Criticism of Sam Harris' The Moral Landscape

Commentary on The Moral Landscape: How Science Can Determine Human Values by Sam Harris.

... How could we ever say, as a matter of scientific fact, that one way of life is better, or more moral, than another? ...

I will argue, however, that questions about values—about meaning, morality, and life’s larger purpose—are really questions about the well-being of conscious creatures. Values, therefore, translate into facts that can be scientifically understood: regarding positive and negative social emotions, retributive impulses, the effects of specific laws and social institutions on human relationships, the neurophysiology of happiness and suffering, etc.

This is incorrect because well-being is itself value-dependent: what a person values affects which physical states constitute good/high well-being for that person. Studying these scientific facts – just like studying economics – helps people figure out what to value by helping inform them about the consequences of various choices and actions, but it can't directly tell them what to value or what goals to have in life. That requires moral philosophy. Omitting moral philosophy leaves no way to connect facts with values.

One plan could be to claim moral philosophy as a part of science (because the laws of physics determine the laws of computation which determine the laws of epistemology and the foundations of moral philosophy may be from epistemology). But that's not what Harris is saying. He thinks he can directly connect facts to values.

Also, even if something can be studied scientifically via a lengthy chain of relevancies, that doesn't mean that's the best way to study it. Science and reason aren't equivalents, one can do rational thinking outside of science. For moral philosophy, you'll learn more if you think about it rationally and directly than if you try to figure it out via the scientific study of physics (which would be a reductionist approach).

Cancer in the highlands of New Guinea is still cancer; cholera is still cholera; schizophrenia is still schizophrenia;

Harris doesn't understand schizophrenia. Schizophrenia is not a disease like cancer or cholera, for it's a social judgment that cannot be detected at autopsy or by other scientific methods.

Either it's intentional, off-topic activism, or Harris is so ignorant of the issue that he chose this example while trying to choose an uncontroversial example. Both of those possibilities are bad.

And if there are important cultural differences in how people flourish—if, for instance, there are incompatible but equivalent ways to raise happy, intelligent, and creative children—these differences are also facts that must depend upon the organization of the human brain.

This misses the point. There are cultural differences in how people judge flourishing, in which life outcomes they value.

Also, lots of cultural differences are due to context, not value differences nor brain differences. E.g. there is more flourishing-via-camel-breeding in areas where camels live, and kids riding on camels is a larger part of good parenting in those areas.

In principle, therefore, we can account for the ways in which culture defines us within the context of neuroscience and psychology.

But the presence of camels in the area affects culture – and how many people it defines as camel breeders – but isn't neuroscience or psychology.

While the argument I make in this book is bound to be controversial, it rests on a very simple premise: human well-being entirely depends on events in the world and on states of the human brain.

That's literally true because the states of the human brain include value judgments, too, not just the kinds of things mentioned above like being happy or having a retributive impulse. But that doesn't mean that studying brains is the best way to learn about moral philosophy, just because there's a connection doesn't mean one should take the indirect route. It's good to be aware the indirect route exists because it may be relevant to some arguments, but there's nothing wrong with the direct route (I think Harris believes there is something wrong with doing moral philosophy directly, which is why he prefers this more indirect way of trying to approach moral issues.)

Earlier I quoted Harris:

Values, therefore, translate into facts that can be scientifically understood: regarding positive and negative social emotions, retributive impulses, the effects of specific laws and social institutions on human relationships, the neurophysiology of happiness and suffering, etc.

Did he really mean something more like the following?

I don't think he meant that. Studying the human brain because it physically contains information about values – just like a book – doesn't appear to be the project Harris has in mind. So I think we disagree. I think Harris is incorrectly trying to claim his value judgments about certain emotions and psychology states as a part of science, not saying that value judgments are recorded as physical information in brains (which is also true of books, which I think he views differently than brains).

A more detailed understanding of these truths will force us to draw clear distinctions between different ways of living in society with one another, judging some to be better or worse, more or less true to the facts, and more or less ethical.

I think it's disgusting and revealing that Harris wants to use the authority of science to "force" people to think in certain ways, rather than to persuade them to.

There are, for instance, twenty-one U.S. states that still allow corporal punishment in their schools. These are places where it is actually legal for a teacher to beat a child with a wooden board hard enough to raise large bruises and even to break the skin.

This is factually false (in 2010 when the book came out). The part I object to is about raising large bruises and even breaking the skin. Some of the 21 states referred to do not legally allow that. Here's an example of contradicting information, from Time:

What is Harris doing by including factually false information in his book? What's going on? One of this themes is scientific rigor (which he's bad at in his own scientific papers), but he's not being rigorous in his claims. Either research corporal punishment adequately or don't write about it.

In fact, all the research indicates that corporal punishment is a disastrous practice, leading to more violence and social pathology—and, perversely, to greater support for corporal punishment.[4]

As much as I despise corporal punishment, I don't trust Harris' claim about the state of scientific research. So I checked the cite and it's just one long paper which criticizes corporal punishment in US schools. That can't be adequate for Harris' claim about "all" the research because it's not a survey of every piece of research in the field. It's not a survey at all, and doesn't tell us what even 20% of the research says, so reading Harris' "all" as an exaggeration of "most of" won't fix this error. Harris is trying to deny that people disagree with him, which is false and nasty (you should refute opponents, not deny their arguments exist). He does this by citing a paper that argues for his position but doesn't actually try to survey what everyone else is saying.

Further, the research doesn't indicate it's a "disastrous" practice because what is a disaster is a value judgment, which is outside the scope of any current empirical research (and this isn't even brain research, which is the type of research Harris thinks can tell us about morality). You can research how wounded students are in practice, or the severity of wounds permitted by law, but that kind of research can't tell you what wounds or lack of wounds would be a disaster or otherwise deviate from the moral or good life.

Papers like this often include value judgments which aren't labelled appropriately. It's common to either include philosophy arguments in papers as if they were part of science, or to sneak in philosophy conclusions without arguing them. E.g. this paper says "Fortunately, the practice of government-executed corporal punishment has been declared unconstitutional." But what is fortunate is a value judgment which the research doesn't determine (the research is relevant information to help us make this value judgment, but that's different than the research itself being able to conclude that this is fortunate in the way a physics paper can reach a conclusion about gravity.)

Similarly, the paper says, "A wealth of scientific research demonstrates that corporal punishment of children damages them cognitively, motivationally, physically, psychologically, and emotionally." No it doesn't because "damages" is a value judgment – parents differ regarding what kind of child they want to have. I think there's a truth of the matter and some parents are mistaken, but my knowledge of that comes from rational argument, not from scientific research. Regardless of what future brain research may reveal, today's corporal punishment research is not capable of telling us what science says we should value, it only aids us in choosing our values by helping us better understand the consequences of actions.

The "research" paper concludes with blatant political activism, not science:

Agree or disagree, that's not empirical science. My view: I broadly agree that violence against children is bad, and I've proposed a guideline for parents: never do anything to your child that would be a crime to do to your neighbor. But I disagree with the author's perspective on government, which I want to be more limited. I think the government should stay out of science, parenting and education. (I have logical arguments regarding these beliefs, which we could discuss in the comments below, but I don't claim they are the outcome of scientific research.)

I think the example about corporal punishment is representative of how Harris (and many other authors) incorrectly use research, facts and cites.

And so it is obvious that before we can make any progress toward a science of morality, we will have to clear some philosophical brush. In this chapter, I attempt to do this within the limits of what I imagine to be most readers’ tolerance for such projects. Those who leave this section with their doubts intact are encouraged to consult the endnotes.

