The Science of Bad Philosophy
A Review of The Moral Landscape: How Science Can Determine Human Values by Sam Harris
By Joseph Bingham
Sam Harris's announced project, in the subtitle of his most recent book, The Moral Landscape: How Science Can Determine Moral Values, is to show how science may determine moral values. One immediately suspects a misuse of "determine"; Harris may be taking the plausible position that science can explain why we tend to adopt the moral opinions we adopt. This is not the meaning of his subtitle. Instead, he claims to explain how science may in fact tell us what we ought to do, in an objective moral sense. This makes the book sound interesting; how does Harris propose to accomplish what philosophers and scientists alike tend to agree is impossible? The answer is that despite his claims, he does not propose to accomplish it at all. The subtitle is misleading, and while it might make the book less laughable—who knows what Harris might have accomplished had he actually undertaken to explode the is-ought distinction?—it makes it much less interesting. Harris in fact does not, as he purports, derive ought from is, or (objective) values from (purely descriptive) facts. Rather, he assumes utilitarianism, and moves on to describe how science may play a role in utilitarian calculations, something which it is probably safe to say that not even the densest person alive has ever doubted.
To say Harris "assumes utilitarianism" is not precisely fair; Harris does present an argument for utilitarianism, but it is not an argument from science, it is a dialectical argument that utilitarianism is self-evidently true, because it is the only possible way to speak in moral terms. Imagine, he says, a state of the universe in which every conscious creature endures constant, excruciating misery. Surely we can all agree this state of the world is "bad" in some moral sense. According to Harris, the moral senses of "bad" and "good" can only be understood in relation to conscious "well-being," which he uses interchangeably with "happiness" and "satisfaction." All of these terms are understood by Harris to be forms or degrees of certain chemical brain states.
Harris's is not the crudest conceivable utilitarian account, but this may only be because it is so hazy that it is completely impotent. He fudges between act and rule utilitarianism—he suggests that rules (such as "do not lie," "do not procure judicial execution of an innocent man") are simply useful heuristics to be employed in an act utilitarian calculus ("What will lead to the greatest aggregate pleasure? Usually not procuring the execution of an innocent man, therefore I will err on the side of not procuring this innocent man's execution." [These parenthetical examples are mine, not Harris'.]). In some cases he suggests that our wrong (by his account) moral intuitions and the distress they cause us should be taken into account as part of our utilitarian calculus. For example, the average person's response to the trolley-and-fat-man problem is obviously irrational (according to Harris), but the distress it causes a person to murder the fat man is an aspect of well-being that should enter the calculus of whether the act is moral. This is rather a silly nuance, since even in the trolley-and-fat-man problem itself (let alone, say, the case of a nuclear bomb or other vast danger) it does not seem to me conceivable that the distress to the protagonist would amount to five conscious lives worth of disutility. Harris's utilitarian formula will likely still seem repugnant to most people, even once he adds their repugnance into the formula.
Harris also fails to explain whether his utilitarianism is purely quantitative or qualitative. He is unclear about whether the difference between lower and higher pleasures, or "deeper satisfactions," as he calls them, is one of kind (and presumably accordingly in the associated sort of brain state, or type of chemical reactions) or merely intensity (in which case pleasure would presumably be associated with a single type of brain state which was more or less present on some particular dimension). The second account is wildly implausible. The first account, which holds that pleasures differ in kind, with deeper pleasures like those stemming from intimacy or Bach being a different sort of pleasure than lesser pleasures like those stemming from PixiStix or cocaine, presents many other damning problems for the utilitarian account of well-being. More importantly, the first account does not seem to get Harris where he wants to be. For his "science of morality" to be possible, the brain states that morality requires us to maximize must be quantifiable. But it will be impossible to quantify different sorts of brain states in relation to one another unless we rely on some subjective measure like self-reported happiness. This may mean the measure of morality Harris believes exists objectively (independent of our subjective judgments about it) likely exists only in the form of the objective reality of our subjective judgments themselves about our own brain states.
Harris does not seek to define well-being beyond saying it "relates to" the brain states of conscious creatures (a low bar indeed). But there are more fundamental and obvious problems with his account than simply his inability to say what he means by the concept on which his entire book hinges, even in the most general sense. The most basic problem is that his argument for utilitarianism is that it is self-evidently true, even though to most people it is self-evidently false.
Consider Harris's thought experiment, described above, about the world in which everyone suffers the worst possible misery. One need not think about this for long to see problems with Harris's argument from intuition. Suppose half of the conscious creatures in the world suffer the worst misery of which they are capable and the other half of the creatures enjoy a moderate amount of delight, which is derived entirely from a distilled sadistic pleasure taken in the misery of the other half. To Harris, this world would presumably be self-evidently better than the worst-possible-misery world; to most of us, it is at least a tricky question, because we have a sense that sadistic pleasures are not good in the way innocent pleasures are. Another obvious and oft-cited thought experiment is to imagine the Matrix; is there any reason to take the redpill and escape the Matrix? To most people, it is obvious that it is better to take the redpill, but this seems completely inconsistent with crude experience utilitarianism. There are sophisticated responses to the Matrix question (Nozick's "experience machine" in the literature), but Harris offers no account which leads one to think the bluepill could be anything but perfectly moral and rational—Cypher, even if his means were unscrupulous, apparently had the right idea about maximizing his own well-being.
