Well, actually, they got Tim Williamson (Oxford) to do it. And now Alex Rosenberg (Duke) comes to the defense. Thoughts from readers?
I first met Professor Hoy when I returned to UC Santa Cruz in Fall of ’92 to finish my undergraduate…
34 responses to “NY Times Attacks Philosophical Naturalism!”
-
A vote for Rosenberg.
-
Tim Williamson's argument is quite modest – it invests in a fallbilist approach that seems much more empirically justified than the Rosenberg position – he argued in 'The Philosophy of Philosophy' that it's philosophically interesting to raise the question as to whether everything is in principle knowable. Rosenberg is dogmatic in refusing to even consider the issue. Perhaps like those utilitarians who argue that a good utiltarian should reject utiltarianism on utiltarian grounds, the naturalist needs to similarly reject naturalism.
-
Make that two votes for Rosenberg.
His point about the death of the concept of purpose is for the purpose of anchoring his position dead-on. (Rosenberg: Go ahead–make my day.) On the other hand the in-and-out-and-in-and-. . . status of the concept of force in physics does make one more reflective about how minds mirror reality in reliable ways.
Math concepts are a naturalistic problem for sure. Functional mapping of what's thought with what seems to be with what we can do works well enough at least for overall confidence if not thorough understanding.
My solution is mostly Rosenberg titrated with a pinch of Williamson. Maybe a healthy pinch–but I have to go with Rosenberg overall.
-
Hmmm. So it seems that Williamson thinks the expression "the scientific method" can imply something excessively narrow, while the correct attitude is the inherently open-ended "scientific spirit." But I am not sure why he thinks a naturalist is committed to the former rather than the latter, or even why he thinks the former is always intended to be more narrow than his conception of the latter.
On the other hand, Rosenberg does seem to be advocating something narrower than Williamson, but if history and literary criticism are his only examples of Williamson's excessive open-mindedness and tolerance, I really don't see the excess.
Still, ultimately it's striking how little they disagree about. Perhaps the issue is really Williamson's framing of the dispute? Many are quite concerned that when a serious thinker says "naturalism is a dogma," the statement is likely to be endlessly quoted by believers in much sillier things than literary criticism, people with various superstitions that Williamson isn't intending to defend. There may be something to that, but it's not as if Williamson's essay is particularly unclear on such matters, and I tend to doubt that any amount of careful phrasing is ever going to provide sufficient defense against a determined quote-miner.
-
As a historian, I find it interesting that Rosenberg avoids Williamson's point about history, which is surely a discipline with a good claim to intellectual rigour, and a discipline productive of knowledge. How exactly is valid historical knowledge to be reached apart from through archival research? Maybe archival research is compatible with Rosenberg's naturalism (as it is with some forms of methodological naturalism), but he should have said something about this. Having a pop at trendy movements in literary criticism seems to me just an easy way to duck the problem. And the description of literary criticism as fun and 'not knowledge' is pretty crass. Scholars of renaissance English can surely understand Shakespeare's works to a much more exacting standard than can casual readers, because they have a much greater knowledge of literary history and of the linguistic and ideational context in which Shakespeare wrote.
-
I'm not sure that Williamson means to attack a position so much as to resist a slogan or label, as either too vague to be helpful or when made sharp enough to have some bite as outstripping its justification. That's roughly how he casts his idea at the start of his piece.
-
Must naturalists assume there is a single scientific method? I rather think naturalists can accept the vagueness of scientific methodology and define their position in more negative terms: as the rejection of supernaturalism. That seems to historically be the case. Naturalism is the categorical rejection of a certain sort of argument, and not a categorical exaltation of another.
Also, why must naturalism be a claim about all kinds of knowledge? Naturalists can believe, for example, that learning how to tie your shoes is not something you need science to accomplish. Some kinds of knowledge can be procedural, and not a matter of scientific discovery. Mathematics can be of this kind, and not a problem for naturalists at all.
-
Alex Rosenberg writes: "Naturalists have applied [the theory of evolution] to reveal the biological nature of human emotion, perception and cognition, language, moral value, social bonds and political institutions."
"Reveal" is a success verb. Is it really true that evolutionary theory has *revealed* the *nature* of all these things? Surely the best we have in many of these areas are evolutionary stories, consistent with known facts, according to which this or that trait *might* have evolved. But how many relevant facts that could overturn these theories are unknown and will probably never be known? Lots. And lots. It's one thing to defend naturalism as a research program. It's another to overinflate its accomplishments.
With respect to moral value in particular, the best thing that evolutionary theory can give us is a theory of how we come to care about, or approve of, this or that. But it's pretty clear that many of the things many come to approve of (e.g., revenge, honor killings, genital mutilation) are wrong, and many of the things many disapprove of (e.g., homosexuality) are not wrong. Unless you adopt a kind of Humean sentimentalism or other philosophical theory linking moral value to moral valuing, evolutionary theory isn't going to get any traction. (And if you end up with an objectivist version of sentimentalism — ideal observer theory or somesuch — evolutionary theory can't help you.) And how exactly are you going to defend such sentimentalism? By testing it in the lab? No. By looking at philosophical arguments for and against.
Let's by all means adopt naturalistic accounts of natures in those fields that call for them. But it is indeed a form of dogmatism to insist that absolutely everything has the kind of nature that calls for a naturalistic account.
-
It seems that Williamson attacks a straw man. I’ll put a couple of reactions, but I don’t have time to deal with everything. First: “If we say it is the world of matter, or the world of atoms, we are left behind by modern physics, which characterizes the world in far more abstract terms.”
This seems an ambiguity fallacy since the word “characterize” has two different meanings. One is to "describe" something. The second meaning is “be typical or characteristic of”.
The second meaning suggests that modern physics view of the world is something like Pitagora’s “all is number”, like they are some kind of idealists.
I believe the first meaning is true about physics, it only uses abstract terms to describe the world. But all sciences use abstract terms to describe their object. If I use abstract terms to describe a cow, it does not follow that the cow is just an idea.
Williamson seems to use the first meaning of the word characterize to convince us of the second meaning. The argument seems like the following analogy:Premise one: Peter characterizes his friend John in abstract terms.
Conclusion: Therefore John is an abstract idea. -
One straw man is: “What, for a start, is the natural world? If we say it is the world of matter, or the world of atoms…”
I don’t have the impression that naturalists describe the natural world as a bunch of atoms. Rather they say that something exists only if it can have causal effects on something else. That’s the condition for something to be knowable and observable. Every measure instrument shows us a causal effect of something else. When we look at pictures of the results of collisions of accelerating particles, we only see that those particles had some causal effect on the sensors.
For naturalists, ideas or “abstract terms” would not have any causal effect on other stuff, therefore they don’t exist. -
“Anyway, the best current scientific theories will probably be superseded by future scientific developments. We might therefore define the natural world as whatever the scientific method eventually discovers.”
This seems a new straw man. It seems to suggest that superseding scientific theories talk about different things than their predecessors. But that is not what naturalists claim. They claim only that new theories are just a better description of the same thing that was described by their predecessors. Naturalists expect that new theories are an qualitative and quantitative improvement in knowledge and that we will eventually know one day everything that we can know. Thus some theories will not be superseded by others. -
Williamson is right. Much of what we know (I would say, most of what any one of us knows) is not known through science. A great deal of my knowledge about the contents of my own mind was gained without the benefit of science. Similarly, I rely, for my life, on knowledge of the thoughts, feelings, and motivations of people around me, and science has nothing to do with it. Likewise with culture.
As has been noted, Williamson attributes more than is necessary to naturalism. It need not be taken so categorically. But that is how Rosenberg defends it: "[N]othing that revelation, inspiration or other non-scientific means ever claimed to discover has yet to withstand the test of knowledge that scientific findings attain." But this fails to delimit the subject matter appropriate to natural science. Consider: science has not been around forever. People had a great deal of knowledge before the Renaissance; they had a great deal of knowledge before the birth of Thales. Had they not, they all would have died. Natural science has corrected many falsehoods and added enormously to our epistemic scope, but it has not supplanted all other forms of learning.
Finally, Rosenberg fails to appreciate Williamson's metaphysical realism: to be a realist about the world means to accept that there may well be (and almost certainly are) many things that we will never, and could never, know. Science's success in testing the various laws that have been proposed to explain observational regularities has no bearing on this point.
-
I'm not sure Wiliamson is interested in those straw men at all. It seems perfectly sensible for him to worry that alternative research programmes to naturalism are being ruled out too easily. I think that even as a claim of metaphysics the issue is a live and serious one. So, for example, Roy Sorensen has worries that there is a bias against non-material existence and naturalism seems guilty of that. Non-existing things are never going to be subjected to scientific cause and effect are they? For example, the elephant not in my room is not subject to any kind of physics. Holes, silence and shadows are all non-material and so they too, presumably, are non-naturalistic. Williamson is surely right to note that current metaphysics is quite capable of presupposing all sorts of weirdness – including supernaturalism if it wants – McGinn's 'mysterons' are supposedly inaccessible to any type of scientific analysis and so are both epistemicaly hostile and maybe metaphysically non-natural too. I guess Wiliamson's point is just a gentle reminder that the proper spirit shouldn't rule out any avenues for serious thought a priori.
-
Rosenberg, by default.
-
"People had a great deal of knowledge before the Renaissance; they had a great deal of knowledge before the birth of Thales. Had they not, they all would have died."
Does the fact that a representation is survival-conducive make it knowledge? Do Mother Nature and Father Epistemology care about the same things?
-
One reader of the NYTimes piece by Rosenberg is right to ask what it means when Rosenberg talks about "getting things right." But applying the same question to Williamson results in even more perplexity.
In saying, "getting things right," Rosenberg means "right" in any of the ways that science can be successful: a valid proof, a successful prediction, a clarifying interpretation or model, a deep explanation, production of a working technique or machine, the achievement of a goal, etc.
In contrast, Williamson concludes by pointing to the "spirit" underlying science, and somewhat rhetorically says naturalism may fail in its attempt to "condense the scientific spirit." Yes, perhaps there is some spirit that philosophers can hope to embody or enlighten others about. The history of philosophy has many inspiring examples. But this is even murkier than naturalism about what it means to "get things right."
As a naturalist I do not share Rosenberg's fear of meaninglessness. Human imagination is prolific. I'm afraid that the meanings we create may easily end up disconnected with how we can provide well-being for ourselves and others.
-
Johnathan M. Cohen
I'm not so sure Rosenberg isn't only defending the negative kind of naturalism that Jason Streitfeld describes.'Non-scientific means' does not have to mean 'means not actually used in science', but could be interpreted more strongly as 'means that science couldn't possibly use'. The examples that Rosenberg uses suggest that the second interpretation is the right one, and if so he needn't deny that the things you talk about are knowledge.
-
I'm not sure Williamson is right, but I suspect Rosenberg is wrong. Allow me a few grand, unguarded remarks.
Consider those normatively structured phenomena that are tracked by the concepts of warrant, justification, intelligibility, coherence, etc. And suppose (metaphysical) naturalism is true. If such phenomena aren't natural, they aren't real, and so the possibility of science is mysterious. Naturalism can't abide this, since the content of its thesis depends upon the possibility of science.
So such phenomena must be real, thus natural, and thus the possible subject matter of some scientific theory or other. But what sort of scientific theory would this be? For example, what scientific theory will treat of the intelligibility of nature? I would say that the intelligibility of nature is a precondition of scientific theorizing, not the subject matter of this or that theory – and so not natural.
Am I off the mark?
-
The two are working at cross-purposes. Williamson wants to maintain a dose of skepticism about the reliability of human undertakings according to scientific method. Rosenberg wants to demonstrate that the fallibility of experts shouldn't reflect negatively against the method. Absent clearer definitions of terms they both use–spirit, knowledge, criticism, confidence, etc.–it's hard to measure one against the other. These terms have plain meanings, but Williamson and Rosenberg both nuance those meanings. I do think Rosenberg concludes on an unfortunately sour note. His naturalism would not treat anything as "fun." Nor do I see a threat to the meanings we cherish from a mere demonstration of their false appearances.
-
I don't take Rosenberg to be responding to Williamson, though perhaps he was trying. I take Williamson to be claiming:
1. Naturalism is not a well defined position.
2. More inclusive ways of defining it render it nearly contentless.
3. More restrictive ways of defining it render it dogmatic.Rosenberg seems to be responding to the claim that naturalism is dogmatic, and wishes to defend naturalism against that attack. But I don't take Williamson to have attacked in that way. 'naturalism is dogmatic' only appears as the consequent of a conditional in Williamson, at least the way I read him. The antecedent of that conditional was 'naturalism rules out mathematics, history, literary criticism, etc. as knowledge generating'. But Rosenberg does not say they are not knowledge producing. He is explicit that naturalism has to find a place for mathematics, and that would only be imperative if it is a knowledge producing discipline. For the other disciplines he only says they might not be knowledge producing (I too find his and Williamson's inclusion of history odd, since I am more convinced of the naturalistic bona fides of history than I am of theoretical physics). So Rosenberg is defending naturalism against a charge Williamson never asserts, but which he only mentions would be valid if naturalism implies things that Rosenberg says it does not.
On the whole I think I agree with Williamson. I know of no clear statement of ontological naturalism that isn't buck passing to the extent that I don't know what I would be agreeing to if I accepted it. That doesn't mean that I have no scruples when it comes to ontological commitment. I try not to believe in things that I think science couldn't explain or endorse, but then again I try not to tell scientists in what direction their investigations will take them, so that is easy to do.
-
Williamson suggests that naturalism is too obscure to really mean anything, but that's only because he frames naturalism as a claim about 'the natural world' and 'the scientific method'. Naturalists don't have to use those terms. I would have thought that what really unites naturalists is a certain kind of attitude to science: in short, the attitude that science is the best guide we have to what the world contains.
This is compatible with plenty of non-scientific disciplines generating knowledge, and it seems to me that this is where Rosenberg misses the mark. For example, history clearly generates knowledge. But if historians went around reifying entities that have no place in a scientific worldview (e.g., if they reified 'historical forces'), we would rightly be sceptical.
-
A vote for Sam Rickless, especially his third paragraph exposing the ambiguity of "moral value" in Rosenberg's "the biological nature of … moral value." No biological investigation of our psychological states of moral valuing can show that there isn't a truth about what's morally valuable, which is what ethical non-naturalists have been concerned to assert. Or at least it can't show that unless it's assumed from the start that what science doesn't discover can't exist. But Rosenberg's list of supposed naturalist successes is supposed to be establishing that thesis, not just assuming it.
-
Joseph Streeter: "As a historian, I find it interesting that Rosenberg avoids Williamson's point about history, which is surely a discipline with a good claim to intellectual rigour, and a discipline productive of knowledge."
Rosenberg has already stated: "History is bunk"
http://onthehuman.org/2009/11/the-disenchanted-naturalists-guide-to-reality"Having come this far, scientism has the resources to explain the frustrations and the failure of the social sciences and history, and it provides a firm basis on which to establish reasonable expectations about the prospects for the human sciences, qua sciences.
The nature of meaning and its at-best merely instrumental grasp on real events in our brains and in the world gives scientism manifold reasons not to expect history and the historical versions of the social sciences to provide anything more than diverting stories, post hoc explanations and very short term expectations about the human future. But there is a much deeper reason to be pessimistic about the uses of history: reason enough to conclude that Santayana’s or Churchill’s reasons for taking history seriously—to know the future–will never be borne out."
As I've said in the past, I'm never sure what philosophers mean by the terms "humanism" and "naturalism." To a historian, Humanism originates in the renaissance with a reengagement with the past. To a philosopher it centers on the Enlightenment and an optimistic sense that the past is something that can be left behind. Similarly, naturalism has become synonymous with scientism, while Santayana was once considered to be a naturalist. Is he still?
A recent example of why history is useful
http://crookedtimber.org/2011/09/16/money-sex-economics-and-stuff/ -
I second the vote for Sam Rickless' remarks, especially if we can add a little 'neo' to the "Humean" in that third paragraph…
-
“the scientific method…It involves formulating theoretical hypotheses and testing their predictions against systematic observation and controlled experiment.”
Not necessarily. I could count all the swans in the world and conclude that all the swans in the world are white. And I’m doing science. I don’t necessarily predict anything or make controlled experiments. Sometimes science is just collecting data, after all science is about knowing stuff. Therefor mathematics can be seen as just collecting information. The same is the case for history.
-
A minor point: Patrick mayer writes "I am more convinced of the naturalistic bona fides of history than I am of theoretical physics". I'm pretty confused by that: if an extensive and continuing track record of quantitative, empirically confirmed novel predictions doesn't count as solid evidence of naturalistic bona fides, what does?
-
I am probably guilty of building too much into my definition of naturalism, and certainly guilty of ignorance of the physics beyond a Popular Science level. I was treating naturalism as incompatible with idealism of any type. It seems to me that lots of physicists seem comfortable with the mind dependence of certain properties of electrons and other small stuff. I do not understand why this doesn't count as contemporary physics endorsing limited idealism. I have spoken to philosophers of science who do not think that is true, and that the seemingly idealist commitments of quantum mechanics are just on the surface, and I defer to them on this issue, because I do not know enough of the physics to give much credence to how it seems to me. But nonetheless it is true that it is not obvious that physics is not committed to some amount of idealism. So it is not obvious to me that physics is thoroughly naturalistic (maybe I am thinking of naturalism too narrowly), and it is obvious to me that history is or can be one in a thoroughly naturalistic way.
