In the comment thread of our previous post, we’ve been looking at Sam Harris’s claim that there can be a prescriptive natural science of human morality, one that uncovers objective normative truths. This would rebut, it seems, the idea that there are no “oughts” in nature.
People do want there to be absolute moral truths, and many people feel that there can’t be any such truths in the absence of God. Without an absolute authority to appeal to, moral disagreements cannot be resolved, the center will not hold, and society will disintegrate in relativism and nihilism. Plenty of folks will tell you that this is already happening.
But grounding moral truth in divine command — the whim of God — has its problems too. What if God commanded us tomorrow to fry our children in hot oil? We would mutiny in moral revulsion. It seems, then that there is some deeper moral truth to which we expect God to adhere, and with which we would still comply even if He didn’t. So why not cut out the middleman?
What, though, can moral “truth” be under a Godless, naturalistic worldview? Answering this question has has been one of the most pressing challenges for our current crop of atheist intellectuals, many of whom are scientists and philosophers working in the relevant fields of biology, evolutionary psychology and cognitive neuroscience. It’s pressing because before they can wean our civilization away from religion, as they would like to do, they need to come up with a new story about the foundations of morality. People still want their moral truth.
There are at least a couple of ways it can go. One way is to say that yes, there are moral facts, and they are simply “there”, existing alongside other Platonic abstracta like mathematical truths. It is a mathematical fact that any angle inscribed in a semicircle is a right angle; in the same way, it is simply a moral fact that torturing children is evil.
The problem with this (even if we accept the actual existence of Platonic abstracta, which I’m reluctant to do) is that we have objective ways of verifying mathematical facts: if you deny that any angle inscribed in a semicircle is a right angle, I can offer a simple proof. (I can also suggest that you just go out and start inscribing; you’ll soon see that you always get a right angle.)
But we have nothing like that for moral truths. If you and I disagree, say, about whether bestiality is morally wrong, it seems there’s no authority to which we can appeal the matter, other than social convention, religious dogma, and the subjective moral intuitions of our own consciences.
The angle that folks like Sam Harris and Steven Pinker are working these days is to claim that there actually is a naturalistic foundation upon which moral truths can rest. It is nothing more or less than the “fact of the matter” about which codes of behavior best enable intelligent social primates like us to live together successfully in thriving and happy groups. In other words, there are accessible natural facts about which moral systems maximize our fitness.
But wait — is that it? We want bedrock moral truth here, not just what happens to optimize our Darwinian outcomes.
Well, how deep do we really need to drill? Let’s say that reciprocal altruism is such a powerful fitness-maximizer that any intelligent, social life-form will either hit on it and thrive, or miss it and dwindle into misery and extinction. In other words, it is a basic fact of the Universe that a moral system based on reciprocal altruism is the optimal, most fruitful strategy for generating successful and well-functioning social groups.
“That’s all well and good,” the absolutist might say, “but all this is awfully contingent. What if it maximized our fitness to pick a child at random once a week, roast him over a slow fire, and eat him? Then we’d see that as an obvious moral ‘truth’.”
“But you see,” comes the reply, “it seems that the world is so constituted that doing what you suggest doesn’t maximize our fitness. Instead, it turns out that there are other things that clearly do — things like the Golden Rule — and with careful study we can learn what they are, with enough certainty that we can say with confidence that any society that does roast and eat its children is making a factual moral error.”
“Well”, says the absolutist, “that may be, but even so you still haven’t got me an ‘ought’ from an ‘is’. Why should some brute fact about the superior Darwinian fitness of altruistic groups mean that we ‘ought’ to be altruistic? Even if we do develop a metrical science of human well-being, and learn very clearly what types of behavior increase and diminish it, all we’ve done is establish some facts about social animals. But there is still nothing in Nature that makes it a moral fact that we ought to encourage behavior that will maximize human well-being; that is still just a subjective valuation on our part.”
