We tried to write an academic article in a day. You won’t believe what happened next.


In June 2106 I co-organized a workshop at the University of Southampton with Justin Murphy. Our objective was highly unconventional: we invited the participants to try to write an academic paper together in a single day. The product of that workshop has now been accepted for publication at International Studies Quarterly. You can access a pre-print version of that paper here. What follows are my reflections on the process of writing that paper, from the initial inspiration for the workshop through the final acceptance in a peer-reviewed journal

About five years ago Isaac and I were commiserating over beers at a conference about the inherently alienating experience of being academics. I was just coming to the end of the rather exhausting and demoralizing tenure track process at the University of Oklahoma, and Isaac Kamola had just landed a tenure track job after what was a long hard job search in the post-financial crash academic hellscape of a job market. We weren’t quite veterans of the academic conference circuit, but we weren’t exactly rookies either. And we were both struck by a similar observation – the typical conference experience involved spending the weeks before a conference frantically assembling a paper to circulate to a panel and discussants who were too busy with their own projects to pay much notice. Then you had to spend the days prior to your panel stressing about a presentation, only to give it to a room of a handful of disinterested colleagues. Maybe (if you were lucky) you would get a question or two in the 10 – 15 minutes left in the session that sort of kind of touched upon what you said in the paper, but was really about what the question asker was interested in – before everyone dashed off to the next event on the conference schedule. Then you were stuck with the puzzle of what to do with this paper you had bashed out under such tight time constraints – I guess send it off to a journal and hope that reviewer two was feeling kind that day? And then repeat. It was alienating in the way that Marx spoke about alienation – the papers we produced both felt deeply personal, yet stopped being our property. They ended up as things to submit to journals, hopefully get published somewhere, so we could put them on our CVs, maybe if our institutions were feeling generous they would pay the Open Access fees, otherwise they would be pay walled, owned by some multinational publishing conglomerate who would send threatening letters to us if we posted a version too close to the published version on Academia.edu or some other academic repository service. If we did this enough times, and our department was nice, we might get tenure. Which is certainly better than what the factory workers that Marx was thinking about when he coined the term alienation got, and better than most people in precarious employment get today, but really is a strange way to go about writing things. It is also alienating (and Marx foresaw this too in his concept of alienation) because the conference experience, the job market, tenure, and peer review are all competitive experiences. Certainly debate and critique are important avenues for testing and confirming ideas, but somewhere along the way these processes have become institutionalized, to the point where questioning in conferences often aim at destroying the scholar, not just the scholar’s idea, and gate keeping is not so much about maintaining quality as it is about making sure the right people, with the right connections, and the right views get and maintain access to the increasingly scarce resources like tenure track jobs and publications in high quality journals. This institutionalized adversarialism turns scholars against each other in subtle and not so subtle ways, making the production of knowledge isolating (especially if you work in a field such as theory where the norm is solo-authored work), often feeling like nobody cares about a paper you’ve poured your soul into for months, perhaps years, on end.

            Out of that shared sense of alienation came an idea: what if we flipped the whole conference on its head? At the time the idea of flipped classrooms were all the rage – the idea was to post lectures online and then have the students do the homework in class time while the teacher came around and offered support. What if rather than coming to a conference with a pre-written paper to share, people came with nothing but themselves, met some fellow scholars, and together they tried to write a paper from scratch? What if (and at this point we were many beers into our conversation) we got a group of people together and we agreed to write a paper in a day?

            Time passed, I moved to England. I went to a BISA in 2015 with Alex Prichard and we had a very similar conversation about how alienating conferences are. It triggered my memory of the conversation with Isaac, and I told Alex about our idea. He was enthusiastic about the idea – I believe his exact words were “That’s fucking brilliant! You should do it!” So I emailed Isaac the next day (I still have the email). And we agreed we should give it a go. Our first attempt was to pitch the idea to the International Studies Association as a workshop associated with the ISA annual convention. Isaac and I spent a good month on the proposal, and it was promptly rejected. Undeterred I went to our director of research, told him my idea, and he said he would offer us some funding for the project if we ran it at Southampton.  Now with some money to spend we set out trying to organize a one-day workshop. At this juncture a surprising obstacle appeared – we had a very hard time recruiting people to work on this project. We received many polite declines to our invitation for the workshop – perhaps everyone was busy that day in June. But my strong suspicion is that our initial proposal came across as bonkers – “Would you like to come to Southampton and try to write a paper in a day with us?” In the end we managed to get seven people to commit to coming to the workshop. The invitation was deliberately open ended – we did not want to specify in advance what the paper would be about. Instead we wanted each scholar to write a short 500-word statement about what they were interested in at the moment. We also asked them to share any bibliographies they may have for a shared Zotero folder and encouraged each person to bring an artifact.  That was it. We wanted everyone to bring what they had, and then we hoped a conversation would ensue that would lead to the production of a paper. At the last minute two of our participants pulled out for personal reasons. So in the end it was just Isaac, myself, Justin Murphy, Alex Prichard, and Ilan Baron. Alex from the get go saw this as anarchism in practice – and there is some truth to that. Ilan was “baffled but intrigued”, Justin was sold on the “a paper in a day” premise. I just wanted to see if it would work.

