You know I'm not a big fan of group work in schools, either - it gives a reliable illusion of collaboration or soft skills building with precious little objective payoff (except perhaps for the teacher...)
What would it take to nudge things back into the solo author direction at this point? Anything institutions can do or just hope the winds blow back the other way? (e.g. maybe will happen naturally as AI takes over assistant type stuff?)
I was going to include a paragraph about group projects at school because everyone’s experienced how shitty they are. It’s a perfect microcosm of the problems inherent to collaboration.
On solutions, it’s tough to outline something concrete. More autonomy to individual scientists (including PhD students and postdocs—in line with what I wrote in my previous essay) would help. Stricter policing of authorship criteria by journals would help. Less dogmatic belief in collaboration from funders would help.
Ultimately I think it stems from a psycho-cultural malaise involving a lack of self-confidence and courage, and a deep societal suspicion of excellence which then breeds shame among those capable of excellence.
Personally, I’m going to just try to lead by example and see if I can construct a blueprint for others.
Another problem in teams is that those who hold de facto clout in a research hierarchy (whether as part of the formal hierarchy or not) have a de facto freedom to confuse their heuristics and intuitions on the one hand with evidence and scientific principles on the other, while those who lack clout can either play along, or take a principled position and thus end up not being a co-author.
In design and operation teams, those at the bottom of the totem pole can suck up and snitch (and claim credit for others' work) to climb, or 'lead from below' while taking flack to keep the operation going. Of course, the latter option provides no guarantee against the boot.
Interesting piece and gave lots to think about. Was curious about the Craig Venter omission on HGP?
I sometimes wonder whether author count is an all that reliable way to look at this. In the early Royal Society, there were large teams of instrument-makers, technicians, etc.. who ran experiments. Today, we would regard many of them as scientists or co-authors, but, because of the conventions of the Society, they were largely excluded from credit in their time. Point being, part of the historical increase in team size reflects changes in who we regard as a scientific contributor.
Same if you compare something like CERN to Uraniborg. If you don’t know the story, Brahe was given an island by the King of Denmark and basically made all its residents servants to build an observatory. The villagers didn’t have much of a choice, but it was still a huge collective scientific endeavor. CERN puts far more of that work inside the boundary of legitimate scientific contribution, and I don’t think we should ding it for that.
I find myself wondering whether the problem is really collaboration, or the culture of competition and individual achievement in which we practise it. Why have we constructed a scientific culture where other scientists can become obstacles to our own recognition rather than people with whom knowledge is made? Why must ownership and recognition be what motivates us, rather than participation, shared curiosity, care, or helping someone else have their Eureka moment? Individuality and collaboration don’t have to be opposites. IMO, genuine collaboration is what nurtures originality
I don’t think we’ve constructed a culture that enforces dilution of credit—I think that’s just an arithmetic reality. It’s just true that, if you have three other collaborators on a project, you will feel less personally responsible for each component of it.
“Why must ownership and recognition be what motivates us …”
-> I don’t know, but it just is the case. This is human nature. Academics try really hard to pretend that they are above human nature, but we aren’t. A pragmatic acknowledgement of incentives is required if we want to produce as much new knowledge as possible.
I do sort of agree with what you’re getting at though. There are loads of examples of duos or trios of scientists working together. This can work really well if they spend a lot of time physically together to the point that much of the interpersonal friction dissolves. It also requires compatible personalities, which is hard to engineer or select for.
But in general, a solo scientist doesn’t have to live in a cave away from all other scientists. I think feedback and frequent conversation about science with colleagues is super beneficial. I just wouldn’t call this “collaboration” necessarily.
Fair enough. In my idea of collaboration, I would count all this dialogue too! I agree that the kind of collaboration you’re describing is not necessarily something that benefits science, but I think the successful duos and trios you mention are a perfect example of why the problem may be structural. It seems less about the number of people and more about the nature of the relationships.
Human nature also includes belonging, care, curiosity and collective achievement. I think we should strive for a research culture where people have collective responsibility without individual responsibility disappearing. It doesn’t have to be arithmetic. You can feel deeply responsible for something you don’t exclusively own.
Of course we care about recognition and ownership, but I’m not sure the extent to which we compete for them is independent of the work culture and incentives we create. Institutions can’t erase self interest or the desire for status, but they can amplify those motives, reward them, or hopefully leave more room for other motives too.