To which our Sam Harris might answer:
“So what? If you can know with certainty that one direction leads to a maximally miserable world, and the other to a maximally happy one, what deeper moral ‘truth’ do you need? Your insistence on supernatural bedrock is nothing more than an intellectual fetish, of no practical value. Why should anyone care about that?”
So there it is, if I understand it all correctly: there are natural moral facts, because there are optimal strategies for maximizing well-being — strategies that are optimal for any intelligent social creature. And this is different from simply positing moral facts as Platonic objects, because we can actually use the tools of natural science to discover what these optimal moral systems are.
Are you convinced? I’m not — but I will say it is certainly an interesting argument, and not without its merits. And again, to be fair, I haven’t even read Sam Harris’s book yet, so I should be careful about putting words in his mouth. I do think I get the gist of it, though; these ideas have been in the air for a while now (I even had a chance to discuss them briefly with Steven Pinker up in Wellfleet a couple of summers ago, after a talk he gave for our local library, and what he said then was pretty much what Sam Harris appears to be saying here).
There are a lot more questions to ask, and soft spots to poke at, but it’s late now; they can wait for another post.
24 Comments
Harris’s hypothetical reply, above, isn’t far from philosophical Buddhist axiology, which would take phenomena as a given, then speak in terms of upaya (expedient or skillful means) when talking about human conduct. Some forms of conduct are more skillful in certain situations if they (1) reflect an awareness of the situation and (2) respond to it accordingly, in such a way as to minimize suffering. Unskillful conduct results from inattentiveness and/or attachment, and generally creates greater suffering. In all cases, skillfulness and unskillfulness have to do with one’s mindful responsiveness to a given situation.
Example: without even worrying about “oughts,” we can learn very quickly that repeatedly pressing our cheek against a hot iron isn’t conducive to the reduction of suffering. Because of how we’re wired, our need to minimize suffering can be taken as a simple given; the why of the matter is little more than an amusing distraction (cf. the Buddhist story about the man shot with the poison arrow who spends his time asking questions — Who shot me? What sort of poison was used? — that are irrelevant to his need to extract the arrow if he doesn’t want the poison to kill him).
In philosophical Buddhism, then, right and wrong are less about finding absolute cosmic principles than about skillful conduct in any given situation. “Follow your situation,” the Korean Zen Buddhists say. Everything unfolds from the central premise that sentient beings suffer, but prefer not to suffer. (Either Wright or Pinker alludes to the idea that we humans are wired in such a way that we’re never satisfied with our lot. That dissatisfaction is dukkha, suffering.)
NB: I say “philosophical Buddhism” because living Buddhism, taken as a whole, offers a more complex, even contradictory, outlook than I’ve hinted at here. Traditional Buddhism, for example, contains concrete precepts, and even though many Buddhists understand those precepts to be subordinate to the upaya dynamic, a great deal of ceremony has accreted around the notion of “taking the precepts” as a way of demonstrating that one belongs to the Buddhist community. Buddhism also makes reference to what seem to be eternal cosmic principles, like emptiness, impermanence, and no-self.
Malcolm’s dissatisfaction with the dialectic he sketches above is well-placed, because Harris is looking for “natural oughts” in all the wrong places. At least that’ my take on things.
As I’ve said in previous discussions of this topic, I don’t think that morality is likely to have evolved via natural selection. Instead, I think that natural selection provided us with a bunch of capacities, one of which is the ability to employ abstract reasoning which, quite incidentally from the perspective of natural selection, turned us into moral creatures. As I understand morality (not very well…), it involves acting according to rules (principles, maxims) that would apply to any reasoning creature in the relevant circumstances instead of rules that reflect individual preferences. Why? Because reasons are impersonal things; i.e., reason doesn’t privilege my, or your, or his or her interests. That, to me, is the beginning of ethical wisdom. That’s also why I think the ought of morality is either identical to or a close kin of the ought of reason. And notice, the ought of reason strikes those who understand the reasoning as compelling. Not a bad place to start in trying to understand the nature of morality.
