The Philosophers' Cocoon

A safe and supportive forum for philosophers.

Owned & Moderated by Marcus Arvan (University of Tampa)

Contact: marvan@ut.edu

Blog mission & moderation policy

IMPORTANT NOTE on how to ensure anonymity in commenting
Job-Market Reporting Thread (2025-26)
Job-Market Discussion Thread (2025-26)

recent posts

about

Writing sample: published or unpublished?

In our newest "how can we help you?" thread, E asks:

Suppose you're choosing between two papers to use as a writing sample. One is the best paper you've written to date. It was published a few years ago. The other is a new manuscript. You have evidence that it's a good paper. Which should you choose? Does it matter?

Excellent questions! Offhand, I would think that one should simply use one's best paper, as the presumptive point of a writing sample is to display you doing philosophy at your best. Then again, I can't help but wonder whether some search committee members might prefer unpublished work for a couple of reasons. First, if the work is really excellent, then it may convey that your best work is ahead of you, rather than in the past. Second, it might excite the committee about where your research is heading and demonstrate that you have excellent work that is likely to be published once you are hired.

Long story short, I'm not sure. What do you all think? It would be great to hear from search committee members what they prefer and well, as well as from job candidates who used already published work (how well did that approach work?).

Posted in

14 responses to “Writing sample: published or unpublished?”

  1. Peter

    I think you gotta go with published haven’t you? Sometimes, at the short listing stage, you get a chance to submit another piece of work – and maybe that’s an opportunity to be more future looking?

    Reply
  2. SLAC Associate

    I’d suggest using the published piece for the writing sample and saving your exciting new manuscript for on-campus interviews.

    Reply
  3. Slop

    Those who say published can you give reasons? I am having trouble seeing what the reasons would be to submit something published but older over newer (and very good!) but not yet published.

    Reply
  4. Cecil Burrow

    The reason to prefer the published paper is that it probably better and more polished than the unpublished one. If that’s not true, then of course go with the unpublished one.

    Reply
  5. slop

    Thanks. I agree that all things considered you should probably submit your best work, though there are other considerations (I was assuming the question came from someone who thinks the new thing is also best, but I realize now that might not be). For example, is the older paper on a topic that you are going to continue to work on? Does it fit with the job ad (or the intellectual bent of the dept you are applying to) better or worse than the new thing? Is the new thing closely tied to your dissertation, and is the old thing something you developed pre-dissertation? I think a lot of committees want to see things that are at least somewhat connected to your main projects.
    If it’s for a job with a PhD program, at an R1, or that you have a reasonable expectation will look at the research side of your application fairly rigorously, I think I would suggest the new paper (but only once a bunch of people have signed off on it being at least as good as the old one). That’s because, at least at my R1/PhD-granting program, when we review job apps, once we narrow things down quite a bit to the top 30-ish candidates, we tend to read some of their work that they didn’t submit. You’ll get more of a spread of what people read if you submit things in your application that aren’t just available online. I’m not sure how common this practice is but I don’t think we are probably especially unusual in trying to look at people in a more holistic way once we are trying to narrow down the longer list to an interview list.

    Reply
  6. New Assistant Prof

    I got a TT, state school RI job with an unpublished writing sample and a different unpublished job talk paper.
    My experience is that you should submit the paper that best shows you are a good philosopher in the relevant area, including, potentially, teaching areas. That might be published work, but it might not be.

    Reply
  7. David O. Brink

    I agree with Peter, SLAC, and Cecil. Include the published piece as part of the application and save the unpublished piece for requests to see more work or a possible job talk. Like most advice, it’s ceteris paribus. This assumes that the published paper falls squarely within the advertised AOS, is not too old, is representative of OP’s interests, and appears in a respectable venue. There are several reasons for this advice. OP describes the published paper as their best work to date. Also, as Cecil notes, published papers tend to be more polished than unpublished drafts. Further, there’s the fact that the published piece has survived peer review, and so is an actual and not merely potential publication, and is nicely formatted. These factors can weigh consciously and/or subconsciously with hurried readers.

    Reply
  8. re slop

    I am not a committee member but I have heard this reasoning: places with smaller faculties that lack someone in your subfield might prefer the published piece because it’s been vetted by subfield experts. (We are bracketing, for now, the assumption that peer review functions well…).
    But for places with larger faculties, then you might go ahead with the unpublished piece, if you think it’s really just as polished as the published work.

    Reply
  9. Bill Vanderburgh

    Published, unless the unpublished one is closer in area to what the job is searching for. (Committees are looking for reasons to exclude files. Something that has passed peer review raises fewer doubts, especially when the committee has little expertise in the subspecialty, i.e., almost always.)

