Why I think Sam Harris is wrong about morality
Several commenters have said I should not just critique the excessive certainty of the New Atheists. I should respond directly to Sam Harris’s Moral Landscape Challenge. I should say why I think the argument he makes about a science of morality are wrong. (Harris argues that what is right and wrong can be determined scientifically, just as we can determine truths in the natural sciences). Fair enough. So this morning I submitted the following text as my entry in his challenge.
————————————————
I see two principles errors in The Moral Landscape.
1) The claim that well-being can be measured in an objective way that is similar to measurements in the natural sciences.
I am active in positive psychology. I believe that well-being can be measured. But there is no one measurement, and no way to aggregate measurements in a way that removes the need for philosophical discussions about what matters to people in a particular culture and era. The form of measurement Harris seems to assume is a utilitarian approach: sum up all the well-being of each person at each moment, perhaps estimating it by taking brain scans. Whatever way of life leads to the maximum objective well-being across all people is morally better than one that leads to less.
This approach to measurement is similar to the approach that Dan Kahneman took in the 1990s, when he thought that happiness was the area under the curve, when you graph out a person’s well-being from moment to moment. The more moments spent in high well-being, the better. But Kahneman eventually renounced this view, recognizing that experienced happiness is not the only criterion, and that the duration of experiences does not matter in a linear way. To give one example, which life would you rather lead:
Life #1: You have an easy life, full of intense pleasures, and you are very happy for your first 60 years. But at age 60 you take stock of your life and spend your last 20 years unhappy, with a sense that your life has been a waste.
Life #2: You work hard, have repeated failures, and are rarely happy for your first 60 years. But at age 60 your hard work finally pays off, you have great accomplishments, and you spend your last 20 years happy, with a sense that your life has been profoundly meaningful.
If we stipulate that the total well-being experienced in life #1 is twice as high as in life #2, would it then be wrong to choose life #2? According to Harris’s logic, a society that pushes people toward life #1 is morally superior to one that pushes people toward life #2. It has a higher peak on the Moral Landscape, even though most people would probably choose life #2.
There is no single metric of well-being; there is no way to eliminate the need for reflection, or for philosophers. There is no way to turn values into the sorts of non-contested facts that we find in the natural sciences, where variables of interest can be measured with no need for cultural or historical knowledge.
2) The claim that moral facts are non-anthropocentric facts.
The philosopher David Wiggins (1987) distinguishes between “Non-anthropocentric” and “anthropocentric” facts. (This is similar to Locke’s distinction between primary and secondary qualities). Facts of chemistry, physics, and other hard sciences are non-anthropocentric. They do not depend on any aspect of human nature. If intelligent aliens had come to visit the earth long before humans appeared, they would have found that the earth is the third planet from the sun, and that copper is a better conductor of electricity than is aluminum.
Anthropocentric facts, in contrast, are only true given the kinds of creatures that we happen to be, due to the twists and turns of our evolutionary history. Examples include the facts that sugar is sweeter than ascorbic acid, and that extended solitary confinement is painful. Those are not just my personal opinions; they are facts about sugar and isolation.
Harris is asserting that correct moral claims are non-anthropocentric facts. He is asserting that if intelligent aliens came to Earth today, they could in principle judge the moral worth of human societies, as long as they learned about human brains and could take accurate measures of well-being.
But moral facts are anthropocentric facts. If intelligent aliens came to visit, we can have no confidence that they would reach the same moral conclusions that Harris reaches, based on his utilitarian ethos. Perhaps these aliens evolved by cloning rather than sexual reproduction, and, like the Borg on Star Trek, are concerned only about the strength of the collective, with no concern for individuals.
Even within the category of anthropocentric truths, there are subtypes. Perceptual claims are generally (though not always) true across cultures and eras. Because of our shared evolutionary history, it will be an anthropocentric fact everywhere that sugar is sweeter than ascorbic acid. Yet many other anthropocentric facts are emergent –– they emerge only when people interact, in a particular cultural or historical era. Prices are a good example: It is a fact that gold is more valuable than silver. That is not just my opinion. Other examples include judgments of humor, sexiness, or deliciousness. Some comedians, fashion models, and restaurants really are better than others. Standards emerge at particular times, and the aggregated judgments of experts are actually ratings of quality. There are facts, but they are very different from the facts of chemistry and physics. We might call such facts “emergent culture-specific anthropocentric truths.”
