The Order of Things

by Henry Farrell on February 25, 2010 · 10 comments

in Political science

Scott Jaschik does a piece covering the recent debate in PS on co-authorship, and coincidentally reminding me that I’ve been meaning to do a post on this for the last few weeks. In particular, I wanted to note David Lake’s piece, suggesting that we should throw off the tyranny of alphabetical ordering of co-authors’ names on articles, and adopt a new convention under which the names are listed in the order of relative contribution, but with the senior scholar (in a situation where there is a clear gap between a senior scholar and a junior scholar or junior scholars) coming last, where appropriate.

My first reaction is to note that John, Eric Lawrence and I have adopted a far superior ordering system for a forthcoming article, in which the authors are listed in the order of their familiarity with the intricacies of Big Ten Football.[1] My second and more serious reaction – I wonder whether the ‘senior author goes last’ convention which, as Lake notes, is common in more scientific disciplines, is mostly a product of different funding structures. In the hard sciences, as I understand it (or possibly misunderstand it), the last author is not (as in Lake’s ‘senior scholar as mentor’ model) necessarily actually involved in the writing of the article at all. He or she may simply be the director of the lab where the research was done, or the PI on the grant. His or her job was acting as the funder or rainmaker for the project. Thus, the informational content of the last author position is quite specific – it tells you who secured the money and resources. This may make it more difficult to transplant the convention to political science, where large grants are the exception rather than the norm. As Lake correctly notes, name order conventions are often the product of power hierarchies (Lake of course is best known for his study of hierarchy in the purportedly anarchical international system), and it may be hard to transplant conventions from one system – with a particular set of power relations – to another. I suspect that another of Lake’s proposed innovations – explanatory footnotes which explains who did what – might be more viable with a bit of a push (it would also have higher informational content).

fn1. Read the next issue of PoP if you think I’m bluffing. John and I are also going to be collaborating on a new piece, where I am proposing that we list the authors in the order of their relative dweebishness. Some might think that this is a stratagem on my part to gain the coveted first author position. These are the people who do not know the full story of John “Songbird” Sides’ carefully occluded past history.


Matt Jarvis February 25, 2010 at 11:57 pm

Just a reminder that I need to stop coauthoring with Justin Buchler and work more with John McNulty…..

Joel February 26, 2010 at 12:39 am

you could go with whoever has the highest level magic user …

Todd Coleman February 26, 2010 at 12:47 am

I can’t speak for other “hard” sciences (and I believe there is variation even within physics), but particle physics and mathematics use the alphabetic order convention.

Forrest February 26, 2010 at 12:59 am

Thanks to perception bias, assessing relative contribution is never easy. Is the person who analyzes the data; collects the data; comes up with the ideas; funds the work; or writes the results in such a way that other people can understand it making a bigger contribution? Everyone knows how much time they put in and inevitably overestimates their relative contribution. My preference is to list junior colleagues first since they derive the most benefit from any exposure and since I like my coauthors. After this, I prefer to go alphabetically to discourage people from trying to make something out of the ordering. Since my last name always starts with an M and the two most common letters I coauthor with are are B (Binder and Bailey) and S (Shipan, Smith, Sigelman, Spriggs), it is clear that I am not making this suggestion out of self-interest. It is simply the right thing to do. Trust the middle of the alphabet.

Erik February 26, 2010 at 5:31 am

I prefer alphabetical too for the same reasons as Forrest and my name starts with a V! (Perhaps the existing ambiguity and the V make me a more attractive co-author than I otherwise would be).

John Sides February 26, 2010 at 10:11 am

Ascending or descending order of dweebishness, Henry?

Erik, let’s start a co-authored project. “S” is getting me nowhere fast.

Mike Bader February 26, 2010 at 11:36 am

Henry, this is a really interesting post and I can offer some insight to the public health model that sits between the natural-science/medical world and the social sciences. You are correct that the senior author position is driven largely by funding mechanisms, but there are logical reasons for doing so that I think can be adapted to political science and other social sciences (like sociology, which is my discipline).

Usually, the PI writes the first paper or couple of papers that come out of a given grant as the first author. Then, after those initial papers are published that establish the new line of research, the PI takes the senior author position as the last author. In my experience, however, the senior author has usually done a great deal of work on each paper, if not directly then certainly indirectly. For example, often much of the conceptual work in manuscripts is pulled from the grant itself (not plagiarized, but developed from the background section of the grant) and often work-groups or labs have regular meetings where papers are discussed. So it might be true that the PI never writes a single word in a given article, but in reality their influence is more than simply acquiring the funding.

On the other hand, public health also has some projects that don’t fit well inside of the typical medical model. In these instances, when a paper is more conceptual or lays the groundwork for a junior scholar to start developing their own funding stream, then the PI will often minimize their role by either not occupying the senior author position or, less frequently, not being included on those papers.

In my experience, it has been far more favorable of an experience because the guidelines are well-known and the conventions are pretty standard compared to my experiences of co-authoring in sociology where there is a lot more time and angst involved with deciding authorship order.

Paul Camp February 27, 2010 at 5:54 pm

I don’t know what scientific disciplines you are thinking about, but in all the ones I’ve worked in, the senior ego comes first, followed by alphabetical. This is so the resulting paper will always be referred to as Bigshot, et. al.

lylebot February 27, 2010 at 7:57 pm

In my field we order authors by relative contribution. This is not that difficult, as most papers have only two or three authors, and most of the work will have been done by one of them. Any difficulties can be resolved through negotiation.

Does it have problems? Of course, but so does straight-alphabetical. I would argue that being able to tell who did the most work at a glance is pretty valuable when you have to get a sense of someone’s achievements based on reading a CV.

At any rate, no system is going to be perfect. It’s amusing how everyone argues that the system they use is necessarily the best, though.

bbnck March 1, 2010 at 8:41 am

As someone who works at the intersection of math and experimental sciences, I’ve had experience with both conventions: alphabetical (usual for math papers) and magnitude of contribution (usual for the natural sciences.)

I favor the alphabetical. It’s not always straightforward to determine who did the most work on a paper. Sometimes one person only contributed one small portion, but it was the key part. Sometimes the less senior contributor (student, postdoc) was actually more important than his mentor. When (as in math) the key work in the paper is thinking, not performing experiments or writing code, and when much of the thinking is done in conversation, it’s fruitless to try to decide whose thoughts were better or more important.

Having to hash out “who’s more important” can lead to bad blood between colleagues that lasts for years.
I personally hate jockeying for status. Give me an arbitrary system any day.

(Then again, my name is at the beginning of the alphabet.)

Comments on this entry are closed.

Previous post:

Next post: