Morality, Health and Sam Harris

9 02 2011

There’s a lot to like about Sam Harris‘ views on morality. In fact, I suspect that even his most vocal critics agree with him on a vast majority of what he has to say. His advocacy for a scientific engagement with morality is warmly welcome, as is his commitment to go beyond the old God versus no-God debate to suggest a positive agenda to build a secular morality devoid of supernatural meddling.

But there’s one sticking point  – one to which Harris continues to apply glue – and one against which people like myself and Russell Blackford continue to rebound. That is Harris’ commitment that science can describe morality all the way down.

Harris suggests that science doesn’t stop at the descriptive waters edge, but that it extends as far as being able to establish our fundamental values. His brand of bald naturalistic realism is not only extreme but, in my opinion, overshoots his objective. And in doing so receives criticism that distracts from the merits of his view.

Harris, being understandably tired of seeing empirical research sidelined in moral enquiry, is keen to challenge the dogma that reinforces the descriptive/normative gulf. His main targets are Hume’s is-ought gap, along with Moore’s naturalistic fallacy.

These are commonly interpreted as saying that science can tell us how the world is, but it cannot tell us how the world ought to be. Science can tell us what things cause pain, but nothing it can say about particles and nervous systems can tell us that pain is bad. Science can tell us how zygotes develop into embryos and on into foetuses and then into people, but it can’t tell us where to draw the line in terms of what counts as a fully-fledged moral agent worthy of a right to life.

Harris suggests the is-ought gap and naturalistic fallacy have gone too far in declaring that empirical research is entirely separate from moral enquiry.

And he’s dead right.

But instead of accepting a modest victory, and reintroducing science into moral philosophy, he seeks an unconditional surrender from moral philosophers, and in doing so (to push the metaphor) he suffers a Pyrrhic victory.

It’s not Harris’ emphasis of science that is the cause of so much philosophical criticism directed against his theory, it’s insistence on a particularly extreme brand of realism.

In metaethics – that branch of moral philosophy that doesn’t seek to actually guide moral behaviour, but which seeks to understand what terms like good and bad mean – there are two broad schools of thought. In the blue corner, moral realists suggest that moral statements – things like ‘capital punishment is wrong’ – are the kinds of things that can be true or false. Deeper in the blue corner, naturalist moral realists suggest that the things that make these moral statements true are natural facts about the world.

In the red corner are moral anti-realists, who argue that moral statements aren’t true or false, but that they’re either assertions about how we want people to behave (non-cognitivists), or they’re just wrong entirely (error theorists).

Sam Harris is advocating a particularly uncompromising brand of naturalistic moral realism. But this stance is deeply problematic. And posts like those of Russell Blackford outline in exhaustive detail why.

Even Harris’ response to Blackford suggests that he hasn’t fully grasped how extreme his view is, and that he doesn’t need to push that far in order to win a place for science at the moral table.

In his response, Harris likens morality to health.

Clearly there are scientific truths to be known about health – and we can fail to know them, to our great detriment. This is a fact. And yet, it is possible for people to deny this fact, or to have perverse and even self-destructive ideas about how to live.

The hope is that we will all concede that health is something real, and there are real facts discoverable by science that tell us what is and what isn’t conducive of health. Yet health can remain a somewhat elusive concept – it’s a property of living things, but is not directly observable, for example – but it’s still real. And it’s transparent to scientific enquiry.

The thing is, Harris isn’t the first to draw a parallel between morality and health. Another moral realist, Paul Bloomfield, has dedicated virtually his entire 2001 book, Moral Reality, to fleshing out this analogy and using it to defend moral realism.

Moral properties have the same ontological status as healthiness or other biological properties. If being invisible to observation and not playing a role in causal explanations are not problems for realism when concerned with being alive and healthy, then these cannot be problems for being good or bad or getting it right or making a mistake. If we are committed to realism about other properties with these attributes (these properties of properties), then moral properties having these attributes cannot be a problem for realism with regarrd to them. So we arrive at a form of moral realism that takes its lead from the metaphysics of physical health. (Bloomfield, 2001, p.28)

The analogy is intriguing. But it’s ultimately a dead end in terms of justifying the kind of naturalistic realism that Harris (and Bloomfield) advocate. However, the reasons why it’s a dead end are illuminating.

The analogy, if it holds, doesn’t actually reinforce full-on realism. If anything, it supports a kind of anti-realism. And I believe if Harris were to embrace the moral-health analogy in this anti-realist way, his entire approach to morality would be stronger – and people like myself and Blackford would only be making noise about how good it is instead of spending the bulk of our words on our disagreements.

The thing with health – and the problem with the health-morality analogy – is that one can have exhaustive knowledge of the facts about health, and the facts about what one ought to do to be healthy, but they can still rationally (if imprudently) say “so what?” There is nothing about the facts of health that compel one to pursue health, nor that condemn that individual as being irrational if they don’t value health and go off and guzzle doughnuts.

Likewise, someone could have exhaustive knowledge of all the facts about wellbeing (assuming it’s even possible) and still say “so what?” Yes, this person is perhaps acting imprudently, but there’s nothing rationally binding about these facts that commit them to pursuing wellbeing.

Ultimately, we need to decide that wellbeing (or whatever we peg morality to) is worth pursuing. And science alone can’t deliver this practically-motivating punch.

The facts about health and morality might be real. But the the lynchpin of morality – the commitment to pursue health/wellbeing – has no foundation in scientific fact, only in our own agreement to pursue them.

The thing is, this isn’t even remotely a practical problem for morality. This is because, as a matter of empirical fact, virtually everyone is already pursuing health and wellbeing. We’re social animals, after all. Being animals, we pursue health. And being social, we pursue morality. And if everyone is pursuing these things, there are things they ought to do, and things they want others to do.

Morality is really a deep kind of prudence: if you value wellbeing, do X, and demand that others do X too.

And as virtually everyone values wellbeing, then all’s good. For those who don’t value wellbeing, and who by doing so threaten the wellbeing of others, are kept in line by those who do value wellbeing. Hence, moralising, norms, laws, social contracts etc.

And if one day everyone decided not to pursue health and wellbeing, we’d probably all die out. And we’d die fools (if rational fools).

So you can be an anti-realist at the bottom – suggesting that pursuit of health or wellbeing is not written into the laws of nature – and be a realist up from there, and you get a moral system that is basically identical to the one Harris is advocating.

Science doesn’t have to dictate morality all the way down in order for it to play a pivotal role.

Were Harris to adopt such an anti-realist position, he’d gain much and lose very little. He’d still have a strong case for the role of science in moral enquiry, and he’d still be able to build a robust moral system that promotes the values he wants. He’d have to argue (philosophically) about wellbeing, and whether there are any better ways of defining morality, but that’s an argument into which he’d be welcome.



17 responses

9 02 2011
James Gray

I reviewed Bloomfield’s book a while back and the disanalogy between health and morality is that morality is “overriding.” That doesn’t mean that Bloomfield used the analogy in a fallacious way because a lot of his discussion was about how to get an ought from an is. I think his argument is incomplete, but his points are relevant. One main point with the analogy is that we can get an ought from an is from our goals whether they are moral or not. The missing ingredient in morality seems to me to to be the greater importance found in intrinsic value.

10 02 2011

You’ve a typo in the first para: “his his” should be “is his”

10 02 2011
Mark Sloan

Tim, I think this lays out nicely your (and my) complaints and disappointment with Sam’s position. Yes, the cause I think we all support, “a positive agenda to build a secular morality devoid of supernatural meddling” might be much advanced if Sam could be convinced to modify his apparent claim that “increasing the well-being of conscious creatures” is somehow magically binding on people regardless of their values, needs and preferences. That is, moral standards and values are, for the most part, best understood as useful strategies for attaining the goal of better meeting people’s individual long term needs and preferences related to living in groups.

However, I take issue with the idea that this means “Science doesn’t … dictate (a) morality all the way down.” The evidence is excellent that science defines “a” morality all the way down to the fundamental nature of the universe in the form of “Unselfish behaviors that increase the benefits of cooperation in groups are moral behaviors”. Of course, we ‘ought’ to adopt and practice such a moral principle not as an imperative ought (which does not exist) but in the form of a prudent ought, the moral principle expected to best meet our long term needs and preferences.

So science does define, in my opinion, a moral principle, “all the way down to the fundamental nature of the universe”, it is just not a moral principle that is magically binding on people to adopt and practice it.

Conveniently though, this particular moral principle is not only the one science defines “all the way down to the fundamental nature of the universe”, it is arguable the most prudent choice, of all available alternatives, for adoption and practice even when we think, in the moment of decision, it will not be. It is arguable the most prudent choice since it 1) defines moral behavior in terms of benefits, 2) has higher consistency with existing moral intuitions and cultural moral standards, and 3) is cross species universal from the beginning of time to the end of time.

It is a moral principle that was available when the fusion fires of the first star lit (though there was no biology to take advantage of it) and will be available tens of billions of years after the sun expands to a red giant and swallows the earth.

10 02 2011
Sean Carroll

Found this via Russell Blackford. I think you are right on point. I said pretty much the same thing here:

but you’ve laid it out more clearly and precisely.

10 02 2011
Richard Wein

Hello. I’m new to this blog. I came here via a link from Russell Blackford’s blog. Very interesting article. I hope it’s OK for me to offer some robust criticism. ;-)

I can see that you have more chance of persuading someone out of moral realism if you offer him a half-way house on the way to anti-realism. But attractive though it may be, I don’t think your compromise is rationally tenable.

You rightly said at the outset that metaethics “seeks to understand what terms like good and bad mean”. And you proceeded to define moral realism and anti-realism in terms of their different stance on the meanings of moral statements. In particular, you defined moral realism as follows:

“…moral realists suggest that moral statements – things like ‘capital punishment is wrong’ – are the kinds of things that can be true or false.”

So far, so good. But later you seemed to forget all about moral statements, and didn’t explain how your discussion related to them. Eventually, you summed up with this passage:

“So you can be an anti-realist at the bottom – suggesting that pursuit of health or wellbeing is not written into the laws of nature – and be a realist up from there, and you get a moral system that is basically identical to the one Harris is advocating.”

What is this anti-realist “bottom” that you’re referring to? If it’s some moral statemement like “well-being is morally right” or “we ought to pursue well-being”, then a moral anti-realist considers these statements not to be true, as you’ve said. In that case, your compromise position seems to involve deriving true moral statements from untrue ones. If the “bottom” is not a moral statement at all, then what is it and how can you derive true moral statements from it? Or does your compromise position involve some different sense of what it means for a moral statement to be true, and if so what is that sense which is “basically identical” to Harris’s and yet different?

I think the vagueness of your argument is preventing you from seeing the flaws in it. Which is the same criticism I make of Harris.

10 02 2011
Richard Wein

P.S. From reading some of your other posts it seems that you’re more of a moral anti-realist than you appear in this one. I suspect that here you were bending over backwards to find some agreement with Harris. I think you bent too far, and it’s a mistake to think that a rational metaethical position can deliver anything like the kind of moral system that Harris wants, or indeed that _I_ want but realise I cannot have. What’s more I don’t think Harris is going to buy that he can give up moral realism and still have most of what he wants. What moral realists (and I) really want is to be able to say, “That’s wrong!” in a moralising tone, and have it be an objective truth.

10 02 2011
Tim Dean

Hi Richard.

To clarify where the anti-realism fits in, I’m suggesting that when someone says ‘you shouldn’t kill innocent people’, one can ask ‘why?’ They might say ‘because human life is worth preserving’, and the response might be ‘why?’ And so on.

And as you go, some layers of ‘why’ might mention natural facts. I agree with Harris that science plays a pivotal role in helping us understand those facts, or showing us when we have some fact mistaken. Pivotal.

But what I’m also suggesting is that when you get as deep as one can get, right down to the very roots of morality, the answer will not be natural fact. It will be something like ‘I value wellbeing’ or ‘I value pursuing my interests’ (depending on your persuasion), and this statement won’t be the kind of thing that is true or false.

That’s the anti-realism. The injection of the value premise into the argument is not something binding, it’s not an explicable natural fact, and it’s not revealable through science. We chuck it in ourselves.

However, once we’ve debated that root clause and come to some kind of agreement – which is an ongoing field of philosophical debate – science and natural facts are very important to work our way back up to the ‘thou shalts’ and ‘shalt nots’.