Harris is hostile to philosophy. That's notable because the book consists almost entirely of philosophy (or at least non-science, like politics, which is a sub-field of philosophy that we often don't call philosophy, and which requires philosopihcal methods to think about well). This is typical: people study science and then do philosophy, but don't do it very well because they haven't studied philosophy adequately (often because they dislike philosophy and don't think it's valuable, which is often because most philosophy is bad – but people's philosophical intuitions, learned in childhood, aren't very good either and it's necessary to find or create good ideas about how to reason).

But this notion of “ought” is an artificial and needlessly confusing way to think about moral choice. In fact, it seems to be another dismal product of Abrahamic religion—which, strangely enough, now constrains the thinking of even atheists. If this notion of “ought” means anything we can possibly care about, it must translate into a concern about the actual or potential experience of conscious beings (either in this life or in some other). For instance, to say that we ought to treat children with kindness seems identical to saying that everyone will tend to be better off if we do.

But what constitutes being "better off" depends on what you want, so this does nothing to address the is/ought problem – it just moves the problem from "ought" to "better".

The person who claims that he does not want to be better off is either wrong about what he does, in fact, want (i.e., he doesn’t know what he’s missing), or he is lying, or he is not making sense.

Right, because "better off" means "better off according to your own values", so it's best for you no matter what you value. But this doesn't address the is/ought problem or the problem of determining what to value.

Imagine if there were only two people living on earth: we can call them Adam and Eve. Clearly, we can ask how these two people might maximize their well-being. Are there wrong answers to this question? Of course. (Wrong answer number 1: smash each other in the face with a large rock.)

Harris is appealing to widespread moral intuitions and common values in our culture, not actually scientifically establishing anything about moralitty. He just thinks it's obvious (which is what the phrase "of course" means), but appeal to obviousness isn't a method of science (it's a mistaken method of philosophy).

while there are ways for their personal interests to be in conflict, most solutions to the problem of how two people can thrive on earth will not be zero-sum. Surely the best solutions will not be zero-sum.

I believe this (the non-existence of conflicts of interest) because of non-scientific arguments put forward by liberal political philosophers like Ayn Rand and Ludwig von Mises. But Harris is just saying things like "surely" instead of relating it to scientific facts, so it's a bad argument.

While this leaves the question of what constitutes well-being genuinely open, there is every reason to think that this question has a finite range of answers.

No, logically there are an infinite range of answers to that question. I have no idea how Harris decided it's a finite range. For example, one could value there being exactly 3 paperclips in the whole universe. Or 4. Or 5. So you can see that, as a logical matter, there are infinite potential answers. Most of the logically possible answers are dumb, but dumbness is a matter for fallible, rational, critical discussion.

Let me simply concede that if you don’t see a distinction between these two lives [descriptions of lives that almost everyone in our culture, including Harris, considers especially good and bad] that is worth valuing (premise 1 above), there may be nothing I can say that will attract you to my view of the moral landscape.

Basically, Harris is admitting he lacks arguments about his main thesis. If you don't already agree with him about some of the main issues, he doesn't know what to do. He doesn't have a logical way to connect values to science, he needs you to share existing intuitions about morality with him.

Personally, I agree with him about the distinction (I disagree with the altruistic attitude, but it's still way better than rape, violence, and being hunted through a jungle by would-be murderers). I do believe that my view is rationally defensible, but I do not believe that my view of this matter is a part of science.

Harris, by contrasts, seems to think his view is not rationally defensible in full, because he thinks there may be "nothing" that he could say to persuade someone who doesn't already agree with parts of it.

It can be useful to say, "Here are arguments for conclusion C that use P as a premise, so if you already agree with me about P then I think you should agree with me about C too." But the book doesn't present itself as merely doing that – as building some additional moral ideas on top of common, existing moral ideas. Harris claims to be able to put morality on a scientific footing and otherwise deal with fundamental and foundational issues. But his book openly concedes it can't do that.

Science simply represents our best effort to understand what is going on in this universe, and the boundary between it and the rest of rational thought cannot always be drawn. There are many tools one must get in hand to think scientifically—ideas about cause and effect, respect for evidence and logical coherence, a dash of curiosity and intellectual honesty, the inclination to make falsifiable predictions, etc.—and these must be put to use long before one starts worrying about mathematical models or specific data.

The book seems to argue that there is a connection between empirical science – like brain scans – and values. But then Harris says actually he doesn't have any clear definition of science. If one is willing to include "respect for evidence" within the domain of science, then of course science can tell you about values – it can tell you to respect evidence (respect is a value judgment). Similarly, honesty and curiosity are moral issues. But for some reason Harris doesn't conclude, "Morality precedes science and moral values are needed before you can do science successfully, so trying to scientifically establish moral values is pointless." (To give credit: the need for moral values before you can do science was told to me by David Deutsch, years before The Moral Landscape was written.)

Broadly, if you think rational philosophy is a part of science, because you think science refers to all our best efforts to at rational understanding, then of course moral philosophy (being a field of philosophy) is part of science. But that's bad terminology (our culture usefully distinguishes physicists from reason-oriented philosophers), and it's not actually Harris' point.

Conclusions

The book is sloppy, and the thesis is misconceived because of Harris' mistaken attitudes towards science. It's unnecessary to claim everything as part of science. Reason isn't limited to empirical matters. He should study epistemology and understand reason correctly, rather than trying to use science as his only rational tool.

Everything about human beings physically exists, so technically physics research (including its sub-fields) can investigate any aspect of human beings. Further, human brains are computers which operate according to the laws of physics (which determine the laws of computation), and so physics is relevant. But that isn't Harris' thesis. And even granting all this, science wouldn't simply determine values on its own, and supercede philosophy, because we need epistemology in order to judge which science and applications of science are correct. (What I think is that science is relevant in many ways to thinking about morality – it's useful – but not that science can determine morality.)

Harris doesn't know how to scientifically determine which physical states of human beings to value and consider to constitute "well-being". He thinks that brain scans will help with this, but such scans can never tell us that the brain scan results we label "happiness" are scientifically good things (the "happiness" label is not science, it's not an observed fact, it's philosophy and value judgment which is open to rational discussion).

And science isn't the best way to learn about people and their actions or values, even if it would work. For more explanation, see the criticism of reductionism in chapter 1 of The Fabric of Reality by David Deutsch, or ask a question in the comments below. And as Popper explained, we can start anywhere – conjecture anything in any field – and approach it critically. We don't have to focus on building up to the ideas we're interested in starting from foundations that are difficult to argue within our current culture. We can just learn about morality directly with guesses and criticism – but Harris doesn't know that, so his book isn't very good. For example Harris writes:

Many of these people also claim that a scientific foundation for morality would serve no purpose in any case. They think we can combat human evil all the while knowing that our notions of “good” and “evil” are completely unwarranted.

Harris thinks that if you don't have scientific arguments, your conclusions are "unwarranted". This is a major error which is corrected by Critical Rationalism. Harris' problem is he has no idea how to defend reason itself without using science, not just when it comes to moral values but also for anything else (e.g. politics, economics, logic, math or epistemology). There are many valuable areas of human knowledge which are predominantly not understood in a scientific way, but which are rational nonetheless. Reason is actually about error correction, not about empiricism nor about using justifying authorities like science. Authority is actually the arch-enemy of reason, so Harris' book is actually, by accident, an extended attack on reason, because the essence of his project is about justifying moral claims (all justifications are appeals to authority, sometimes in disguise) rather than about thinking critically to try to correct errors and thereby improve our ideas. (In fairness, he's not alone here, and he's not unusually bad. These kinds of mistakes are common when one doesn't understand Critical Rationalism adequately, and we live in a world where fewer than 100 living people have that knowledge. What I dislike is the lack of Paths Forward – ways for Harris' mistakes to be corrected.)

PS I have not read the whole book. If I missed a part which addresses one of my criticisms, please let me know in the comments below and provide a quote.

Related Post: Sam Harris vs. Capitalism