On scooping

During a weekend conversation with Vancouver’s inarguably most famous female canoeing british blogger, spanning a range of diverse, controversial subjects (including, and not limited to, hockey and lazy people), the issue was broached of scooping in science, especially in her field, namely biology.
According to Wikipedia‘s definition, scooping takes place “when a scientist or research group publishes their findings first. Being beat to the punch in this regard renders one group’s work redundant, and is regarded as a very undesirable outcome […], particularly since in some cases a single paper can include years of work and can qualify the scientists for competitive prizes”.

I think that there is more to scooping than what suggested by the above definition. Science is a competitive enterprise. If a number of research groups separately and independently attempt to find the answer to an outstanding scientific question, one of them will typically reach that goal before the others, by sheer ingenuity, hard work, luck, or any combination thereof. Such an outcome is common to many a human activity, and it does not in the least devalue the effort of the “losers”.
In science, it is always necessary for more than one investigator to reproduce the same results independently, and even though more honour will be bestowed upon the first who made the discovery, proper credit will also go to the second (and third and fourth) who confirmed it, especially if using different methods or approaches.

The word scooping implies that the scientist or group that first arrived at the desired conclusion, did so also by taking advantage of crucial information that at one time was not in their possession, but rather in the hands of competitors, information that leaked out possibly due carelessness or maladroitness of the latter, or in some cases unfair practices on the part of the former.
Let us get one thing straight: No talk of “espionage” should be taken seriously within academia. The fact is, sharing of information with the outside world is an inevitable consequence of the very way in which the scientific enterprise operates. Any suggestion of withholding cogent information from the competition, in order to avert scooping, is nonsensical and should be firmly dismissed. Scientific progress requires collaboration, openness and communication. It will be a sad day for science when we all stop talking to each other for fear of being scooped.
Not only is anyone entitled to ask questions at conferences, anyone presenting work in public is expected to answer pertinent questions in full and convincingly. Of course, people in the audience could get a clue, use the answer to their own advantage, boost their own research effort based on a hint obtained at a conference, with time achieving a leadership position within the community, a position perhaps previously held by the presenter. Again, that is simply a fact of life — it is what competition is all about, and as long as it remains fair, most of us regard it as positive for science as a whole [0].

But there are illegitimate practices into which some scientists can and do unfortunately engage, in order to gain unfair advantage over competitors. Perhaps the best known case has to do with anonymous peer review of manuscripts submitted for publication in scientific journals.
Imagine yourself as a peer reviewer, with a manuscript sitting on your desk, submitted for publication to a prestigious journal by your fiercest competitor, on a hot subject into which you and your group have been working for years without making the progress that you had initially hoped for. The manuscript contains novel, important findings. Based on your experience, you can foresee that its authors will receive a lot of accolades, possibly even prestigious prizes.
The manuscript describes how your competitor has been able to overcome one specific hurdle that has stalled your students and postdocs for months — it was a simple idea, a little fix, but for some reason you could not think of it. Damn !
You know, you are sure that, if your group implemented that very same simple fix, you could get the same results of your competitor, perhaps go even beyond what (s)he did. You think of how great it would be for you, your postdocs and your students, if it were your group to publish those findings first….

Ooh, now, the temptation of putting that manuscript at the bottom of a pile of stuff with a “to do” label on it, is really strong. After all, you are, and will remain anonymous to the author of the manuscript. The editor of the journal will not be on your case if it takes a couple extra week, maybe even a month, to submit a review. That will give you plenty of time to do the same experiment (after all, the size of your own group is twice that of your competitor), write a manuscript and submit it for publication. Naturally, you will make sure to request that your competitor not be selected to review it (something that (s)he obviously forgot to do, or was not allowed to do by the editor)…
You know that the reputation that you have come to enjoy within your community is higher than that of your younger, and lesser known competitor. Nobody will believe them when they will cry foul, contending that they had obtained that result before you… the community, in its wisdom, will simply dismiss them as crackpots, and laugh at them… as for their students and postdocs, you can always hire them in your group — heck, they are good

According to Cath, and other friends of mine who do research in the life sciences, the above scenario is far from rare. People worry about it all the time. But in physics, I think it is fair to say that it is almost non-existent. Why ?
Manuscripts submitted for publication in physics are typically distributed around, at the time of submission, as preprints. Preprints constitute essentially an official record of completion of a piece of research work, available to the public before its publication on an actual journal. Electronic online archives such as ArXiv render the distribution of preprints straightforward, but the procedure of sending preprints to colleagues at other institution is decades old. I remember as a graduate student making dozens of copies of my first submitted manuscript, and sending them to a list of persons recommended to me by my major professor.

