Should one only publish one's best work?

We have discussed publishing strategies before on this blog, mostly in reference to which kinds of journals to submit to in order to be competitive on the academic job-market. However, there is another, broader issue that most of us encounter from time to time, and which I have been struggling with a bit lately: how should one decide which of one's papers to try to publish?

Given that we are human beings, it is highly unlikely that all of our work is of the same quality. Some of the papers (and books) we write are better than others, in some cases by leaps and bounds. And, of course, sometimes one writes something terrible (or, at least, I do!). Worse still, we also seem quite fallible evaluating the quality of our work--and those around are aren't necessarily the best judges either. In graduate school, some of the most "famous" papers we studied were profoundly flawed. In one case, I remember a fellow student, asking in class of one very famous paper, "How did this ever get published?" I recall everyone in the class--including the instructor, also famous int the field--being just as baffled about the paper's status as that student. And yet, despite its very serious flaws, it was a very famous paper that ended up establishing its author as a top-figure in the philosophical subfield. And it's not the only case like this that I can think of, by a long shot. Many profoundly influential works, both in philosophical history and the contemporary scene, also appear to be profoundly flawed--while many far less flawed (but perhaps less interesting?) papers seem to go relatively undiscussed. One reason for this could be because flawed but interesting works open up more ground for fertile discussion--e.g. refutations, attempts to clarify and fix the work's flaws, etc.--than less flawed, but less ambitious works.

Anyway, as a researcher trying to decide what to publish, these are some of difficulties one faces. One knows, or has reason to believe that:

1. Some of one's works are likely to be far better than others.
2. Some of one's works may even be terrible.
3. One is likely to be highly fallible in evaluating the quality of one's work.
4. Others around oneself are likely to be highly fallible in this regard as well.
5. Some deeply flawed works turn out to be influential.

How should respond to these facts, if indeed they are facts? Very early on in my career (as a grad student), I remember hearing from quite a few people that one should only try to publish one's best work. The rationale, if I recall, is that one should want to present oneself--particularly early in one's career--only in the best light, as publishing bad work could be disastrous for one's career. However, I think this may be too simplistic. And indeed, it conflicts with some important findings in contemporary psychology about what is known as "regulatory focus" (I thank my spouse, Maryana Arvan, for drawing my attention to this work).

Regulatory focus theory in psychology tells us that there are broadly two orientations that we can approach life-tasks with:

• Promotion-focused orientation: where one focuses primarily on hopes and accomplishments (i.e. achieving possible good outcomes), rather than avoiding bad outcomes.
• Prevention-focused orientation: where one focuses primarily on avoiding bad outcomes, rather than facing risks for possible good outcomes.

Notice that the advice discussed earlier--the advice I heard a lot as a graduate student--is highly prevention-focused. It says, in essence, "Make sure you publish only your best work, and don't dare publish bad work!" Now, in the abstract, this might sound like a rational strategy--for the very reasons I was given in graduate school: publishing bad work can "make you look bad." But alas, this is not merely an abstract issue. People who want to publish are human beings--and, as I will now explain, empirical findings in regulatory focus theory suggest that prevention-focused strategies can be disastrous if imposed upon people who more naturally approach life-tasks from a promotion-focused orientation. Allow me to explain.

Early in my career, when I adopted the prevention-focused orientation I heard advocated around me (viz. "Only publish your best work. Don't publish anything that could be bad!"), I found doing philosophy a rather miserable experience. I was doing philosophy in fear. I worried so much about avoiding saying stupid things in seminars, or writing stupid papers, that I lost my passion for philosophy: I found little joy in it. And, here's the thing...empirical psychology suggests this is what tends to happen when a person with one regulatory orientation (promotion-focus) is required to adopt the opposite orientation (prevention-focus):

"The decision has positive value when people attempt to attain their goal in a way that fits their regulatory orientation and it will have negative value when people attempt to attain their goal in a way that does not fit their regulatory orientation.^[4]...In the broad sense, for people to appreciate their own lives, they need to be satisfied and "feel right" about what they are doing, and the way they are doing it.^[7] If it is not satisfying, it is known as "non-fit", and they will not reach their desired goal." (my emphases)

In other words, when you try to teach a naturally promotion-focused individual to approach a life-task in a prevention-focused manner, that will tend to discourage the individual for pursuing the task at all. Which is precisely what happened to me in grad school. The more "careful" (prevention-focused) I tried to be, the less I enjoyed doing philosophy, and the less hard I worked. It was only once I was out of grad school, and adopted a much less risk-averse approach to writing and publishing (see "Trust thyself, let it flow, throw it all at the wall, and see what sticks"), that I began to enjoy philosophy again...and publish some work I am proud of.