    Reply
  10. slop

    I find some of these comments puzzling in the sense that committees also have a CV to look at, so they know that the candidate has x number of papers that have passed peer review and been published. I guess I don’t really understand why anyone would (all things being equal) prefer a published to non-published writing sample if the candidate had, say, five publications in peer-reviewed journals. The CV would give the evidence that the person’s work is publishable/rubber stamped by others, no? Isn’t the function of a writing sample so a committee can directly evaluate the candidate’s actual work? (Otherwise, why not just use the CV alone?)
    (By the way I don’t mean “puzzling” in a critical way–I couldn’t find a better word that made clear that I wasn’t trying to come off as obnoxious about this, I just am genuinely interested in what the norms are, why they are there, and what explains away my apparently wrong intuition here.)

    Reply
  11. Andrew

    I was a bit surprised to see the number of people who picked published. My first instinct was unpublished. My reasoning might be completely flawed but “published” does not equate to “best” or “polished” because in my experience publication can be a pretty messy business. An unpublished work can still be improved on whereas a published version is what it is and the author can’t make it better.
    Those emphasizing the opposite seem to be focusing on the value that the committee might attach to something in print as a demonstration that you can get things out there. But shouldn’t including it in the CV be enough to show that?

    Reply
  12. Daniel Weltman

    I might be very weird but my own view is that to think one ought to submit one’s “best” work or to think that something that has made it through the scrutiny involved in peer review is likely to have improved is to have a lot of misunderstandings about how philosophical work is judged. There is very little consensus when it comes to fine grained judgments like “is paper X better than paper Y.” If you have published 5 papers and send them to 5 people you will probably get 5 completely different rankings of which is best and why. Often the things you do in order for a paper to get published will make it better in the eyes of reviewer #1 and reviewer #2 and worse in the eyes of 90% of readers (and maybe worse according to your own judgments).
    Thus I think it’s kind of a hopeless task to pick one’s best paper, where “best” picks out what the hiring committee thinks. And if one wants to send one’s best paper by one’s own lights, one might as well trust one’s own judgment rather than the judgments of one’s peers, since depending on what journal you’ve submitted to and how lucky you’ve gotten, sometimes the people doing peer review are clearly in a worse position to evaluate the paper than you are. (Sometimes they are in a better position, and their comments make the paper better, but then hopefully you are able to recognize this, and then your judgment ends up aligning with those of the peer reviewers, and so once again it makes sense to trust one’s own judgment.)
    That doesn’t help at all when it comes to whether to submit published or unpublished stuff. I don’t really have any sense of which is better or even if it makes a difference.
    My suspicion is that for some members of the hiring committee, it will improve their estimation if you submit something published, because this shows you can publish stuff. (This is of course ridiculous, as slop and Andrew have pointed out, because they have your CV. But we are humans, not robots, and we evaluate people in ridiculous ways.)
    For other members of the hiring committee, it will improve their estimation if you submit something unpublished, because this shows you’ve got more ideas than just your published material. (This also ought to be irrelevant insofar as you mention future work in your research statement, but see above.)
    And perhaps the most common situation for most members of the hiring committee is that it’s irrelevant: whether the work is published will not play any role in their judgments.
    I am pretty confident the three-way spread I’ve just described is a pretty good description of how things work. I am much less confident that I have the category sizes correct: that is, I think category #3 (it doesn’t matter) is the biggest, but I might be wrong.
    Perhaps ideally there should be a poll on the topic. Have people vote: “if you’re on a hiring committee, would you prefer to see a published or unpublished writing sample?” with the options “published,” “unpublished,” “it’s irrelevant,” and “other.” I’m not sure I have enough philosopher Facebook friends to run a helpful poll but I guess I’m curious enough that I might try it just to see what happens. If I get any results I’ll post them here.

    Reply
  13. additional

    I agree with the most that has been said. To add a little additional observation, arguably, the work that some committee members find exciting > work that everyone finds good. The former will push hard to get you shortlisted or get the offer. The latter is unlikely to push hard if your profile have some shortcomings. This was my experience as a job candidate and a committee member. But this is only useful if you have some ideas about the search members.

    Reply
  14. early career

    For what it’s worth, I always attempted to send out unpublished writing samples whenever possible, and gave 6 job talks and got 4 job offers in 2 job market cycles (including at a few Leiter-ranked places). This also accorded with advice my advisers at my grad program gave me. Their reasoning was that a job application is a place for search committees to see new work, that you’re continuing to produce, and that you have an active research pipeline beyond what you’ve already gotten out. Given their reasoning, I think a piece published “a few years ago” (meaning it would have been accepted, and written, likely a few years before that) might count against you in certain departments. There’s simply no other part of the application for the search committee to vet the quality of your pipeline beyond what’s already publicly available. Also, I think the number of people answering “published” may be discounting the degree to which newer work should, in general, be better than older work for ABDs or early career people, so that unpublished work may very well be better than work that’s already come out, despite the lack of peer review.

    Reply

Leave a Reply to Cecil BurrowCancel reply

Discover more from The Philosophers' Cocoon

Subscribe now to keep reading and get access to the full archive.

Continue reading