No evil scientist can really scoop a competitor, at that point. In a preprint-free world, they will simply claim not to have been aware of the existence of the work of the other group, at the time they submitted their own paper. But in the presence of the record provided by a preprint, that becomes very, very difficult [1]. The other group will always be able to point to its own preprint and claim priority.
Why is such a simple remedy not adopted in the life sciences as well ? According to Cath, there is no well-established tradition in her field, there are no suitable on-line archives (enjoying the same reputation and acceptance of ArXiv), and nobody is going to want to be the first. But even in the absence of a tradition, why not send dated preprints, with official university cover, to groups that could conceivably try to engage in unfair competition ? it would seem like a simple, obvious way to cover one’s back…

I am wondering whether the strong connection between research in the life sciences and industry, far stronger than in physics, might have something to do with that… but I shall defer comments to those who know more than me.

Notes

[0] Anyone answering a question at a conference in a way similar to “I am not at liberty of disclosing that information, which at this time should be regarded as exclusive property of our group, and as such classified”, would simply look silly and raise legitimate suspicions, at least in my discipline.

[1] Impossible ? No, not quite. Everything is possible, with humans. But unlikely enough that one ought not waste one’s time worrying about it, yes, it is.

Tags: ,

15 Responses to “On scooping”

  1. Schlupp Says:

    Ad [0]: Happens in more applied branches of our own trade, as soon as the information might be related to anything patentable. It is a standing joke, because, yes, it does tend to render conference presentations somewhat pointless. So, your last paragraph might have some truth, but then, the use of the arXiv or similar is not all that uniform in physics and mathematics either, so I am not sure.

    Another thing one should keep in mind when accusing people of unfairness in “scooping” is that it is very easy to overlook a pertinent paper, which should have been cited but was not, just by bad luck and without bad faith. Especially if its authors didn’t put it on the arXiv, in which case they have nobody to blame but themselves..

  2. Cath@VWXYNot? Says:

    Dude! You scooped the post I was going to write about how nice it would be to have such an accepted archiving system in biology!

    Also: kayaks, not canoes. Canoes are for Manitobans.

    Better response coming, I just have to check how much of my detailed thoughts I’m allowed to disclose.

    • Massimo Says:

      Dude! You scooped the post I was going to write about how nice it would be to have such an accepted archiving system in biology!

      Bull ! I didn’t scoop anything, you didn’t archive your post and did not send me a preprint, so I shall simply deny, deny, deny…

  3. Schlupp Says:

    Not much of a discussion around here, I see. Massimo, what do you think? If either of us mentions “washing machine”, will this get the comments pouring in?

  4. El Charro Says:

    Horrible last two weeks, hadn’t had a chance to check blogs until now.

    Scooping… oh boy, every time we have an invited speaker that works in our area (and here I don’t mean the exact same research, it’s more like we work with carbon nanotubes, they work with high-temperature superconductors but thus we are both in the condensed matter area) my advisor sends us an email saying that we should be very careful with our words. Not give anything out but still sound interesting. WTF?

    I think in this case the extra fear of being scooped is due to my advisor being an assistant professor, with only a few students working in the biggest projects he has thought of so far. I guess if we get scooped then that would be a big hit for the chances of getting tenure, at least soon. Plus, I would like to be the first one in the world to publish the stuff I am working on, might make my future just a bit brighter.

    The weird (to me at least) part is that our biggest scooping threat comes from my advisor’s postdoc advisor. I understand that, given you were trained in their group, you’ll probably end up working in the same general area, but I would think that there should be some kind of professional courtesy towards a former student/postdoc and refrain from stepping in his/her shoes at least for a few years.

    I think the intro song for Monk could *almost* apply to academia… it’s just a jungle out there!

