34

Assume that Jane Doe has published a paper in 2010 where she has developed a model or a theorem or a similar result, let’s say, that it relates to growth.

Now assume that Jane Doe is writing another paper in 2015, where she refers to the model/theorem from her paper in 2010.

Is it acceptable for Jane to write something like the following?

Doe’s growth model (2010), implies that ...
Doe’s growth theorem (2010) implies that ...
The Doe growth model (2010) implies ...

  • 1
    As you can see from the answers and discussions about it, you might want to add your field and if double blind reviews are common there or not. – Dirk Jan 11 '18 at 9:55
  • 7
    'Doe's growth model' could be 'the growth model used/created by Doe', so could be read as not assigning a name but referring to posession. On the other hand, 'the Doe growth model' leaves no question that is is assigning a name. – Jessica B Jan 11 '18 at 14:09
  • 25
    "Our previous growth model (2010), implies that..." – DBB Jan 12 '18 at 1:32
  • 1
    @DirkLiebhold, I wasn't asking because I was planning to do this. I was asking merely out of general interest. (I'm not going to publish anything). – user56834 Jan 12 '18 at 4:32
  • 2
    I see this now and then. I do feel mildly annoyed that the author's own name passes without special comment among the other references, but I also see it as the convention. Could be different depending on the field. – Luke Sawczak Jan 13 '18 at 6:27
74

This is, unfortunately, a case where English grammar can be tricky and exactly how you phrase things is going to matter.

It is often seen a presumptuous to name something after yourself: "Newton's Laws" and "Hawking radiation" and "Rayleigh scattering" are retrospective judgements of significance by the community. Claiming a similar name is an assertion in advance that your work will be at a similar level of significance.

The problem, then, is that if you say "Doe's growth model", it is ambiguous whether that is intended to be a construction like "Hawking radiation" or whether it just means "the growth model that happens to have been developed by Doe."

This potential problem can be avoided by rephrasing to avoid the parallel construction: "the growth model by Doe et al (2010)", or better yet, "the growth model presented in (Doe, 2010)." The point here is that the significant item under discussion is the growth model, not Doe, and the sentence should be phrased to make that as clear as possible.

Finally, note that these sorts of phrasings can work both for double blind review and for review where the authors' identities are known: even when the identity of the author is known, the important thing should be the relationship between the work, not the fraction of authors that are shared between two papers (unless you are specifically trying to talk about independent co-discovery).

  • 5
    "Doe's growth model (2010) implies that ..." definitely doesn't read to me like "The Doe growth model (2010) implies that ...". Can you really read it as a name rather than a mere reference? – Mehrdad Jan 11 '18 at 22:21
  • 5
    @Mehrdad It's less so than "The Doe growth model," but still potentially ambiguous given its parallel to a construction like "Newton's Laws." Since there is an equivalent sentence without any potential for confusion, I recommend making the safer choice. – jakebeal Jan 11 '18 at 22:47
  • 2
    Ahh... I would substitute that example for "Hawking radiation" in your answer; it's a lot more convincing. – Mehrdad Jan 12 '18 at 2:04
36

No, this would be odd and appears presumptuous.

In your own paper you would write something like "Our growth theorem (Doe 2010) implies..." or "Our previous growth theorem (Doe 2010) implies..."

It is up to your peers to refer to it as "Doe's growth theorem".

  • 2
    Alternatively "the author's growth theorem" (if you take the point of view that using "we" is including the reader, you can't really say "our theorem"). – user9646 Jan 11 '18 at 8:12
  • 8
    Why not refer to it as "theorem 7" in "Doe et. al. ? – user21264 Jan 11 '18 at 11:48
  • 2
    @Magicsowon The underlying question of this solution is: can references be treated as grammatical objects in a sentence? In some fields, the answer is yes, in some fields no, and in some fields in depends on the reviewer. – lighthouse keeper Jan 11 '18 at 12:26
  • 1
    On the other hand, I did not know that a theorem was called "Doe's Theorem" because I learned it from Doe's textbook where he unpresumptuously did not call it that. – Keith McClary Jan 11 '18 at 19:15
15

Yes. It is even mandatory sometimes.

When preparing a paper for double-blind peer-review, references to the authors' prior work must be anonymized,* hence, it is perfectly reasonable to write "Doe's growth model" and "Doe's growth theorem." Moreover, it is expected that you do so.

