Sam Harris, Science, and Morality

Sam Harris gave a TED talk in 2010 in which he argued that science can–and should–be used to define morality and ethics. His argument essentially boils down to this: moral decisions are decisions made about facts. The more we know about the world, the more facts we have about it and the better and more sophisticated our understanding of those facts, the better decisions we can make. Therefore, morality should be guided by science (and presumably not religion) because it is the scientific process that allows us to test which ethical decisions work well, and which are deficient.

At its core, I don’t disagree with this argument. For example, if we want to help children grown up healthily, I think it makes sense to research nutrition, to see what foods tend to help children grow quickly and healthily. Such an approach would be broadly scientific, and it’s hard to argue with. But it also seems clear to me that Sam Harris both misunderstands the traditional “science can’t define an ethics” argument and is overly credulous when it comes to science’s general merits. The presentation video is below:

First off, though Sam Harris seems to think that he is debunking the argument that science can’t provide the basis for ethics, he actually never discusses it. He is either woefully, even shockingly unaware of what the real debate is about, or is being disingenuous in his presentation. The claim that ethics and morality are beyond the realm of science is a claim about the foundations of ethics, not its application. It’s one thing to say that “assuming that X is good, science can help us achieve X”. I think this is a pretty uncontroversial statement. But what if someone questions the goodness of X? Sam Harris argues that science can help us to figure out how to help conscious beings live more fulfilling lives. But why is helping conscious beings live fulfilling lives good?

This may seem to be some sort of trick question, but it’s not. Harris simply assumes an ethical system, and then argues that science can help us to apply that system–and he’s right. But he completely ignores that science is essentially agnostic when it comes to the basis of ethics. Why is it wrong to kill a person? Science can help us develop better ways of saving lives, of fighting disease, perhaps even through psychology it can help us to deter people from attempting harm one another. But what does empiricism have to say about why it’s wrong to kill someone in the first place? The traditional approach has been to build on some sort of pragmatic or utilitarian philosophy, but again these simply assume the right- or wrongness of given activities, and argue how best to organize human activity so as to maximize the former and minimize the latter.

Science simply gives us data on causality and being: it helps us understand why and how A leads to B. But ethics is interested in whether A leading to B is good, bad, or neutral. And this is always subjective. Is my taking $100 dollars from your wallet a good or a bad thing? Obviously that depends on who’s asking. I may argue it’s a good thing; you would probably argue it’s a bad thing. Most observers would probably agree with you. But what if an extremely poor man stole $100 dollars from Jamie Dimon to feed his starving children? Dimon might actually argue that this is still a bad thing, but I imagine that most of us would not agree. Could science ever provide a definitive answer to this? It might very well provide better systems for preventing theft, it could also provide better thieving systems. But could it provide conclusive reasoning for deciding what actions are bad?

In short, why is it good to help conscious beings thrive? Viewing such thriving as a good thing is an act of valuing. I value conscious beings, but not for truly, or directly, rational reasons. I am delighted to see a dog playing, for example, but it’s not because I somehow think that the dog’s play will indirectly help me. It’s an emotional response, deeply complicated. Now, it is true that that science, in the form of psychology or neurology, could help explain why I find a dog’s playfulness enjoyable, and why I might choose to, for example, build a dog park, or rescue a dog from the pound. But is my choice to do so the right choice, from some sort of factual, objective standpoint? What does that question even mean? How would we determine the rightness or wrongness of my rescuing a dog, from a scientific standpoint?

The point of arguing that science cannot speak directly to ethical issues is to make it clear that science cannot give definitive answers as to why something is ethical or unethical. It can help us achieve a more ethical society (at least in theory) but science could be marshaled to defend any number of ethical systems which would conflict with one another. Science itself is ethically agnostic, because it objectifies the world. Science analyzes things into their components to understand them. Ethics is a subjective process in which unified wholes are valued for a complex set of reasons; different subjects value differing things, and there is no objective way to prove or disprove either valuation. Although science certainly can explain how the valuing occurs, it can’t comment on whether the valuing is good or bad, right or wrong.

Harris doesn’t seem to grasp this, which is amazing, because this is really fundamental philosophical stuff. He wouldn’t have to open the Bible or any other religious book to explore this conundrum; Sarte or Nietzche would do just as well. The disconnect between the world seen as an object and the world experienced as a subject is probably the oldest problem of philosophy, and one that still dominates it. That Harris could spend years writing about religion an ethics, and seemingly never come to understand this, is quite amazing and perplexing. But his position also belies a subtler, but still significant confusion.

Harris seems extremely confident that empiricism–science–will allow humans to build a better and better world; he seems to believe in the inevitability of human progress: as we learn more about the world, we can manipulate it into a better and better place for us to live. The evidence suggests, though, that science has had a much more equivocal impact on the world and on human life. Science has, on the one hand, brought us vaccines, and sanitation systems, and medical intervention, and increased food production, and all sorts of creature comforts. This can’t be denied, and let me be honest: I’m sitting in a heated room, typing on a computer. I have refrigerated food here, and all sorts of books, food, clothes, etc. that were shipped here on technologically advanced ships, trucks, and trains. I’m not a Luddite, and I’m not here to say that science is inherently evil.

But science has also brought atom bombs, machine guns, mustard gas, mercury poisoning, and global warming. It’s not some unalloyed good; progress isn’t guaranteed just because we are applying science to our problems. In fact, it could be argued that though science has improved the lives of a relatively small number of fortunate people, on balance it’s proving to be a growing catastrophe for life in general. This remains to be seen, though. Many people hope that we can use science applied through technology to address the problems caused in the past by science applied through technology. “Green” energy sources, for example, can hopefully be deployed to replace fossil fuels. I hope they are right, but I have to be honest that I’m not particularly confident. “Green” technology may prove to be extremely damaging to the environment; let’s remember that when people started burning coal on a large scale in the 19th century, they had no idea it would lead to the problems we now face. Manufacturing millions of solar panels and wind turbines will involve vast mining operations and the expenditure of huge amounts of energy, and their deployment into the environment may prove to have unforeseen negative consequences.

Of course, perhaps not. I’m not trying to define a wholesale anti-scientific pessimism, but I do think we should be aware of the limitations of our knowledge and the real possibility of serious problems arising from the solutions we are so enthusiastic about today. Ultimately, this credulous approach to science is very much an ideology; some have called it “scientism”. It boils down to a fervent confidence bordering on faith (though they would hate for me to use that word) that human beings, through the application of reason and empirical investigation, can fully understand the world, and apply that understanding through technology to master the world as on object. I am decidedly unconfident about our odds here; as we just discussed, our history suggests that science’s advances nearly always come with huge downsides, major vulnerabilities. I don’t think we are as in-control as “scientismists”, as it were, would have us believe.

And, interestingly enough, this gets us back to the subject-object dichotomy discussed above. A highly credulous view of science ultimately depends on a fully object-focused view of the world that is reductionist and even mechanical. Such a view is less and less capable of making effective predictions as more and more complex systems are added to what is being observed. We are coming to find that the earth, as a biosphere, is far more complex, and sophisticatedly balanced, than we realized before. The argument that we can simply apply our ever-increasingly knowledge to the objects before us and increasingly develop a more convenient environment runs into the real experience of humans, that as we manipulate the biosphere to garner given benefits, real costs are extracted, though often in hard-to-predict ways, and often on people who were not involved in the development of the original technology (i.e. these effects are often “externalities”).