I’ll be very busy all day, with few opportunitites to join in, but briefly:
Very interesting points, Kevin. I know that Sam Harris takes Buddhism very seriously, and it may well be what got him looking at the problem from this angle.
Bob, we do disagree here about two things.
First, I think that our moral faculties are the result of evolution by natural selection, even if only because selection pushed us onto a “fitness peak” representing the advantages for social creatures that morality offers.
Second, I question the idea of “oughts” of reason; I’d say, rather, that reason merely offers an analytical method, and requires normative “oughts” as external input if we are using it to decide what actions to take. Reason gives us “if X, then Y”, but can’t by itself give us “we ought to do Y”.
Malcolm – Briefly…
I’m not sure what you mean to include in the category of ‘moral faculties,’ uless you mean the various social instincts. I don’t view these as particularly “moral” since they tend to be present as well in lots of species that we are not inclined to view in terms of moral agency. Add the capacity for reason to the mix, and you’ve got creatures with the capacity for moral agency. But I doubt that it was the capacity for moral agency that drove the evolution of the capacity for reason. I view the former as most likely an accidental by-product of the development of the latter.
As for the “ought of reason,” I view this in terms of the compulsion one experiences to accept the conclusions of valid inferences. I certainly didn’t choose to accept that 1+1=2 — having understood ‘1’ and ‘+’ and ‘=’ and ‘2,’ the idea that 1+1=2 forced itself on me. That’s a “natural ought” if anything is.
FWIW, I should add that I see some connections between my views about the role of reason in making us moral agents and the Buddhist ideas about “skillful conduct” that Kevin outlined. In particular I would interpret ‘attentiveness and/or nonattachment’ in terms of sensitivity to the interests of all, not privileging one’s own interests.
By our “moral faculties” I mean our built-in apparatus for developing emotional affinities and aversions to various social behaviors (i.e. conscience). I think we have a general-purpose “module” for this, much like we do for the acquisition of language. And like language, which varies widely but within the bounds of a more universal human grammar, the range of moral codes within which our moral faculty operates is both flexible and broadly constrained. I see no reason to think that any of this is not the product of evolution and natural selection.
I understand the comparison you are making between morality and reason, but I don’t find it as apt as you do. Yes, reason commands our assent, but reason is self-contained; it’s a rule-bound process for manipulating propositions. All that it commands is our belief that its output is valid, given the input; but it says nothing about what we ought to do with its results. For that we need valuation, which is independent of reason. The way I see it, reason alone is not, and cannot be, normative.
Morality, by contrast, is all about behavior, about what we ought to do; it’s nothing BUT normative. Given valuations as inputs (e.g. it would be good to ease X’s suffering) we can apply reason to moral decisions (X is obviously tired, and suffering from it; my getting up and giving her my seat will help me achieve my goal of easing her suffering, therefore I ought to do so). But reason only tells us that IF we give X our seat THEN her suffering will be eased; it says nothing about whether we ought to give up our seat. We still require the prior moral intuition, the normative valuation, that tells us easing X’s suffering would be a good thing that we ought to do.
Perhaps we agree more closely than it seems; you and I often can’t quite get our terminology synced up. So I wanted to be as clear as I can about how I see this.
Malcolm –
I think our disagreements are real.
Dogs, even cats(!), show emotional affinities and aversions to various social behaviors, but few of us would treat them as moral agents. While I don’t doubt that the capacity to develop such affinities and aversions is the product of evoultion by natural selection, I don’t think this qualifies as a moral conscience.
I don’t think you understand the comparison I make between morality and reason. You say that reason doesn’t tell us what to do with its results AND that it commands our assent. Commanding our assesnt IS telling us what to do with the results, namely assent to them. And the point of having a notion such as ‘assent’ is that it tells us something about overt behaviors; i.e., if you assent to a proposition, you will then behave as if it were true.
Also, I don’t see reason as telling us simply “what follows” from valuations given as inputs. As I’ve indicated above, I see reason as the source of whatever universalism we find in morality.
So, our disagreements are real.
Bob, I agree that reason and morality are similar in the way they can command our assent; we just “see” that 1 + 1 is 2, in the same way that we just “see” that it is morally wrong to torture innocent children. It is rarely if ever a matter of voluntary choice; the “output” of both reason and moral intuition have the feeling of facts. (I think this is why it is so hard for people to imagine that moral “facts” might be anything but objective truths, and insist on grounding them in God or some other ultimate.)
But, at the risk of repeating myself, I think that’s as far as reason takes us: given valid premises and reasoning, we reach a trustable conclusion. Reason is just a process of transforming propositions.
There is I think, an enthymeme lurking in your statement “you will then behave as if it were true”. For example, reason might tell me that it is raining outside:
1) The weatherman predicted rain today;
2) I hear a tapping on the roof;
3) The tapping sound I hear is just the sort of sound rain makes.
Knowing all this, it’s a fair conclusion that it’s raining; indeed I will be compelled to believe it in just the way we’ve been talking about (assuming I have further reasons to believe my house is not being sprayed with a hose, etc.). Will I “then behave as if it were true”, and take an umbrella when I go out? Likely I will, but not as a matter of pure reason. Reason commands only that I assent to the conclusion that it’s raining — but it’s only if I also place a higher normative value on keeping dry than getting soaked that I will feel that I ought to take an umbrella. You’re sweeping that under the rug, I think.
In itself, there is only one compelling aspect of reason: that we make doxastic acceptance of its conclusions. (To call that an “ought” is, I think, already a bit of a stretch; it’s not that we “ought” to accept reason’s output, it’s simply that we do, willy-nilly — and if we do think we “ought” to accept the results that reason hands us, that is itself a normative position, based not on reason, but on our wish not to seem mad, etc.)
Morality, it seems to me, goes further: not only do we simply “see” moral truths, in the same way we “see” the conclusions of reason, but moral insights and conclusions are also intrinsically normative — whereas reason is only extrinsically so, and needs to import some normativity at its input to have any in its output.
Excellent points Malcolm, et al.
For my part, I worry less about foundations, than about radical contingency of moral truths. It seems to me that when we verbally express what we take as being a moral truth, we either are making an assertion about the moral valence of some particular concrete action in a given concrete (actual) situation, or, more commonly, we are making an assertion about the moral valence of some type of action in a type of situation. But in classifying actual situations and actions in this way, we must invariably fail to have access to an abundance of quite possibly morally relevant information in any actual situation of that type. Simply saying, Thou shall not kill, doesn’t cut it; it is too crude a moral rule.
If what is in fact most relevant to our moral reasoning is information falling below the threshold that our faculties avail us, then whatever moral truths there may be may simply be inaccessible to us; on the other hand, with the growth in our knowledge and faculties, it seems that our moral responsibility grows: with knowledge comes sin, perhaps, for which we were thrown out of the garden.
Malcolm –
I’m still hung up on the role of reason in morality, so some questions…
Do you think that possessing social instincts is sufficient for moral agency? If not, what else is necessary, and why?
Focusing just on deductive reasoning, if reason is just a process of transforming propositions, does it matter whether the transformations preserve truth? If it does matter, isn’t reason inherently normative?
Bob,
Penetrating questions. I’m swamped at work again, so can only offer very brief placeholders for now, but will return later so we can continue in more detail.
No.
Moral responsibility increases according to the agent’s freedom of choice. That raises the question: is the concept of “freedom” compatible with a naturalistic worldview? I think it is, but it’s a huge topic, and a difficult one.
Yes — with the caveat that because we are only imperfect, “good enough” reasoners, we shouldn’t assume that it always does. See here.
I don’t see how it can be, as I tried to illustrate above.
Malcolm, old chap,
Morality, as both Mr Aristotle and Mr Koepp say, is bound up with our rational nature. It is a part thereof.
Herein lies much of the problem:
“By our “moral faculties” I mean our built-in apparatus for developing emotional affinities and aversions to various social behaviors”
No wonder you find (objective-rational) morality a difficult thing to grasp: you are an emotionalist! Since you take morality in the first place to mean something all to do with feelings and subjective states, so you are never going to find a ground for “morality” in a nature conceived in opposition thereto. The very terms by which you bifurcate your conception of the world forbid the one to be reconciled to the other. But this is a very modern misconception of morality (and of reality in general). We now live in a world that has lost its rational moral traditions, and has sunk into emotionalism. (MacIntyre sums up the situation brilliantly in After Virtue.) The bifurcated conception of the world – a mechanical nature on the one side, and a somehow purposeful and rational mankind on the other – just means that there is a gulf in that conception between nature and mankind, as though they were in fact of two seperate worlds, a conception which is somewhat embarrassing for the so-called naturalist who holds it. One good thing to be said about the eliminative materialists is that they stand true to the creed and do not fudge it.
On another point: if your conception of God is of a being which has whims, then it is certainly a poetic, if not a crude one. A straw God for an unpoetic atheist, one might say. Also, “divine command” is perhaps not what you think it is. In the context of classical theism, the Euthyphro dilemma is a false one, and hasn’t troubled anyone therein for a very long time. See Edward Feser for more details. You might also find his latest post interesting.
But anyway, why bother with Sam Harris, when you neglect Aristotle? He really is worth the effort. The old Greek fella, that is.
Deogolwulf! Good to hear from you. Welcome back!
I must be brief:
You are quite right that I place the normative burden of valuation on the emotions; it’s a fair cop. But valuation is what the emotions are, no? We love this, hate that, fear this, desire that: they are valuations all — and without affinities and aversions, dreads and desires, loathings and longings, whence normativity? (I know already the answer, originating with the Greek fella, that I expect you will give; but whether the view I am defending here is one we moderns have “sunk into”, or is instead a way of understanding that we have “risen to”, is itself a valuative dispute for which I am afraid we’ll find no higher court of appeals.)
You misrepresent my views, though, when you suggest that I conceive a bifurcation of the sort you describe. I see no “opposition” between Nature and morality; indeed, given the obvious fact that morality exists (else it would never occur to us to have this conversation!), how could a reasonable man of naturalistic bent imagine such an opposition to exist? Likewise rationality.
If anything, I’d say Harris et al. are trying in an interesting way to address exactly this issue.
As for “whims” of God: I confess that I did hesitate at my keyboard there. But a God with whims is certainly no straw man as far as a great swath of religion is concerned, including for example most forms of Islam (a God who cannot have whims is arguably diminished, although I realize that is also a controversial point; indeed, what about God isn’t?) — but we needn’t quibble about precisely which God it is I don’t believe in; I’m afraid I don’t believe in any of them.
Finally: an “unpoetic” atheist? You cut me, sir, to the quick.
I was tempted several times, to the point of starting to compose a message, to bring Aristotle and Kant into the discussion. But I didn’t want to “stack the deck” in my discusssion with Malcolm. Since Deogolwulf has done half the introductions, though, I’ll climb aboard. I view A and K as the two giants of ethical theory, and note that both insisted that morality is a form of practical rationality.
It did occur to me to suggest to Malcolm that he is taking an emotivist line, which has also been called the “boo/hurrah theory” of ethics. But I didn’t do so because, in previous discussions, Malcolm has eschewed any positive theory and taken a nihilist position. I don’t know what he believes about the nature of morality.
I do hope nobody’s getting the idea here that I am plumping for a simplistic, straw-man view of human morality that is nothing more than a linear emotional checklist of independent items, in which reason plays no part: “Abortion? Boo! Death Penalty? Hurrah! Theft? Boo! Honest toil? Hurrah!” No.
Obviously all of our attitudes and volitional acts are the result of an interplay between reason and emotion, with emotion providing valuative checks as they are needed for the inputs, intermediate results and predictions of reason. As distinct from mere social instincts, the higher-level “oughts” of a rational human being are a complex and continuous dynamic balance, in which reason and emotion each play a necessary, and very different role.
