Scientism, Evolution and the Basis for Morality
Cut, jab, thrust, confusion! That seems to be the spirit of an ongoing exchange between Michael Ruse and Jason Rosenhouse of Evolutionblog. It started with scientism, the term (often used in the pejorative) applied to the notion that science is the best/only way of knowing the world. It then shifts to a somewhat complex (but useful) discussion of moral knowledge, moral absolutism and the slippery slope into moral subjectivism.
The discussion is useful precisely because it’s complex and irresolute – and that’s precisely where the debate lies at the heart of naturalistic ethics today. In delving to this depth – a more arcane depth than most public commentators would delve – we can get to some of the most pressing and important questions in ethics.
First, a word on scientism: I do firmly believe that science is the best tool in our kit for understanding the natural world. But it’s a limited tool. As they say, science is a wonderful tool, but a terrible master. Let’s not wander into the fallacy of assuming because it can’t do something then therefore that thing doesn’t exist.
Ultimately, I take a pragmatic stance on knowledge, and on the utility of science. We’re confronted with phenomena, we organise and structure that phenomena and posit theories to explain it and make predictions about future phenomena. Science has been very carefully developed and refined to be exceptionally good at this task – and if you care about explaining and predicting phenomena, then science beats all comers, especially any brand of revelation.
But that’s not all there is to knowledge. As Ruse points out, there are questions about this method itself, or about how the world can be such that science even works. Science can’t answer those. And that shouldn’t worry us a jot. That’s what philosophy is for.
On morality, and the concern expressed by Rosenhouse over the slip towards subjectivism: there are two brands of moral subjectivism, and his conflation of the two is, I believe, the source of his concern.
The first brand of subjectivism is at the level of moral norms. A norm subjectivist will look at a particular moral norm – “murder is wrong” – and take it or leave it according to his or her subjective whim. That’s an unsettling notion. But, thankfully, almost no-one holds this brand of mad-dog subjectivism to be true.
Instead, we want to say that moral norms are binding in some important way precisely because we don’t want people presuming they’re justified in abandoning a norm just because they haven’t yet had their morning coffee and they’re not in the mood. But attempting to secure that binding force is a tremendous challenge, and it’s that endeavour that Ruse and Rosenhouse (and legions of moral philosophers) spend a great deal of words trying to solve.
The problem is no matter how deep you delve, it appears there isn’t a rock solid ‘real’ foundation to morality in absolute terms. It can’t be found in God (for a bunch of reasons, oft discussed); it can’t be found in descriptive or empirical facts alone (for reasons Hume points out); it can’t be found in non-natural facts (because they’re ontologically dubious and have a queer relationship to the real world, as Mackie states); it can’t be found in evolution (because we’ve evolved tendencies that are generally agreed to be good and bad).
So some have taken to abandoning the rock solid foundation. As Ruse states, there are several arguments in this tradition, one being a social contract theory, another being a kind of psychological sentimentalism. The latter says that morality isn’t rooted in objective moral facts but in deeply held sentiments of right and wrong, sentiments Ruse things are effectively universal thanks to evolution.
This approach does effectively constitute a kind of evolutionary ethics, but it’s not as bad as Rosenhouse thinks when he says this:
Ruse just got through telling us that you cannot derive ought from is, but isn’t he doing precisely that in these final paragraphs? It looks to me like he is pointing in some way to the facts of human psychology and to the vagaries of our natural history as the justifications for our moral beliefs.
This is because Ruse is a ‘non-realist’ (or an ‘anti-realist’) and has given up seeking a rock solid ‘real’ foundation to morality. There’s no is-ought problem for anti-realists, because they’re not even seeking to build a solid bridge between is and ought. Folk like Ruse (and Hume) take the sentiments as being the foundation of morality and note, with interest, that these sentiments come from evolution – but they don’t claim that their evolutionary source justifies their moral force. That would be making the is-ought fallacy.
This is where the second brand of subjectivism comes in. Instead of being a norm subjectivist, you can be a moral subjectivist. You can acknowledge that there is no binding, logically necessary or factually obligatory reason to be moral, but you can choose to be moral. And there are plenty of good non-moral or prudential reasons for doing so, such as that social living benefits us, and it’s a darn sight easier to live socially when there are rules of conduct. So you be moral.
It’s a subjective choice – made implicitly by almost everyone – and it’s pretty hard to choose not to be moral because all those other people who have (implicitly) chosen to be moral will probably not like your amoral behaviour and will try to persuade you in the strongest possible terms to be moral.
And once you’ve chosen to be moral, that binds you to playing by the rules of the moral system you’re in. Like when you agree to play a game of cricket, you can’t just go around breaking or conforming to particular rules, or making up new ones, willy nilly. If you did that, in some important sense you wouldn’t be playing cricket. And the other cricketers would certainly look upon you with great scorn and disapprobation.
So once that core subjectivism is out of the way, then moral discourse can kick on arguing whether murder really is right or wrong, with the implicit addendum “in this moral system to which we’ve agreed to adhere”.
OK, sure. Say you’re with my until now. The question becomes: how do you argue about whether murder really is right or wrong if there is no really is to the matter? What criteria do we apply to judge its rightness or wrongness?
This is where there social contract comes in. The (implicit) contract is ultimately entered in to for our own self-interest. We want to live socially and benefit from social interaction, and that’s why it’s a good idea to make that prudential decision to play the moral game. If that’s the case, then we can evaluate particular moral norms according to how well they make that social interaction possible, maintain social stability, prevent cheating and disruption, and how well that benefits us. Sometimes we have to sacrifice some interests – such as the interest to murder people we don’t like – in order to serve others – such as our interest in not being murdered by people who don’t like us.
It’s subjectivist, it avoids the problem of finding a rock solid core to morality, but it’s practically binding in the way that Rosenhouse wants when he says:
In my day-to-day life I am an unabashed moral absolutist. Some things are just right and others are just wrong, and if you disagree with my judgments than I will unleash upon you a barrage of stern looks and disapprobation. I simply regard it as obvious that people have certain obligations to one another, and I really have no desire to debate the matter.
And, while it’s slightly different from Ruse’s rendering, it’s not that far off. Evolution also comes in because evolution has shaped our interests (biological, psychological, social etc) that underpin our reasons for being moral, and evolution has also pre-equipped us with a bunch of emotions and cognitive faculties that work towards making social life possible.
These faculties aren’t foolproof nor all we need in order to be moral. The cultural innovations of creating behavioural norms, morality, laws etc are far more effective at maintaining social behaviour and serving our interests than our evolved heuristics and proclivities alone. But, you know what? Evolution provided us with the plastic, conditional and rational faculties that enable cultural innovations as well. Evolution has its mits all over this story, at least in terms of providing the building blocks, but it’s up to us to build the moral house.
So there you have it: an anti-realist subjectivist evolutionary ethics that avoids the curly problems of finding the non-existent rock solid foundations of morality yet allows for ‘binding’ norms that we can’t subjectively escape. I don’t doubt many will disagree with plenty of details, but I do believe it goes some way to resolving the confusion between Ruse and Rosenhouse, at the very least.
9 Comments
JW Gray · 22nd December 2011 at 5:54 pm
I consider myself to be a moral realist, but I don’t think that means anything so strange about what “norms” are. The anti-realist explanation for norms can be compatible with moral realism. What is the best “realist” explanation/justification for “norms”?
Mark Sloan · 23rd December 2011 at 3:18 am
Tim, I was wondering what you might think of the Ruse – Rosenhouse exchange.
My diagnosis differs from yours. I see the chief miscreant, the chief source of both Ruse’s and Rosenhouse’s confusion, as a terribly misleading definition of “realism” that leads both sides to different logical dread ends.
According to Richard Boyd (according to Wikipedia) moral realism means that:
1. Moral statements are the sorts of statements which are (or which express propositions which are) true or false (or approximately true, largely false, etc.);
2. The truth or falsity (approximate truth…) of moral statements is largely independent of our moral opinions, theories, etc.;
3. Ordinary canons of moral reasoning—together with ordinary canons of scientific and everyday factual reasoning—constitute, under many circumstances at least, a reliable method for obtaining and improving (approximate) moral knowledge.
I like this definition as is. It describes a subject fully accessible to science. What category of things called ‘moral’ does science have access to? Science has access to all past and present enforced cultural norms (virtually all of which advocate forms of altruism) and our biology that motivates altruism. That is close enough to what morality ‘is’ for me.
The terribly misleading part of the common definition of realism is the follow-on claim of what I call ‘magic oughts’ that are binding on people regardless of their needs and preferences. Since neither Ruse, nor Rosenhouse believe in ‘magic oughts’ they both claim to be anti-realists. Ruse, who knows better, even seems to take a perverse delight in saying things like “Morality is a delusion” when all he means is that there are no ‘magic oughts’. It follows (I assume Ruse thinks) that if morality is a delusion, how can science tell us anything useful about the contents of a useful delusion?
If we separate the existence of ‘magic oughts’ from that of Boyd’s moral facts, then Ruse would, so far as I know, be a moral realist. He understands as well as anyone that morality, as an evolutionary adaptation, is as fully accessible to science as understanding metabolism and how brains work. Ruse has been trapped in a logical dead end by Including a requirement for the existence of ‘magic oughts’ in the definition of moral realism.
My cure for his, and Rosenhouse’s, confusion is to give up the idea that there are ‘magic oughts’ and define realism as specified by Boyd above. Then everything falls into place.
But if there are no ‘magic oughts’, how can we rationally justify the burdens of moral claims as is critical to knowledge from science becoming culturally useful?
That is easy, you rationally justify them by instrumental oughts.
Now I will replay my standard spiel that I also offered as a comment to Ruse’s second post on scientism. Please pardon the repetition, Tim. I am not sure what parts you do and do not agree with.
If you desire X (say well-being over a lifetime), then based on science Y (based on morality being an evolutionary adaptation to increase the benefits of cooperation in groups), you ought to do Z (perhaps act altruistically to increase the benefits of cooperation in the groups you belong to).
Much of our sense of well-being is an evolutionary reward selected by its ability to motivate behaviors that increased the benefits of cooperating in groups for our ancestors. As far as I can tell, something like “Altruistic acts that increase the benefits of cooperation in groups are moral” is the unchallenged winner of reference moral principles for groups to enforce as their instrumental choice with the goal of increasing durable well-being.
It reassures me that the above is on the right track because it also appears to be empirically true that virtually all past and present enforced cultural norms (no matter how diverse, contradictory, and bizarre) advocate altruistic behaviors that increase the benefits of cooperation in groups.
For example, “You ought to accept the burden (act altruistically) of resisting any urges to rape children, steal, lie in court, or murder or infringe on the rights of people to life, liberty, and the pursuit of happiness”.
The diversity and contradictions of cultural moral standards are mainly due to 1) differences in who is in the in-group and collects most of the benefits and who is in the out-groups who are exploited and 2) different ‘flags’ of membership in an in-group such as circumcision, not shaving beards, and some really bizarre stuff – the book of Leviticus provides some prime examples.
Why should I accept the burdens of the above moral principle and its heuristics (moral standards such as “Do unto others as you would have them do unto you”) when, in the heat of the moment of decision, I expect doing so will be against my best interests? I will (almost always) accept those burdens because I put little faith in my imperfect prediction capabilities of what will be in my long term best interests, but, based on careful consideration, a lot of faith in the wisdom of the ages.
It should not be a big surprise that the underlying principle defining what morality ‘is’ (as a matter of science) is also the instrumental choice for an culturally enforced moral standard that is most likely to produce durable well-being.
Tim Dean · 23rd December 2011 at 8:25 pm
James, as far as I understand, my anti-realism is constituted by my rejection of three things:
1) the existence of ‘moral facts’. So a rejection of a special class of facts that are only about morality, such as facts about what constitutes ‘the good’. There are certainly ‘real’ natural facts about a token case of murder, and that’s all there is. None of these facts are distinctly ‘moral’ facts (thus rejecting the kind of realist naturalism Moore opposed), and there are no additional non-natural ‘moral’ facts (thus rejecting non-naturalism).
2) a rejection of intrinsic value, which is really just rejection of some mysterious thing that distinctively moral facts could be based upon. I believe all values are instrumental and contingent.
3) a rejection of internalism. I don’t believe someone is behaving fundamentally inconsistently or irrationally in saying “murder is wrong” and then murdering someone in a non-chalant manner. Not that I’d expect people to ever do this, but someone who is emotionally stunted yet capable of understanding the concept of behavioural rules could conceivably be capable of doing so. Or they could say ‘slavery is permitted’ in culture X, but still be entirely opposed to it today.
With that said, you’ve mentioned before that I could have some realism in my view. Given that’s your forte, can you tell me where it might slip in?
JW Gray · 24th December 2011 at 6:05 am
I can’t remember what I said about realism being possible in your view, but I am wondering what exactly the debate consists of concerning “normative facts” — the sort you reject. The difference between realists and anti-realists isn’t obvious regarding normative facts. Where does that debate take place? What exactly do normative realists say about normative facts that anti-realists reject?
Mark Sloan · 25th December 2011 at 4:26 am
Tim, I want to take the liberty of answering your question addressed to James.
I see your position as chock full of realism as the word is commonly used outside of moral philosophy.
I understand, or at least suspect, that you hold the following to be likely ‘true’ facts about morality which are provisionally true in the normal sense of science. These are roughly in order of likelihood.
1) No source of rational justification exists for accepting the burdens of acting morally regardless of personal needs and desires.
2) There is a rational justification for almost always accepting the burdens of acting morally in order to achieve certain personal needs and desires such as maintaining a sense of durable well-being. This is an instrumental ought.
3) Our biology that motivates altruism (such as empathy, loyalty, conscience) and enforced cultural norms which advocate altruism are evolutionary adaptations that exist because they were selected for by increased benefits of cooperation in groups.
4) Our biology and moral intuitions, as molded by our cultures, provide emotional motivation to act morally, as we individually understand morality, independent of any rational justification.
5) It appears to be empirically true (true in the provisional sense of science) that there is an underlying principle for virtually all past and present enforced cultural norms. That principle is “Enforced cultural norms, commonly called moral standards, advocate altruistic acts that increase the benefits of cooperation in groups”.
6) It also appears to be empirically true that much of our biologically based sense of durable well-being (a sense of belonging and the security that belonging brings plus pleasure in the cooperative company of family, friends, and community) evolved as a psychological reward to increase the benefits of cooperation in groups.
7) The above is consistent with the philosophical view about morality called “anti-realism”. (This, IMO, requires a really misleading and unfortunate definition of philosophical “anti-realism”.)
JW Gray · 25th December 2011 at 5:53 pm
The distinction between realism and anti-realism is not clear. Perhaps Tim’s view could be considered to be a form of realism. He thinks we can rationally discuss right and wrong. He thinks facts concerning morality are based on the empirical world — what we desire etc. He doesn’t accept “queer” moral properties, but not all realists would accept queer properties either.
Some moral realists are reductionists and think words like “right” and “wrong” refer to facts about us (what we desire or a “social contract.”) However, social contract theories are considered to be forms of archtypical forms of anti-realism thanks to Hobbes.
Kevin · 26th December 2011 at 1:10 am
I agree with Rosenhouse’s posting on the “Basis for Morality”, except for the second and third last paragraphs, where he uses expressions such as “no doubt bequeathed to us by natural selection”. If he could reduce that assertion to a mathematical formula, perhaps it would be easier to follow. At the moment, I am afraid it reads like a statement from Ovid’s Metamorphoses.