As for Harris's engagement of the relevant literature, it is nearly non-existent. Any reader with much interest in utilitarianism will scan the index for Peter Singer, the most famous and controversial utilitarian today. He is cited for, by my count, one uncontroversial point. Of course, Harris does not face the problem most utilitarians and others who closely link a human being's (or other creature's) moral worth with cognitive capacity would face, which is eagerness to distance themselves from Singer's morally abhorrent advocacy of infanticide and so forth. Harris is himself perfectly willing to adopt abhorrent conclusions, although—since this is, after all, a tract—he does not go out of his way to point them out. I confess that I have not read Harris's previous books, and after enduring this one, it is unlikely I shall ever bring myself to do so (you may learn more and better philosophy from the editorial page of the Wall Street Journal, at the expense of less time and fewer brain cells, and without the guilty conscience that comes from lining Harris's pockets). But Harris's earlier work has by his own account made explicit his support for torture and preemptive nuclear strikes against Muslim nations. This of course follows from his moral perspective (one hesitates to call it a theory, given its underdevelopment). For whatever it's worth, Oxford philosopher G.E. M. Anscombe gave her thoughts on Harris's ilk in 1958: "[I]f someone really thinks, in advance, that it is open to question whether such an action as procuring the judicial execution of the innocent should be quite excluded from consideration—I do not want to argue with him; he shows a corrupt mind." Harris, the reader discovers, is open to this and more.
A reader with any interest in epistemology (Harris's epistemology reeks of logical positivism, which is to philosophy something like what geocentrism is to astronomy) will scan the index for Plantinga, one of the most respected epistemologists of the last six decades, the most formidable critic of philosophical naturalism, and the most potent defender of the possibility of warranted Christian belief. It is, of course, not present. Harris never engages seriously with sophisticated arguments—in fact, he goes out of his way to disclaim them; he notes that although he has read some ethical theory, his own theory is highly original and arrived at independently. The reader is relieved.
A Harris book would not be complete without a tangential tantrum about religion, to which he dedicates a chapter. Much of the chapter is spent sneering at Francis Collins, erstwhile director of the Human Genome Project, current Director of the National Institutes of Health, and an evangelical Christian. An undefended philosophical naturalism, which Collins of course does not share, is a premise of the chapter, much of which is dedicated to Harris's quoting Collins saying things like that atheist materialism is to be resisted, then by turns gaping, presenting sarcastic strawmen, and showing off his incomprehension of Christian thought. Harris exclaims that one cannot distinguish the writing of British physicist and theologian John Polkinghorne from "an extraordinarily patient Sokal-style hoax." Of course, when one finds a passage incomprehensible, it is possible that the passage is actually meaningless. It is also possible that one has simply not comprehended the passage. Harris frequently mistakes the one for the other, strengthening the sense that he employs an implicit logical positivism.
There are innumerable silly things throughout the book, too many to do it justice in a short review. A personal favorite: an early endnote, Harris defines "science" as "our best effort to form a rational account of empirical reality," not to be distinguished from a more general concept of "facts." For example, the fact that JFK was assassinated, he writes, is a scientific fact as he will use the term science. Of course, under this bizarre and idiosyncratic definition of "science," essentially every intellectual inquiry is a science, including philosophy and religion. It is typical of Harris, though, to misuse terms in this way. He regularly elides relevant distinctions—between pleasure and satisfaction, between showing the existence of moral reasons for action and persuading someone to act morally, between something's existing and its being quantifiable, between religion and metaphysics, and—the ultimate conflation on which the entire book pretends to be based—between facts and values. If you find a discarded copy, read the book for a laugh; if you don't, read some of the many philosophers Harris should have considered reading before undertaking this project.
1. The problem compares two cases: (1) You see a trolley is about to hit and kill five people. You can throw a switch to divert the trolley to another track, but doing so will result in its killing a single person on the other track. (2) You see a trolley is about to hit and kill five people. The only way to stop it is to push the fat man in front of you to block the trolley. Nearly everyone would throw the switch, but would not push the fat man.
2. Two commonly raised objections are (1) the variety problem and (2) the fungibility problem; this view requires that (1) there is a single quality intrinsic to all pleasurable experiences, from singing to canoeing to kissing, and (2) that ten thousand years of the mildest possible pleasure, such as having one's back gently scratched, may be better than, say, twenty years of happy marriage, because the net quantity of the single quality which all pleasures have in common is greater. On this view, there are no "higher" or "lower" pleasures except in the degree of a single type of pleasure they produce.