            The structure of the workshop was pretty simple: each person presented their statement, we listened without critique the first round, and then we had a brainstorming session to try to narrow down to a common topic. We zeroed in on a shared suspicion about the liberal peace hypothesis, and we had Steven Pinker’s The Better Angels of Our Nature as our foil. Interestingly we all had a sense that Pinker was wrong, but none of us could fully articulate why we thought he was wrong.  We agreed to free write a critique of Pinker – we took about 45 minutes to do that. Then we shared our ideas with the group, and then agreed upon a rough outline and argument. We divided the paper up into 5 sections and each of us agreed to write one of the sections. By 4pm we met to assemble our drafts and we had a 6200 word document that largely followed the outline that we had agreed upon at lunchtime. To be honest it was not terribly good and it definitely read as if five different people had written something and slapped it together, but it wasn’t terribly bad either, and it did say something that was new to us. We went off to dinner and agreed that as a next step each of us would take the paper for an additional day over the summer, rewrite it, and try to shape it into a workable draft. Over the course of the summer we managed to do this. We would circulate the paper around from writer to writer, revising the document, until we had something that we felt was ready to share. Over the next month each collaborator took their turn with the paper and by late July we had a 12,000 word draft that we were ready to share. At this point we debated what to do, and agreed that we should recruit a reader for feedback. Jonneke Koomen had offered to read a draft and give us comments. So we decided to take her up on the offer and offer co-authorship in exchange for her participation. Going back through the emails, there was a bit of debate about this point (co-authorship). Although when she came back with those comments everyone was eager to add her to the group.

In the initial drafting phase there were really three intense periods: the day of the workshop itself, the month after the workshop, and then the period after Jonneke’s comments when we revised the paper for submission. In my spare time I coach junior curling (the boys team I coach is four time England Under 21 national champions). I’ve taken several coaching courses over the year and one core concept in coaching is “team dynamics”. In short the coach’s job is to get the team to function well as a unit. It is a delicate balancing act where each person’s contributions have to be recognized, but you also have to make sure that each person is doing his or her role, not everyone can do everything, but everyone has to do something, and each something has to be unique, and help build to the overall goal.  As this project progressed, I was struck by how we became more and more like a team. One of the best definitions of a team I’ve learned over the years is “A group of individuals coming together to achieve a common goal”. What began as an experiment in changing the conference experience morphed into a goal of getting a published article. Another key concept I’ve picked up from coaching is the idea of a team dynamics performance wheel. Basically a team needs to go through four stages in order to function as a team. The four stages are forming (the process of getting together, and getting to know one’s team mates), storming (the process of fighting both about what the team is trying to accomplish, but also what the acceptable behaviours are), norming (whereby the team establishes explicit norms about what are acceptable behaviours), and performing (the team follows its self-imposed norms and in so doing achieves its goals). A lot of sports films follow this formula. I also see an implicit democratic theory in the performance wheel, and an implicit constructivism. Perhaps I should have spared myself the seven years of graduate school and done a three-day coaching course instead.

I digress into sports coaching to note that over time we started to follow this performance wheel. To my mind there were two critical junctures for the group. The first, and to my mind most important, was the commitment in the month after the workshop to complete a round of revisions that turned a rough draft into a paper we could share with a colleague. From there we were within spitting distance of a draft article. Oddly though as we got closer to publication the work seemed to increase, not decrease. Towards the end of the process Justin quipped to me “How to write a paper in a day, and then spend two years re-writing, and re-writing, and re-writing”. And this points to the second critical juncture. As we got the readers and editors reports back we often had very different interpretations of what to do. Oddly our worst dustups happened after the two conditional acceptances. If I look at the disagreements they really focused on two issues – one was workflow and the other was voice. Workflow consisted of who was going to do what next. Early on we had a discussion about this, and settled on a circulation queue as our basic norm. We would generate either by Skype or email a list of things that needed to be done to the document, and then would circulate the document from person to person, then review and progress the document to the next stage. That got us all the way to a submission to ISQ without many problems. It was after we received the R and R from ISQ that things became more difficult. First there were vast differences in how we interpreted the decision letter. This prompted a meeting at ISA and then a long Skype call to agree on revisions. We eventually broke this deadlock by creating a spreadsheet that listed each revision request in the decision letter and then assigning it to a specific author. Rather than trying to rewrite the whole thing together we divvyed it up into specific tasks to revise individually and once the whole draft was completed we gave it to Ilan and Jonneke who read through the text line by line.

