Authoring a PhD

Robert J. Sternberg
4
Good headings should accurately characterize your text. In a
very few words they should give readers a helpful advance 
idea of what is to come in each section or subsection, and wher-
ever possible what your substantive argument will be. Devising
effective headings is a difficult art that needs sustained atten-
tion from authors. You can tell that the task is complex because
in the business world there are highly paid professionals who
do nothing else, people like advertising copywriters, newspaper
or magazine sub-editors, and Web-site designers. Intellectuals
tend to make fun of many of these groups and to see their 
outputs as non-serious. But the job they do is not as easy as it
looks.
Consider the following problem. It is 1989 and the
Communist Party of Czechoslovakia has renounced its previous
‘leading role in the organs of the state’, bringing to an end over
40 years of one-party rule and state socialism, and opening the
way for democratization and a transition to a capitalist econ-
omy. You are working as a sub-editor for a right-wing British
tabloid newspaper, the 
Sun
, whose daily audience of 4.3 million
8 4
◆
A U T H O R I N G A P H D

---

readers is mainly preoccupied with soap opera stars, footballers
and the nude pin-up girls on page 3 of the paper. None the less,
your editor has decided to lead on the historic Czechoslovakia
story to please the right-wing proprietor. You are told to devise
a front-page headline, to take up two-thirds of the page, but to
use
no more than three words
, and four syllables (given 
Sun
read-
ers’ limited attention span and linguistic competences). How
are you going to get the essence of the story across within these
limitations? This is a genuine question, and I would encourage
you to get pen and paper now and try to come up with your
own answer. In the notes for this chapter I have printed the
brilliant solution that the 
Sun
actually went with.
5
The paper’s achievement in this case was to give the essence
of the whole story in its headline. Of course, tabloid newspa-
pers have to try harder to grab readers’ attention than most
writers of doctorates. As a thesis author you can allow some-
what more words and many more syllables into your headings
than the 
Sun
. But the basic goal, of putting the message in the
shop window, is just as appropriate for doctoral work. Taking it
to the limit here, one approach much used in fairly short busi-
ness and government reports is to use narrative headings and
subheadings, which give a mini-précis of what each section or
subsection covers. This style has a lot to commend it. Yet it is
rarely used in PhD dissertations, mainly because it could get
very wearing if repeated over a long text. Headings and sub-
headings in doctorates, and in journals and books, are normally
much shorter, ranging from one or two words at minimum up
to seven or eight words at maximum. Headings for main sec-
tions only might be a bit longer if they have two parts separated
by a colon. However, subheadings should always stay quite
snappy (on one line, without parts). None of these limitations
is inconsistent with trying to get as much of the text’s key 
message as possible into the heading or subheading.
There are four common general failings in how PhD and
other academic authors title their chapters and sections:
(i)
Non-substantive headings
do little or nothing to cue read-
ers about the line of argument you are making. People often
choose headings which consist only of vacuous verbiage or are
very formalistic. Some are process-orientated or refer only to
the methodological operations you carried out, rather than to
O R G A N I Z I N G A C H A P T E R O R PA P E R
◆
8 5

---

your findings. Some are completely vague. Others tell readers a
little about what topic is being covered, but give no clue about
what the author wants to say about the topic, what position is
being argued, or what the ‘bottom-line’ or conclusion of the
argument may be. This problem is far and away the leading
defect with headings in academic theses and publications, espe-
cially when authors are using an analytic pattern of expla-
nation. Poor headings often feed into mismanaging readers’
expectations, because authors choose very grand or sweeping
subheadings to caption small subsections, feeding a sense of
disappointment amongst readers. To pick up cases in your own
work, look through your extended contents page and test each
of your headings for genuine content. Replace those which are
formalistic or process-orientated with something more specific
and substantive.
(ii)