Ask us anything: collaboration agreements

The next question in our ask us anything series is from Christine Rose-Smyth: Collaboration Agreements. Whilst there are many examples of legalese-dense CAs for high level, institution – corporate collaborations, that are aimed at protecting exploitable intellectual property, does the DE team and contributors have advice for the formality one should expect / require in broaching collaborations for ecology research? I’d like to hear not only about within-U.S. academia collaborations, but also internationally, with NGOs, independent scholars and the like.

Jeremy’s answer:

This sounds like a good idea, but I’m afraid it’s not something with which I have any experience. I’ve never collaborated with anyone outside of academia, and never used a formal collaboration agreement with any of my academic collaborators. Though I may start using them within my lab, to help minimize the risk of disagreements about authorship or author order, and to provide a disagreement resolution mechanism. Here’s an example from the American Psychological Association.

Brian’s answer:

It is an interesting idea. Almost all of my science is collaborative so I have lots of opinions about how to do that. But I’ve never seen or been asked to sign a collaboration agreement in advance.  And I don’t think I’m likely to use them in the future.

My main reason is that it is almost impossible to predict the course of a collaboration. You write in it that partner A will write and be first author on the paper. But then partner A is really busy. In a good collaboration partner A volunteers to let partner B lead author. You say that partner A will collect this data in stage II. But stage I was surprising, and that data doesn’t look like a good idea to collect any more. The Prussian general Moltke said some variant of “no plan survives contact with the enemy”. Science is not war, but it is very similar in this sense. You are engaging with something unpredictable and beyond your control. Those types of situations are very hard if not impossible to write into a contract. How many contingency scenarios would you need to be sure the contract covered every possible outcome?

My two rules for collaboration (which I would suggest are an alternative to a contract are):

1. Only begin collaborations with people that you already know and trust fairly well. At a minimum, you ought to have shared several social meals with that person. Know them as a person (married?, partner? kids? pets? music?). And have observed how they treat other people (especially those with less power. I suppose there are scenarios where this isn’t fully viable (e.g. working groups often are required by the funder to assemble people who don’t know each other, but on the other hand its not a coincidence working groups spend a lot of time at meals before real work starts getting parcelled out and assigned).
2. Have a verbal plan and revisit that plan regularly and often. This I would suggest is the main alternative to a contract. Discuss who does what before you start. Discuss it when you start. Discuss it when you get the first results or have personnel changes. Discuss it again. And discuss it again. This feels like a more realistic approach of having a dynamic, ongoing discussion than a written contract. And it is why doing #1 to the extent possible is important.

I suppose written collaboration agreements are a bit like a marriage prenup agreement. Many are going to think that it is a bad sign to not take a leap of trust and then focus on the hard work of communication. Others are going to disagree. And I could imagine some specific scenarios where a collaboration agreement might be important (e.g. if there is a strong power differential between the two parties, and the powerful party has a bad reputation – but is that really a collaboration you want to start? – I would say no, no matter how famous that person is).

Your questions cover a pretty broad range of scenarios. I would say collaborations are easiest when the ultimate goals align (e.g. academics working with academics). For this reason of different goals, I have seen many but not all NGO-academic collaborations involve an agreement. They typically describe resources each side will provide and end deliverables (contents and dates). But not the path to those deliverables. And anytime somebody is giving you money (e.g. state agency, or NGO) something should go in writing (similar in content to the last example). Those are probably the main scenarios I see written agreements as important. But if they are academic-academic collaborations no matter across disciplines or country lines I’m personally going to vote for the marriage model (choose partners carefully, trust, communicate a lot and recognize that everything will change anyway).