So what I’m saying is Harris has only to accept this root anti-realism, and the rest of the stuff he talks about in terms of using science to discover facts about the world is relevant to morality. He just can’t short circuit that first step and peg root value to some fact about the world. Is all.

Now, I should add that I’d like to be a moral realist. In fact, I was for many years, talking a lot about ‘intrinsic value’. But then it dawned on me that the notion of ‘intrinsic value’ is untenable, and I reluctantly turned to anti-realism. But, particularly through reading Joshua Greene’s PhD thesis, I am now comfortably anti-realist, and feel I haven’t lost anything – except for shedding the need to have a solid foundation for morality. A sophisticated anti-realist morality is a morality we would all recognise.

Oh, and for some strange and inexplicable reason, many Australians are anti-realists, error theorists or moral sceptics – Russell Blackford included. So maybe these ruminations carry a particularly antipodean funk that is odious to northern hemisphere sensibilities. But it seems to smell fine from down here.

Hope that clarifies things.

11 02 2011
Richard Wein

Thanks for the reply, Tim. But I don’t think you’ve actually addressed my objections or answered my questions. I suggest you try stating explicitly the premise that you’re starting with, and the moral statement that you’re ending up with. Then I think you’ll see that you can’t get from A to B.

Would you like


In other posts you seem to agree with me that “you shouldn’t kill innocent people” is not a true statement. But then what’s the point in asking “why”?

11 02 2011
Richard Wein

P.S. Oops. I only mean to post the first paragraph of that. The rest were earlier drafts that I abandoned.

11 02 2011
Richard Wein

I should have stated my own position, so you know where I’m coming from. I’m an error theorist (and British by the way).

Though you call yourself a moral anti-realist, you’re position seems more realist than anti-realist to me. It seems like you’re taking a moral realist attitude to morality, and then just adding “by the way it’s not rooted in a solid foundation”.

11 02 2011
Tim Dean

Hi Richard,

Heaven forbid someone mistake me for a moral realist…

Perhaps there’s some equivocation or ambiguity over my use of ‘realism’.

I’m an anti-realist about morality because I don’t think there are moral facts, a la Mackie. But I do think there are natural facts that are crucial in forming a moral judgement.

Just because there are no binding moral facts that compel me to pursue wellbeing regardless of my interests or desires does’t mean there aren’t facts about wellbeing.

Harris, I believe, is trying to stress this point. There are knowable facts about what things raise or lower wellbeing, and these are facts about conscious beings. I’m personally not enamoured by Harris’ definition of wellbeing, but I concede there are relevant facts that contribute to forming a moral judgement. In that sense I’m a realist. Not a moral realist, but a scientific realist. (Actually I’m not strictly a scientific realist, but that’s ot relevant right now.)

So, to go from top to bottom, and back:

‘Harming innocent people is wrong’ is not true or false because there are no facts that make it true.

‘You shouldn’t harm Alice’ is also not true or false, as above. However, there are facts about Alice, about my actions and about whether harm was done that are relevant to this statement.

Hume would be with me all the way, so far. As would , I think.

So I’m a moral anti-realist, but I’m sympathetic to Harris’ scientific realism, and it’s relevance to moral enquiry.

Does that clarify things?

12 02 2011
Richard Wein

Hi Dean,

Thanks for bearing with me. The reason you sometimes sound to me like a moral realist is because you say some things that seem strange for a moral anti-realist to say. But from now on I’ll take you at your word that you’re an anti-realist, and I’ll try to show where I think you go wrong.

Earlier you defined a moral anti-realist as someone who argues “that moral statements aren’t true or false”. Actually, some error theorists would argue that all moral statements are false. Personally, I’m indifferent as whether they’re false or just not true. Still, I’d feel happier leaving out the “or false” from that definition. Also, I hope you won’t object if I make that stronger by changing the “aren’t” to a “can’t be”. That is, it’s not just a contingent fact that there happen to be no true moral statements (but there could be). I would say that true moral statements are a logical impossibility. The meanings of the words make it impossible for moral statements to be true, in much the same way that the meanings of the words “circle” and “square” make it impossible for there to be a square circle.