*MJeffryes better describes the purpose of anonymizing references:

The point isn't to remove the references, but to refer to them in a way which doesn't make it blindingly obvious who the author of the manuscript under review is. If you say 'Our growth model' then it is clear that the author of the cited paper is the anonymous author.


Double-blind review isn't used in all fields. E.g., mathematics, as noted by the L.


Programmer2134 rightly notes that

you can always send "doe's growth model" to the peer reviewers, and then change it before publication to "my earlier model (2010)"

and lighthouse keeper rightly adds

[the above answer doesn't consider] the more involved question of how the final version should look like

In my experience, many anonymized references remain anonymized. I'd speculate that this is possibly because it requires additional time to change them. (Yes, it can be easily achieved with \ifanon Our \else Doe's \fi growth theorem (Doe 2010) ..., but that requires additional time too.)

  • 5
    "When preparing a paper for anonymous peer-review, references to the authors' prior work must be anonymized". This is never the case in the field of Mathematics for example. Actually, I don't understand at all how could this work - if I anonymize references that are required to understand the paper, how can the referee read them? – the L Jan 11 '18 at 9:29
  • 4
    I have no experience with publication, but it seems to me that anonymous peer review should not affect this, because you can always send "doe's growth model" to the peer reviewers, and then change it before publication to "my earlier model(2010)" or something, right? – user56834 Jan 11 '18 at 9:32
  • 4
    @theL The point isn't to remove the references, but to refer to them in a way which doesn't make it blindingly obvious who the author of the manuscript under review is. If you say "Our growth model" then it is clear that the author of the cited paper is the anonymous author. – MJeffryes Jan 11 '18 at 9:32
  • 1
    I see. In any case, at least in math, this is not something being done. The referee is anonymous, but the author is not. Given the arxiv, it is unlikely to be changed. – the L Jan 11 '18 at 9:35
  • 7
    What you're describing as "anonymous peer review" is usually called "double-blind peer review". In all peer review, the reviewer is anonymous. The point of double-blind review is that the author is also anonymous (hence the need to avoid a citation that explicitly says "I am Doe"). And, yes, many fields don't have double-blind peer review. It's often pretty easy to guess who the authors are: a paper that extends three existing papers by Doe is probably also by Doe. – David Richerby Jan 11 '18 at 11:54
3

You can also use "The author's growth theorem" or "The author's growth mode"l if you want to avoid using your name.

1

The best researchers I know find another name for the concept. Tony Hoare often speaks about "assertional logic", while it's "Hoare's logic" since many decades for everyone else, for example. I suggest that you do that same, e.g.,

The Main Growth Model (J. Doe, 2010) implies that self-advertisement need not pay off.

Invent a name. ("Main" above is simply a placeholder.)

  • This reminds me of the Wikipedia article on Paul Dirac: "Most physicists speak of Fermi–Dirac statistics for half-integer-spin particles and Bose–Einstein statistics for integer-spin particles. While lecturing later in life, Dirac always insisted on calling the former "Fermi statistics". He referred to the latter as "Einstein statistics" for reasons, he explained, of "symmetry". – orthocresol Jan 13 '18 at 19:39
1

Writing "Doe's growth model" or "Doe's growth theorem" suggests that the community has agreed to name Doe's results this way. When this isn't the case, drop 'growth' and write "Doe's model" or "Doe's theorem".

This solution works when later works are written alone or with co-authors. It also generalises to the case when the original work was co-authored, e.g., "Doe et al.'s model" can be used by Doe et al. (same co-authors), and by Doe and new co-authors.


This solution was inspired by a previous one, which notes writing "Doe's growth model" or "Doe's growth theorem" is "odd and appears presumptuous" and suggests that "[i]t is up to your peers to refer to it as 'Doe's growth theorem'."

-3

In general no, but it could be.

a) If "Doe's growth model" has become standard nomenclature in the field, it would appear odd to keep referring to it as something else. One should at least add a note saying that this model is well known as "The so-called Doe model (referring to people calling it that)".

b) If Doe made her model along with Cane and Dower, and is now writing a paper with two other people, it would be ok to write in third person: "The growth model introduced by Doe et al. (Doe, 2010)"

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.