Harris, then, misses the mark, I think, both in his basic philosophical confusion, and in his over-enthusiasm for science as a sort of panacea for all human ills. Again, none of this is to say that I do not believe we should apply science to our problems. But, first, I think we have to recognize the science itself is built on a wide base of philosophical assumptions, some of which may prove to be false, and that there are questions that this system of though cannot effectively answer in full. Ethics is perhaps the best example of this; science can certainly  help us to apply our ethical system, but it can’t answer the basic, fundamental, crucial questions at the very core of our ethical investigations. Second, science itself is a bit of a fickle mistress: what it gives with the right hand it takes with the left, and I think we need to be much more cautious with it than many modern science-boosters would have us be. Harris seems to make massive errors in both of these areas of thought, and I am sad to see his public influence continue. It’s especially ironic that he fancies himself a trusted ethicist, considering that he apparently believes that “some propositions are so dangerous that it may even be ethical to kill people for believing them” (The End of Faith, p. 52). Harris really seems to be the epitome of polarized, hyper-empirical “scientismist”: fully confident of his own moral rightness and his capacity to understand anything and everything. He is much more similar to the oppressive religious leaders he is so (rightly) critical of than he seems to realize.

I would submit that the video below, an abbreviated (and wonderfully animated) recording of a presentation that Iain McGilchrist gave to the Royal Society for the encouragement of Arts, Manufactures and Commerce, portrays a much more accurate, useful, and sophisticated view of human decision-making than Harris’.

[UPDATE: this post appears in slightly modified form at the Tikkun Daily Blog]

9 thoughts on “Sam Harris, Science, and Morality

  1. Disclaimer: I’m commenting having seen neither the Sam Harris talk nor the McGilchrist talk (though I fully intend to!), so my comments are only regarding this post.

    You claim that science cannot provide the reason why our ethics have come into existence such that we don’t immediately recoil in disagreement at statements like “helping conscious beings live fulfilling lives is good.” You don’t really go into an explanation of what type of concept might actually explain this, but that’s ok, since in responding to someone else’s work, the burden is not on you to provide an alternative explanation (and it should be pretty clear to most people, including myself, which direction you lean here 🙂 But I am very curious what framework that religious explanations, specifically Christianity, give for the development of human values, so please, do feel free to expound on it at length!!

    There is a field called Evolutionary Psychology* that attempts to explain the origins of human behaviors and make some attempt to even explain our cognitive processes. It has met with some success in recent years (for example, we seem to have evolved cognitive processes specifically to detect cheaters). Curiously, cognitive processes seem to have developed in order to provide shortcuts into actions leading, most of the time, to an evolutionarily productive behavior (not necessarily reproduction, but could potentially be along the lines of a pro-social behavior, etc.) Along with the emerging field called Behavioral Economics, it seems to have identified a number of systematic errors that we are prone to when making decisions.** This field has been met with a lot of criticism,*** basically boiling down to the fact that we don’t know the precise conditions under which humans evolved, and even less so about the evolution of brain development since as a soft tissue it isn’t really preserved in the fossil record. Also there is the inherently problematic fact that these things can’t really be tested in the way that other types of psychological theories can. They can only be tested in the lab to determine that these phenomena exist, but there is no real rigid way to make a strong statement WHY they came about. That would require some type of randomized, controlled experiment, which obviously isn’t practical when we’re talking about things on an evolutionary timescale (ie, hundreds of thousands of years). They may have been purposeful/adaptive or they may have been just not prohibitively costly as to reduce reproductive fitness.

    That was a long-winded way of saying there ARE currently scientific attempts to identify the origins of human emotions, human values, and yes, even ethics, although they’re much less well-established than other areas of science for reasons described above.

    There is also the more philosophical criticism of the field that says that since we know we have these particular cognitive biases/mistakes that we are prone to making, and continue to make even when we are aware of them, it may prove to be impossible to ever know if we’re truly approaching and interpreting the study of our own cognition in the truly objective sense that science requires. Although, I guess this argument could be applied to science in general, which is fairly disturbing.

