Authoring a PhD

note itself
must give full details on first citation of 
a source, covering the same items as required for Harvard 
bibliographies (see above), but with the component items in
a different sequence, as:
W R I T I N G C L E A R LY
◆
1 3 1
8. Terence B. Jones and Arthur Crank, 
One Book Academics: What Goes
Wrong
? (London: Futuristic Press, 1997), second edition. Terence B. Jones
and Steven A. Winge, ‘Deconstructing post-modern writers’ angst’, 
Times
Literary Supplement
, 26 September 2000, pp. 70–1.
9. Terence B. Jones, ‘Academic time-wasting in universities’, 
American
Journal of Scholasticism
(1999) vol. 12, no. 4, pp. 12–71.
The main changes here from the bibliography format are:
the author’s first name and second initial now come before
(instead of after) their surname; and the year date of 
publication moves from early on in the reference to a 
position just ahead of the volume number for journal 
articles, and just behind the publisher name for books. 
In any form of notes system you are duplicating the 
bibliography to a large extent. However, on second or 
subsequent citation of the same source in endnotes it is
possible to reduce the level of detail in referencing, so long
as the source remains unambiguously identifiable. You can
here retain author surnames only, plus a shortened form of
the book or article title, as:
10. Jones and Crank, 
One Book Academics
, p. 87.
11. Jones, ‘Academic time-wasting in universities’, pp. 15–16.
Using a chapter endnotes system involves some inconven-
ience for readers. They have to flip from a note number to the
note itself, which is located either at the end of that chapter or
at the end of the thesis as a whole. ‘Big book’ theses in type-
script and bound as one volume are bulky. So readers might
find it easier to use endnotes located on the last pages of each
chapter, rather than at the end of the whole thesis. But if your

---

thesis gets published as a book readers will face no extra diffi-
culty in using notes placed all together at the back. The great
plus point of endnotes, even more than the Harvard system, is
that it creates a clean-looking main text, with only relatively
unobtrusive in-text note numbers, ideally not too numerous or
overdone.
Footnotes follow the same format for full details and subse-
quent references as in the two endnotes boxes above. But these
citation details are given at the bottom of the same page as the
note number. Footnotes maximize one-stop look-up. Endnotes
are clearly less convenient for readers than footnotes. None the
less footnotes are still a slightly worse system to use for authoring
a thesis, even with the rapid advances made by modern word
processors in handling them. They are somewhat harder for you
to control and keep up to date when you cut and paste text, as you
will have to do extensively. Usually there will be some enhanced
difficulties in maintaining version control between footnotes and
a bibliography compared with endnotes, where all your references
for a chapter are at least gathered together and printed in one
place. And repagination problems tend to increase with footnotes.
Footnotes also maximize the clutter of referencing that readers
see. Especially in PhD dissertations, they often give a ragged and
uneven appearance to your final printed pages, with notes appar-
ently ‘squeezing’ the main text. Because of these and other prob-
lems journals and almost all book publishers have moved away
from footnotes. For instance, if readers are accessing journals on-
line (as more and more are doing), then it is often hard for them
to keep two different-sized fonts on the same page readable. Either
the main text is in focus but the footnotes are too small; or the
footnotes are visible but the main text is then too big.
You will often need to rearrange both footnotes and end-
notes for publication. This task is a very easy one for your word
processor if it entails swapping endnotes to footnotes, or vice
versa. But swapping between notes systems and Harvard refer-
encing is only easy if you are using a citations-handling pack-
age like 
Endnote
. The most difficult rewriting occurs if you are
redoing notes for a journal or a book using Harvard referencing
and have to eliminate subtexts. Using footnotes has its most
likely adverse impact on authors’ intellectual habits here,
encouraging you to create subtexts and then carry on vigorous
1 3 2
◆
A U T H O R I N G A P H D

---

side-shows there. Endnotes have less impact on authors here,
because endnotes are in a much less visible location.
Finally, for completeness, let me mention a newer citations-
handling approach which is even less obtrusive than endnotes.
‘Popular science’ writers follow this style, partly in hopes of
broadening their appeal to readers, especially those outside the
academic community. They provide a full slate of references at
the end of the book, but there are no note numbers or Harvard
references in the text to trigger them. Instead the reference list
gives a page number, perhaps also a line or paragraph number,
and the first few words of a quote or other phrase on that page.
This leads into a full relevant citation in endnote form. This
approach may become more popular in future with academic
books in ‘soft’ disciplines, where authors strive for a better 
literary feel. But at present it would still be an unconventional
referencing procedure to use for completing a PhD.