuses on improving things by
small-scale switches around in the order of sentences or
paragraph chunks. It can also cover more substantive
changes, especially in the beginnings and ends of
paragraphs (remembering the Topic, Body, Wrap sequence).
◆
Upgrade
involves going back from your piece of text to your
original materials and considering whether you can
strengthen the arguments in any way. Can you cite more
scholarly support for points you have made? Or bring 
in additional empirical evidence? Or reanalyse your data to
knock out possible competing interpretations? Can you
extend your key arguments, or develop them in a more
formal or systematic way? You need to be clear when your
approach needs more sustenance and underpinning. But
avoid slipping into ‘thesis paranoia’ by overarguing or
overciting on non-controversial points.
◆
Remodel
refers to a much more radical restructuring of 
a chapter or article, which usually requires a very specific
method, described in the next subsection. Text that is
already in a satisfactory condition may not need full-scale
remodelling. But you will normally have to make radical
changes in at least one or two chapters out of eight, unless
you are a very disciplined and consistent writer.
1 3 8
◆
A U T H O R I N G A P H D

---

Producing a piece of text finished in ‘first draft’ form involves
both your private efforts to generate raw text and improve on
it, and seeking outside commentary and advice. The overall
process can be pictured as having four phases, moving from
most personal and private to most public. Going public with
your commitment, a text that crystallizes your thought and 
for the moment fixes it in one configuration, is a particularly
sensitive stage that needs to be handled carefully.
In Phase 1 you write out a semblance of the argument to an
approximate length of the chapter you are embarked on. This
stage produces raw text, words on screen or handwriting on a
page, arguments played out or attempted, facts marshalled, con-
nections made, positions expounded – but maybe not yet in any
satisfactory joined-up form.
In Phase 2 you stockpile and reassess your text for a while,
looking for ways to upgrade it and tighten it up. After leaving a
short gap (because some time and distance are needed here), you
can review what you have, looking for omissions or inconsis-
tencies, trying to trace the development of the argument and to
see places where moving things around can improve things.
During this shape-up stage it can also be useful to show bits of
text to friendly readers, that is people close to you, such as 
fellow PhD students, friends, relatives, significant others or
lovers. Even people without a background in your topic can be
helpful foils, sympathetic readers who can look at your text dis-
passionately and tell you how accessible or well written it
seems. A trusted, intelligent but inexpert listener can also help
you test your key arguments by letting you say them aloud and
more accessibly. If you are very lucky and get on really well
with one or more of your supervisors, perhaps you may get
them involved in this shaping-up stage. Phase 2 may involve
you in making multiple small revisions as you go along. But it
normally ends with you making a first systematic run through
of your work, inserting additional materials, tying down loose
referencing, moving and reknitting text in an improved 
pattern, and consolidating lots of small upgrade changes into 
a revised form.
In Phase 3 you begin to go public with your text, accumulate
comments, and incorporate them in a more fundamental 
revision or remodelling. In professional contexts you can only
D E V E L O P I N G Y O U R T E X T
◆
1 3 9

---

go forward a certain distance on your own, after which you
need to get some radically different views of what you are say-
ing in order to make progress. Your supervisors or advisers are
the first port of call. One of their primary roles is to look at and
comment on your formal written text. You need to make sure
that they give you effective feedback on your work. Normally
advisers are reassured and even grateful when they get chapters
to look at. It is not easy for them to operate solely at an oral level
in someone else’s research topic. They need your help in the
form of a regular sequence of chapters in order to offer useful
advice and commentary. But supervisors are also very variable
in what they say, for various reasons. Some are famously diffi-
dent or difficult people, like the Oxford philosopher whose
three-word written comment on a student’s painfully produced
12,000-word chapter was: ‘I suppose so’. Different supervisors
also follow different strategies. Some will comment in vigorous
detail on early drafts, where others deliberately stand back for
fear of being too critical of your nascent ideas. Some very well
organized supervisors put their effort in very early on in your
text production process, demanding that you get a near-perfect
chapter draft to stockpile before you can move on to another
chapter. In this perspective, once you have reached the right
‘doctoral’ level in one chapter, it will become easier for you to
deliver subsequent chapters to the same standard. Other advis-
ers (like me) feel that it is only important for you to get a
broadly acceptable chapter draft before moving on, lest you
drag out early writing with perfectionist anxieties and erode
your later research and authoring time. In this perspective,
going from a first full draft of the thesis to a final version of the
text will normally produce so many changes that overwriting
early chapters, before the neighbouring chapters are written,
will too frequently be wasted or redundant effort. The detailed
stylistic and argumentative choices you make in your first two
years’ work are likely to be extensively overturned by more
mature insights and by the alterations inherent in crafting the
thesis into an integrated whole.
Beginning to go public should take other forms than just
showing material to your supervisor, however. Presenting a
chapter in a ‘friendly’ public forum such as a departmental
graduate seminar can be very helpful, even if the audience does
1 4 0
◆
A U T H O R I N G A P H D

---

not include many people who know a great deal about your
topic. The point of these exercises is for you to think through
how your text can be presented and explained to people knowl-
edgeable in your discipline but not in your specific topic. 
The changes that you make in order to mount an effective pres-
entation and the comments that you get back can often be 
very helpful foretastes of how people in your discipline gener-
ally will view your work. Some PhD students resent being asked
by their departments to do regular presentations once or twice
a year to such groups, feeling that so inexpert an audience has
little to say to them about their own specialist research. But at
the end of the PhD other ‘generalist’ audiences in your disci-
pline will make crucial decisions about your future as an aca-
demic, such as deciding whether or not to appoint you to a
university job or to allocate you a post-doctoral grant. It is far
better to have to appreciate early on how the profession as a
whole may see your work – so that you can make adjustments
in the orientation or presentation of your text in time to
improve these later perceptions.
Talking is a basic human art. By it each 
communicates to others what he [or she] knows
and, at the same time, provokes the contradictions
which direct his attention to what he has 
overlooked.