    My main point is that just because empiricism may not have much to say about why it’s wrong to kill someone *RIGHT NOW*, there is the possibility that empirical/scientific approaches could be used to interrogate that at some point in the future. An alternate (read: religious) explanation, such as some innate gift that comes from some other process than just random chance (which is, of course, the basis of evolution), seems intuitively much less likely to me, and one that would require an infinite more amount of prior organization. Though I fully admit I am approaching this with the knowledge that I am susceptible to cognitive biases and faults just like all humans – it seems to violate good ol’ Occam’s Razor. I apologize because I KNOW we have had a very similar discussion in person in the past, but I can’t remember your response. I’d appreciate you telling me again.

    I can’t figure out how to add embedded links, so here’s some messy related reading if you’re interested:



    • Actually, I thought I made this clear, but maybe I didn’t: my whole point in the first half of this post was to distinguish between the “how/why” of ethics and the “rightness” of ethics. I never contested science’s capacity (through, for example, psychology) to explain why humans value, or how that process works cognitively. I did, however, question whether science could ever plumb whether or not a given ethical system was good or bad ethically (that is, in a self-reflective sense):

      “Now, it is true that that science, in the form of psychology or neurology, could help explain why I find a dog’s playfulness enjoyable, and why I might choose to, for example, build a dog park, or rescue a dog from the pound. But is my choice to do so the right choice, from some sort of factual, objective standpoint? What does that question even mean? How would we determine the rightness or wrongness of my rescuing a dog, from a scientific standpoint?”

      So we know a lot about why humans like dogs, about our emotional attachment to them, we know that we and dogs evolved together and affected one another’s evolution. This is all open to scientific inquiry. So we know *why* we might dislike the idea of abusing a dog, for example. If I ask you, “is it wrong to kick a dog?” I think you’d say “yes!” and I’d agree. But can this be, in a final and objective way, proven objectively? You could use psychology and biology to help explain why we feel that way, but feelings and ethics are not identical. Ethics is the argument that certain subjective perspectives are truly, finally correct. But how could they ever be proven? What if someone really enjoyed kicking dogs? Is there anything scientific about arguing that they are *wrong* to do so?

      I think it’s really, really crucial to distinguish between explaining how ethical systems function and the actually right- or wrongness of given actions. I simply don’t think that science can, in any meaningful sense, comment on whether something is good or bad in an absolute sense, even though I think it can explain how we value, and why we do so in an explanatory, evolutionary sense. In the end, for example, science could explain why person A thinks kicking dogs is fine and why person B thinks it isn’t. Could science ever arbitrate between them? Science describes, it can’t really prescribe, because whatever happens is an objective reality which science can explain. But it doesn’t provide moral or ethical force to given actions. Science is analytic, reductive, descriptive, and objective, while ethics is synthetic, relational, prescriptive, and subjective. This doesn’t mean that the two approaches are mutually exclusive, and I freely admitted above that we can and should use science to help *apply* ethics, but I think ethics and science are really two poles of reality.

      • Perhaps another, simpler way of saying this is that a scientific ethics would always have to be completely relativistic; science could never (and this is in direct disagreement with Harris) provide an absolute ethics. It could describe an absolute science of explaining how and why humans create ethics, but it couldn’t comment on whether an ethics was, in any final sense, “right” except by equating the survival/spread of an ethics with its rightness (but surely these aren’t the same; slavery survived for thousands of years and is in fact still alive and well today in various forms–but this doesn’t make it right–right?)

      • Haha, I should have probably rolled these all into one comment; oh well. Just to further clarify: I did not claim above and am not claiming now that religion offers some necessary foundation for ethics. I think that atheists can have a complete and robust ethical system–but I don’t think that such an ethical system would in any way be “scientific” at its roots, though, again, it could certainly be scientific in its application. Such a system would still rely on valuing subjects as subjects for subjective reasons, not objective ones, and one could never “prove” the absolute right- or wrongness of a given ethical valuation, even though scientific proof could be brought to bear on explaining why a person is more or less likely to make a given valuation. Saying something is “right” is not the same thing as saying that it is popular, or evolutionarily advantageous, even if things that are considered right tend to be popular or evolutionarily advantageous. Correlation is not causation, and causation is not identification. It is clear that humans can have incredibly diverse ethical systems, even if there is one science of explaining how we go about making such systems, the systems themselves cannot be justified along scientific lines.

        My point about ethics being non-scientific at its core is about asserting subjectivity over and against objectivity, not about asserting religion itself. Though religion certainly enters the debate, I definitely think non-religious people can have complete and robust ethics–but they will always still be fully subjective–and I think this is good, not bad. I do also think that there is an absolute ethics, but this ethics will be understood, be reached, through radically affirming the reality and value of subjects as subjects, that is, subjectively, not via trying to “objectify” subjects or subjective valuations. Again, this is not to claim that nothing objective can be said about such valuations. But here my thought would need substantial fleshing-out–I’ll save this for a later post.

  2. “I simply don’t think that science can, in any meaningful sense, comment on whether something is good or bad in an absolute sense, even though I think it can explain how we value, and why we do so in an explanatory, evolutionary sense.”

    This is related to the point I’m trying to make – I don’t see why anything has to be good or bad in an absolute sense. In fact, it seems more likely that concept doesn’t really exist. (like Shakespeare, “Nothing good or bad, but thinking makes it so.”) So what, exactly, is ethics then? It is increasingly starting to seem like ethics itself is just some epiphenomenon of cognitive shortcuts and/or evolutionary drives that is likely manipulated at will by individuals and society. And as such, its origins AND prescriptions would be amenable to scientific inquiry. My point is that, our ethics are coming from somewhere, and there is no reason right now to think that at some point in the future systematic inquiry might yield some objective insight into its origin.

    An individual’s ethical system isn’t scientific in itself, but I think we can agree its origins can and have been examined scientifically. It seems like you’re trying to make a case that there is some sort of ethics that exists outside of what humans have created through individual cognition and society-building, and as such, isn’t able to be understand from a scientific framework. Is that correct?

    “I do also think that there is an absolute ethics, but this ethics will be understood, be reached, through radically affirming the reality and value of subjects as subjects, that is, subjectively, not via trying to “objectify” subjects or subjective valuations. Again, this is not to claim that nothing objective can be said about such valuations.”

    So what you’re saying is that in order to understand the absolute ethics (which I’m assuming means some universal, absolute prescription for Right and Wrong behaviors/thoughts/etc.), we would have to rely ONLY on our own subjectivity (ie, intuition), fully knowing that we as humans make systematic, predictable errors when we do this very thing?

    I value subjects for subjective reasons, but the reasons why I value them can possibly be objectively understood, if not at the present moment, then possibly as research techniques improve.

    • My point is that the process we use to explore ethics is the opposite of that which we use in science. My impression is that, for you, “science” is just shorthand for “the way things actually work in the material world” (though of course I could definitely be misunderstanding you), whereas I’m using it in the more restricted sense of the epistemological process which relies on inductive reasoning and confirmation through experimentation, and which is based on certain assumptions, such as the homogeneity of space and time, logical inference, etc. Maybe some would prefer to use the term “empiricism” here. For this discussion, I’ve been using science to refer to this epistemological method, not as shorthand for any sort of assumed ontological or material reality.

      My point is that ethics does not use such a reasoning system, though I’m not arguing that it is totally separate from it, but rather that if science is analytical, reductive, and inductive, ethical reasoning follows the opposite paths: it’s synthetic, relational, and deductive. This doesn’t mean that it’s somehow opposed to science–I’m not suggesting that ethics just flies out of the sky for no reason–but that it’s at the opposite pole of human reasoning. I don’t think, for example, that there is any final sort of scientific answer to a question like “is it right to kick a dog arbitrarily?” If someone *does* kick a dog for no reason, science could, at least in theory, explain why, but this isn’t, in any way, an answer to the question above. Explaining how or why is not the same as talking about rightness or wrongness. These are distinct categories, I think.

