“Encouraging” solo authorship would be a really bad idea

Image credit: S. Heard.  Hand models: Ken Dearborn, Allyson Heustis (thanks!).

I just read an intriguing opinion piece, Garsten et al’s (2015) “Single authors: an exterminated race”, which argues that “the scientific community could benefit from encouraging solo authors”. By all means read the piece (it’s a short and easy read); you may agree. I don’t, and here’s why.

Garsten et al. begin with the familiar observation that the fraction of solo-authored academic papers has been declining for a long time, especially in the natural sciences. (I’m certainly no exception: in the last dozen years I’ve published just two solo-authored papers, and I don’t see many more on the horizon.) For multiply-authored papers, the average number of authors has been rising too (sometimes to silly extremes, with hundreds or thousands of authors*). There are many reasons for this, as Garsten et al. correctly point out, including increasing complexity of scientific work and (less admirably) author-list inflation arising from our metric-obsessed attempts to measure researcher quality through publication counts.

So far, not much to argue with. But things get interesting when Garsten et al. move from description to prescription, arguing that solo-authored papers are superior to multiply-authored ones**. They claim that “single authors papers [have] qualities unobtainable in multi-author papers”, in that “they present an opportunity to publish opinionated and creative thoughts unbound by compromise.” They go on to suggest that solo-authored papers (but not multiply-authored ones) can produce “highly creative and laudable individualistic approaches to fundamental questions”.

This is an interesting and important claim. It would be difficult to put it to quantitative test, although one might ask whether papers later cited as paradigm-shifting have shorter authorship lists than the average for their field and time. I’m not going to do that analysis (because I have a day job!), but it would make an interesting science-studies PhD thesis***. Instead, I’ll turn to personal anecdote . My own experience is that multiply-authored papers can be more creative, not less, than solo-authored ones.

In fact, greater creativity is one of the things that draws me to coauthorship. Throughout my career, I’ve published multiply-authored papers that were far better than anything I could have written alone. Sometimes that’s because a team can cover more methodological ground, combining ideas from different fields to make an advance no single author could have made alone (as with the paper I blogged about here, or more dramatically, with my papers on the diversity of T-cell response to immune challenge). Other times it’s because ideas come out of two authors tossing ideas around, or arguing about them, maybe even before they know they’re going to coauthor a paper (as happened with my paper on the heritability of speciation rates). Many times I’ve had a conversation that went like this:

Author A: I think thing X.

Author B: You mean thing X1?

Author A: Well, no, I meant thing X2. But your X1 idea is better. Or is X3 better still?

Author B: Ooh, yes, I hadn’t thought of X3 – that’s brilliant. Could we test it with experiment Y?

And so on, until in the very best collaborations the team ends up with idea X7 being tested with experiment Y4b and nobody can quite remember whose idea anything was. That’s where creativity comes from – the jostling, tumbling play of ideas that ensues when two or more scientists think usefully different things. That’s why I coauthor papers – not because I can write them more easily, or because I can publish more of them, but because they’re better. More interesting, more integrative, and yes, more creative!

Now, Garsten et al. are quite right that we should make sure we give solo authors and coauthors appropriately weighted credit for their contributions (although defining “appropriate” credit is not trivial). But that doesn’t mean thinking solo authors, or their solo-authored work, are necessarily superior. Coauthorship brings wonderful opportunities; let’s not “encourage” each other to miss them.

© Stephen Heard (sheard@unb.ca) April 28, 2015


*Yes, ATLAS detector collaboration, I’m looking at you.

**They even suggest that PhD students might be required to publish a solo-authored paper as a condition of graduation. Here’s Alex at The Lab and Field explaining why that’s not a good idea.

***To be published solo-authored, of course.

Advertisements

20 thoughts on ““Encouraging” solo authorship would be a really bad idea

  1. ScientistSeesSquirrel Post author

    Thanks, Jeff; that link is temporarily down, but I’ve added two more. And thanks for linking to your post. (You are right about the author Van being made up: see Vincent, T. and J. Brown. 2009. Thomas L. Vincent: a game theorist for all seasons. Evolutionary Ecology Research 11:137–138.)

    Liked by 2 people

    Reply
  2. jimwoodgett

    The point they missed is that *some* types of scientific paper are indeed suited as single authors. These are usually hypothesis or isolated experimental result studies. But most experimental work rests on the shoulders of many. Moreover, taking this to the ridiculous limit, one would have a single author paper with no references cited. The authors do have a few good points. There is a temptation to add authors to bulk up a study. There is also a problem with attribution with many-authored papers but these are natural aspects of science. The high-energy physicists are well ahead of us (as usual – e.g. Arctic). What I would like to see is more single author papers from graduate students (i.e. without their supervisor). Yeah, not holding my breath.

    Liked by 3 people

    Reply
    1. yannponty

      Single authors papers by grads routinely happen in Maths/CS (and more generally in theory-centered fields). I’ve seen it a lot in France in the past (but I believe it is declining), where academics/supervisors are tenured much earlier in their career, giving them less drastic incentives to systematically put their name on student’s contributions. Also, single-author publications are very positively perceived by hiring committees, and this gives mentoring academics indirect incentives to “do the right thing” (so that their student will get hired in more prestigious institutions, indirectly impacting their own career).

