Team science is slow science

With Times Higher Education writing about citation gaming and hyperprolific authors (surely not unrelated) I hope we can save some of our attention for what Uta Frith and others have called slow science. On that note, consider this: Team science is (often) slow science.

Recently two team science projects I’ve been involved in since the early 2010s resulted in publications: a book on recruitments, edited by Simeon Floyd, Giovanni Rossi en Nick Enfield; and a paper on sequence organization led by Kobin Kendrick. Some of the first results for both projects were presented at the 2014 International Conference on Conversation Analysis (ICCA) at UCLA, and it is so great to see them out now.

The Recruitments volume address the question of how we use language to get others to do things. It presents the results of collaborative research on 8 languages around the world and is published as an open access book by Language Science Press, linguistics’ most visionary open access publisher. This volume has a long history, and we’ve written about the background and methods of the project for the ROLSI blog. In a thread on Twitter, Nick Enfield sets out some of the key findings:

The sequence organization paper studies a basic aspect of how social action is organized in everyday language use in 12 languages around the world. It is published in Journal of Pragmatics, and first author Kobin Kendrick sets out the main findings in his own Twitter thread:

Our team work on repair —as part of Nick Enfield’s ERC grant— was similarly systematic and slow-paced; the special issue of Open Linguistics we edited is in many ways the sister to the recruitments book now out. From the start in 2010 it took us several years of intensive work before the first publications started coming out. The recruitment and sequence organization projects, which got off the ground a little later, had the additional challenge of an increasingly distributed team of collaborators (to the point that no one currently has the same affilation they had when the projects started).

This kind of systematic comparative work, which takes years to carry out and bring to fruition, is perhaps the antithesis of the hyperprolific output valued by bean counters. In this lies both a risk and a reward. The risk is that contributions can take years to become visible, which is especially tricky for early career researchers. The reward is that results tend to be solid and substantial. We need institutions & funders that don’t reduce us to output counts, and instead help us manage the risks and reap the long-term rewards of team science and slow science.


This post originated as a Twitter thread. The Frith paper mentioned in the first paragraph is:

2 thoughts on “Team science is slow science

  1. The Fast and Highly Cited
    Couldn’t agree. Now we have fast science and rapid citation.
    After Hyperprolific authors, now these authors come with hyperinflated citations.
    There are at least two hyper researchers that publish a paper every 2 days and got a highly cited paper every 2 weeks.

    Is it possible to publish so rapidly and got highly cited very quickly? Maybe if one published 100 papers a year and cited his pervious ones in every single paper, they got a lot of citations quickly as well.

  2. Your blog (the links to which I have removed because it seems you advocate for this as a method to become highly cited, making it spam in my judgement) shows that the main way to become hyperprolific AND highly cited is to simply cite your own work, creating a boring and uninformative feedback loop. This is not science but self-congratulation. This only underlines that we shouldn’t pay attention to hyperprolific authors, and that we should focus more on people who produce careful work without feeling the need to cite themselves every other paragraph.

Leave a Reply

Your email address will not be published. Required fields are marked *