      As for there being no absolute ethics: I would agree that one descends down the rabbit hole with hypothetical ethical questions, e.g. “Is it OK to torture? What if there’s an atomic bomb in Manhattan, and you only have 10 minutes to find it, and you have captured the man who deployed it?” etc. etc. Arguing that there is an absolute ethical system is not the same as arguing that one could actually give absolute, general moral injunctions. I think this is, in fact, the distinction between morals and ethics: the former generally rests on absolute statements about broad categories: “don’t steal, don’t kill, don’t lie”, etc. But an ethical system gives ground rules, as it were, and then gives both the freedom and responsibility to the individual or group to try and navigate the real world while employing that system.

      The problem with the “there is no absolute right or wrong” is, of course, that I don’t think many people actually believe this. I certainly don’t think you believe this. I think that you and I both have strong feelings about social and economic justice, and honestly I think it’d be disingenuous for you to argue that you think that it might actually be morally or ethically acceptable to punch a homeless man in the face and take what little change he has on him. I think you think that’s wrong, period, and that something *real* is communicated when we express that. My position that this real thing is not verifiable by science is part-and-parcel of my argument that science does a great job of exploring objects but can’t see subjects at all.

      This isn’t a condemnation of science, it’s just a delimiting of it. I’m not anti-science, as I took pains to communicate in the original post. But I think it has limits. As I mentioned in a comment above, I don’t think those limits are bordered by religion as such. I think atheists can admit and engage with subjects-as-subjects. I think, actually, that religion is a specific approach to grappling with subjectivity, but I definitely don’t think it’s the only way. I think that perhaps you are suspicious that my delimiting of science here is a false-flag or trojan-horse operation to sneak my creed into the debate–but it’s not. I’m interested, first and foremost, in building a study of subjectivity-as-subjectivity, and I think that can only occur if we admit that science is constitutionally unable to investigate this, precisely because it assumes the objectivity, or object-ness of something, before it even begins to investigate it. I do think people are objects, but I also think they are subjects, and I think these are two poles of one reality. So whatever science and ethics say, they definitely shouldn’t talk past each other. But they will also have to accept the delimiting presence of one another. Note that I think this applies equally to ethics as it does to science.

  3. You’re right, I think we may have been using the word “science” to mean different things. To quote Lincoln in the movie, “What a joy it is to be comprehended!”

    Unlike, I’ve noticed, many atheists who engage with religious people on matters of ethics, morality, etc., I don’t approach the discussion trying to pin religious people down and hold up a mirror to their own misguided thinking. I think the reality is much more complicated and interesting than one party having some delusion that only needs Truth exposed.

    But that being said, I am intensely interested in how religious people’s beliefs influence their thoughts and actions. I know that religion does inform your opinions about this matter, and I’m interested in how. That’s the only thing I was trying to communicate. I know that some people can be dismissive of religious explanations if they are presented as such, so I understand your desire to build your argument without immediately being dismissed by some non-religious people. I do not, and never have, thought your work and your thoughts and your writing were only a reaction/parroting/re-wording of religious principles.

  4. You need to attend more closely to what Sam Harris says in his introduction. What does it mean to say that something is ‘good’ if not that it promotes the wellbeing of humans. This is an elective, subjective statement but one that we all share… unless you are from ISIS.

    • The issue I attend to in my post is not whether Harris has a definition of moral goodness, or even whether that definition is correct. Rather, my point is that Harris’s definition is not (and cannot be) considered scientific, which raises problems for his basic philosophical position, as he wants to claim that science can be used as the sole adjudicator for all truth claims.

      You say above that Harris’s definition of moral goodness is that which “promotes the wellbeing of humans.” It’s worth pointing out that such a definition is exceedingly vague and offers no process for determining what to do when different people’s wellbeing are in conflict. But I will leave the ethical analysis for another time. The point here is this: such a claim is obviously not scientific: it cannot be turned into a hypothesis, nor can it be tested or falsified in any way.

      So it seems that you are actually making precisely the point I made in the piece: that Harris’s moral reasoning is in direct contradiction with his overall philosophical position.

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 )

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