Also, Bob, it has never been the case that I have eschewed any positive theory. Morality is a central feature of human existence, and it wants explaining.
Edward Feser:
This is actually strikingly parallel to the approach that Harris, Pinker, et al. are taking, only without the bits about God. By substituting a few words, they could be made quite nicely congruent.
Malcolm – I should have qualified my statement and I didn’t. You have eschewed a positive theory of substantive or normative morality, but obviously have not eschewed positive theory at the metaethical level. And it is primarily the latter that this conversation addresses, so I misspoke.
Now it appears you may not be an emotivist who thinks the normative force of moral judgments resides entirely in subjective emotional affinities and aversions to various social behaviors. Now it appears reason has a necessary role to play, albieit “with emotion providing valuative checks as they are needed for the inputs, intermediate results and predictions of reason.” I don’t understand the picture of moral consideration that you are sketching here. (I chose ‘consideration’ as a more neutral term than either ‘deliberation’ [too rationalistic to be fair to you] and ‘engagement’ [which sounded a bit pretentious].) What sorts of checks are provided by emotion at the different stages in this process?
Those of us who see the capacity for reason as the ground of morality would probably be inclined to say that it’s reason that provides checks on emotional affinities and aversions. We typically don’t deny the moral significance of emotions, but believe they need to be modulated by reason before they can be actual moral sentiments. And that seems more than plausible if the phenomenology of moral considertion is any sort of guide in these matters.
Well, Bob, just the sort of thing that I mentioned in brief in previous comments. If I have, say, an emotional aversion to the suffering of others, then that valuation can serve as input to, and interact with, a process of reason — like this:
That’s rather an exaggeratedly cartoonish and robotic example, of course, but it illustrates the point: without the emotions to weight the choices, there’d be no basis for any decision. We would just have a set of options, and nothing to make us prefer any one of them over any of the others. In negotiating the complexities of the real world it can take an awful lot of reasoning to work through the implications of all the various choices so as to present them for evaluation, so there must be an enormous amount of interplay between reason and emotion. But they have different jobs to to; both are necessary for sophisticated moral agency. I do not deny the role of reason in human morality.
Anybody here read Bernard Lonergan — he of the “transcendental imperatives” (be attentive, be intelligent, be reasonable, be responsible — derived from his four-step paradigm about how the mind works: experience, understand, judge, decide)? He’s a Catholic thinker who did a lot of work in cognition, though it was never obvious to me that he was in touch with neuroscience when developing his own theories about the mind.
Malcolm –
If we are going to stay focused on the metaethical issue about the ground of moral norms, your illustration leaves unanswered the obvious question of whether the aversion reported by Emotion is a moral aversion. That it is proffered in response to a request from reason is at least a good sign.
Kevin –
It’s been a long time since I even heard mention of Lonergan. But, yes, that does ring some bells, especially the appeal to transcendental arguments. I also remember that he knew Kant “inside and out.”
Kevin,
No, I’ve never read Bernard Lonergan, but I’m sure I ought to have.
Bob,
You’re right, it’s important to keep that in view. I’d say that at the very least, such direct emotional responses provide the necessary valuative pillars on which the apparatus that makes our final moral choices must stand.
Many of those moral choices are bafflingly complex, requiring that we weigh a great many conflicting valuations, and that we assess a tangled decision tree with great subtlety. Often the moral “output” — the final “ought” — is far from clear, given the interdependent ramifications of all the possible courses of action whose potential consequences and effects must in turn be evaluated. The ridiculously simple example I gave is obviously just a toy version of all that.
But are the sentiments reported by Emotion themselves “moral” aversions? Well, sort of. They’re the normative raw materials of the finished product; they’re the entirety of the valuative input data that Reason has to work with in making our final moral choices. And Reason often plays a very small role in shaping a particular moral stance — but Emotion‘s valuation is always necessary.