I believe that moral truths are of this sort. This still makes it possible to critique practices in other cultures. All cuisines are not equal – French cuisine was better than 1950s American, and Julia Child offered Americans a way to improve. Similarly, a culture that oppresses categories of people against their will is worse than one that does not. Massive human rights violations, in which large numbers of victims are crying out for foreign assistance, can justify a military response from other nations. But the fact that humanity has reached that point is an emergent fact about modernity and our changing moral standards. I do not think it was morally incumbent upon ancient Rome to stop human rights abuses in neighboring kingdoms. Moral facts are not eternal and universal in the way that non-anthropocentric facts are. But if moral facts are unlike facts in the natural sciences, then Harris’s attempt to collapse the fact/value distinction fails.
26 Comments
Trackbacks/Pingbacks
- The Righteous Mind Challenge | The Righteous Mind - […] Here is my entry in the Moral Landscape Challenge, where I say why I think Harris’s claims about morality…
- Why I think Sam Harris is wrong about morality | The Righteous Mind | SelfAwarePatterns - […] via Why I think Sam Harris is wrong about morality | The Righteous Mind. […]
- Las dos Españas | La nodriza de las hadas y el rey carmesí - […] resistentes al acuerdo racional. Quizás porque, como recuerda Haidt, se trata de argumentos “antropocéntricos” que no se pueden resolver…
- Four Lame Responses to Sam Harris’ Moral Landscape Challenge @ Helian Unbound - […] his essay, Why I think Sam Harris is wrong about morality, Haidt comes up with two objections to Harris’…
- Anthropocentric “Facts” and the Illusion of Value @ Helian Unbound - […] of facts as “anthropocentric” and “non-anthropocentric” in his refutation of Sam Harris’ scientific morality. The terms were coined by…
- The Atheist Evangelist | Handle's Haus - […] a work of propaganda politicized advocacy, which I think is revealing), and This View of Life gives these four…
Regarding the first criticism, Harris makes appeals to the analogy of “health” in the medical sciences. What makes a person “healthy” is comprised of many metrics, and it is very hard to be precise about what “health” is. And yet we can still scientifically study how to improve it. Similarly, I think Harris could acknowledge the ambiguity of “well-being” and its many metrics, and still claim that we can scientifically study the relative peaks and valleys of “well-being.”
Regarding the second criticism, can you provide some support for the claim that Harris thinks moral facts are non-anthropocentric? That doesn’t strike me as something Harris would claim, but I may be forgetting something.
It doesn’t matter how many metrics you have. If you have 100 people using the same metrics, you’ll have 100 different results because some metrics may be more important, subjectively speaking, in some cases than others. So now you have the issue of determining the relevant weights of each metric which is going to bias the results.
So, in the first objection, regarding the question of “Life 1” versus “Life 2,” Haidt unconvincingly suggests science is unable to tell us which is better, and therefore philosophy is indispensable… so, for the sake of argument, the question that remains is, what can a philosopher offer to solve that constructed dilemma? Anything? No.
So, perhaps we should instead try to improve the science that might lead to an answer! I think that is a pretty obvious conclusion…
Then, the second objection, Haidt inexplicably conflates Harris’ clearly delineated “conscious creatures” with the obviously narrower category of “human beings”(… sugar is sweet for more than just human animals, duh, and solitary confinement is a bitch for most creatures, to be sure…) but, why does Haidt make such a sophomoric error? Did Haidt not read the book he’s criticizing? That would seem to be the case, but of course, it’s an obvious immediate fail for this half-assed argument.
The Borg aliens of Haidt’s imagining would not be making decisions about OUR well-being if they are following their own collective ethos, so naturally they would not be authorities about OUR morality. Yeesh… this is 101 level stuff. Hardly worth the pixels, here.
