The Science of the Good

The Science of the Good September 12, 2019

Can science determine morality?

Since the Scientific Revolution, there have been various attempts to ground human morality in science. The goal to discover moral laws in the same way that science has discovered physical ones is a Promethean project that continues to occupy thinkers and writers to this day. Most notably, Sam Harris wrote a book called The Moral Landscape: How Science Can Determine Human Values that defended his moral realism with appeals to science. Considering Harris’s defense of torture and his backslapping of racist pseudoscientists, we may wonder whether he has applied the scientific method rigorously enough to his approach to morality. However, his perspective is shared by many leading lights in science writing; people are convinced that there can be a “science of right and wrong.”

But the quest has been in vain, according to James Davison Hunter and Paul Nedelisky. Their book Science and the Good: The Tragic Quest for the Foundations of Morality examines the problems in the project.

Levels of Findings

First off, the authors admit that the search to find the science of morality is a worthwhile cause. Social conflict, discrimination, and oppression are moral matters, and discovering what’s right and wrong in them could also uncover solutions to these intractable problems. The authors claim that their issue is in the way the search has been defined. There seem to be different categories of findings we’re being told to expect from the research program:

[S]cience might tell us something about the moral realm in different ways. Consider three: The most interesting way—always the highest aspiration of those who have sought a scientific foundation for morality—would be if science could settle longstanding moral questions. Call this level of scientific results “Level One.” Level One results would provide specific moral commands or claims about what is genuinely valuable. They would demonstrate with empirical confidence what, in fact, is good and bad, right and wrong, or how we should live.

“Level Two” findings, while falling short of demonstrating some moral doctrine, would still give evidence for or against some moral claim or theory. For instance, if there was empirical evidence that virtue theories of ethics [“theories that emphasize the role of character and virtue in moral philosophy rather than either doing one’s duty or acting in order to bring about good consequences”] were false, but the evidence fell short of settling that this or that moral claim was correct, that would constitute a Level Two result.

“Level Three” findings would provide scientifically based descriptions of, say, the origins of morality, or the specific way our capacity for moral judgment is physically embodied in our neural architecture, or whether human beings tend to behave in ways we consider moral. Evidence for these sorts of views doesn’t tell us anything about the content of morality—what is right and wrong—but they speak to the human capacity for morality and in that sense are interesting. […]

In fact, nearly all of the actual science attempting to deal with morality lands at Level Three findings.

This is no mere quibble. In terms of scientific success, this would be like not being able to tell us what the Sun is, but rather examining the neural and physical basis of the human capacity for perceiving the light and heat of the Sun. As the authors admit, these are interesting findings in themselves; however, they tell us nothing in terms of what’s right and wrong.

Critique of Pure Nihilism

This bait-and-switch is made necessary by the consensus among these science writers that there can’t be moral facts in the same way as there can be empirical facts. “In a world where physics fixes all the facts,” says one writer, “it’s hard to see how there could be room for moral facts.” It seems to me that this puts the cart in front of the horse: if physics can’t detect something, then it doesn’t exist. Couldn’t you say the same thing about any non-empirical cultural construct? Couldn’t there be a mode of inquiry appropriate to moral facts, the same way that we use different modes to study living organisms rather than black holes? If our most reliable methods of inquiry aren’t equipped to detect moral facts, then in seeking a scientific foundation for morality we need to either redefine science as just a vaguely coherent process of reasoning, or redefine morality to mean something very different than what’s right and what’s wrong. We end up with a moral nihilism that renders the quest irrelevant even before it’s started.

The Light’s Much Better Here

Hunter and Nedelisky conclude that one major obstacle to the program of the moral scientists is the “effective absence of any working awareness of or engagement with history, culture, or political economy.” The idea that we can examine morality as a neurological phenomenon, or as the outcome of selective battles and differential reproductive success, ignores so much social and cultural context that it’s absurd:

Invariably, the science of morality is directed toward unearthing universally shared moral principles. These are ethical generalities that take shape as moral-philosophical abstractions. The evidence used to address this moral reality purports to be species-wide, whether it is drawn from data from neurochemistry, the evolutionary record, or public opinion surveys.

This is fine as far as it goes, but it barely scratches the surface of morality as it exists in the lives of individuals, groups, communities, and nations. While it may be possible to speak of universal moral principles, nearly all of what we actually know and experience of morality exists only in its particularity: in a bewildering array of complex and contradictory moral traditions, stories, and ideals—made all the more complex by race, ethnicity, gender, region, political economy, and history. This empirical complexity arouses little curiosity from the new moral science. It is as if the best way to address empirical difference were to ignore it altogether.

In other words, it’s odd to think that we’re studying a phenomenon if we’re saying so little about how it functions in the world. Unlike the other objects of scientific inquiry, moral decision-making wouldn’t exist without humans and human societies. The moral scientists assume that the study of morality can be conducted in the same objective, dispassionate manner we use when we’re studying planets or molecules:

There is a temptation to assume that the world of science or the world of philosophy is the world, and that the pursuit of truth somehow rises above the push and pull of social life and the swirling dynamics of the historical moment. what is arguably missing within this discourse is the caution that comes from reflecting on its human and thus contingent nature. One rarely finds in the philosophical or scientific literature any self-consciousness about the ways in which science and philosophy are embedded within a social and historical context and thus operate alongside a range of exogenous—that is, nonphilosophical and nonscientific—social dynamics. […] Although [the epistemologies of philosophy and science] aspire to an ideal of objectivity, the history of science offers a practically endless catalogue of researchers seeing and understanding what they were socially conditioned to see and understand.