Now your position seems to be that we can derive moral statements from other moral statements (together possibly with non-moral statements). Perhaps we can derive the statement “murdering Alice is wrong” from the statement “murder is always wrong”. Perhaps we can derive the statement “you ought to give to charity” from the statement “you ought to maximise well-being” plus the non-moral statement “giving to charity tends to maximise well-being”.

Even if we accept that such derivations are legitimate in some sense, why would we be interested in them? We would be deriving untrue statements from other untrue statements. Derived moral statements are just as untrue as non-derived statements. In any case, we can derive any statement we like by starting from a suitable moral premise; and since all moral premises are untrue, we have no basis for insisting on one premise rather than another.

And if you accept that moral statements are not just untrue but incapable of being true, taking them as premises becomes an absurdity. Logical deduction from a logically impossible premise makes no sense.

As far as I can see, the only way to engage in rational “moral” (scare quotes) discourse is to adopt a fictionalist stance. That is, to pretend that moral properties exist and can be attributed to certain things. We can then say that if certain sets of things have this fictional property, so do others. But then we’re not talking about real moral properties. We’re talking about fictional pseudo-moral properties. (By a moral property, I mean a property corresponding to a term of moral judgement, like rightness or oughtness–the property possessed by something we morally ought to do.)

Such a fictionalist approach may be pragmatically useful. If you think that realist moral discourse has a desirable effect on people’s behaviour (perhaps including one’s own), you may well want to encourage it and engage in it. But when we’re in full truth-seeking philosophical mode, we need to recognise that such discourse is about a fiction, and not mistake it for a genuine moral discourse.

In your latest comment, you wrote: “I’m an anti-realist about morality because I don’t think there are moral facts, a la Mackie. But I do think there are natural facts that are crucial in forming a moral judgement.”

The first sentence here is fine. I think we can treat “moral statements can’t be true” as effectively equivalent to “there can be no moral facts”. But the second sentence is problematic. If moral statements cannot be true then there can be no rational way to judge the truth of such statements. So nothing can be (rationally) crucial to such judgements. It’s only when we adopt the sort of fictionalist stance that I’ve mentioned that we can make rational judgements about “moral” statements. But then we’ve recognised that we’re not making genuine moral judgements but judgements about fictionalist pseudo-moral statements.

You wrote: “Just because there are no binding moral facts that compel me to pursue wellbeing regardless of my interests or desires does’t mean there aren’t facts about wellbeing.”

Now you’re talking about a non-moral property, well-being, not about a moral property.

You wrote: “‘You shouldn’t harm Alice’ is also not true or false, as above. However, there are facts about Alice, about my actions and about whether harm was done that are relevant to this statement.”

In what sense are those facts relevant? They can’t be relevant to the truth of the moral statement, since it can’t be true.

You wrote: “Hume would be with me all the way, so far.”

I’m not so sure. Wait a moment while I pull him out from behind this sign, and we can ask him. ;-)

12 02 2011
Tim Dean

Hi Richard. Sounds like you’re selling Joyce-brand fictionalism. I’m sympathetic to that view, but it’s not quite my own. I’ll stick up a more detailed post about where I stand on metaethics that hopefully will clarify why I subscribe to error theorist, but still believe you can talk moral judgement.

22 02 2011

It will be something like ‘I value wellbeing’ or ‘I value pursuing my interests’ (depending on your persuasion), and this statement won’t be the kind of thing that is true or false.

Why can’t this statement be true or false? Isn’t it an empirical fact whether someone values wellbeing or not? We may at some point be able to verify this with a brain scan, for instance.

Just like you say in the original article:

This is because, as a matter of empirical fact, virtually everyone is already pursuing health and wellbeing. (Emphasis mine)

Wouldn’t we – at least in principle – be able to study scientifically what people value, and use that to inform our moral decisions?

22 02 2011
Tim Dean

Hi Deen. I should clarify my language. I mean that it’s not objectively true that wellbeing or interests ought to be valued.

But you’re correct that we can determine (in principle) what people really do value.

However, we would be unwise to simply look at what people value, and then suggest that whatever that is is morally justified. I might value pillaging, for example…

22 02 2011

I might value pillaging, for example…

Yes, but you likely also value not being pillaged, like pretty much everyone else :)

22 02 2011
Tim Dean

And from thence morality blooms…

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: