Only Naturalists Should Worry About Evolutionary Debunking Arguments

Thank you again, Cocoon, for hosting me as your first featured author. I’ve enjoyed—and profited from—the discussion in the comments section of my last post. In this final post, I’d like to cut a new path through some issues I’ve been thinking about lately. Specifically, I’ve been thinking about the connection between the problem of contingency for religious belief and evolutionary debunking arguments.

These two types of arguments have a lot in common. They both start with observations about the contingency of our beliefs on a certain topics (religion or morality), observations about the method by which these beliefs were formed, about how the content of these beliefs was the result of historical accident. And then, by some steps that are under-described in the literature, we’re meant to get to a skeptical conclusion.

Another commonality: evolutionary debunking arguments seem to rely variously on a sensitivity principle, or a safety principle, or a non-accidentality principle. Here’s one example of each variety.

Justin Clarke-Doane (2012):

The claim that we were selected to have true moral beliefs has counterfactual force. It implies that had the moral truths been very different, our moral beliefs would have been correspondingly different—that it would have benefited our ancestors to have correspondingly different moral beliefs. Accordingly, the key implication of the claim that we were not selected to have true moral beliefs is the negation of this counterfactual.

For Clarke-Doane, the skeptical worry is that facts of evolution show that this counterfactual is false: had the moral truths been very different, our moral beliefs would have been correspondingly different. And that counterfactual is so crucial for knowledge—the skeptic says—that once we realize this, we can no longer sensibly claim to have any knowledge of mind-independent moral facts. Hence the epistemic challenge to moral realism from evolution: our moral faculties—the processes by which we arrive at our moral beliefs—lack an important virtue. And that virtue seems to be sensitivity.

And here are E.O. Wilson and Michael Ruse (1985) arguing for moral skepticism:

Suppose that, instead of evolving from savannah-dwelling primates, we had evolved in a very different way. If, like the termites, we needed to dwell in darkness, eat each other’s feces and cannibalise the dead, our epigenetic rules would be very different from what they are now. Our minds would be strongly prone to extol such acts as beautiful and moral. And we would find it morally disgusting to live in the open air, dispose of body waste and bury the dead. Termite ayatollahs would surely declare such things to be against the will of God… Ethics does not have the objective foundation our biology leads us to think it has.

The idea seems to be that, had things gone differently in our evolutionary history—and they easily might have—the method by which we arrived at our current moral beliefs would have produced radically different moral beliefs. And so, imagining that we “rewind the tape,” so to speak, and rely on the same method of moral belief formation that we actually used, we’ll see that we easily might have arrived at moral beliefs that, by our own lights, would have been false. And that, one might think, shows that our moral beliefs were not formed safely, and so do not count as genuine knowledge.

Finally, here’s Sharon Street (2006) putting the evolutionary challenge to moral realism in terms of luck and coincidence:

[T]he realist must hold that an astonishing coincidence took place—claiming that as a matter of sheer luck, evolutionary pressures affected our evaluative attitudes in such a way that they just happened to land on or near the true normative views among all the conceptually possible ones.

And Matthew Bedke (2009) has it that “cosmic coincidence” is a defeater for intuitive non-naturalism; like Street, he too seems to gesture toward this non-accidentality, anti-luck condition on knowledge. We have good textual reason, then, to suspect that accidentality is the “active ingredient” in evolutionary debunking arguments.

So these authors seem to think that a successful evolutionary debunking argument will have roughly this form:

1. [Insert facts of evolution here]

2. Therefore, my moral beliefs were not formed sensitively, or safely, or non-accidentally.

3. Therefore, my moral beliefs don’t count as genuine knowledge.

Of course, this argument’s success will depend on what we insert into that first premise! Exactly what are those facts of evolution that entail that our moral faculties are insensitive, unsafe, or accidentally true if true at all? That's been left unclear in the literature.

But we can still make progress in evaluating this argument. For reasons we considered vis-à-vis the problem of contingency for religious belief, we can say three things for sure. First, as I argued previously, that inference from 2 to 3 is dubious, so the moral skeptic will have to fashion another version of the evolutionary debunking argument. Second, If the facts of evolution that are meant to carry us to premise 2 are merely facts about the contingency of our evolutionary history, facts about how easily our moral sense might have delivered beliefs that would be false by our own lights had we evolved differently, etc., then the first inference will be vulnerable to the objections we discussed last time.

Third, it looks like the problem of self-defeat will be less pressing for the moral skeptic than it was for the religious skeptic. Everyone—except Phil H in the comments! ;-)—has beliefs on religious topics and therefore will have to answer to the problem of contingency for religious belief. But genuine moral skeptics might plausibly withhold judgment on any moral topic, and they might claim that their skepticism resulted not from the use of their moral faculties but by the use of their rational faculties, faculties that aren’t undermined by the facts of evolution.

And moral realists can’t avail themselves of that exemption. Unless of course they think their moral faculties just are (among) their rational faculties, or that our moral sense is analogous to our rational faculties in crucial respects.

And some philosophers do think that. We might call them “Rationalists,” and they think our moral judgments do not rely crucially on any mental intermediate, any sentiment, seeming, “gut feeling,” “affect-laden intuition,” etc. Rather, we just see the truth of certain moral propositions in an immediate, non-inferential way, as we just see that e.g. Gettier’s Smith had JTB but no knowledge. Philosophers we might call “Sentimentalists” think otherwise: our moral judgments do rely crucially on some mental intermediary.

I can see—if a bit dimly—how Sentimentalists might be threatened by evolutionary debunking arguments. For them, moral judgment relies crucially on a report or representation from our evolved moral sense. And this evolved moral sense was not aimed at the truth or anything necessarily connected to the truth. We can’t make moral judgments without this moral sense, and yet we have no reason to trust it, given those facts of evolution. We’ve got no dispute-independent reason to favor the deliverances of our moral faculties over conflicting deliverances from others’ moral faculties (or the conflicting deliverances our moral faculties easily might have given had we evolved differently). Sentimentalists, it seems, are in trouble from an evolutionary debunking argument that doesn’t rely on controversial claims about sensitivity, safety, or non-accidentality, but only rather more plausible conciliatory claims from the epistemology of disagreement.

But Rationalists are not threatened—at least not on pain of self-defeat for the skeptic. The skeptic’s rational faculties are not so tainted by the facts of evolution that he cannot run through and accept this evolutionary debunking argument. (Otherwise, self-defeat.) So, if Rationalism is right about moral psychology, why can’t we say the same about our moral faculties? That seems to be an acceptable way out of the evolutionary debunking argument. Either the skeptic self-defeats, or Rationalists are off the hook.

And there’s a third option for moral psychology. Maybe our moral judgments do rely on some intermediary (some report, some representation), as the Sentimentalists say. Yet maybe the source of this intermediary was not the product of unguided evolution, showing no concern for truth. Maybe, as “Divine Revelationists” hold, the source of the reports on which we base our moral judgments is some supernaturally-endowed conscience, divine testimony, or the like. In that case, it’s hard to see how the facts of evolution could undermine those moral beliefs.

So that’s what I’m thinking these days: Sentimentalists should worry about evolutionary debunking arguments, but Rationalists and Divine Revelationists need not worry.

One more upshot: For obvious reasons, Divine Revelation is incompatible with philosophical naturalism. For less obvious reasons, naturalism doesn’t play nice with Rationalism. (Accepting Rationalism about moral psychology would be, for many, to just give up on the anti-a-priori, demystifying project of naturalizing moral psychology.) So naturalists are pushed in the direction of Sentimentalism. And non-naturalists are not; non-naturalists can embrace Rationalism or Divine Revelationism with smiles, open arms, and back-pats.

But recall that Sentimentalists should worry about evolutionary debunking arguments. So I’m inclined to conclude that only naturalists should worry about evolutionary debunking arguments.

This is very much a work in progress for me, so I'd greatly appreciate any feedback or comments that come to your minds. Thanks in advance! :-)

Comments

[Nick Smyth]

Hi again Tomas,

Thanks for this post. My response to it will be much shorter than my previous comment: as you say, surely the rationalist can only stop worrying about evolutionary debunking if rationalism is a *true* or *justified* picture of moral psychology, right?

Put another way, it is compatible with my experience of "just seeing" moral truths that I do not "just see" them, that they are deeply conditioned by affect. Hume's "calm passions" are still passions. Furthermore, I know of no rationalist who has produced solid empirical evidence for the hypothesis that our moral judgments are not influenced by affect. (Singer and Greene claim to show that consequentialist judgments are produced by reasoning, but they do not address the question of where the value-judgments themselves come from... weighing outcomes 'rationally' still requires substantive value-judgments, for example concerning the badness of suffering.) If my understanding of the literature is correct, there is strong evidence is that affect is crucially involved at many stages of moral judgment.

So, no philosopher or psychologist has actually shown, empirically, that this strong form of rationalism about moral judgment is true. This is the basic problem with this response: merely staking out a dialectical possibility is not in and of itself a counter-argument. Surely, that can be done in response to any philosophical argument!

[Brad Cokelet]

I find the claim that 3 does not follow from 2 cold comfort. If my moral beliefs were not formed sensitively, or safely, or non-accidentally, that is bad news. For example when it comes time for people to justify punishing those they take to be morally bad, they will need to be able to justify this belief and I worry this will be hard to do if their beliefs are not formed sensitively, or safely, or non-accidentally.

My basic worry is that we quite often want our moral beliefs to have forms of epistemic standing that outpace lucky, insensitive, unsafe, knowledge.

[Dustin Locke]

Thanks for the interesting post! I have a comment about the following.

"Rather, [rationalists believe that] we just see the truth of certain moral propositions in an immediate, non-inferential way, as we just see that e.g. Gettier’s Smith had JTB but no knowledge. Philosophers we might call “Sentimentalists” think otherwise: our moral judgments do rely crucially on some mental intermediary."

I can see two ways of understanding the rationalist view as you've described it. (A) The view is that we form our moral beliefs as a direct response to *perception* of the moral truth. (B) The view is that we form our moral beliefs simply because we are directly disposed towards having such beliefs (and not because we are caused to have them by some other psychological state--e.g., emotion).

The trouble is that (A) strikes me as question-begging against (1) of the evolutionary debunking argument and (B) strikes me as offering no help against the evolutionary debunking argument.

But I confess that I've never really understood rationalism or the rationalist response to the evolutionary debunking argument. Maybe you can help me? (Full disclosure: I do have a dog in this fight. http://www.cmc.edu/pages/faculty/dlocke/docs/DarwinianNormativeSkepticism.pdf)

[Ambrose]

I wonder about the distinction between "sentimentalism" and "rationalism". Why couldn't it be that (a) we experience/know/perceive basic moral truths by a gut feeling/affect-laden intuition, etc. and (b) it is exactly those gut feelings (or whatever they are) that constitute "just seeing" (or rationally intuiting) those same facts. In other words, why can't feelings be identical with reasons, emotional faculties with rational ones?

[Nick Smyth]

Tangentially, Dustin: yours is one of my favorite papers on this topic. The section on naturalism and "granting the defeater" neatly articulated ideas I had been trying to get into a seminar paper (20 woeful pages trying to say what you said in a couple of paragraphs).

[Tomas Bogardus]

Hi Nick,

Thanks for the comments! They’ll certainly help as I revise this paper I’m working on.

You said:
>> surely the rationalist can only stop worrying about evolutionary debunking if rationalism is a *true* or *justified* picture of moral psychology, right?>>

I suppose that’s right.

>>I know of no rationalist who has produced solid empirical evidence for the hypothesis that our moral judgments are not influenced by affect…. If my understanding of the literature is correct, there is strong evidence is that affect is crucially involved at many stages of moral judgment.>>

I guess we’d have to get into the nitty gritty on those studies you’re thinking of. I’m skeptical that we currently have instruments with sufficient resolution to show not only that moral reasoners are experiencing sentiments/feelings/etc., but also that their moral judgments are *based* on those sentiments/feelings/etc., that the moral judgments *relied* on those sentiments/feelings/etc. How could we possibly show that, especially given our limited understanding of the brain and the crudeness of e.g. fMRI?

>>So, no philosopher or psychologist has actually shown, empirically, that this strong form of rationalism about moral judgment is true.>>

I’m worried you’re reasoning this way: the rationalist has no (or little) empirical evidence in favor of his view; therefore, he can’t justifiably believe rationalism. I don’t think that’s a good inference. I agree the rationalist would be much better positioned if he had, in hand, some compelling empirical evidence for his view. But he may be well-enough positioned if he has other kinds of evidence for his view, e.g. philosophical arguments or introspective evidence. That may provide enough justificatory “oomph” for him to reasonably deny certain premises in evolutionary debunking arguments.

[Tomas Bogardus]

Hi Brad,

Thanks for the feedback. I share your concern!

The concern was:
>>My basic worry is that we quite often want our moral beliefs to have forms of epistemic standing that outpace lucky, insensitive, unsafe, knowledge.>>

Yes, that would be very nice if we could have that. But, if evolutionary debunking arguments steal that from us, one possible response is to accept that *knowledge* is good enough to justify e.g. punishment and reward. That's not so crazy. If I *know* you deserve a reward, why shouldn't I give it to you?

And recall the Rationalist and Divine Revelationist can reasonably claim to have the sort of knowledge you’re looking for, even in the face of evolutionary debunking arguments.

So I suppose you have three options: settle for lucky/insensitive/unsafe knowledge when doling out rewards and punishments, or stop doling out rewards and punishments, or become a Rationalist or Divine Revelationist and dole out rewards and punishments on firmer grounds. I’ll take door number 3. :-)

[Tomas Bogardus]

Hi Ambrose,

You asked:
>>I wonder about the distinction between "sentimentalism" and "rationalism". Why couldn't it be that (a) we experience/know/perceive basic moral truths by a gut feeling/affect-laden intuition, etc. and (b) it is exactly those gut feelings (or whatever they are) that constitute "just seeing" (or rationally intuiting) those same facts.>>

I’m not sure I understand the suggestion. The “just seeing” locution is meant to capture an unmediated access to the truth of certain propositions. You ask why we couldn’t “just see” basic moral truths *BY* gut feelings/sentiments/etc. I’m not sure how to read that “by” there in a way that doesn’t end up in a kind of indirect perception, and so not “just seeing.” Saying “I see/hear/know BY…” is, I would have thought, a standard way to talk about indirect seeing/hearing/knowing. E.g. I hear the mail carrier is hear BY hearing the dogs barking.

I’m also not sure what it would be for gut feelings to “constitute” just seeing moral truths. I would have thought just seeing that p is a relation between a subject and a proposition, whereas gut feelings aren’t. Could you elaborate a bit? :-)

[Tomas Bogardus]

Hi Dustin,

That’s an interesting suggestion you’ve got there (and a nice paper; thanks for linking to it).

You asked what I meant by “Rationalism”:
>>(A) The view is that we form our moral beliefs as a direct response to *perception* of the moral truth. (B) The view is that we form our moral beliefs simply because we are directly disposed towards having such beliefs (and not because we are caused to have them by some other psychological state--e.g., emotion).>>

I wouldn’t put it exactly like (A), but I think that’s very close to what I meant. I guess it depends on just what you mean by “perception” there. I meant a kind of direct perception, not something like visual perception that operates via representations.

>>The trouble is that (A) strikes me as question-begging against (1) of the evolutionary debunking argument>>

Hm. Well, Rationalism is just a view, not an argument, so I’m not sure how it could be question-begging. What did you mean exactly? You think it’s sort of dialectically unsatisfying to respond to evolutionary debunking arguments by showing how some of their inferences fail if Rationalism is true? Why is that unsatisfying? I find it pretty satisfying. Maybe I’m just easier to please. ;-)

Please say more!

[Ambrose]

Hi Tomas,
I'm equally puzzled by your way of framing these ideas :)

Of course, you can't literally "see" anything by having a gut feeling. Just as you can't hear it by seeing it. I guess if I were to put my proposal more precisely, it'd be this: for various moral properties, a person can DIRECTLY perceive that things have those properties (or don't have them) IN having the appropriate "gut feeling" or emotion or moral intuition or whatever it is. So i did mean to align the idea with notions of "unmediated access" and such like... Maybe "by" was not a familiar way to express the point.

Now the substantive question is how a feeling (or whatever) could constitute a case of direct perceptual access to something. I admit it's not clear to me how that might work. But that's probably just because it's never clear how direct perceptual access works, though it had better work somehow or other (on pain of extreme skepticism). For example, in the case of seeing -- that is, "directly" seeing, literally seeing with your eyes -- how exactly is it that some entity or fact or whatever is present to your mind IN the act of seeing it? I have no idea. Still I figure that happens all the time.

You suggest that what we're talking about here is a relation of a subject to a propositional content, e.g., 'John just sees that murder is wrong'. In that case I think there is a problem of mediation. The wrongness of murder is not a proposition, if it's anything, so what John 'just sees' in that case will not be the thing it's supposed to be. The wrongness will be known or perceived BY his awareness of some proposition (or its truth). That's not what the realist wants, I'd think.

So instead I'd suggest that feelings and the like could be more like (literal) seeing: there are properties in the world, such as shape, that you directly perceive by seeing them, and other properties, like wrongness, that you directly perceive by feeling them (e.g., feeling what-it's-like to do wrong or be the victim of it). Then propositional knowledge, say the knowledge that murder is wrong, is built on that kind of acquaintance with the worldly facts or properties the relevant propositions are about.

Don't know if that helps...?

[Dustin Locke]

Nick and Tomas,

Thanks for the kind words about the paper. Nick, it only took me about two years to figure out what I was trying to say in those two paragraphs. :)

Tomas,

I guessed that you meant something like (A), and yes, I didn't mean for the "perception" to be taken to literally. I suppose the basic idea is that according to (A), our moral beliefs are formed as a result of more or less immediate causal contact with the moral truth. It seems to me that to assert this is just to deny the (alleged) evolutionary facts stated in (1). I take it that the key (alleged) evolutionary fact is that our moral beliefs are not formed by more or less immediate causal contact with the moral truth, but are instead the product of a causal chain that traces back through our evolutionary history. So if your point is that a proponent of (A) will deny (1), then yes, that seems right to me. But then the question is whether they have good reasons for doing so, or whether, instead, the empirical facts make denying (1) implausible. (I suppose that this was roughly Nick's point in the first comment above.)

[Phil H]

I definitely do have moral beliefs, so the previous arugument doesn't apply here! I actually agree with all of this. I'm not sure if this is helpful, but I wrote t I thought was a fun argument about it (for Sam Harris' competition, no less!)

"In fact, I worry that our intuitions could easily lead us astray on questions like this. I’ll pose a thought experiment to illustrate. Humans are social animals – overdeveloped chimps. As our technology improved, our instincts drove us to increase our population density. But what if consciousness had emerged in a being that was not social, but territorial? It could be a shark-type predator, or a plant that grows until it crowds out competition. Imagine that this creature has a strong preference for solitude. As develops more technology, it uses its extended technological reach to compete better, driving population density down.

What does flourishing mean for this creature? It seems to be driving itself into a moral (and evolutionary) dead end: without cooperation, could it develop science? Technology? Cultural achievements? Its “good” or “flourishing” seem morally impoverished. But for the creature, a cooperative way of life might seem like the worst possible situation. I have no answer; I am utterly unequipped to make moral judgments about this kind of being. But then, are my instincts about what constitutes good and bad simply expressions of my species’ biological and evolutionary imperatives? If my moral instincts about the absolute good or absolute bad are inapplicable “sharks”, then perhaps my instincts are also wrong about my own species."

[Nick Smyth]

Hi Tomas,

"I’m worried you’re reasoning this way: the rationalist has no (or little) empirical evidence in favor of his view; therefore, he can’t justifiably believe rationalism. I don’t think that’s a good inference."

I agree. But let's distinguish a rationalist theory of justification and a rationalist theory of moral psychology. Empirical evidence may not bear on the truth of the former, but it certainly bears on the truth of the latter. And, so far as I can tell, rationalists need the latter thesis in order to stop worrying.

[Tomas Bogardus]

Hi Dustin,

Thanks for elaborating. I'm getting clearer on what you had in mind.

>>I take it that the key (alleged) evolutionary fact is that our moral beliefs are not formed by more or less immediate causal contact with the moral truth, but are instead the product of a causal chain that traces back through our evolutionary history.>>

I'm not yet seeing why those are incompatible. I thought the facts of evolution would include merely that our moral *faculties* are "the product of a causal chain that traces back through our evolutionary history," as you say, i.e. the result of evolution via natural selection. (Maybe "my moral faculty"--singular--is misleading and instead I rely on a *bundle* of faculties or abilities to feel and judge.) I thought the Rationalist could accept all that and hold that one of the faculties/abilities evolution (or God) gave us is this faculty of direct perception. It's sort of hard to see how that could have happened, on naturalism, but I suppose that's part of why naturalists don't like Rationalism. And I don't think any of the science precludes the Rationalist view. Has science disproved that? If so, how?

