Some academic journals request that all references to publications by the author(s) are anonymized by replacing them with “Author A”, “Author B” etc. At first sight, this seems quite reasonable and in support of double-blind peer review. However, this approach is flawed. Unless we write things like “how I showed previously (Ruedin 2017)”, including a reference to my publications does not actually tell the reviewer whom I am (e.g. “as Ruedin (2017) showed”). If we then use “Author A” etc., we indicate to the reviewer that this really is one of my publications. For anyone familiar with the literature relevant for the paper (as the reviewers probably should be), the effort to hide the identity of the author actually makes it clearer.
Let’s take the perspective of the reviewer. I get a paper, not knowing who wrote it. If it’s a paper from a conference I attended, I probably know it (assuming I attend relevant panels), and any effort to anonymize are in vain anyway. If it is by someone working on similar issues, I might guess — but typically I don’t try, because it should make no difference to my review who wrote the paper. Seeing lots of references to say Smith doesn’t actually mean it’s a paper by Smith. It could also be: (a) a post-doc or PhD student of Smith, (b) someone hoping that Smith would be a reviewer, (c) someone who found the work of Smith quite useful, (d) that Smith reviewed the paper at a previous journal and asked insisted to be cited, etc.
So how to anonymize properly? Avoid (a) references to unpublished output of yourself (“Ruedin, 2017, unpublished manuscript”), (b) constructs like “how I showed previously (Ruedin 2017)”.
Added 12 October 2019:
I come across this ever so often when doing a review, and there are even journal guidelines giving bad advice on how to blind a manuscript for double-blind peer review.
A good guide can be found here at Oxford Univeristy Press.
What gets me every time is violations of this:
Do not employ (Author 2016) and similar devices.
While it may look like a good way to anonymize, it actually encourages guessing who the author may be more than anything I know. Often it provides good clues, especially if we can read the titles in the reference list. The funny thing is that as a reviewer, typically I do not want to know who you are (at least not until I have completed the review).
The guidelines cited above are pretty clear:
do cite yourself if it is relevant
use appendices/supplements to describe methodology
do not use (Author 2019)
do not cite unpublished work
Published 21 October 2017