In short, although an "only publish your best work" approach may be the most rational approach in the abstract, in terms of actual human psychology it might not be the most rational approach--as such a conservative, prevention-focused approach may undermine the motivation of naturally promotion-focused individuals. At the same time, a less risk-averse approach does have its risks. Nothing I say above should be taken to discount these risks. If one publishes some bad work, it could indeed be held against you by search committees, tenure committees, etc. But, even here, it is not obvious to me that the risks necessarily outweigh the potential rewards. In many areas of human life--business, professional sports, etc.--there do indeed seem to be two poles: people who succeed through adopting conservative, risk-averse (prevention-focused) approaches, and those who succeed through less-conservative, risky (promotion-focused) approaches. And indeed, in sports, everyone, as they say, "has a bad game" here or there. Even the best players (Peyton Manning, Tom Brady, Joe Montana, Cam Newton, etc.) have bad games now and then. The important thing, as far as their professional reputation is concerned, is to have many more good games than bad games. Might the same be true in academic philosophy?

Anyway, I am not saying one should adopt one of these approaches (risk-taking vs. risk-averse) over the other. Truth be told, I am writing this post because it is an issue that I am currently struggling with! Should I only aim to publish "my best work"? The fact is, I'm just not certain what the right answer is. There are risks all around. If one sits on ideas for too long, worrying whether they are good enough to publish, other people could publish them before you (this has happened to me before!). On the other hand, if one is not careful enough, one could publish embarrassing things (I will leave it to readers to judge for themselves whether this has happened to me;). My point in writing this post is not to advocate for one answer to the question over the other. It is to simply raise the question for discussion, drawing attention to some of the many complexities (and uncertainties) involved in answering it, including the relevance of contemporary psychology and individual differences of personality orientation.

In any case, what do you all think?

Comments

[Kristina Meshelski]

If publishing is a means for specialists to share research findings, converse about them, and advance human knowledge in general then you should publish what you think other researchers need to read. This may or may not be one's best work, it really depends on the state of the field. Also publications are used in different ways in practice- an article can be very necessary to publish in the sense that people will immediately read it and change their own research in response, or it can be useful for teaching, or it can be used to translate one sub field for use by another sub field... For me it really isn't about which work is the "best", it's about what will others find most useful.

[Kristina Meshelski]

At any rate, it is less anxiety producing for me, and probably makes me a better philosopher, to think of myself as trying to be of use to other researchers (and maybe to the public) than to think of myself as constantly being judged by other researchers. Sometimes hard to do, but that's what I try to do.

[Marcus Arvan]

Hi Kristina: Thanks for your comment! I think the general attitude you're espousing sounds healthy. Conceiving oneself primarily as trying to be of use to other researchers--rather than as constantly being judged by them--seems like a promotion-focused approach (which might, for that reason, be good for those of us who are promotion-focused!).

At the same time, I can't help but wonder whether it largely (if not entirely) pushes the central practical question, "What should I try to publish?", back a step. After all, suppose your aim is to publish stuff that might be of use to other researchers. How are you to judge *that*?

Here again, it seems one could adopt a conservative, prevention-focused approach (viz. "Other researchers are only likely to find good work useful, so I'd better not try to publish stuff that isn't really good!"), or a riskier, promotion-focused approach (viz. "I don't know *which* of my works will be useful to others unless and until I publish them, since reviewers and readers are the ones who will judge whether that work is useful to them!").

Do you think conceiving yourself as a researcher in terms of "use to others" tips the balance in one direction over the other here (i.e. a less conservative approach to publishing)? On the one hand, you could adopt the conservative approach of deferring to others' judgments (i.e. ask other specialists you know whether your work is useful). Yet, I have worries about this approach. In many creative areas of human life (film, literature, philosophy, music, etc.), "experts" have a pretty poor record spotting how useful a piece of work is in advance. J.K. Rowling's Harry Potter book was rejected by almost every publisher, Gottlob Frege was totally unimpressed by Wittgenstein's Tractatus, etc. One can of course adopt a conservative approach of deferring to others, but in that case one runs the risk of not sending out work that *might* be useful on the basis of a few people who think otherwise. On the flip side, one could adopt a much liberal approach of deferring to one's own judgment of what is likely to be "useful"--but there are risks there in the other direction: one could be deluded, thinking a new paper one has written is likely to be "useful" (when, let's say, it is actually terrible). So, here again, even if one simply conceives oneself as aiming to be "useful" to other researchers, one runs into the same problem again, no?

[publishing person]

Marcus,
There are many reasons to publish. Generally, one should publish high quality work only. And I would say that people should try to publish in journals that bound to be read. Publishing in obscure journals counts for very little, and such papers are seldom read (and then seldom cited).
With that said, many of us have other incentives to publish. I work at an institution that, like many in the US, assess merit and give raises annually (when they give raises), and one of the criteria used in determining who gets what sort of raise is publications. There is thus an incentive to publish annually. This sort of incentive cuts two ways: it does encourage one to finish projects that may otherwise just lay around (and should be finished), but it can also encourage excessive publication (and publication of weak papers).
And, sometimes I have agreed to contribute a paper for an invited volume (special issue or edited book). I sometimes agree to do so, even though I may have other things I would rather work on. I may want to contribute a piece because of the other people involved.
But one thing your post asks us to reflect on is why we are bothering to publish. I find it extremely rewarding, but it is most rewarding when the paper is of a very high quality, and it is received by others as such.

[Trevor Hedberg]

I've seen many discussions about this topic before (both on this blog and elsewhere), and there's always one thing I can't pin down: what exactly do we mean by "bad work" in this context? The plausibility of a maxim to avoid publishing bad work is going to hinge pretty significantly on just what's being defined as bad work. Here are some possible definitions:

Bad work = "work that is judged by your peers and/or mentors to be bad"

But philosophers disagree so frequently in their judgments that some are likely to find your idea promising while others think it isn't. Thus, since it will be hard for you to identify your bad work, it will be hard to avoid publishing bad work under this criteria unless you publish very little (i.e., publish only those things garner extremely widespread support from your colleagues and/or mentors).

Bad work = "work that is, by your own lights, not the best work that you can produce"

If this is the standard for bad work, then there are two problems with trying to follow the maxim. First, you can underestimate the quality of your own work (or overestimate the significance of an objection to it) and thereby misjudge the merits of your work. Second, this standard makes the imperative to avoid publishing bad work too demanding: why is every paper you produce required to be your absolute best for it to be worthy of publication? Surely there are some good ideas that advance debates and are worth publishing even if they aren't the best or most original ideas that the author ever had.

Bad work = "work that is below some rough threshold of excellence, either by your own lights or by the consensus of your peers"

This is my best effort at generating a plausible definition of bad work, but there is still some vagueness in this definition because it doesn't specify what the threshold of excellence is supposed to be. Is it the top 10% of one's work? The top 25%? The top 50%? Does the threshold vary from person to person? Might there be some experienced and talented philosophers who are justified in publishing nearly everything they write whereas certain other philosophers should exercise more discretion? Is this threshold established primarily by the status of the journal in which one is able to get the paper published? I don't have first answers to these questions, but I think we've got to make an effort to answer them if we're to understand what is meant by an imperative to avoid publishing bad work.

[Pendaran Roberts]

Worrying about whether something you've written is really good enough to be published is a good way to insure you publish little.

Whenever you get an idea that seems to you to be a novel contribution, write it up the best you can, and send it somewhere. Revise the manuscript based on the rejection (you'll most likely be rejected), and then send it elsewhere.

Often referees' comments will evidence that they don't really understand what you're saying. If so, make efforts to be clearer.

Often referees' will have objections. Even if you think they're dumb, try to accommodate some of them and respond to them. You are likely to get one of the referees again.

But you do have to draw a line somewhere. If a comment is really over the top or really misguided, ignore it. I've often done this.

If you have a novel idea and you keep working on it, improving it, and revising it based on comments, you'll eventually get published.

Don't worry about having been rejected a lot. It's a lottery. Even good papers that end up being published in top journals may be rejected many times.

My paper 'An ecumenical response to color contrast cases' was rejected 6 times before being published by Synthese. In fact, it was rejected by far worse journals. http://philpapers.org/rec/ROBAER-2

In sum, I wouldn't worry about what is good enough to submit. Do your best to write novel and interesting stuff then submit it. See what the referees say, then revise.

Let the peer review system decide what should be published. That's how I've gone about it for the last few years, and it's worked pretty well for me.

Not to say the whole process doesn't suck!

[Kristina Meshelski]

Hi Marcus, and sorry for my late reply! Yes, I admit this does push the question back a step. However, I find it easier to be an objective judge about whether my work is useful to others than to be an objective judge of whether my work is *good*. I think I sometimes over-value my work and sometimes under-value it, depending on my confidence level at the moment. Not that I am very sure about my work's usefulness, but it seems to make the issue more tractable.