Tim Dean · 26th December 2011 at 5:47 pm
Mark, I would agree with all those tenets, with some minor changes to details and some definitions, and perhaps some mention of implicit social contracts. But that does broadly captures my moral view.
James, perhaps I’m mistaken in thinking you once mentioned that my view might be compatible with a form of realism. And I agree wholeheartedly that the realism/anti-realism distinction is by no means clear cut.
My objection to moral realism is the notion of uniquely prescriptive ‘moral facts’ which are distinct from non-moral descriptive facts. I think there are only non-moral facts – including facts about cooperation, psychology and human interests – and our contingent desire to pursue those interests.
If one wanted to call facts about cooperation, or facts about how to behave socially in order to further our interests as social creatures, as ‘moral’ facts – which would constitute a weaker sense of ‘moral fact’ – then perhaps I could be a realist. But I don’t believe there are any objective facts that are intrinsically prescriptive or binding or motivating.
If that clears things up…
Mark Sloan · 27th December 2011 at 7:02 pm
Tim, your position is not as clear as I would like. Perhaps you could clarify it relative to the following.
Relevant to the discussion, I have been rethinking my previous rejection of morality, defined as an evolutionary adaptation, being intrinsically binding and motivating.
I am considering three ways that “evolutionary morality” is intrinsically binding and/or motivating for human beings. As I understand you and I agree, this evolutionary morality is defined as something like “Altruistic acts that increase the benefits of cooperation in groups are moral”.
First, as a claimed descriptive fact, virtually all past and present enforced cultural norms (no matter how diverse, contradictory, and bizarre) appear to be heuristics (rules of thumb that usually achieve a goal but are fallible) advocating altruistic strategies for increasing the benefits of cooperation in groups. For example, “Do unto others what you would have them do unto you” compactly advocates the altruistic game theory strategy indirect reciprocity. “Do not steal or kill” advocates altruistically accepting the burden of not stealing or killing regardless of how much you want to.
The altruistic strategies from game theory such as indirect reciprocity and not acting selfishly (which are winning strategies under certain common conditions) are as intrinsic to our universe as mathematics. Therefore, the above evolutionary morality definition is as intrinsic to our universe as mathematics. Mathematics is binding on all of us in the sense that we are stuck, regardless of our needs and preferences, with 2+2=4. In the same way, when deciding what cultural norms we will enforce in the many groups we all belong to, we are stuck with this intrinsic definition of morality regardless of our needs and preferences.
We can propose that other underlying principles, perhaps Kant’s categorical imperative, be culturally enforced in the groups we belong to, but this ‘moral standard’ will not be ‘moral’ in the same sense that virtually all past and present enforced cultural norms have been moral. Such other underlying ‘moral’ principles will be something else, from a different category of ethics.
Second, it also appears to be descriptive fact that our biology that motivates altruism such as empathy, loyalty, and conscience was selected for by the benefits of cooperation they produced in our ancestors. We are stuck with this biology, and the motivations it produces, regardless of our needs and preferences. We have other biology that motivates ‘immoral’ behavior. So our ‘moral machinery’ biology is not absolutely binding or motivating in terms of controlling out behaviors, but it still binding and motivating in the same limited sense our other biology is.
Third, since this evolutionary morality definition is the underlying principle of virtually all enforced cultural norms, heuristics for it will be culturally enforced regardless of our personal needs and preferences. That is, these heuristics for this definition of morality are binding, regardless of our needs and preferences, in the sense that they are culturally enforced.
While I can argue that this evolutionary morality is in some sense both binding and motivating, I do not see that this makes it necessarily prescriptive – what people ought to do. It still seems to me that the rational justification for accepting its burdens can only be an instrumental ought. It is odd to me that a definition of morality could be intrinsically binding and motivating for human beings, regardless of personal needs and preferences, but not intrinsically prescriptive.