Meet the Professor - Professor Duncan Purves

In this interview with Professor Duncan Purves of the University of Florida, we discuss reasons for pursuing philosophy, important works, and Professor Purves's research on the environment and artificial intelligence. Read more for a full transcript.

Transcript

Arjun Verma:  Hi everyone. Welcome to all of our viewers. I want to first introduce our guest today: Professor Duncan Purves. He is a professor of Philosophy at the University of Florida where he focuses on issues of environmental ethics and artificial intelligence’s use in war and policing. Thank you for agreeing to speak with us today, Professor Purves.
 
Dr. Duncan Purves: My pleasure.
 
Arjun: Okay so to begin our interview, I’d first like to ask a couple of broad questions about philosophy generally before we dive into some specifics about your work. Why do you think that people, students in particular, should pursue philosophy?
 
Purves: One reason to pursue philosophy is that, if my experience as a student and later a graduate student was any indication, I think that it shows a lot of students who didn't get this experience in high school that there is an entirely different set of questions that you can ask about the world and the nature of the world, the nature of our values, the foundations of our politics, that we didn't even realize you could ask. And I remember when I went to college that was what struck me most about my first philosophy classes - that I didn't even know we could ask these questions, and not only that, but philosophy provides you a unique way of trying to answer those questions. And I think it actually provides a kind of methodology that allows you to make progress on those questions, and maybe not arrive at a perfectly decisive answer, but at least give you a sense of how you might try to answer some of those questions. I think, more instrumentally too, it provides some tools for thinking more critically about some of your other coursework; some of the topics that you might encounter in an economics, political science, or a sociology course. I think it allows you to think a little more deeply about any sort of topic that you might encounter in your academic pursuit. Then lastly, it's a pretty well-established fact at this point that philosophy students perform the best on certain standardized tests including the LSAT. I think there are lots of good reasons to get into philosophy.
 
Arjun: Yeah that's those are definitely a lot of really good reasons. How did you specifically get interested in philosophy? What was your first introduction?
 
Purves: When I was a freshman in college, I took an introduction to ethics course in a big lecture hall. It was 8 a.m in the morning which was tough for a freshman in college and that course was actually terrible. That was not why I got into philosophy, but that was my introduction. I won’t say who taught it, but it was just not very interesting… But the second course that I took was a political philosophy class with a Professor named Brian Kierland and this is at the University of Missouri - I think he's at Boise State now - but that class sort of blew my mind for lack of a better word… The course focused on contemporary - by contemporary I mean 20th century theories of political authority and political obligation – so, why are citizens obligated to obey the laws of the state, and what gives the state authority to coerce its citizens into obeying those laws? One of the philosophers we spent a lot of time on was Robert Nozick, a famous libertarian political philosopher, and I just found both Professor Kierland's engagement with the material, but also the material itself, just so fascinating. I don’t know if you’ve read Nozick’s stuff, but his arguments are so clear. He doesn’t use any sort of fancy language. He’s very straightforward, but his arguments are so clear and so compelling. You’re sort of left at the end of every chapter wondering how you could possibly refute what he has to say, right, and yet he’s arguing for these pretty surprising conclusions about political authority. I just found that class fascinating and that's when I realized that philosophy was up to something really interesting and something that you couldn't do in any other discipline - no other disciplines were even trying to answer these questions - so that's when I changed my major at the end of that semester.
 
Arjun: I definitely read some of Nozick’s evidence for debate rounds and things like that and I’ve had great success with them. For other books, publications, papers or podcasts – what would you recommend that people who are interested in philosophy or want to pique their interest read, or listen to?
 
Purves: I’m just going to provide you with a list of some of the ones that I thought resonated most with me in grad school. Most of these are focused on ethics, although not all of them, so I actually have them right here.
One book that I think is great just for a first-year college student who just wants to get a sense of what kind of questions philosophers are asking is a great book by Thomas Nagel. Thomas Nagel is now retired, but he has had a lot of influence in all sorts of areas of philosophy, including philosophy of mind, epistemology, ethics so this book's just called “What Does It All Mean?” It's a tiny book - it's super short - but he talks about a quick survey of different issues and philosophy. How does one know anything the problem of other minds, the mind body problem, what do words mean, free will, right and wrong, all this stuff; it's just a great short survey.
Another cool book that I think is interesting is called “Living High and Letting Die” by Peter Unger - I'm not sure if you've read this one but if you're familiar with Peter Singer's argents from “Famine, Affluence, and Morality,” the conclusion of his arguments is that we should all be giving a lot more to help people in need than we actually do. Peter Unger takes us to the absolute extreme and argues that we should not only should we give be giving most of what we have to helping others in need, but we should probably be stealing from other people to do it. So he takes it to a very logical extreme and it's just a great piece of provocative analytic philosophy, so that's squarely in sort of applied ethics.
This book is probably a little more challenging technically; it's a little more theoretical, and it's a lot more in engaging with sort of very contemporary cutting-edge questions in ethical theory. This is a book by Ben Bradley who's a professor at I think Syracuse University called “Well-being and Death.” This book is really in my wheelhouse. A lot of my research is in this area, and basically he's trying to provide an answer to two questions: one is what is it that makes our lives go well for us and the other question is what is it that makes our deaths bad for us. And he gives sort of a nice systematic set of answers to those questions. It's a great book. He's a super clear writer. He's very careful. It's just a great display of philosophical talent.
Another book is “Ethical Intuitionism” by Michael Huemer. He was one of my advisors in grad school. He is somewhat of a – maybe controversial figure's too strong – but he's a pretty s staunch libertarian, so a lot of his political philosophy is in that vein. He argues against gun control, although not in this one. I think he argues against the minimum wage. He also argues against any sort of immigration restrictions, so he's a really interesting guy, but in this book he's concerned with meta ethics. He's defending a view that he calls ethical intuitionism and that's basically the view that there are objective moral facts about the world and we can come to know what those moral facts are through a faculty that he calls intuition. It's just great. He spends a lot of time trying to refute objections to this view because a lot of people are very skeptical about this view, but it's a great if you're interested in that question how can we come to know anything about moral properties or moral facts what are moral facts. He gives, he tries to give some answers to that those types of questions.
I got two more. Another book is “Naming and Necessity” by Saul Kripke. This is another short book, but it's maybe one of the richest pieces of philosophy that you could ever read. It's in the area of philosophy of language, so he's interested in questions like what do what do our words mean and how is it that we refer to things using them. When I talk about Arjun how is it that I am talking about you and not some other person. He has a great set of arguments. An interesting feature about this book is that apparently this entire book is just a transcription of a series of lectures that he gave and he gave those lectures without any notes. This is just a book that he had in his head that he just produced over the course of a series of lectures. Saul Kripke is now I think retired, but he was definitely an important philosopher in the area of philosophy of language.
Then the last one is just a another short book called “The Problems of Philosophy” by Bertrand Russell. This book is primarily interested in knowledge, the question of how epistemology right how it is we come to know things about the world,  but it's just a nice. It's very short again. He's a very clear writer, and he sort of gives a little account of how it is we come to know anything in this in this short and interesting book, so that's it.
 
Arjun: Those definitely sound very interesting, and I’d love to take a look at them. Now that we've discussed a little bit of other people's work and other people's research, I wanted to delve a little bit more into your work and your research. I know that you've done a lot of work on the environment and its ethics and its value. Could you briefly summarize your research into that?
 
Purves:  My work on the environment - these days I’m doing less stuff in that area -but most of my work in environmental ethics concerns questions about intergenerational justice. My dissertation topic was something that's called the non-identity problem, and the non-identity problem is a problem that was discovered or articulated in the early 70s, really 60s and 70s, by a few philosophers that seem to kind of come upon it independently. But the basic problem is this: you know a lot of us think that we should curb our carbon emissions, we should drive less, we shouldn't pollute, we shouldn't throw trash on the ground, and a big reason for this is that it's going to harm future generations. We should do it for the sake of future people, but the non-identity problem arises for this natural intuitive position, because, as it turns out a lot of the actions that we perform now are going to affect not only the quality of life of future people, but also who comes to exist at all. right A simple example: if somebody decides that they, for the sake of the environment, they want to start taking the bus to work instead of driving their car, they have to leave a few hours earlier every day or a few minutes early every day or maybe 30 to 40 minutes early, depending on the bus route to get to work on time and the same for going home. So their entire weekly schedule changes, and what ends up happening is that this person as a result of this choice ends up- suppose that they and their partner decide they want to conceive a child, well as it turns out their conception schedule ends up being very different than what it would have been had they not made this choice to start taking the bus to work. Now what this means is that in this scenario, a different sperm and different egg combine at a different time than they otherwise would have, and so insofar as our identities are intimately connected to the particular sperm and egg that we are conceived from and the timing of our conception, and then all of the experiences that result from that event, it's very likely that person has a different child than they would have had, had they not started taking the bus to work. Now that's just a simple example, but imagine all of us or a national policy change and an environmental policy change that changes the way that all sorts of people go to work go to the grocery store. This can affect where people move, who they meet, who they conceive children with, and so on right for generations. Well the thought is what looks like a simple environmental policy will have this result of changing the identities of many future people. Why does that matter? Well, why it matters is that if what we're trying to do is we're trying to avoid harming future people with our environmental policies, one thing we have to reckon with is that those policies are not only going to change the quality of life of future people, they're going to change who exists .So the people who will exist if we have a very wasteful environmentally unfriendly policy going forward, there becomes a question about whether those people can now say you have harmed me through your wasteful actions, because had we adopted a more environmentally friendly policy it's very possible those particular people wouldn't have existed at all. They’re arguably not worse off than they would have been had we chosen the environmental conservation realm so that's the problem. How do we explain why it is we should be adopting sort of environmentally friendly policies and so on, in light of this fact that the future people who will exist may not exist if we adopt environmentally friendly policies? How do we explain our reason to save the environment in terms of future people? That’s the basic puzzle, and in in my work, I try to give a sort of solution to that question or a solution to that problem and answer that question.
 
Arjun: That makes a lot of sense, and it sounds very interesting. I know you've also done a lot of research into the ethics of artificial intelligence. How do you think AI and all of its manifestations will challenge our traditional conception of what ethics means?
 
Purves: There are a couple of different answers there. Far down the road, perhaps many years from now, I do think that artificial intelligence might pose a challenge for our concepts of personhood or what it is to be a moral patient or a moral agent - that is, what it is to be someone whose interests matter for their own sake that we should consider for their own sake. it's certainly possible and there are plenty of TV shows that are sort of giving us a little glimpse of this future already, but  it's very possible that we're going to at some point achieve a level of artificial intelligence, general artificial intelligence, where we actually have to ask whether or not this thing has consciousness, has feelings, has beliefs, has desires. I do think that the development of AI will eventually challenge our concepts of a person and a moral patient in that respect. I don't think we're anywhere close to that now, but someday we will… The more interesting features of artificial intelligence for philosophers and ethicists are actually some of the ways that computer scientists and statisticians are actually thinking about concepts like fairness. What you're seeing a lot more of and what I've been thinking about in my own work is, you're seeing a lot of computer scientists try to develop ways to measure fairness or to quantify fairness in artificial intelligence systems. A lot of the interesting work right now is sort of trying to understand what computer scientists have in mind when they develop these measures of fairness and what it is, what sort of moral features of the world these measures are supposed to be picking up on. So you know apparently there are at least 11 different measures of fairness in AI systems that computer scientists are working on, but I think a lot of those are sort of underexplored, the sort of the moral grounds, the moral foundation of those measures is really underexplored right now, and they matter a lot because there have been prominent examples like in the Pro-Publica article called Machine Bias that revealed apparent racial disparities in classification of defendants by an AI system. So this system apparently disproportionately mistakenly classified black defendants as high risk of re-offending compared with white defendants, and this seems to be unfair but according to other measures of fairness that computer scientists use, the system actually wasn't unfair. So disentangling this stuff is really important because it's going to determine our policy going forward for AI systems so that's something that I'm really interested in right now.
 
Arjun: Another manifestation that I know you do work into is like autonomous weapons, autonomy, and predictive policing. How do you think we can ethically design these systems or even whether they can be actively designed to ensure fairness, to ensure justice, and all these important values?
 
Purves: It's a great question. One core requirement of an AI system that's being used in the public domain particularly in criminal justice is that it treats everyone equally before the law. Equal treatment before the law is taken to be an ethical and also legal requirement of our criminal justice system. One thing that you need to watch out for in the predictive policing case is any sort of inequality with respect to the way that the law is being administered. So if a predictive policing system exacerbates that inequality or any kind of inequality then that is probably a moral reason to revise the system or to stop using it. I do think that one problem there is that it's not exactly clear what equality before the law means in the case of predictive policing. I think that's somewhat underexplored. Another key component that I've been working on recently of AI systems in criminal justice in particular, they need to protect or promote the trustworthiness of our criminal justice systems. We need to make sure that we design AI systems so that they do promote trustworthiness, rather than undermine it. For example one concern that people raise about AI systems is they're opaque or they're inscrutable - you can't know exactly why they're making the determinations that they do - and I think you know one problem with this feature of these systems is that it does call into question their fairness. If you can't dispute the way they make a decision because you don't know how, then you can't dispute unfair treatment. But also I think the mere opacity of these systems, the inability to sort of understand why they do what they do, I do think it actually undermines the trustworthiness or it can undermine the trustworthiness of the criminal justice system that uses them. I think that's a big problem because I think one of the pillars of the legitimacy of our criminal justice system is that it's trustworthy so anything that undermines its trustworthiness can undermine its legitimacy.
 
Arjun: Thank you so much for that insight into your research. It was really interesting and really insightful. So now let's take a step back and as we conclude, I was just wondering who was your favorite philosopher, who is your favorite philosopher, maybe not the one you agree with the most.
 
Purves:  That's a tough question to answer. So I would say that as far as shaping my scholarly career, Derek Parfit was probably the most important, most impactful. He wrote the book that I didn't show you because I don't think it's actually very friendly reading for like an undergraduate who wants to get into this stuff, but he wrote a book called “Reasons and Persons,” and that book basically laid the groundwork for what became my dissertation and a lot of other people's dissertations. He was really important in my career - I'm not sure that he's my favorite philosopher, all things considered.  I actually think Michael Huemer -who wrote this book “Ethical Intuitionism” - he's up there for being my favorite philosopher simply because his work is so challenging, and he argues for so many conclusions that I disagree with but I'm unsure how to dispute them. As far as applied ethics goes or practical ethics, which is a lot of what I do, David Boonin at the University of Colorado has to be up there as well, and of course my dissertation advisor Alastair Norcross would feel left out if I didn't include him on that list, but that's just a few a few folks that I think have been particularly influential in my career.
 
Arjun: It’s really interesting how like philosophy builds upon itself when people create new innovations and everyone has something new to offer. Thank you for speaking to me with today Professor Purves.
 
Purves: My pleasure.