In conclusion, as a non-scientist and non-philosopher, I really have to wonder how folks like Haidt remain gainfully employed after parading their cluelessness and carelessness like this in public. Talk about playing small-ball… a swing, and a miss. Back to the dugout…
As for your first comment you say “science might lead to the answer” But you don’t say how.
It seems to me that in the hypothetical we can assume all the empirical data is in and we have analyzed it. We are still stuck with the question which can not be answered by doing another scientific study or test. IMO that is a refutation of the core of Harris’s claim.
You quickly claim that philosophy can’t answer the question. But have you thought about it?
I’m not sure how the fact that sugar might be sweet to other creatures refutes Haidt’s point.
“The Borg aliens of Haidt’s imagining would not be making decisions about OUR well-being if they are following their own collective ethos, so naturally they would not be authorities about OUR morality. Yeesh… this is 101 level stuff. Hardly worth the pixels, here.”
If both parties can not reach the same conclusions based on observation then it seems neither is doing science. Again that seems to be a direct refutation of Harris’s claim.
I think Haidt can do better than this. This response honestly seems sloppy and like Haidt hasn’t given it his best and all.
I don’t think Sam argue #1. Sam doesn’t really get into the knitty-gritty of that sort of stuff. His book is more principled and philosophical than that, although it alludes to paradoxes and difficulties with utilitarian positions in general. The choice between life 1 and 2 is difficult. But what if we added a third option, a life with 80 years of misery? Sam’s point would be we could say that life is objectively worse than the other two. Some times it will be easy to tell which two states of the world are better, and some times not, and I think Sam is in full agreement with Haidt there.
#2 is the interesting bit of this essay, but is that really the biggest issue with the book?
There are other issues that are more on my mind. I think I can see how you can get an ought from an is with regards to myself. Why should I prefer a happy life over a life with the worst possible misery? Is it against reason to chose the miserable life? It’s certainly against human nature. Any non-crazy human being would prefer a happy life over a life of the worst possible misery. All else equal that means I prefer happier states of the world, any non-crazy human being does. That means facts about my well-being will be normative for what I should do. I can derive ought from is.
But why should I care about other people? The argument from Sam is that any sensible human being will prefer a happy state of the world over a state of the worst possible misery for everyone. But that includes yourself. What if you exclude yourself? Well, you have people you care about in the group, and you don’t want them to experience pain. But what if we where talking about aliens living on a planet a billion light years away? I guess even then, only psychopaths would be truly indifferent to their faith. But what if these aliens weren’t conscious? What if they couldn’t experience pain or pleasure? Would we have any preferences for them then? The argument from Sam is the only reason we would care about them, was if they were capable of such experiences. If they where mindless zombies there would be no moral choices to make with regards to them.
Okay, so I have established I have a preference for well-being over misery for myself and others. But how do I weigh those two concerns? What if I care 99 % about my own well-being and only 1 % about others? That’s not inconsistent with the argument so far. I don’t think these arguments get you to true utilitarianism at all, weighting all life equally. As people we prefer our own well-being very strongly over that of a stranger. I guess this sort of thinking leads to different morals for different people. Some people weigh the happiness of others more than others, and they should act more selflessly, because that’s how they can maximize the well-being in their life.
When it is discovered that the religious reach a higher state of well-being, I look forward to New Atheists submitting to lobotomies and ECTs until their defective brains are cured of the problem that causes them to be less happy. Or we could simply drop them off rooftops, which would certainly cause my sense of well-being to spike.
Saying this as a sympathizer with the new atheist movement: It has been pretty well documented religious people are happier than the non-religious within a country. It’s certainly true in the US, and it’s not difficult to see why. There might very well be a positive effect from believing in something larger than yourself, but surely there has got to be a positive effect to being part of a religious community.
There is a but though. You might be better off being religious as an individual, but on a nation level, the less religious, the happier a nation seems to be. Quality of life improves as a nation grows less religious.
In econ language – being religious has negative externalities. With the religious mindset comes intolerance. With it comes a culture of science distrust. With it comes tribes of us and them, people who share my religion and those who don’t. That can be politically toxic and negatively effect government effectiveness, as well as lead to intra-state conflict. (Case in point the US, although it’s difficult to say just how detrimental the effect of religion is on US politics. It could for example be argued the historically weak welfare state in the US is down to race issues not shared by the rest of the West. Politically the US has somewhat diverged from the rest because of that.)
Overall effects of religion on well-being are not so clear. You can definitely argue the new atheists are wrong to be so certain religion is a net negative. I’m from Scandinavia though, and in a lot of respects I think the world should follow us, and I think a big reason we are the way we are, is because we are the least religious nations on Earth. If we are to live under the same roof, tolerance is so important, but religion simply comes in the way of it.
The mistake you make here is in assuming that people who worry about who is religious and what they are saying and how intolerant they are aren’t any happier than the religious people who worry about opposing beliefs, etc. I see no real evidence that Scandinavia (a terribly depressing region that I would prefer to have nothing to do with) has higher well-being, unless binge drinking is somehow evidence of that.
As a second criticism, you conflate cultural and religious conflict. Your vision of tolerance involves reducing all culture everywhere to a bland harm-based morality. I suspect this will not max out well-being, but instead make a small sliver of (mostly white) elites happy while the rest of the world suffers a misery of authoritarian PC rule.
Can you point to a source that proof your claim “You might be better off being religious as an individual, but on a nation level, the less religious, the happier a nation seems to be. Quality of life improves as a nation grows less religious.”
When you are looking at that data sheet, please ensure the governments of the nation states you are looking at are not fully infiltrated by extremists (religious/atheists). Because, I am sure you want to be fair to those everyday folks, and you wouldn’t want to categorize a blue collar worker who believes in god to an extremist who is out to cause havoc.
That would be like concluding a group, a society, or nation’s harmful decisions on its atheist leader. Not, fair right?
Point is, any person, religious extremist or an atheist, can cause the same level of havoc if they can unite people (intelligent or unintelligent).
What is the easiest way to unite people?
Use a highly sensitive topic (works better when combined with the socio-economic conditions)!
What are highly sensitive topics?
Let’s see… colour/race, nationality, religion, sports team, size of nose (yes, who knew even this could come into play).
So it is not fair to pin religion as a whole, when its an individual taking advantage of the sensitiveness of the topic for their self benefit.
You can, ask why did the people follow an extremist? It’s hard to tell when you are part of the environment. Like a mouse in a maze… Like banks and bubbles… examples can go on…
Let’s not forget, the other sensitive topics mentioned above have also caused deadly outcomes in recorded history.
Therefore, stop blaming religion for things individuals are causing for self benefit.
:{P
Wireless Printers – Business from Anywhere!
The above response starting with, “Can you point to a source that proof your claim “You might be better off ”…” was pointed @Frode not at Udolpho
Udolpho, first of all, your comment doesn’t even have anything to do with the article, but forgiving that, where’s your evidence that the religious systematically reach a higher state of well-being?
Sociological evidence appears to falsify your claim. For example:
http://moses.creighton.edu/JRS/2005/2005-11.pdf
“Data correlations show that in almost all regards, the highly secular democracies consistently enjoy low rates of societal disfunction, while pro-religious and anti-evolution America performs poorly.”
Maybe when atheists do show less well-being, it’s because ignorant bigots like yourself suggest that they get lobotomies and be murdered.
“There is no way to turn values into the sorts of non-contested facts that we find in the natural sciences”
But if those values lead our species (and other lifeforms along with it) towards survival or extinction, does that not result in a non-contested fact that we can measure? This is the eventual bridge between facts and values, the ultimate bridge between is and ought. Naive evolutionary thought once led to naturalistic fallacies by touting racial superiority and competition in the survival of the fittest, but our now more mature understanding of what actually survives in evolutionary systems (cooperation, progress, adaptability, robustness, diversity, etc.) can help us determine the morals we really ought to value—those that lead to the long-term survival of life. There will be conflicts between individuals, society and ecologies over short and long-term preferences, but that is where philosophers and scientists can join hands to help show us the way. And if we can’t figure it out ahead of time, then we know from evolution that limited experiments of trial and error are the best way to proceed.