            We submitted the revised version in August 2017 and received a conditional acceptance in November 2017 (almost a year from the initial submission). Oddly this is where I think things started to break down from a teamwork standpoint. Normally a conditional acceptance is the end of the process. It is normally a matter of some light revising per editor guidelines and then on to the proof stage. I think two things happened at this stage. First our conditional acceptance was almost entirely about the voice of the manuscript. Because we had six different authors, it was very clunky in terms of writing style. It often shifted tone and style, and we also had significant differences in writing cultures. The US in particular favours an active voice, strong sign posting writing style, whereas the British style can be a bit more literary. The reviewers and editor also flagged us for being too much of a literature review. Part of this is writing style. I suspect part of this is also because we had six different authors bringing their respective literatures to the article that it led to more summaries of arguments than average. We oddly spent more time rewriting for style than we did on the first two drafts for substance. We also almost missed the submission deadline as we struggled to revise the paper. There were whole new arguments introduced into the conditional acceptance draft and the entire text was reworked one more time. We thought at this point we were done, but only two weeks late we received a second conditional acceptance, which I had personally never experienced before.  At this point I got very frustrated – I think we all did. The requests from the editors were very clear, but there was disagreement amongst us about how to implement them. In the end we opted for a dictator – i.e. one person should have final say about how the manuscript would be shaped. It is interesting that a project that began as a collaborative experiment in knowledge production ended in dictatorship (although the democratic theorist in me would point out that for the Romans the dictator was an office established during a time of crisis to protect the republic from threat either internal or external). I was appointed the dictator by the group and in the end Ilan and I got into a workflow where we would call each other on Skype and work during the day. By our count it actually took us 50 hours of writing to revise the manuscript. We then handed it to Jonneke to line edit, and asked the others for comments. This time the manuscript was accepted without condition.

            In the end this approach succeeded beyond my wildest expectations. I never would have imagined generating an article in one of the top journals in the discipline out of a one-day workshop. In some ways it did save time at the beginning. I would say from the workshop through the initial submission I worked less on the article that I have on any other. So the division of labour was very effective at this stage. Even the R and R seemed to go fairly smoothly. It was only as we neared final acceptance that the workload seemed to ramp up, and I think this is the product of two things. First I think detail work is best left in the hands of one person. If I were to do it over again, I think I would have gone to the dictator model earlier. I would recommend that the group just appoint one person to write through the whole manuscript in a single voice and style, and then to have a second member proof read closely. I think the multiple inputs at this stage really led to diminishing returns. Part of this may also have been the result of the editorial team at ISQ are sticklers for good writing, which is not the case at every academic journal. In the end I think the results are really good. It is the best written manuscript I’ve ever published, and that goes down to the collaborative effort. If you have six scholars all going over the same script, it ends up being very clean in the end. But part of writing style also goes down to authorial voice, and each writer is going to have a different voice.

            A few more technical things we learned along the way. Version control became a very serious issue early on as we began circulating drafts. Eventually we settled on a system of a shared Dropbox folder with each author saving a new version with the date of revision in the title. We also established a clear circulation order and timeline. I think that helped everyone stay on point. There was a bit of shared monitoring in the sense that if author 1 had completed his or her tasks and turned it over to author 2, author 2 felt responsibility to build on author 1’s work and the pressure of needing to get revisions to author 3 by an agreed date. The other big issue was workflow and citations. People have very different ways of writing, and working with six different styles was challenging at times. I tend to free write very poor first drafts and then clean up as I go. Others like to outline and elaborate on an outline. A bit of discussion at the outset about writing process and how we would proceed might have helped. We ended up settling on modifying word documents because that was the one software system with which everyone was familiar, but some effort at the front end to learn about a collaborative writing system may have saved time long term. Second people had very different systems for citations. Some of us use Zotero, others just write citations in manually. In this project we ended up with over 180 citations, which would be unwieldy and a lot of work if we were entering manually. But again I think some time spent on learning a shared reference manager system would have saved time at the end.

            These reflections are largely about process. That is what I did not anticipate about this project: that the problems would not be in writing a paper, but in managing the writing styles of six busy academics. The writing itself was actually very easy. I still have the original draft article from the workshop and all the main ideas are in it, just not clearly expressed or very polished. But the ability to get six people together to come up with an idea worked very easily. I can’t imagine ever having written this paper myself. There are certainly parts of it that contain ideas I’ve been thinking about, but there are other sections about authors I’ve never read or that do technical work that I’m incapable of doing. It is a weird experience to read an article and see it as mine yet not quite mine. Several times – after the article had done a full round or circulating with my peers –I’d open up the file and say to myself “Damn, who wrote this? This is amazing!” There is no doubt in my mind that the outcome, by collaborating with five other very smart colleagues, was far greater than anything I could ever have achieved myself.