passage you cite whatever edition of the work they are using.
Legal case referencing also has its own forms. And citations of
documents in historical archives should also follow referencing
and numbering conventions, often ones particular to that
archive. In all such cases you should preferably use a
convention that is already well established in your discipline
for the Harvard in-text reference, explaining what you are
doing on first use for that source. The idea here is to maximize
the ability of other professionals to retrieve and check the
documents or other material that you cite. If no convention
exists for your source then establish your own rule clearly on
first use, giving readers a brief reminder about it later on when
needed. Primary sources that are constantly referred to can
also be abbreviated, so long as you explain the shortened form
used to readers on first use and include it in the glossary of
acronyms. For example, a reference to John Locke’s Treatise on
Civil Government, Book 4, Chapter 3, section 6, might appear as
(Locke,TCG, IV.3.vi). This kind of abbreviated reference is
perfectly neat to use many times over in your text, but is also
accessible enough once you have explained the convention
being used. Where your thesis revolves centrally around the
use of a set of primary sources, then it is often useful to
discuss them in a Research Methods Appendix, and this is
a good place also to explain the referencing conventions you
have followed.
◆ Unpublished and un-indexed sources, such as documents
located in a depository that is not a well-organized historical
or other archive with retrieval numbers, can be handled in a
similar way in the Harvard in-text reference. Establish and
explain your own referencing or naming convention as for
primary sources above. Include a set of convenient
abbreviations, ideally acronyms that will be intuitively
understandable (as with the Locke reference above).
◆ In-text references for interview material are also sometimes
cited as a problem for Harvard referencing, but are in fact
straightforward to handle. ‘On-the-record’ interviews should be
cited in a similar way to primary sources, by establishing a
convention including the interviewee’s surname, the fact that it
was an interview and the interview date, as: (Smithers, interview,
26 October 2000). Your Research Methods Appendix should then
WRITING CLEARLY◆ 129