Nice update of the arguments against utilitarianism in your essay, but I think we can do better than that now. See my essay on Evolutionary Philosophy for what I mean (http://is.gd/IzLP3e). I’d love to hear your feedback on it. You’ve obviously given this a lot of research and thought.
You cannot scientifically determine right and wrong. At most, all you can do is argue what feels right to you. What is best for the most people is not a measure of right vs wrong. The Nazis believed they were doing what was best for the human species. They cannot be scientifically condemned.
Ooh… it appears as if Haidt doesn’t respond to criticism, even on his own blog. So much for intellectual honesty.
I guess this is just link-bait: mention Harris’ name, and watch your hits rise!
Such a big reputation, to be built on such a feeble foundation… I guess that’s why “argument from authority” is a classical fallacy.
I think you’re severely jumping to conclusions here, Ryan. Although I would enjoy more responses from Haidt, his lack of comments does not mean he is intellectually dishonest. Some people have other priorities, or may simply be busy. Some comments are not even worth responding to, because they are very low in quality. Making a blog post does not obligate anyone to respond to the comments.
Further, to conclude that Haidt is simply link baiting for hits is itself intellectually dishonest. You seem to have jumped to the most negative interpretation in an unwarranted way.
Thank you, Nolan,
this is right. I can’t even answer all of my email, so i must prioritize. I do check comments periodically, but can’t reply to all.
–jon haidt
Let’s do a test. We get Sam Harris and put Nolan in one room and Jonathan Haidt in the other. Then we let Harris (standing in the hallway) decide which room he would rather spend the day in. Let’s even assume that Harris is well rested (doesn’t plan to sleep) and is intellectual engaged (wants to spend his time fruitfully). Which door would he choose? I suspect that the choice Nolan thinks that Harris would make is not the same choice that anyone else thinks Harris would make. And I bet that Harris and Haidt would emerge from their room many hours later having had a pleasant, courteous, and intellectually stimulating time. Happily, Nolan would emerge from his room convinced that he had spent his day with the smartest, cleverest, most insightful person he had ever met.
So, I came across these writings after I had submitted my own response to Sam’s challenge. My primary critiques were very much in line with Haight’s, or at least my understanding of what he’s saying. I think some of these ideas are being dismissed a little too cavalierly in these comments.
Even accepting as fact that “well-being” has ‘good’ states as well as ‘bad’ states does not imply that you can objectively establish one state as ‘better’ than another, any more than you can objectively establish which person is ‘healthier’ than another, or even which object is ‘bigger’ than another. In order to make such judgements, you need to decide on a scalar metric which encapsulates all the dimensions of the domain at hand. And, since each domain is multi-dimensional and has orthogonal dimensions (i.e. independent from one another), deciding on such a metric is more than a simple matter of obtaining an ‘answer in practice’. In other words, until you decide on a scalar metric for comparison, you can’t say that an anaconda is bigger than a cat (since it’s shorter in the vertical axis), you can’t say that a pony (a small horse) is bigger than a bullfrog (a big frog), and you can’t state as ‘obvious’ that morality based on superstition is ‘worse’ than a morality based on maximizing the ‘well-being’ of conscious creatures. Good luck on getting a consensus on the one true measure of well-being, and further good luck on providing an objective, axiomatically derived basis as to why your measure is ‘better’ than other measures.
Now, does this mean that the scientific method cannot be applied to questions of morality? Of course not. Moral choices are weighed all the time with the available empirical evidence and our increasingly scientific understanding of the world. It does mean, though, that Sam Harris overreaches when he asserts that the one true measure of ‘well-being’ can be objectively determined, even in principle.