-
Patrick: ok, I see where you're coming from now (I hadn't quite realised how you were taking "naturalism"). But I'm with the philosophers of science you've spoken to: no-one I know in contemporary physics talks about electrons etc that way. (Maybe a few did in the 70s.) Plenty of physicists flirt with (relatively crude) instrumentalism, but none are idealists.
Having said that, does naturalism rule out the mind-dependence of the physical world a priori, or just because it's an overwhelmingly unsuccessful research strategy, empirically speaking? I'd have thought the latter, but it might be a matter of definition rather than substance.
-
In the context of Williamson and Rosenberg's pieces I think naturalism is supposed to be a matter of method rather than doctrine. I.e. about endorsing whatever theories the hard sciences (perhaps some of the soft sciences as well, however you want to draw the distinction) produce and confirm and being skeptical of a priori philosophical inquiry, rather than a metaphysical doctrine which is in conflict with idealism, supernaturalism and so forth about the basic ontology of the world. So I don't think the question of whether physicists think that some of the properties of quantum objects are mind-dependent (I happily admit to knowing nothing about physics and have no idea whether this is the case or not) is relevant to whether physics or history is 'more naturalistic' in the sense being discussed.
-
Ah, I see that David Wallace has made the same point at the same time.
-
A vote for Williamson
-
Not to clutter up the comment board with a tangent, but my thoughts about naturalism are influenced a lot by its use in philosophy of mind, and how some naturalists in that sub-discipline respond to the argument that because many and perhaps most previous scientific results, accepted by the scientific community at large, have been overturned by subsequent scientific results, we should not give current scientific theories any privileged status in setting our basic ontology. While it seems to me there are all sorts of interesting ways of responding to that argument, some people working in the philosophy of mind fall back on a definition of naturalism that doesn't claim that our ontology should be determined by current scientific results, but rather that our ontology shouldn't include certain kinds of entities, absent some reduction of those entities to more homely objects, events and relations. So then naturalism turns into an injunction not to invoke unreduced, or at least non-supervening mental, or unreduced or non-supervening theological, entities (Papineau and Loewer both seem to do this). Thorough-going Idealism seems to be a view which denies that the mental supervenes on anything non-mental, so it would be ruled out. I certainly think such a definition ought to be defended by appeal to past successes of physics in, for example, explaining all phenomena by appeal to some fundamental set of forces (to rip off Papineau) or neuroscience in explaining mental behavior by appeal to perfectly ordinary physical and chemical laws and phenomena. An apriori defense wouldn't strike me as the least bit compelling.
As for whether there is a substantial disagreement here, I doubt it. I am happy to admit that this might not be the conception of naturalism at play in the debate between Williamson and Rosenberg, and if it isn't then my comment about history being more obviously naturalistic than physics was out of place. But at the very least history seems to be pretty naturalistic, and naturalistic enough that we shouldn't lump it in with deconstructionism.
-
Williamson: 'Naturalism tries to condense the scientific spirit into a philosophical theory. But no theory can replace that spirit, for any theory can be applied in an unscientific spirit …'
Jenkins: 'An initial point to note about the array of naturalisms is that while ‘naturalism’ often refers to a doctrine, it does not always do so. Sometimes, for example, ‘naturalism’ refers to a stance, an approach, a methodology or an injunction (e.g. an injunction to use a certain kind of methodology). Maddy, for example, characterizes naturalism as ‘an approach’, namely that of ‘working within science to understand, clarify and improve science’ (2000, p. 108). Sometimes naturalism is characterized by an attitude of ‘respect for the findings of science, and for the progress that has been made by science’ (Nolan 2005, p. 10). In other senses of ‘naturalism’, mere failures to do something may be sufficient for being a naturalist. For instance, mere non-belief in certain things appears to suffice for a kind of naturalism mentioned by Sider (MS, chapter 13).
Even when we focus on the doctrines, there are very substantial differences …'
(From my 'Naturalistic Challenges to the A Priori', commissioned for this collection: http://philpapers.org/rec/CASTAP-2, draft available at http://carriejenkins.co.uk/WorkInProgress.aspx.)
-
Sorry, my links seem to have picked up some punctuation. Remove it for working links!
Leave a Reply to Aaron Lercher Cancel reply