      Like

      Reply
  3. Michael Wiener

    Among the best papers I’ve ever read, quite a few are single-authored. However, these papers tend to come from brilliant loners (or at least a brilliant people from a different field). The papers I’m thinking of seem to come completely out of left field and change the direction of research for years afterward. If researchers choose to write single-authored papers, this will not make them more brilliant or more capable of contributing ground-breaking ideas. Just as it makes no sense to buy an expensive car in the hopes of becoming as rich as your neighbour, it makes no sense to write a single-authored paper in the hopes of becoming smarter.

    Liked by 3 people

    Reply
  4. Tobi

    I think the increasing rarity of solo authorship is a reflection of the emerging complexity of scientific mode enquiry. Some of the earlier work involving solo authors are largely descriptive and others are discipline-specific. Those type of inquiries are not popular anymore. Now, we talk about integrating molecular and climatic information, linking genes with ecosystem function and so on. The skills required for these lines of inquiry rarely reside in the toolbox of one individual. I don’t know a lot of climate scientists with molecular background. Hence, the increasing collaborative work.

    While I subscribe to the cliche that two heads are better than one, I worry that the extent of collaboration that we are now seeing may be weakening the healthy competition that has propelled science to where we are now. I sited with a group where the discussion seems to be more (not intentionally) about making everybody happy than upholding good science. I also worry about researchers coming together (again not by intention) and perpetuating dogma. All of these are potential dangers of unchecked (tall order) collaboration. I’m not sure to what extent these are happening, in collaboration, we certainly compromise on our good insights or disagreement or subscribe to the pessimism of others that certain grounds cannot be accomplished or not important. But history has shown us time and again that solitary insights can be durable.

    Lastly, regarding grad students publishing solo papers, this is not something that is not achievable. I know a lot of humble researchers that will credit the success of their labs to the creativities of their grad students. The problem is that grad students typically don’t have strong writing skills (of course there are exceptions) while under training and peer review process can be cruel to the rookies. I read a blog post where a reviewer mentioned that he rarely recommend that MS submitted by grad students be accepted and on occasions where he did, he gave the authors a needless tough time. If we are judging MS by the originality and creativity of the authors alone, then publishing solo papers can be readily achieved by many grad students.

    Liked by 2 people

    Reply
  5. Jeremy Fox

    I don’t see much evidence that collaboration for collaboration’s sake is causing people to pursue collaborative work and publish multi-authored papers that they could equally well have pursued and written themselves. Nor do I see much evidence that people are systematically choosing to pursue collaborative projects when they could be pursuing superior projects themselves. But then, I’m not sure how you could possibly obtain such evidence. You’d have to look at not just whether people work independently vs. collaboratively, but somehow evaluate the quality of science they produced, *and* the quality of science they could have produced had they chosen to work in some other way than the way they chose.

    I think we should, and mostly do, favor good science, rather than favoring science that was produced in a particular way (e.g., collaboratively vs. not):

    https://dynamicecology.wordpress.com/2015/03/12/does-the-competitiveness-of-academic-science-devalue-or-inhibit-collaboration/

    Liked by 2 people

    Reply
  6. Manu Saunders

    Thanks for another thought-provoking post! I totally agree that co-authorship is a source of creativity. But, at the same time, I think that single authorship can often inspire discussion. Are multi-author papers more likely to conform to contemporary thought because the aims/message have to agree with more people? Are single-author papers more likely to present new concepts or ideas, whether they be misguided or not, that are essential to the development of ecology (and science in general)? Who knows? Rather than *encouraging* either way, I think it’s important to encourage researchers to do both – showing they can publish as a team AND as an independent researcher is better than showing they can only achieve at one method!

    Liked by 2 people

    Reply
  7. EnviroSolutions

    I am incline to agree with Garsten since the best academic papers I have read are single authored, but then again my conclusion might as well be premature since I am now getting into reading more coauthored/ multiple author research papers for broader perspective. I am currently undertaking the write up for my masters degree.

    Liked by 2 people

    Reply
  8. PJ Parsons

    Ova my 25 years in academia, I’ve observed several people promoted through the ranks — all of them with the same papers as their usual co-authors. Perhaps they should have shared one promotion?

    Liked by 1 person

    Reply
  9. Tana Daily Telegraph

    Reblogged this on Tana Daily Telegraph and commented:
    Much as I would support solo-authoring of papers, it’s worthy noting that collaboration is just as good as putting two heads together in the quest for breakthroughs of any form. Collaboration brings in diversity of ideas and approaches. Collaboration brings in innovation as opposed to invention. I am, indeed, a strong believer of the mantra “two heads are better than one”, or was it intended to be “two functional lobes of brain are better than half”. Probably.

    Like

    Reply
  10. The Last Wilderness

    One might fall back to that adage which says two heads are better than one. I’m an aspiring researcher and even though I’m not publishing yet, I have benefited tremendously from discussions regarding ideas. Also, collaboration with multiple scientists is an important aspect of interdisciplinary studies.

    Like

    Reply
  11. Pingback: Why grant funding should be spread thinly: some followup | Scientist Sees Squirrel

  12. Pingback: Does mega-authorship matter? | Scientist Sees Squirrel

  13. Pingback: Are “side projects” self-indulgent? | Scientist Sees Squirrel

  14. Pingback: Why I assign the group work that I hated as a student | Scientist Sees Squirrel

  15. Pingback: Negative results, conferences, and the open file drawer | Scientist Sees Squirrel

  16. Pingback: The coauthors I’ve never met | Scientist Sees Squirrel

Comment on this post:

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