>>So if your point is that a proponent of (A) will deny (1), then yes, that seems right to me.>>

I should have been clearer about this in the post: I imagined the Rationalist denying the inference from 1 to 2, i.e. accepting the facts of evolution but denying that those facts entail our moral beliefs were formed by faculties that are unsafe/insensitive/accidental. Maybe some of our moral beliefs were formed that way, if we relied on sentiment; those beliefs may well be undercut by these evolutionary debunking arguments. But some of our beliefs were NOT formed that way, according to the Rationalist. And those beliefs won't be undercut by evolutionary debunking arguments (at least, not any I've seen).

[Tomas Bogardus]

Hi Nick,

>>But let's distinguish a rationalist theory of justification and a rationalist theory of moral psychology. Empirical evidence may not bear on the truth of the former, but it certainly bears on the truth of the latter. And, so far as I can tell, rationalists need the latter thesis in order to stop worrying.>>

Sure, I agree with all that. Empirical evidence could bear on the Rationalist view of moral psychology. I'm just questioning whether any empirical evidence actually casts doubt on the Rationalist view of moral psychology. Showing that sentiments are floating around when subjects make moral judgments is insufficient, and even showing that *sometimes/often* subjects *base* their moral judgments on these sentiments is insufficient, since Rationalists claim only that at least *sometimes* our moral judgments are not *based* on those sentiments.

And I'm questioning whether, if there were no *empirical* evidence in favor of Rationalist moral psychology one could not be justified in believing it. I'm suggesting that there are other kinds of evidence that could justify Rationalist moral psychology--e.g. philosophical arguments and introspective evidence.

[Dustin Locke]

Tomas,

I see. Well I think that philosophers like Street are pretty clear that they don't merely think that evolution has given us a moral faculty (the use if which brings us into contact with moral truth), but rather that evolution has given us a moral faculty that tends to produce certain kinds of moral beliefs (e.g., the belief that the fact that doing something would kill your children is a reason not to do it). In other words, evolution has given us a "belief loaded" moral faculty, not a "truth detecting" moral faculty. This is what I take to be the key (alleged) evolutionary fact in (1). So I think the rationalist you have in mind is really just denying (1). And again, the question becomes whether she has good grounds for doing so.

[Tomas Bogardus]

Hi Dustin,

Thanks for the additional feedback. This is all certainly helping!

>>I think that philosophers like Street are pretty clear that they don't merely think that evolution has given us a moral faculty (the use if which brings us into contact with moral truth), but rather that evolution has given us a moral faculty that tends to produce certain kinds of moral beliefs (e.g., the belief that the fact that doing something would kill your children is a reason not to do it).>>

Just curious: do you know where Street says something like this?

>>In other words, evolution has given us a "belief loaded" moral faculty, not a "truth detecting" moral faculty.>>

I still don't get why evolution couldn't have given us both. You say Street thinks evolution gave us a moral faculty that *tends* to produce certain kinds of moral beliefs--it's "belief-loaded," you say. That's an alleged fact of evolution, you say. But couldn't a "truth-detecting" moral faculty tend to produce certain kinds of moral beliefs like that? I don't see how that's incompatible with this alleged truth of evolution. :-/

[Dustin Locke]

Thanks for taking so long to reply, Tomas. I've found this discussion very helpful as well!

I'm thinking of Street (2008), although I don't think you will find that contrast drawn as explicitly as I have drawn it here.

I'll also just add that I don't think Street or anyone else thinks that this view follows from the *mere* fact that we are evolved creatures. It's not like: "We evolved, therefore our moral faculty is content-loaded." Rather, I think that Street and others think that this view is just the most plausible given all the more particular facts about the kinds of creatures we are and the environment(s) in which we evolved.

[Dustin Locke]

Oop! That should be "SORRY for taking so long to reply, Tomas"! Haha.