the mad dog naturalist
Alex Rosenberg interviewed by Richard Marshall.
[Photo: Nate Glencer]
Alex Rosenberg is the mad dog proponent of nice nihilism who broods on the implications of naturalism. He is always thinking about the relationship between science and religion, science and its laws, reductionism, Dan Dennett, the philosophy of biology, about why scientific realism is better than instrumentalism, giving the atheist a guide, why it’s ok not to have freewill, why Fodor was wrong about Darwinism, why economics is mostly mathematical politics and is improving but still faces the reflexivity problem, about how biology is growing in importance in the social sciences and about analytical metaphysics and recent disputes. This one bites!
3:AM: What made you become a philosopher?
Alex Rosenberg: I made a conceptual error. I was studying physics. The mistake was also one Kant made: thinking that a physical theory was not adequate unless it was necessarily true and made things intelligible. It was reading Hume’s Treatise that disabused me of my mistake. But by that time it was too late to go back into physics. And I turned out to be better at philosophy.
3:AM: You tend to work in three areas, metaphysics – especially the nature of causality , the philosophy of economics and the philosophy of biology. If we start with the latter, we get involved in some pretty big issues straight away, such as what kind of a scientific theory is biology, what’s the status of claims about biological laws and so forth – amplified in importance I guess by the current state of the Darwin wars over in the US. So before discussing those issues, is self-censorship and anxiety over the religious lobby a genuine difficulty these days, to the point where it is destructive or at least distorting the way philosophers and scientists go about their work?
AR: Multipart Answer:
Religion and Darwinism.
Fortunately, even in my benighted country the religious opponents of Darwinism have no appreciable impact in academic philosophy. The same thing can’t be said for the Ayn Randians, interestingly. Several philosophy departments in Brian Leiter’s top 10 or 15 have actually taken money from right wing foundations in exchange for treating her views with respect. But the closest this has come to happening in the philosophy of science and in particular biology, is the willingness of philosophers to repackage their research interests in terms that make them eligible for funding from the Templeton Foundation. We’ve had much more trouble from the likes of Fodor and Nagel than from the Discovery Institute. And so far as I can see there has been absolutely no self-censorship among philosophers of biology in the United States on any matter that might pain a fundamentalist or evangelical Christian. And I think the same thing goes for the biological scientists I know. They may be spinning their research a little more towards its pay-off for improvements in medical care, but that’s a response to politics and budget cutting, not no-nothing bigotry.
Biology as a science, natural selection as its laws?
So, the questions raised in the philosophy of biology reflect a purely intellectual agenda, not a political one at all.
What are theories in biology like, and how do they differ from those of physics—well that isn’t really a very hot questions in the subdiscipline. This is partly because the nature of theory in the physical sciences is a bit indeterminate these days. Think about recent work on the nature of scientific laws, causation, counterfactuals. Among philosophers of science there’s no consensus about which of these three concepts is primary and how to build up the others out of it. In the absence of agreement about physical law, it’s hard to take sides on how biological laws, theories, causal claims, differ from those of the harder sciences. Now, I think the only laws in biology are the laws that govern the Darwinian processes. But the trouble with that view is the evident difficulty of actually stating those laws in a way that will secure agreement among philosophers. Philosophers of biology have the same problems understanding models in biology as they do about laws: are they like models in physical science? They may look like them, but they have almost no empirical consequences for precise observation the way models in physics do. The most important of them are just mathematical truths, according to the most influential of philosophers of biology, Elliot Sober. So, how do they explain contingent mattes? Not a rhetorical question, but one to which there are multiple incompatible answers.
Now, none of these disputes actually has had any ramifications for the question of whether biology is an empirical science or not. The creationists are still stuck on Popperian unfalsifiability, but not one interested in the nature of biology is prepared to use that hoary old notion as a litmus test for whether it’s a science or not. Of course it’s a science. Just look at what we’re discovering every day about mechanisms.
I’ve always been interested in reduction and reductionism. Laying cards on the table, I’ve been arguing for reductionism in biology for a very long time. I was doing this even when I was offering accounts of why the positivist and post-positivist accounts of reduction in terms of theoretical derivation could not be satisfied—owing to supervenience of multiple realized macromolecular arrangements. Then, and today, I was still trying to figure out how reduction proceeds. I know it proceeds, because I see it reported every week in the pages of Nature and in Nobel Prize winning research. But obviously, as in the philosophy of mind and psychology, the reductionists are in the minority and most people are physicalists but antireductionists. How to reconcile physicalism and antireductionism remains a vexed question in biology, in psychology, and of course among metaphysicians as well.
But philosophers of biology have stopped trying to nail down any reduction in the field. They no longer seek to define the gene in terms of DNA. We’re more interested in questions of whether the gene/DNA make a unique contribution to heredity and development, or whether the latest science shows that nucleic acids are on a par with a lot of other factors in transmission of traits between generations and development of traits within generations. You’d think these are empirical questions for molecular biologist, but the molecular biologists answers turn on concepts like information, causation, and reproduction that are rife with ambiguities that can easily shift the answers biologists give. Philosophers are sensitive to the ambiguities and we take sides on the questions ambiguously expressed.
One reason I am so interested in matters of reduction is that I think that most of the evidence that decides biological claims is molecular. Just consider what DNA evidence—genetic and somatic–is telling us about systematics, phylogeny, development, or neuroscience, etc. One area that I follow closely and has the greatest implication for human evolution, human culture and human prehistory is the sequencing of the genomes of the four different Homo species that have been alive together on the planet in the last 60,000 years. Every few months more comes out from the Max Planck institute in Leipzig that help us understand differences between these species and scenarios that led to the extinction of three of them and the domination of one. I think and I hope that we are going to be able to shed increasing light on problems about human cultural evolution from these sources. That’s a tangential reason I remain interested in reductionism.
The agenda of the philosophy of biology
But of course these interests of mine are not on the front burner in the philosophy of biology. The questions that are important in the discipline are the ones that are on the intersection of interests among biologists and philosophers—the perennial interest in the units and levels of selection, and its upshot for altruism and cooperation, the evolution of language—where Brian Skyrms work has been of great importance in vindicating David Lewis’ ideas of how conventions spring up, and of course the distinction between natural selection and drift—are they distinct forces, are they forces at all. People are also very interested in drift and the role of probability in evolution. After about 50 years of serious work by a lot of very smart people, we still don’t have much agreement on how best to state the theory of natural selection. But one thing we don’t have in the discipline is any disagreement on the role of the theory in organizing biology and on its explanatory status in the domain of biology.
3:AM: You wrote your book ‘Instrumental Biology’ to argue that biology should be seen as a useful tool for coping reality rather than an account of reality didn’t you? Could you explain why you think this and why it’s better than realism?
AR: I am a scientific realist. I believe the world is literally (approximately) the way the best scientific theory says it is. I am also a reductionist. I think the complete explanation of natural processes is one that adverts to physical processes only. When I wrote Instrumental Biology or the Disunity of Science I despaired of providing an account of the process of natural selection that would show how it could be, let alone was, in fact, reducible to a purely physical process—the operation of physical laws at all. Since biology is organized by the theory of natural selection, as a realist I couldn’t accommodate it without the required reductive explanation to physics, except as a useful instrument. Like the rest of natural science there is in fact a good deal of biology that is a matter of useful instruments, harmless idealizations, good approximations. I tried to marshal those features of biology into an instrumental account of the science as reflecting our cognitive limitations in the face of huge biological complexity.
But about 10 years later I figured out that the process of natural selection is nothing but physics, or maybe chemistry. Darwinian processes are just the results of the 2d law of thermodynamics operating around here. I figured out how natural selection is reducible in principle to physics, that is, how, as I say in a later book, physics fakes design.
So, I was able to give up instrumentalism about biology and I did so in another book, Darwinian Reductionism, or how to stop worrying and love molecular biology.
Thus, short answer to your question: instrumentalism about biology isn’t better than realism. It’s worse than realism and I don’t recommend it.
3:AM: Ok, excellent. But were you always a realist about physics?
AR: I am strongly committed to realism, and by a combination of arguments—including Boyd’s no miracles arguments and Leplin’s novel prediction arguments. But I recognize that the agenda of realism still has some important work to do. Two of the obstacles we need to deal with are finding a realistic interpretation of the most fundamental of physical theories—quantum mechanics, and dealing with the increasingly sophisticated versions of the pessimistic induction of Larry Laudan, for example ones due to Kyle Stanford.
On balance realism seems to me the most well grounded view about scientific theory.
Are physicists realists? Well, you can do great physics and not be a realist. When I listen to Leonard Susskind, it’s clear to me he’s no realist. He treats physics as an input-output information processing device. And I don’t think we’ll be able to make physics safe for realism until we have agreement from physicists about a realist interpretation of quantum mechanics.
3:AM: Is there no prospect of any kind of nomological universality in biology? I guess this is the issue about reductionalism and anti-reductionalism again isn’t it?
AR: Beyond the laws governing the process of blind variation and environmental filtration that produces adaptations in the biological domain, I doubt there can be nomological universality, strict laws. Why? Simple: selection for effects is blind to structure. There is always more than one way a given function or adaptation can emerge. Indeed there seems to be more than one way it has emerged, if you consider the redundancy of the genetic code. So, both the antecedents and the consequents of biological generalizations are multiply realizable, and that makes the regularities gappy at best. All that actual and physically possible multiple realizabilty at the basement level means that at most we get ceteris paribus generalizations at every higher level of organization. And despite Jerry Fodor’s bluster, these ceteris paribus laws in biology and the special sciences, don’t support counterfactuals and can’t be improved into laws. Since they aren’t laws they can’t be reduced to laws either, though their approximate truth can be reductively explained.
3:AM: Your position has been called a ‘mad dog’ naturalist position – by yourself amongst others I think. And your happy to be labeled scientistic, which has been a term of abuse for some! In the Moving Naturalism Forward conference Dan Dennett and others thought your view too austere, missed out too much, such as freewill. Why do you say they are wrong and that your position is the better one? And what do you say to someone like Tim Williamson who says you can’t explain mathematics from a naturalism like yours?
AR: In The Atheist’s Guide to Reality I tried to coopt the word ‘scientism’ and to argue that science can answer the persistent philosophical questions that trouble people, including the nature of reality, the purpose of life, the existence of a soul, the grounds of morality, whether we have free will, and the meaning of human history. Most of the answers science gives to these questions are unpopular and people neither understand them nor want to hear them.
In arguing that science has to reject almost everything common sense tells us about reality and our place in it, I am going against a Naturalistic tradition whose leading figure is the sainted Daniel Dennett. Dan has done more than any one to advance the naturalistic program of giving answers to the persistent questions that reconcile common sense—the manifest image, in Wilfred Sellers’ words—with science. But a lot of other philosophers have helped to try to advance this agenda. I honor them all, but I deem their program a failure by the standards that they set themselves. Two examples: Naturalists either violate Hume’s dictum is/ought dictum or reject it without good arguments. Sam Harris wrote a whole book steadily doing so, The Moral Landscape. Intentionality—the aboutness of propositional thoughts: a half century of the philosophy of psychology and we still haven’t figured out how it is even possible.
In The Atheist’s Guide to Reality I try to explain why the reconciling program of naturalism won’t succeed, and how we can live with that fact.
As for free-will, if you can find the spot in the web-cast of Moving Naturalism Forward [Day Two, Afternoon, Second Session: Freewill/Consciousness] where Dan and I discuss free will, he comes as close as I could get him to agreeing that the concept of free-will is just a useful instrument for beneficent social control, not a fact about reality.
I’ll come back to Williamson and maths at the end if I may.
3:AM: Why do you say it’s ok not to have freewill?
AR: Since we don’t have free will, it’s OK to say so, insofar as speaking the truth is OK.
It might not be OK to say we don’t have free will if that seriously undermines the benevolent social institutions that employ the words ‘free will’ in ways that contribute to general human welfare (if we are in favor of fostering human welfare, as I am).
So, you might think there is a cost benefit problem here: speaking the truth vs. unraveling human society. But there isn’t. We wouldn’t be able to give up using the words ‘free will’ the way we do in a million years. Mother nature, a.k.a. natural selection, long ago turned us into creatures that explain and predict the behavior of others in terms that carried along notions of responsibility, praise blame, will, etc. We wont be able to shake these notions in even geological epochs. The belief in free will is like the belief in sunsets.
3:AM: In the rather heated response to Jerry Fodor’s provocations about natural selection your response was one of the few that recognized that he was onto something. I want to quote you: ‘ His modus tollens is a biologist’s and cognitive scientist’s modus ponens. Assuming his argument is valid and the right conclusion is not that Darwin’s theory is mistaken but that Jerry’s and any other non-Darwinian approach to the mind is wrong. That puts Jerry in good company, of course: Einstein’s.’ I don’t know if you’d agree, but it struck me that many of the responses to Fodor’s argument got it wrong about why he was wrong (if he was). Why do you think Fodor wrong but in an interesting way?
AR: When Einstein developed the “spooky action at a distance” objection to quantum mechanics in the’30s (the EPR thought experiment), he had no idea he was actually formulating the idea of entanglement and that his objection when tested would vindicate quantum mechanics 40 or 50 years later. When Fodor argued that natural selection can’t see properties, and can’t produce organic systems, for example brains—that respond to, represent, register properties, he thought he was providing a reduction ad absurdum of Darwinian theory (the way Einstein thought he was providing a reduction ad absurdum of quantum theory.) The confirmation of Bell’s inequalities theorem by D’espagnat’s experiments turned Einstein’s reductio into a modus tollens. I believe that Fodor’s attempted reductio of Darwinian theory is a modus tollens of representationalist theories of the mind, theories that accord to the wet stuff, to neural states what Searle calls original intentionality. It’s an argument for eliminativism about intentional content. So Fodor is totally wrong abut Darwinian theory, but his argument shows that we Darwinians (and all the physicists if I am right that Darwin’s theory is just the 2d law in action among the macromolecules) have to go eliminativist about the brain.
3:AM: When it comes to economics you wrote a book posing a question that I guess still needs asking so I will – is economics mathematical politics, or science of diminishing returns?
AR: Large parts of economics remain what I disparagingly called mathematical politics—especially those slavishly devoted to providing unique, stable general equilbria. Nothing more fully vindicated the view I defended in Economics—mathematical politics or science of diminishing returns than the Chicago school’s response to the sub-prime mortgage melt-down: “real estate bubble? What bubble?” If you are a regular reader of Paul Krugman’s blog in the NY Times, you will know exactly what I mean.
But since I wrote that book, indeed even before I published it, things had begun to change in economics. Experimental economics, evolutionary game theory, behavioral economics, began to make a difference. Economists ceased to turn their backs as utterly as they had for the previous century on psychology and especially cognitive social psychology. So, economics is a distinctly more empirical discipline than it was 30 years ago. The trouble is that no one has yet been able to integrate the developments of evolutionary, experimental and behavioral economics into the equilibrium approach which his still the obsession of most of the parts of the discipline that distribute prestige.
3:AM: Does philosophy constrain economics?
AR: Does philosophy of science constrain economics? No. Otherwise Amartya Sen would be even more influential than he is among economists.
Does political philosophy constrain economics? It must, especially libertarian political philosophy. That’s the only explanation I can see for the strength of commitment of so many economists to the free market in the face of so many and many glaring examples of market failure—the uses of market power to move the economy away from allocatively efficiency towards the rent-seeking of the 1 %.
3:AM: Should we understand economic theories as part of the biological sciences, in the way that, say, some xphi philosophers are looking at moral theories? You said once that you’ve been unsettled by much of the theoretical activity in economics since you wrote ‘Microeconomic Laws’ but you’ve stuck with some version of understanding economics in the same way as theories in physical, biological and behavioural sciences. Are you still sticking?
AR: Economics is a social science, social sciences are the study of one particular species (occasionally in comparison with other primates). The study of any biological species should proceed in accordance with well established principles of biological theory and method. Case closed.
The harder question is whether we are likely to produce any solid predictively useful theory about human behavior by a relentlessly evolutionary or biological approach to social behavior. I don’t know the answer to that question, though I think there is already some good institution design that exploits findings about cooperation. But I do know that there is no alternative theory to a biological one that could give us a better handle on human behavior. Note that a biological or Darwinian theory of human behavior is emphatically not a genetic theory that treats much of anything important about culture as hard wired. A biological theory is one that seeks the explanation of cultural regularities—local equilibria, if any—as the result of blind variation and what Donald T. Campbell called, selective retention, or what I call environmental filtration.
3:AM: Is economics unimprovable? And what implications follow in terms of our attitude to economic theories and predictions? Should we listen to economists?
AR: There doesn’t seem to be any a priori barrier to improvement in the predictive precision and explanatory power of economics as a research program. What worries me is the reflexivity problem…Suppose big data mining enables us to produce a really predictively reliable macroeconomic theory or even a micro-theory that some one can use to trade in financial markets. It won’t be kept secret long enough to exploit because incentives to sell it will be irresistible, and employment of it will change the behavior the theory systematizes…end of predictive power. Doubtless some wag will point out that I have just made an economic claim that is predictively reliable.
We should listen to economists who advise us about institution design. They understand that such institutions must be incentive compatible and must protect us from Hume’s knave and Hobbes’ fool.
3:AM: You’ve updated your text book on the Philosophy of Social Science that you first wrote back in the 1980’s. As much as anything, the changes in the status and condition of the social sciences over that time are fascinating and stark – I guess economics is now much more seen as a social science when before perhaps it liked to think itself a cut above. So perhaps you could sketch what you’ve seen as the interesting changes over the time you’ve been surveying this scene? And is philosophy also a social science, or is there something different about philosophy as you see it?
AR: My answer to earlier questions tell you a lot about what I think has changed over the 25 years and 4 editions of The Philosophy of Social Science. The book contains a fair amount of continental philosophy of social science, but that’s largely a matter of demand in the market, not of any importance I attach to the Geisteswissenschaften tradition in the subdiscipline. The main thing that’s changed in the real social sciences since the first edition of the book is the ever increasing role of biological approaches to behavior and the sophistication of the theories that have been advanced in these domains. In the text I try hard to set the stage and identify the underlying assumptions of such theories and protect them from jejune objections.
It’s true that my own department of philosophy has moved to the social sciences division at Duke (or at least half way there), but that reflects the discomfort academic analytical philosophy has with the humanities, not the fact that we are more closely connected with the social sciences. I do think we are equally closely connected with physical science and life science–just look at the work in philosophy of physics and biology. The agenda of philosophy as a second order discipline or one that concerns itself with abstract and general problems that intersect with those of scientists seems to me to be the right one, and I think that has characterized philosophy since Plato.
3:AM: Your last book asked atheists and naturalists to face up to reality rather than just endlessly engage with the opposition. What were the main points you really wanted to get people to understand when you wrote that – and having lived with the responses over the last year or so have you found people have taken up your ideas or found ways of ducking some of the hard issues?
AR: The Atheist’s Guide to Reality made the case to those who trust science (and are in consequence atheists) should also trust science’s answer to the persistent philosophical questions that keep us all up nights, even when those questions are disconcerting or at variance with comfort or consciousness or common sense. Science is much more reliable than common sense since it is common sense recursively reconstructing itself on firmer foundations, and by now it has shown us that conscious introspection is a wholly unreliable guide to almost anything, including the nature of cognition, emotion and sensation.
What is clear to me about the reception of The Atheist’s Guide was first how hard it is to get nonphilosphers to understand the problem of intentionality and aboutness, second how much harder to understand the eliminativist solution to the problem, and most all, the degree to which our emotional attachment to narratives—stories with plots, good guys, bad guys, agents with motives—gets in the way of our understanding science and applying it to these persistent questions.
Now I know how Berkeley must have felt when Dr. Johnson refuted him by kicking a stone, especially when I read the puerile self-refutation arguments against my eliminativism.
Of course a few theists have identified The Atheist’s Guide to Reality as correctly identifying the implications of demonic materialistic naturalism. But I am actually surprised by how few have done so. It’s no tribute to the intelligence of the rest of them.
And of course naturalists have pretty much ignored the arguments for the same reason. It gives naturalism a bad name with the public, whom they hope to win over to a humane and civilized point of view. I wish they had been able to succeed in reconciling science and the manifest image. Maybe they yet will. I doubt it. If they do most people won’t notice, at least where I come from.
3:AM: What do you make of the recent disputes between philosophy and science, as around the Nagel and Krauss books – and do you think disputes in philosophy between naturalists and analytic metaphysics though less well known equally important?
AR: I think of analytical metaphysics as very general, very abstract physics. A good example that vindicates this work are the papers, published and unpublished, of Laurie Paul.
The recent disputes between philosophy and science are like the ancient ones, for example Zeno’s paradoxes. In philosophy the surest route to immortality is to provide a logically impeccable argument from true premises to a conclusion that is known, usually by science, to be false. We prize such paradoxes above all else, and celebrate their originators. The last 400 years have given us a lot of reason to believe that the mind is the brain, that it couldn’t be anything else. That’s what makes the qualia arguments so interesting. Start from something we know for certain—by conscious introspection, as Descartes did, and by valid reasoning conclude that physicalism is unintelligible, as Nagel did in “What it’s like to be a bat.” Then when you realize that the choice you face isn’t just first person certainty vs. neuroscience, but first person certainty vs. all of science, if you are really confident in your penetration, write a very short, uninformed book like Mind and Cosmos: Why the materialist Neo-Darwinian Conception of Nature is Almost Certainly False. (I recognize that my detractors will accuse me of writing an uninformed book, but at least it was a little longer than Nagel’s).
Of course since we philosophers see things more deeply than mere scientists, we are likely to become angry when scientists claim to be able to solve philosophical problems or better, to dissolve them. That’s why a hard headed philosopher, and a really nice guy, like David Alberts went out against poor Laurence Krauss, merely a rather good physicist who thought he had an answer to the question of why is there something, rather than nothing. Krauss didn’t realize that smart philosophers could poke holes in any argument a quantum physicist gave for why the cosmological argument Is invalid. So he had to be brought up short, and quickly too.
One of the things I tried to make clear in The Atheist’s Guide to Reality is that for many of the persistent questions that keep people up at night, we should trust science, and not clever arguments build by smart philosophers out of premises that merely feel right to us, because though the reasoning is good, the premises are probably doubtful. Even when we can’t refute these arguments, we have 300 or 400 years of science’s success to tell us that there is something, somewhere wrong with them. That doesn’t mean that all questions philosophers have are addressed by the sciences. I think in fact the greatest achievements of philosophy are the formulation of questions that haven’t occurred to scientists and that the sciences cannot yet answer. These questions serve as milestones marking scientific progress.
One set of questions that philosophy wrestles with and no science does, are the questions about the epistemology and metaphysics of mathematics. These are the hardest questions for naturalism. For without some nominalist strategy for turning them into empirical claims or analytical truths, we need to come up with an epistemology that reveals how we known necessary a priori synthetic truths. And we naturalists don’t have any such epistemology. But then tu quo que, neither does any nonnaturalist have a good handle on the nature of mathematical knowledge and mathematical reality. Tim Williamson seems to allow the whole question about whether science is the only kind of knowledge or not to rest on whether it can make sense of the philosophical problems about mathematics. I want to allow that the whole question about whether we will ever make sense of mathematics to rest on the success that science has had so far making sense of everything else.
3:AM: And finally, for readers here at 3:AM, are there five books other than your own that you’d recommend we read to get further into these issues?
AR: I’m afraid 5 is too few. I am going to give you 7, though only two are by contemporary philosophers:
Dan Dennett, Darwin’s Dangerous Idea—Darwin for philosophers
Jared Diamond, Blood, Germs, and Steel—Darwinian cultural evolution done right
Sean Carroll, From Eternity to Here—why the universe—especially time–is nothing like we think it is
Sean Carroll, Endless Forms Most Beautiful—how physics fakes design
Paul Churchland, Plato’s Camera—eliminative materialism without tears
Lucretus, De Rerum Natura—a philosopher who had it all pretty much right (by accident) 2000 year ago, and said so in verse.
ABOUT THE INTERVIEWER
Richard Marshall is still biding his time.
First published in 3:AM Magazine: Friday, July 19th, 2013.