You know I'm not a big fan of group work in schools, either - it gives a reliable illusion of collaboration or soft skills building with precious little objective payoff (except perhaps for the teacher...)
What would it take to nudge things back into the solo author direction at this point? Anything institutions can do or just hope the winds blow back the other way? (e.g. maybe will happen naturally as AI takes over assistant type stuff?)
I was going to include a paragraph about group projects at school because everyone’s experienced how shitty they are. It’s a perfect microcosm of the problems inherent to collaboration.
On solutions, it’s tough to outline something concrete. More autonomy to individual scientists (including PhD students and postdocs—in line with what I wrote in my previous essay) would help. Stricter policing of authorship criteria by journals would help. Less dogmatic belief in collaboration from funders would help.
Ultimately I think it stems from a psycho-cultural malaise involving a lack of self-confidence and courage, and a deep societal suspicion of excellence which then breeds shame among those capable of excellence.
Personally, I’m going to just try to lead by example and see if I can construct a blueprint for others.
Another problem in teams is that those who hold de facto clout in a research hierarchy (whether as part of the formal hierarchy or not) have a de facto freedom to confuse their heuristics and intuitions on the one hand with evidence and scientific principles on the other, while those who lack clout can either play along, or take a principled position and thus end up not being a co-author.
In design and operation teams, those at the bottom of the totem pole can suck up and snitch (and claim credit for others' work) to climb, or 'lead from below' while taking flack to keep the operation going. Of course, the latter option provides no guarantee against the boot.
Interesting piece and gave lots to think about. Was curious about the Craig Venter omission on HGP?
I sometimes wonder whether author count is an all that reliable way to look at this. In the early Royal Society, there were large teams of instrument-makers, technicians, etc.. who ran experiments. Today, we would regard many of them as scientists or co-authors, but, because of the conventions of the Society, they were largely excluded from credit in their time. Point being, part of the historical increase in team size reflects changes in who we regard as a scientific contributor.
Same if you compare something like CERN to Uraniborg. If you don’t know the story, Brahe was given an island by the King of Denmark and basically made all its residents servants to build an observatory. The villagers didn’t have much of a choice, but it was still a huge collective scientific endeavor. CERN puts far more of that work inside the boundary of legitimate scientific contribution, and I don’t think we should ding it for that.
I find myself wondering whether the problem is really collaboration, or the culture of competition and individual achievement in which we practise it. Why have we constructed a scientific culture where other scientists can become obstacles to our own recognition rather than people with whom knowledge is made? Why must ownership and recognition be what motivates us, rather than participation, shared curiosity, care, or helping someone else have their Eureka moment? Individuality and collaboration don’t have to be opposites. IMO, genuine collaboration is what nurtures originality
I don’t think we’ve constructed a culture that enforces dilution of credit—I think that’s just an arithmetic reality. It’s just true that, if you have three other collaborators on a project, you will feel less personally responsible for each component of it.
“Why must ownership and recognition be what motivates us …”
-> I don’t know, but it just is the case. This is human nature. Academics try really hard to pretend that they are above human nature, but we aren’t. A pragmatic acknowledgement of incentives is required if we want to produce as much new knowledge as possible.
I do sort of agree with what you’re getting at though. There are loads of examples of duos or trios of scientists working together. This can work really well if they spend a lot of time physically together to the point that much of the interpersonal friction dissolves. It also requires compatible personalities, which is hard to engineer or select for.
But in general, a solo scientist doesn’t have to live in a cave away from all other scientists. I think feedback and frequent conversation about science with colleagues is super beneficial. I just wouldn’t call this “collaboration” necessarily.
Fair enough. In my idea of collaboration, I would count all this dialogue too! I agree that the kind of collaboration you’re describing is not necessarily something that benefits science, but I think the successful duos and trios you mention are a perfect example of why the problem may be structural. It seems less about the number of people and more about the nature of the relationships.
Human nature also includes belonging, care, curiosity and collective achievement. I think we should strive for a research culture where people have collective responsibility without individual responsibility disappearing. It doesn’t have to be arithmetic. You can feel deeply responsible for something you don’t exclusively own.
Of course we care about recognition and ownership, but I’m not sure the extent to which we compete for them is independent of the work culture and incentives we create. Institutions can’t erase self interest or the desire for status, but they can amplify those motives, reward them, or hopefully leave more room for other motives too.
Anyways, thanks for the nice read!