Transcript of Nicole C. Nelson on 'Ethnographies of Science'
Carmelina Contarino: Welcome back to the HPS podcast, where we discuss all things, History, Philosophy, and Social Studies of Science for a broad audience. I'm your host, Carmelina Contarino, and today we are joined by Dr. Nicole C. Nelson, Associate Professor in the Department of Medical History and Bioethics at the University of Wisconsin-Madison.
Nicole is an ethnographer of science and a familiar face to many within both Science and Technology Studies, and Metascience. Today, Nicole explains how ethnographic studies can help us to make sense of the world and how she uses ethnography to construct the story of science in a way that the published record of scientific articles can't. By immersing herself within the spaces where science takes place, Nicole's research produces a deeper and richer understanding of how and why science is conducted the way it is.
[00:01:04]
Carmelina Contarino: Hi, Nicole. Welcome to the podcast.
Nicole C. Nelson: Thank you.
[00:01:08]
Carmelina Contarino: Firstly, how did you come to STS?
Nicole C. Nelson: So, I was enrolled in this program where I was allowed to design my own major, and as long as you had the sign-off of an advisor, you could make your own thing. So, I made my own thing based on my own interests at the time, which I called genetics and social and political theory; where I took a bunch of courses in molecular biology, and then I took history, anthropology, philosophy, sociology, the whole bin bag of things over there from the humanities and interpretive social sciences. I didn't necessarily have a great plan of what I was doing, but those were the things that I felt compelled to do. And then I was thinking about grad school, and I had a lovely advisor at that point in my career who said, " Well, what you should do is look at universities that will pay you really well and see what programs they have". Funny, super pragmatic life advice, right?
So, by virtue of doing that, I ended up on Cornell's website and was reading through their list of graduate programs and got to science and technology studies and was like, ‘Oh, that's what I've been trying to do for the past four years.’ So, it wasn't a field that I really had much knowledge of as an undergrad, and I kind of lucked into it in a way just because it was at the intersection of my interests. And then once I knew it was a field, then I was dead set on professionalising in that direction because, for me, it seemed like exactly the marriage that I wanted of broad scope approaches, but still thinking about the sciences, and managed to get myself into Cornell's PhD program, went off and loved it.
[00:02:33]
Carmelina Contarino: That aha moment that comes from finding HPS or STS by accident is not uncommon amongst our guests, but what led you to ethnography specifically?
Nicole C. Nelson: Well, that is an interesting question too, because when I applied to Cornell's STS program, at the time, what I had taken most of was history. For me, grad school training did the thing, I guess, that it's supposed to do in the sense that it broadens your horizons quite a bit and shows you different paths that you could travel. And one thing that's maybe unusual about me as an academic is that I'm quite extroverted, as opposed to many academics that are kind of more happy with books, and I sort of realised that the life of a historian versus the life of an ethnographer was going to be sitting in the archives versus talking to people, and at some point, I realised, I am much more constitutionally, suited towards this path of talking to people. So, that and a combination of other things started pushing me down the path of focusing more on ethnography. Although I really see ethnography as one suite of tools that I use, and I still use historical approaches and philosophical approaches as well. But ethnography, in the end, was what I was dispositionally best suited for.
[00:03:44]
Carmelina Contarino: I can definitely relate to that. Can you tell us what is ethnography?
Nicole C. Nelson: For me, the core principle of ethnography is it's about using yourself as a trained human in order to create bridges between worldviews such that those worldviews become sensible to each other. I had a mentor in graduate school who told me that for him, the core piece of ethnography was seeing it as a tool to understand how other people make sense of the world.
That was a really appealing perspective to me because so often you encounter situations in which An action of one person looks incoherent to another person. And then the approach is to try and say, Oh, well, here's where you made the logical error or something like that.
Whereas the ethnographic spirit is really to say, what pieces of information do we need to make this make sense, to make it understandable from my perspective, why you are doing what you're doing from your perspective. So, I see my role as an ethnographer as a kind of translation role, where you're using your positionality in multiple places in order to try and put together sensible explanations for a different audience about why another group of people are operating as they're operating.
[00:04:53]
Carmelina Contarino: So, how do you use those skills in terms of studying or interpreting scientific behaviour?
Nicole C. Nelson: Well, one of the core insights that I learned in my graduate STS training, that is a core turning point in the field, is that The study of science can be like the study of any other part of society in the sense that there's nothing special about scientific activity that makes it outside of other human activity. It obviously requires some skills for you to be able to understand and be fluent in science, just as if you were doing fieldwork in another language, you would need to acquire a language. But a core insight from early STS is Science is not some activity that's a special thing that no one else can actually touch.
We can study science as human behaviour. And so, in a way what I do is I try and understand both the sociological processes of humans interacting in a space, combined with those epistemic interests about the particular concerns of somebody who's trying to make knowledge. And for me, it's the really powerful intersection of those two things that makes STS interesting. Because epistemic stuff on its own, outside of the social constraints, becomes really unrealistic. You know, you can have great understandings of how people are supposed to make knowledge that do not work in practice, and sociological analysis on its own without any epistemic concerns doesn't work either because it makes scientific work seem strange.
None of the world makes sense unless you understand that it's about knowledge production. So, for me, ethnography in a scientific environment is about marrying those sociological and epistemic sensibilities that are marrying the philosophy and the anthropology. History, sociology, sensibilities together.
[00:06:34]
Carmelina Contarino: I'm a fan of using interdisciplinary approaches in research, but what do we actually gain from undertaking ethnographic studies of how we produce knowledge?
Nicole C. Nelson: Yeah, maybe I can best answer that through an example, so I'll give you one from my recent work on reproducibility. For me, as an ethnographer, one of the things that was an immediate question of interest when I started thinking about failures to replicate in the sciences is what was it like to actually experience one of those? And most of the conversation around reproducibility stuff was really focusing on key opinion leaders who were at the PI level or who were like nationally represented figures.
But if you think about who actually experiences physically a failure to replicate its graduate students. So, what do we get from taking an ethnographic approach to trying to understand what is this like for you? Well, one of the things that I found in combination with my team is that you start to see why it's the case that your reproducibility can remain hidden.
So, a classic puzzle for history and philosophy of science is this idea that nobody's really incentivised to replicate. And Harry Collins tells us this in his classic work on replication and changing order that this is a rare activity. Nobody actually does it. But what you see when you look ethnographically is that people do it all the time, in fact.
It's just that the circumstances of their work make it not very visible to the outside world or even to themselves. So, what I found from talking to students, is that, in fact, when they encounter something that's a failure to replicate, most of the time, they look at that, and they think, Ah, shoot, I must have screwed it up.
So, they don't even think about it as a failure to replicate or even what they're doing as replication. They're just learning. And messing it up. And so, when you use that lens, you start to understand, oh, no, it's not that there's very little replication. There's actually a whole lot of replication, but it's the sociological features of the work that make that replication not very visible to the outside world.
So that's an example of where it is that I think you get something different from doing ethnographic work than you get from doing work that is more oriented in history and philosophy of science, where you're kind of taking a big picture view about how ideally knowledge should behave. Ethnographic work really asks you to get down and dirty on the details of like, no, how does it go down? And then you start to see all of these ways, in which it is different than a clean knowledge system perspective would predict that it would go.
[00:08:52]
Nicole C. Nelson: Yeah, it's a great question because I think the answer to that can be directed in multiple ways, and that's one of the joys about being a person who uses ethnography is you can have many audiences. Not that other fields don't necessarily have many audiences, but I see some really ripe ones for doing this kind of work. So, for one, there's speaking back to your peers in the STS and HPS world, whereas in the example I gave before with the Harry Collins argument, you can Take that really good base of ethnographic work that he did and kind of modify it according to the circumstances of how it is that other laboratories work so you can enrich the theory base of your own discipline.
But a thing that ethnography is also very useful for in this sphere is applied work that's aimed at a more policy context where it opens up new places for research or new places for intervention. So, just as an example, if you think about what happens when you do ethnography and you find The real barriers that people are confronted with in their labs, it often gives you a much more sympathetic view of why it is that they're making the choices that they're making.
So rather than looking at them and saying, Man, these guys are real slackers. You know, they're just not doing high-quality science. They really need to get their acts together and do it better. When you go in, and you try and experience the world as a series of choices, as they experience it, you can start to see that they're trading off things that, from their perspective, look logical. And from that, you open up new possibilities for intervention by saying, Oh, maybe they need to see this other piece of information that they don't have and that'll change their mind. Thank you. Or maybe the person who's in charge of the system needs to rework the system such that people can make more logical choices.
So, I like the ethnographic lens because it's fundamentally generative. It's usually trying to go in without a lot of theoretical presupposition. And because it's doing that, it has the habit of kicking up new stuff. And once you have new stuff, then you have many different places that you can work with it and play with it.
[00:10:53]
Carmelina Contarino: Typically, scientific ethnographies are conducted within labs, but are we starting to see that move into other arenas?
Nicole C. Nelson: Definitely. So, if you take a look at the classic quote-unquote classic lab ethnographies, the ones that really sort of gave us this idea of, Hey, we can study science through ethnographic methods. They're based in labs. And those were very convenient units because labs are self-contained. And so, you're able to look at the functioning of a little scientific organism of people because you've got a kind of ready-made unit. But that won't work for answering all types of questions, and there are many types of questions for which you need to be able to see people interacting with each other, or learning from each other, or trying to communicate with each other.
So really, ethnography ideally should be wherever the action of interest is. So, if the action of interest is how does a lab function, Then you need to be within the lab. But if the action of interest is how do we collectively negotiate what it means to be open when we're trying to do open science, that activity happens to an extent in a lab, but it happens maybe more meaningfully in places where people interact with each other.
So, ethnographers have had to get much more creative when doing ethnography of science about thinking about where the activity is. And lab ethnography, in my view, would encompass a lot more than just labs these days.
[00:12:08]
Carmelina Contarino: Is there any value in generalising ethnographic studies beyond a particular lab or scientific episode? Or should ethnographies not be generalised at all?
Nicole C. Nelson: That's always a tricky question. Yeah, because ethnography is a method that is going deep. And so, when you're going deep, you need to be really sure that you've set yourself on a terrain in which going deep will be valuable. Now, sometimes going deep will be useful because you've found a weird lab that will be useful in providing a contrast to what exists in the secondary literature.
But sometimes what you want to do is find a typical lab, one that is representative of a broader set of phenomenon. And in that case, you better make sure that before you set yourself up to really mine that terrain, that you actually have found a representative lab. So, there's value in both the weird cases as well as the normal cases, but where the real difficulty comes Is figuring out which of the ones that you have and not just choosing something because it's convenient, but choosing something because you know, it's going to give you theoretical purchase.
So, it's very easy to show up just because you know your neighbours with someone or they're already friendly with you and use that as your case to generalise out. But a more principled way of approaching it would be to say, well, what do I want? Do I want a contrast to an existing norm? Do I want something that's representative of Labs headed by women, labs in this field, labs in North America? And once you decide the axis on which you hope to be able to generalise, then you have to do your due diligence to figure out whether or not the site that you're going to lock yourself into actually has that generalizability.
[00:13:38]
Carmelina Contarino: You wrote a piece for nature that discusses what ethnographic work actually provides to scientists. Can you tell us about that?
Nicole C. Nelson: In this particular piece, I was taking another question that is a live question these days in the scientific reform field, which is why don't people follow so-called best practices? So, in the clinical sciences, it is absolutely best practice and very commonly practised to blind or mask your interventions so that the person who's scoring something, for example, doesn't know what condition the particular animal or human has been in that they're scoring.
Now, that happens almost all the time in clinical research. But it doesn't happen as often in preclinical research with animals and with cell lines. So now, when I watch people in the reform space thinking about this, what they mostly think about is this incentive story where they say, ah, it's because there's no incentives to make people do the extra work of having to do this masking process. So absent from the incentive, they're not going to do it. And I looked at that, and I thought, Well, that doesn't seem right to me, because most scientists, if you think about what their life-world and their identity is like, they don't wake up in the morning and go, Hey, I'm going to do some shitty science today and cut corners because no one's trying to stop me to do otherwise, from an ethnographic perspective, it doesn't make sense because it isn't compatible with the life world.
It's not the identity that people have. And so, I thought, well, why don't we go talk to people and ask them, why don't you mask what is up here. And from doing that you see all kinds of things that are then potential options for policy interventions. One of the ones that I found most interesting was the fact that in the clinical space, usually scientific teams are made up of, six to a dozen people or maybe even larger.
So, there's lots of opportunities for masking because there's lots of people on the same project. So, one person can collect the data, and then they hand it over to another person to score. Now, in the preclinical or basic sciences, that's much less common. People are working on their own individual projects sequestered in a way in a lab by themselves.
So, if they want somebody else to score their data, they've got a kind of a beg borrow and steal time. And because that person isn't being rewarded in the form of authorship or relief from other work. You can't also really be sure that they're going to score that data well. So, in fact, there's a real problem there, both in terms of how the labour is set up, but also in terms of the fact that if you do mask, it might compromise your data because you have a less interested person doing your scoring.
So, if you see that, you start to see, oh, it's actually not unreasonable that people choose to not do this practice. So this is a long-winded way of saying that what I was trying to argue for in that Nature piece is, rather than starting from the assumption that people are doing irrational things and bad things, and we just need to incentivise to do them other ways, What you can bring is the ethnographic spirit to say, let's try and understand from their own perspective why people do these things and when you see that, you see all these other points for intervention. You can say, well, maybe we need to encourage a more collectivist culture in basic science so that there's more than one person on a project and so that that work can get shared around, and that alone would make masking or blinding much more feasible.
So that's not an issue that was on anyone's policy radar, like let's make a more collectivist culture to have more reproducible science. But when you do the ethnographic work, you start to see that as an option.
[00:17:00]
Carmelina Contarino: Moving beyond policy, what can the general public take from ethnographic studies of science?
Nicole C. Nelson: What I think is most important for the general public to take from ethnographic studies of science is this understanding that science as a professional practice faces a lot more uncertainty than science that you are exposed to in your general education. It's a well-documented phenomenon that insight in most science education, thinking about through primary school and into secondary school, the way that you see science is.
Science is a stabilised body of facts as a set of experiments that are supposed to work in a particular way. And if you don't go further on in your training from that, you don't get to see what kind of uncertainty scientists are actually confronting and why it can be very difficult to make a stable, knowable fact over time.
So the very fact that the general public gets sensitised to this large body of stable facts makes them assume that that's what science is mostly like and that when people have to revise their science or go back on it, or they make the wrong call based on their data, they assume that that might be somehow misconduct or not ideal scientific work because the scientific work that they've seen is all about this certain stable base of knowledge.
So, I think when you, as a general public member who doesn't get to see the inside of lab, get to read an ethnography, it's like you get a little window into what that life world is like. You get to understand why it is that these choices can be so difficult and why it is that knowledge is changing so rapidly from the perspective of the frontier of science. And so even if you never go through that training yourself, reading ethnography can be this educational tool to help you understand what science looks like is an actual professional practice, not as just a stabilised body of facts that you get in everyday education.
[00:18:47]
Carmelina Contarino: Communicating how scientists engage in research is critically important. So, what's your next research project, Nicole? What new ethnographic studies will you be undertaking?
Nicole C. Nelson: What I'm moving to now in my research is trying to do some ethnography of some of the proposed solutions to the so-called reproducibility crisis.
So, for the past maybe four or five years, one of the things that I've been interested in is trying to understand how it is that reproducibility, rigour, replication became these sort of flashpoint words that many different people and disciplines got interested in all at once. So, I've been doing some research there that's a bit more historically oriented to figure out, like, why was this the moment at which this thing come up?
And I've been able to generate, at least for me, some fairly satisfactory answers for that. And so, what I have been moving into now is to think about what are the consequences of the solutions proposed? So, if we accept that science needs to be reformed, and there's a number of proposals on the table for what that should look like, then for me, one of the first questions that comes up as an ethnographer is, what is it like to live and work under those systems?
So, I have been trying to insert myself into a couple of different positions that I think are interesting because of the way that they offer alternatives to current scientific practice. So, just to give you two quick examples of some things that I've been doing, one of them I have been spending some time ethnographically at what's called a remotely operated lab facility for a cloud lab.
The idea with this lab is that it will be a biomedical life sciences laboratory, but one that no one ever sets foot in; you'll interface with it through a computer alone. The idea is that this will be much more reproducible because you have to specify, in code, every single move that you make because every single move is not going to be done by you; it'll either be done by a machine or a remote operator that's a human on the other end.
So, you've got to type out in incredible detail, shake this for 15 seconds. So, in theory, it should give you extremely high reproducibility. Now, in practice, though, what is it like to live that life? It is very different from how life scientists get trained these days, where so much of it is about this manual technical skill, and that's what they get valued for when they go on the job market later on.
Those are their transferable skills that, you know, industry hires them for or that they get a postdoc position because they can master this technique. So, one interesting question for me is, all right, if you adopt this technique of having a remotely operated lab. What's it like to live that life when you're the person that's got to get trained up in that system?
So, there's one little bit that I'm working on. And then another one, which might be well familiar to listeners of this podcast is I've been doing some work with the psych science accelerator. And that, once again, is a really interesting and new model because it's trying to take mainly social psychology as a discipline, which has usually been a smaller lab-based practice and say, let's make it bigger. Let's make it global. Let's involve a network of labs from all over the world. So once again, that's a very different way of doing work that comes with all these problems.
Like if you are an author on a publication that has a hundred other authors, What happens to you when you go out into the world, and you try and explain that value to somebody who wants to hire you? So, for me, it raises lots of interesting questions about what it would be like to live in all the different futures that the reformers are providing for us.
So that is the thing that I'm most interested in these days. And so, if anybody's got a neat little social experiment that they're running and ethnographer to come out and document their work, Tell them to come hit me up.
[00:22:21]
Carmelina Contarino: The idea of remote labs makes me think about the embodiment of knowledge. Do we then end up with more trial and error because that tacit knowledge is lost?
Nicole C. Nelson: You're right on the money there in that the tacit knowledge part of it is one of the ones that I think is most interesting because we have a huge volume of secondary literature that shows How important that kind of tacit training is, but also how important embodied skill is.
And essentially, you're saying, well, what if we just did away with both of those two things? You can make some predictions based on the secondary literature. For example, you can make the prediction that what has happened in other automation settings is that automation doesn't necessarily remove tacit knowledge.
It just displaces it to another location. When ethnographers have done studies of labs where people are using robots to pipette things rather than the tacit knowledge, now being in the literal hand of the person who's doing the pipetting. Now, the tacit knowledge is in knowing.
Which little plates you can put in front of the robot, and how you have to position them so that the robot can pick up the liquid So, it's not like tacit knowledge gets displaced. It just transfers into another form. So now what we can do with that in terms of developing a research question is say, okay, well, these existing studies, they're usually of people who are still at the bench, and they've just transferred from a manual pipettor to a robot, but it's still the scientists interacting with the robot.
But here, what we're proposing is that the scientist never gets to interact with the robot. You know, it's going to be somebody else. It's going to be the employee at the remotely operated lab that's figuring out how to jiggle the plates. So now what we have is a split where the tacit knowledge doesn't just change form; it changes form and gets separated into two different humans.
For me, that gives you a research question where it's like, okay, well, we can predict that tacit knowledge will be a problem here. But what we can't predict is what's going to happen when it gets. Split between two physical bodies at two different locations. I can predict that tacit knowledge will be a problem, but what exactly is going to happen?
Unsure. That's why it's fun.
[00:24:19]
Carmelina Contarino: Does that separation of tasks change the way science is done? And does that mean broader cultural changes for practising scientists?
Nicole C. Nelson: So, in fact, there's a real question here about what culture will evolve. Will it be a culture where doing the science means something more like computer science, where the joy is in hacking a bug and making it work? Or is it going to be something more like bench science today, where we see people kind of getting joy out of the results that they get from a really skilful technique?
It seems more likely to me that it would have to become a culture like computer science because there's much less opportunities for those interactions that happen at the bench to happen when you're interfacing with your science, through a terminal.
[00:24:58]
Carmelina Contarino: Does that start to change the type of science we engage in?
Nicole C. Nelson: Definitely. And it also changes who can engage in the science. For example, one of the things that I think is quite interesting about the remotely operated lab is how do you keep people from doing bad stuff? And I don't mean bad stuff, like nefarious stuff. I mean, literally just experiments that are poorly designed or are going to damage the equipment or something like that. If you think about how you get trained in a kind of classical laboratory setting, you come in and you do very simple work under the supervision of somebody else who's going to like smack your hand away before you break the hundred thousand dollar machine.
But in a remotely operated setting, it's not entirely clear whether or not there are breaks there to prevent you from doing an experiment, which is useless at best or maybe harmful even. And so that presents a very interesting problem basically about what do we do when we have a new group of people who can now access very high-end equipment but don't necessarily have the knowledge of that equipment or the knowledge of biology to be able to design experiments that biologists would consider up to snuff.
So, it really opens up who can do science. But it will create, in my view, probably a lot of problems in terms of how to trust and value that knowledge because that knowledge is being produced in a way that is so different from how it would usually be produced.
[00:26:18]
Carmelina Contarino: And the only way that we can really find that out is by going in and being with the people who are actually engaging in these practices.
Nicole C. Nelson: In other words, by doing ethnography.
[00:26:28]
Carmelina Contarino: Thank you so much, Nicole. It's been an absolute delight talking to you.
Nicole C. Nelson: Thanks so much for the interview. It's been fun.
Carmelina Contarino: Thank you for listening to season four of the HPS podcast. The details of today's conversation, including the transcript and show notes, are available on our website at HPSunimelb.org. You can join the discussion on our social media pages, including Bluesky, X, Facebook and Instagram, and follow us for bonus material and updates from the wonderful world of HPS. We would like to thank the School of Historical and Philosophical Studies at the University of Melbourne for their ongoing support and thank you, our listeners, for joining us. We look forward to having you back again next time.
Comments