As an aside, and just to bring up a rather standard philosophical thought experiment, how would Sam rate the ‘well-being’ of Homer’s land of Lotus-Eaters, where perpetually well-nourished and narcotically-blissful inhabitants lead perfectly happy lives? How about a world where all conscious minds are wired to Virtual Reality simulations, tuned to their specific needs, individually maximizing the happiness of each mind at all times, while sustaining them intravenously? I would submit that following either of these scenarios very far will lead one to conclude that at least some amount of suffering is ‘better’ than none at all, for a multitude of reasons. This leads me to conclude rather emphatically that even if well-being could be measured as a scalar quantity, the maximization of this quantity is NOT what is actually best.
I’ve been waiting for this response. Thank you for taking the time to write it.
I find the first objection quite good. Harris gets out of this, I think, only because of his distinction between what is “in principle” knowable and what is “in practice” knowable, and I’m not sure I can always buy that. My area of science is physics, and so I’m very skeptical about saying anything that can’t be quantified is “scientific” … but I quite like Dr. Haidt’s work and, frankly, it can’t be precisely quantified either, but I find it very compelling.
The second objection I don’t find nearly so compelling. Or, rather, I find it incredibly compelling, but not particularly relevant. The fact that what leads to well-being in one situation is different from what leads to well-being in another situation doesn’t change what the concept of “well-being” means at the most fundamental levels.
In general, this is why I don’t feel that Haidt & Harris are nearly so at odds as they seem to be (and clearly both think they are). The fact that an extreme conservative views authority and loyalty as more morally relevant than I do seems to be closely tied to the fact that the conservative believes that authority and loyalty lead to better well-being in general than I do. Even if you said, “One should always obey their father,” and even would say this if obeying one’s father results in overall suffering, it seems clear this notion is firmly rooted in the idea that, as a general rule, obeying one’s father leads to positive outcomes. Whether or not this is true is indeed an anthropocentric fact, but not one, I think, that science cannot give insight on the validity of.
I would also point out that I see nothing that indicates that Harris thinks that facts about well-being are non-anthropocentric facts. I get the impression that if he were to place them in this framework, he would view them as anthropocentric facts that are knowable – such as the sweetness & isolation examples provided by Dr. Haidt. And some truths about well-being may even be culturally dependent, to some degree. To take an extreme (and conjectural) example, it is possible that someone trained for years in Buddhist meditation techniques would not suffer pain during prolonged solitary confinement, thus invalidating an anthropocentric fact.
I’m unconvinced about point one.
Let’s assume that Life number 2 is in fact generally preferred to life number 1 by most people. Why can we not ask people what life they prefer and build models and theoretical frameworks out of this?
To adjust for this non-linearity, all we’d require is that humans add up ”well-beings” over time and that they would weight these well-beings differently over time. So, for most people, this summation function would increase well-being weight near the end of someone’s life.
This would still be building a theoretical model out of evidence, which, while perhaps not being a full-out science and still retains the flavor of philosophy, is what social sciences like economics generally aspire to do.
I’m not sure I buy morality as a set of “anthropocentric facts”. Consider animal suffering. Surely human appreciation of sentience in non-human creatures requires moral insight beyond “anthropocentric” conceptions. Conscious experience, it seems, extends as a “non-anthropocentric fact,” perhaps closer to a biocentric conception, in which the human species participates and discovers varying ways to suffer and cease to suffer. If we understand objective moral values and duties as emanating from consciousness and sentience, then I’m hard pressed to deny that “intelligent aliens”, having arrived on earth, could not also appreciate, for instance, the immorality of factory farming. The moment that intelligent aliens recognize the moral status of animal well-being we can discredit the “anthropocentric fact” distinction. What do you think?
Life #1 and life #2 aren’t moral issues at all, you wouldn’t call the police and set someone to jail for choosing life #1 or life #2. This article is silly.
Michael, if you think that morality has nothing to do with happiness and well-being, then your argument is with Sam Harris, because he is the one who proposes that morality refers to total, collective well-being. Jon Haidt’s Life #1 and #2 scenarios were designed to show a problem with Sam Harris’s view.