  5. Giovanni Peccati Says:

    On the opposite side of scooping paranoia, it is interesting to note how in (quite pure) mathematics people are right now experimenting with the so-called “Polymath projects”. This is a form of massive blog-based collaboration prompted by the likes of Tao, Gowers, Kalai, for instance.

    See http://polymathprojects.org/

    for examples.

    Basically, the readers of a blog work together on an open mathematical problem they have agreed on. The contributions range from crucial advances to marginal remarks, and the final paper (which is physically written by some of the contributors) is signed as “Polymath” + a link to the blog webpage (this in principle allows to reconstruct the history of the project, as well as to single out the main contributors).

    I wonder whether such a model would be conceivable in other areas of science. I guess that for theoretical physics it should maybe work.

    Ciao

    Giovanni

    • Massimo Says:

      Ciao Giovanni,

      Interesting concept, but I am skeptical regarding its suitability for theoretical physics work. Fundamentally I see two problems:
      1) Theoretical physicists are ego-maniac “empire builders”, always wanting to be in control of their own projects and not really comfortable with the notion of it no longer belonging to them (broadly speaking of course, none of that applies to me — not at all). In this sense, the culture in mathematics may be different.
      2) Theoretical physicists are typically evaluated by their institutions based on the individual scientific contribution. Anyone whose record consists mostly, if not exclusively, of collaborative work may face a harsher review at tenure and promotion time. So, if anything a clear-cut set of criteria to assess individual merits would be necessary.

      Saluti

      • Giovanni Peccati Says:

        Dear Massimo,

        I would have probably said very similar things about mathematicians, had someone explained polymath to me, say, a couple of years ago.

        Most of us mathematicians are of course willing to build our own body of work, with our names clearly appearing on the cover of prestigious journals (plainly, this is a general appreciation that does not necessarily apply to me – said Pinocchio).

        In the case of the Polymath projects I was describing, I believe that part of the succes comes from the fact that they are promoted by some of the very best mathematicians around, widely known in several area of mathematics. For instance, Gowers and Tao are Fields medalists. Collaborating with them, even on such an informal basis, can be a very strong motivation.

        Another important point is that the problems are chosen in such a way that they can be relatively simply stated and understood by a wide audience of professional mathematicians. The first Polymath was about a combinatorial statement (with deep implications) that could be read and fully understood by the whole community. I do not know whether in theoretical physics open problems of the sort exist.

        Cheers, Giovanni

  6. Successful Researcher Says:

    Great post! In mathematics, unlike (theoretical) physics, there still is a noticeable fraction of people not posting their preprints on arXiv. This is probably due to somewhat different scientific cultures of these fields, as discussed at length e.g.
    here
    .

  7. Successful Researcher Says:

    For an interesting discussion of scooping in mathematics see this post at the Secret Blogging Seminar.

  8. Successful Researcher Says:

    Would anybody care to comment avoiding being scooped in the case of grant applications which by definition cover the things one is just GOING to work on and thus cannot post to arXiv?

    • Massimo Says:

      I personally regard this as less likely, for the following reasons:
      1) Program directors are usually not keen on leaving the proposal with a reviewer for too long a time, in part for that very reason.
      2) The work may not have been done yet, but the PI must give some evidence of being able to carry it out. That means that an experimental apparatus must have been built and be functioning, a code must exist and have produced some results, etc. That means that (s)he should have a head start over competitors anyway.
      3) It is usually possible to make sure that the proposal not be sent to direct competitors, namely scientists equipped and poised to work on the very same subject.

      • Successful Researcher Says:

        Thanks Massimo, but I am afraid that the reasons you list are of somewhat limited applicability. About the specific points you make: 2) does not apply to, say, theoretical physics and mathematics; as for 3), if the reviewer is smart, has a sufficiently broad scope (and, of course, is evil enough) nothing really prevents him, at least in the more theoretical fields, from switching to the problems outlined in the proposal he reviewed.
        So if your area is sufficiently hot, you never really have a reasonably complete list of possible competitors.

        Moreover, if the scope of the reviewer is broad enough, the chance of anybody (except for the author of the grant application, of course) smelling something fishy might be fairly small. In fact, I wonder whether there were ever any official investigations for the misconduct of this kind.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: