Thursday, June 19, 2014

Citing Your Own Unpublished Work

In comments here, anon 8:30 asks:
Short version: Can you cite your own unpublished work? 
Longer version: What do you do when you're writing a paper and want to refer to another paper you've written that goes into more detail on a certain point or supporting argument that you don't have time to address at length in the current paper -- but that other paper is unpublished? Are you just screwed?
Short version: yes, you can cite your own unpublished work.

Longer version: even if it looks bad, it won't look bad to anyone, since no one will probably read your paper. Just kidding, kind of. But seriously, I think it's basically ok. At least, it is as long as the thing you cite is eventually published. If I'm reading your 2014 paper in 2014 and I see you do this, I might hold it against you a little, or I might not. Either way, I probably wouldn't think it was a big deal. If your 2014 paper got published, your unpublished paper will probably eventually be published, too. (Of course, I would have more confidence in this inference the farther along you are in your career.)

And if I'm reading your 2014 paper in 2020, and I see that the unpublished paper came out in 2016 (accounting for longish review times and journal backlogs), I wouldn't care at all. I'd figure you'd been working on the two things at the same time, and the one in front of me happened to come out first. No big deal.

However, if I'm reading your 2014 paper in 2024 and I see that the unpublished paper never came out, I might come to have doubts about the substance of that paper. I don't know if those doubts would infect my opinion of the paper in front of me or not. Maybe. Maybe not. Depends on how crucial the point in the unpublished paper is to the 2014 paper, and, like, whether I'm in a good mood, and stuff like that.

One thing I'd be cautious about, though, is that I've had papers change substantially over the course of the refereeing process. It would be a bit of a bummer if published version of the paper you're citing didn't connect with the 2014 paper as well as the unpublished version did.

Additionally, I'm pretty sure I've seen big-name people do this. At least, I'm pretty sure I've seen Mark Schroeder do this. But maybe he's earned the right in a way that us under-laborers haven't. And maybe I'm wrong about this whole thing.

What say you, Smokers?

--Mr. Zero


Anonymous said...

"Short version: Can you cite your own unpublished work?"

Yes. It's been done.

"Longer version: What do you do when you're writing a paper and want to refer to another paper you've written that goes into more detail on a certain point or supporting argument that you don't have time to address at length in the current paper -- but that other paper is unpublished?"

Lots of things to consider here. Are you trying to publish this paper? If so, then sure; you're working good faith that this work will reach print eventually. But ask yourself how far along you are. If the first paper (the one being cited) is out for review, fine. This happens. If it's not out for review - if for some reason you are still working on it before sending it out - consider holding back on the second paper, at least for now. That is, if you are still formulating a point that is important to your second paper - if that first paper has not been reviewed by the academic community in some capacity - it might be best to tackle the first paper first.

"Are you just screwed?"

No, but I would ask what your long-term goal is. From this, it sounds like you are building up to something: Paper 2 depends on Paper 1, and you don't/can't/won't develop the salient points from Paper 1 in Paper 2. OK, fine. But why then are you rushing to get Paper 2 out into the world before Paper 1 is published?

If I'm reviewing Paper 2, and I am not convinced about the point you are using - but not developing - from Paper 1, what is your reply? Do you tell me to trust you, because someday Paper 1 will come out and then it will all make sense? If you reply to me that the relevant paper is under review at X journal, I might be able to live with that, if I can see a copy of the paper you are citing.

A lot of this comes down to the publication status of Paper 1.

zombie said...

I think it comes down, in part, to how much paper 2 depends on the point you're making in paper 1. Is it critical to the discussion, or are you just pointing to somewhere that you have dealt with a small piece of your argument elsewhere? Is there some reason you really can't elaborate on this issue in paper 2? As a reviewer, or a reader, I'd feel frustrated if you made some crucial point that is only supported elsewhere, in a paper I can't read.

I cite myself, but only my published self. I guess I'm not at a point in my career where I can count on my unpublished papers being published. And often by the time I get a rejection, I've usually moved on to another paper, and it takes me a while (years) to get back to the unpublished stuff. Which would make me hesitant to cite it.

And I know of at least one journal in my AOS that won't even allow you to cite accepted, forthcoming papers, so possibly a journal will put the kibosh on your citation.

Anonymous said...

Tacky as hell. Worse, even, that referenceing one's PhD thesis. Superstars can get away with it, but for the average philosopher, no--you just look lazy or that you are someone who claws at self-promotion every opportunity they get.

Anonymous said...

I mostly agree with zombie. I don't think this is bad at all, so long as the point being referenced is not an important step in the argument. If it is a key point, then I think it's a bigger issue, and you probably owe it to readers to sketch the argument of the other paper. But even then, it seems to me that a reference is totally appropriate ("for a more developed version of this argument, see X").

I don't think that the kind of self-promotion involved is bad, as long as the referred to paper is in decent shape. You've got a piece of work coming out soon that you're excited about, and you want to let ppl know about it. That seems just fine to me, especially insofar as you're prepared to share it with anybody who asks in the period between the publication of the referring piece and the publication of the referred to article.

Anonymous said...

I definitely don't assume that a paper which was never published is thereby not as good. There are many reasons people don't even clean up papers for publication - and many more reasons, independent of philosophical merit, why submitted papers aren't published.

I also don't see why there's any difference between citing an unpublished paper of your own and the unpublished papers of others.

With that being said, it's quite frustrating when a paper is cited which is never published, and I do wish that Burnyeat had published his manuscript 'Carneades was not a probabilist', that Code-Laks-Most would publish their translation of Aristotle's Metaphysics, etc.

Anonymous said...

Not all cases are the result of intentional self-aggrandizing. Sometimes you need to cite unpublished work to avoid self-plagiarism. Here's a slightly different example to illustrate this: suppose you're completing a journal article in which you need to cite an argument that you already developed in a book chapter. The book chapter is finished and with the press, but it may not be out before the journal article. In such a case you have a legitimate reason to cite the 'forthcoming' book chapter.

Anonymous said...

FWIW, I always roll my eyes when people cite themselves, esp. when they're more junior.

zombie said...

2:53 -- Why would you especially roll your eyes when junior people cite themselves? Those of us who are junior and work at STEM universities know that the number of citations is counted, right along with the number of publications, when those tenure decisions are made. And self-citing counts. So, go ahead and roll your eyes, but it's a necessity.

Gads, there are prominent people in my AOS who cite themselves dozens of times in a single paper.

zombie said...

2:17 -- Citing something that's forthcoming or in press is definitely fair game. There at least, the reader can find the cited work when it is eventually published. I took the question to be about unsubmitted or unaccepted papers that are not at that stage.

Anonymous said...

I think self-citations of unpublished work are fine as long as they are genuinely relevant and available online (e.g. on a work in progress page on your website). The main thing is not to ask your readers to just trust you - they deserve to be able to check the argument themselves.

Anonymous said...

Why would you 'roll your eyes' at people who cite their own published works? Seems irrational. Apart from the stuff about citation counts, it often just makes sense, especially if you are lucky enough to get several papers accepted that deal with interrelated topics. Given the tight word limits at most journals, it is often extremely helpful if you can say, 'I cannot further discuss this point here, but I examine it at length in xyz.'

Anonymous said...

2:53 here.

I meant that I roll my eyes when people cite their own unpublished work. (I assume citing paper X before it's even forthcoming doesn't count towards citation numbers. Maybe I'm wrong though.)

But I don't want anyone to put too much weight on what I say (thought that you likely would). The practice just reminds me of this:

Anonymous said...

As a reader, I like it when an author cites her own work. If I like what I'm reading, it tells me where I can go to read more.

Anonymous said...

"I took the question to be about unsubmitted or unaccepted papers that are not at that stage."

If this is what we are talking about, then I say don't do it.

Citing unsubmitted/unaccepted papers looks amateurish. May as well cite your graduate seminar papers and personal notebooks.

The whole point of citing work is to tell your reader where to find it. If the work you are citing is inaccessible and/or does not formally exist, then the citation does no good whatsoever.

Anonymous said...

@Anon 5:07

"The whole point of citing work is to tell your reader where to find it. If the work you are citing is inaccessible and/or does not formally exist, then the citation does no good whatsoever."

I'm not sure so that's accurate. After all, it is common (if slightly pretentious) practice to site information gleaned from private conversations in footnotes.

I think the primary reason for citation is to demonstrate familiarity with the relevant literature. Offering people research leads is only a secondary reason. The chief reason not to cite your own unpublished work is, if your work is unpublished it's probably because it is not all that good, or at least it is not yet polished enough. And since you likely lack reputation (only one of us is Kripke after all), there's no reason for anyone to take it seriously until it has passed peer review.

Anonymous said...

Just about anything you do you can do obnoxiously. Self-citation isn't an exception. Here's when I find it particularly annoying:

(i) It cites the author's work on issues that you probably don't need to think to hard about to follow the argument of the current paper;
(ii) It cites the author's work on issues where there are better papers by others that cover the same ground that the author doesn't cite;
(iii) It cites the author's work on issues that maybe haven't been covered by others as well, but doesn't do a very good job citing the work of others in other respects.

I see a lot of self-citation from people that's perfectly fine. You should cite the stuff that you know/should know that's on point and I don't see why you should be excluded. You should get the same care and consideration that everyone should get, though. I only get annoyed when I see authors self-promoting without showing a similar consideration to others.

I get _particularly_ annoyed when people write to you asking you to cite their work when they don't return the favor. Someone emailed me recently to tell me that he had written a paper that addressed some issues similar to issues I was writing about and asked that I'd cite his unpublished work. I didn't mind sharing credit, so I did. Didn't return the favor. I think I have the right to be annoyed about this sort of thing.

Anonymous said...


I think the primary purpose of citations is to tip off people where to look if they're interested in finding more on the topic at hand, or knowing where an idea originated.

If they're really there to test an author's knowledge of the literature in her area, then they should probably be replaced by a better testing system.

If I like a paper, and there's a link to unpublished work by the same author that sounds interesting, I would email the author and ask for it. That seems like a reason to cite unpublished work to me (assuming it is polished/ready-to-go). But after reading this thread, I will think twice about citing any unpublished papers of my own. I didn't realize so many people were put off by it (and I'm still not sure I understand why).

Anonymous said...

In my view people should not ever be citing unpublished work. With published work, this depends on the function being served.

I think it can be OK to cite one's published work if there is a specific reason. This came up for me recently when I had an article I wrote that built upon some previous work. There was a step in the argument that I didn't want to spend lots of time on since I had covered it elsewhere and it would be too long to include. And I didn't want to not mention this fact because a reader might read the submitted article and think I had to say something at that point about the issue. So the only comfortable solution was to mention this and cite the other (published) article. Just don't do this very much or it will look like you're trying to sell yourself and that's not something I favor.

Anonymous said...

Anon 7:21a writes: "I think it can be OK to cite one's published work if there is a specific reason. This came up for me recently when I had an article I wrote that built upon some previous work. There was a step in the argument that I didn't want to spend lots of time on since I had covered it elsewhere and it would be too long to include."

OP here. This is pretty much my situation -- I want to build on / presuppose work that I've done elsewhere. But my problem is that the work is not yet published.

In my case, it is under review somewhere, but who knows its chances of getting published in the near future? (Well, I do -- less than 10%, going by most journals' self-reported submissions/acceptance stats ... :( )

So the gray area here is that _I_ feel like the paper is polished & ready to go, because I'm ready to send it out somewhere, but that's not to say others will agree. Hence the feeling of risk about citing it.

zombie said...

11:08 -- it strikes me that if you are at a point where your papers are truly building on previous work, then you need to get that foundational work published before you keep building on it. As a reviewer, I think I'd reject your paper if a part of the argument was missing, and there was no way for me to see the missing piece.

I.e. this is not so much about whether you can cite yourself as it's about whether you can get published by building on (and merely citing) unpublished work.

I know the review/publishing process is torturously slow, especially if you are early career and really need to add some lines to your CV. If paper 1 gets accepted, good on you, and keep going. If paper 1 is rejected, you might think about finding a journal that can turn it around more quickly next time. Or aim lower on paper 1, so you can aim higher on paper 2 after paper 1 is accepted. Meanwhile, keep subsequent work in your back pocket, so it's ready to go as soon as paper 1 finds a home. (Journals that publish online first could really move the process along. Then you've got a DOI to cite, which is totally legit. e.g. Springer's journals are, IME, very quick about getting accepted papers through proofing and published online first -- literally within a couple of weeks. )

Anonymous said...

7:21 here.

Regarding 11:08, it seems to me a problem with citing unpublished work is that this unpublished work is not "fixed in time." I mean, it could undergo revision or something in the process of getting published and the content of the argument in it might change (someone above mentioned this, I think). This is sort of worrying a bit, and the reason is that it means one is not technically citing a "finished product" but something whose content may be changed. I'm not sure that scholarly work is supposed to be like this. A citation is meant to refer to some material that could be checked for its specific content I would think.

zombie said...

"I think the primary reason for citation is to demonstrate familiarity with the relevant literature."
-True when you're writing your dissertation.
-True if you're unlucky enough to get a referee who will reject your paper because he isn't cited in it (it happens)

But not really the primary reason for citation in a professional paper.

Anonymous said...


What do you take the primary reason for citation to be?

Anonymous said...

The idea that citing unpublished work is meaningless because it lacks evidentiary merit until vetted in peer review should be obviously rebutted by the practice of citing work that is the subject of critique and counterargument. I'm surprised, thus, at the number of comments here that seem to treat a citation as though it were a step in a proof.

zombie said...

I take the purpose of citation to be:
-Give credit where credit is due
-Point to supporting evidence (My work is empirically oriented, so I do this a lot)
-Point readers to other work that is relevant/of interest, or which may be somewhat tangential or beyond the scope of the present paper
-Point readers to other work that augments the present work, or fills gaps in the present work
-References source theory/material

But I don't think it is just to prove you're familiar with the literature. That would make the biblio quite extensive for well-read authors.

Anonymous said...

10:42 writes: "The idea that citing unpublished work is meaningless because it lacks evidentiary merit until vetted in peer review should be obviously rebutted..."

7:21 again. I agree with this. My concern with citing unpublished work is not that it is "not vetted" or something. The problem is that it is not "fixed in time." This is a concern because until it is published there is nothing that can strictly be held against someone in an unpublished article I take it.

Anonymous said...

I have cited unpublished papers of mine a couple of times in the following kind of situation.

Paper X addresses multiple audiences (e.g. technical and general). Question Q is salient but only of interest to the technical audience and it is addressed in Paper Y. Whether or not Paper Y is published, as long as it's available on my site, there is no need to burden paper X with the technical discussion.

I see nothing wrong with this. In fact, I may at some point start self-archiving technical notes that I do not even **attempt** to publish.

As for citing one's PhD thesis, I can see a couple fairly strong reasons for doing so:

1) it's a publication, and until you publish a book it's likely to be your only monograph length publication. You probably had the space (and time) to develop details and extensions you cannot address in an article.

2) your 2014 (say) dissertation might develop an idea that you do not publish until 2019. Meanwhile, XYZ independently published a similar idea in 2017. Citing your thesis is a way of reclaiming the originality of your insight.

Anonymous said...

The primary purpose of citation is to give credit where credit is due. Honestly, philosophy as a discipline is really bad about this---bordering on academically irresponsible.

Anonymous said...

Does anyone recall how many schools interviewed at the APA (as opposed to Skype) last year? I'm trying to decide whether to make plans ahead of time.

zombie said...

6:56: I cannot answer your question about the numbers, but I myself would not make advance reservations (not non-refundable flights, in any case). My experience last year was that all of the schools that interviewed me either were doing first rounds via Skype, or offered me the option of Skype or APA. I chose Skype in every case, and it does not seem to have been detrimental.

Anonymous said...

One think to keep in mind is if you're submitting your paper to a journal for blind peer review, it is very easier for reviewers to figure out who you are--they can just look in the biblio to see whose unpublished recent Ph.D. dissertation is being cited. 9 times out of 10, it is the dissertation of the author. This may prejudice the reviewer (positively or negatively, it depends).

Anonymous said...

Wait, you actually cite your own unpublished dissertation in an allegedly blinded paper for review?

Yeah, that's not a good idea. Someone should have told you.

zombie said...

It is customary, when preparing papers for blind review, to remove identifying citations. You can do this while still making it fairly clear that you've got other published work that is relevant (which you will cite once the paper is accepted).

Unpublished dissertations, I dunno. But I rather think not.

Of course, bigwigs don't bother, and openly refer to their previous work and cite themselves in papers. But the rules are different for them, right?