What’s your opinion? Can science determine morality?

"You'll have to be patient. The Patheos spam filter is a hungry old beast, and ..."

The Science Fundamentalism of King Crocoduck
"My response to this is currently labelled "detected as spam"."

The Science Fundamentalism of King Crocoduck
"I included "narcissism" as a motivating factor for you, remember. And it doesn't really matter ..."

The Science Fundamentalism of King Crocoduck
""Honestly, the reason I wanted to take this to e-mail is because with you holding ..."

My Gruesome Debate With a Pro-Lifer

Browse Our Archives

Follow Us!


TRENDING AT PATHEOS Nonreligious
What Are Your Thoughts?leave a comment
  • My observation is that what is or is not moral is determined by the wiring and chemistry of our brains (which includes both nature and nurture components). An action taken (or considered) fires off a largely autonomous response of “good” or “bad” (or a signal along a continuum between the two). But outside of some very primitive moral strictures that might be substantially innate (related to fairness, in particular), I see no evidence that nature provides any sort of absolute moral strictures. And if they are not provided by nature, they seem largely outside the realm of science to treat. Intrinsically, no stricture is moral or immoral- not killing, not rape, not slavery. These things are accepted by some people and cultures, not by others. The only thing that determines their morality is whether they are cultural norms or not. As such, science might be useful (indeed, has been useful) in teasing out the mechanisms in our brains that assess morality. But not in identifying any strictures that are or are not, in some way, intrinsically good or bad. Those are nothing but personal and social values.

  • I think you’re closer to the mark when you say that morality involves some process of cultural validation rather than when you claim that morality is determined by brain function. It’s an extremely complex matter, no doubt about it. Obviously it can’t be completely determined by cultural consensus, since (as you say) many cultures accepted or accept behavior that we consider unconscionable. But reducing morality to brain function ignores nearly everything we know about the way morality develops in terms of forms of authority, institutions, traditions, and power dynamics. It’s as mistaken as expecting each individual to take responsibility for his or her use of language; obviously the way these phenomena develop isn’t just a personal matter. The contexts in which we define morality have to do with our acknowledgment of the well-being of other agents as well as the effect of our behavior on the environment. By the same token, we shouldn’t mistake the selfish acts of bigoted and cynical people (like those who countenanced slavery for so long in the USA) for legitimate moral reasoning, when they were truly the exact opposite: the refusal to accept the consequences of valid ethical reasoning.

    I was a little puzzled that Hunter and Nedelisky were so dismissive of moral pragmatism—the idea that while it may be problematic to hypothesize universal morality, we can discuss what behavior is good and bad in achieving certain ostensibly positive social objectives—since I’m at a loss to understand why they object to the notion that moral decision making is goal-oriented in the first place. I’ve often used the metaphor of architecture when I talk about morality, because it seems obvious that there’s “right” and “wrong” in constructing a building in reference to the conditions in which it’s being built, the function of the structure, and the materials available. In the same way, we need to assess the goals of a society and the needs of the environment if we’re going to define what’s right and wrong in terms of the behavior of citizens and policymakers, and those may vary greatly.

    Thanks for contributing!

  • Supposedly some characteristics that are universal (kindness, compassion, etc.) exist because of us being social species, have been selected by evolution for its advantages.

  • Sure. But the authors point out that there’s a difference between selected-for traits and moral reasoning:

    But two different senses of altruism are at play in [E. O. Wilson’s] argument—one prescriptive and the other descriptive; one ethically intended and directed and the other biologically elucidated. In biology, altruism means one organism increasing another organism’s reproductive fitness at a cost to its own. This stands in contrast to an ethical understanding of altruism, where the idea is framed in terms of one’s acting with the intention of benefiting another, without regard for the cost to one’s self.

    In evolutionary terms, altruism would have to pay for itself through a corresponding benefit to the organism acting altruistically; in ethical terms, altruism is only altruism if there’s no concern for a return on the altruistic behavior.

  • Anthrotheist

    I suppose that I see two fundamental issues with scientific study of morality.

    First, science is generally understood to be descriptive; it attempts to describe and explain phenomena that are observed through controlled interaction (i.e., experimentation). Morality on the other hand is by definition prescriptive and proscriptive, not descriptive. The point that the authors make about all of current science focusing on neurology and evolutionary biology illustrates this firs issue. It also points to the difficulty of the social sciences; what is the point of a statistical observation of social phenomena by sociologists without any usable lessons that can be gleaned from the effort? The advance of physical sciences allows us to more powerfully manipulate the physical forces and objects in our universe; I think it is right to be disturbed by the notion of following that same motivation for advancing social or moral study.

    The second issue I see is that morality is more like weather than it is gravity. We understood evaporation, vapor pressure, air density, thermal convection, and other principles of material science long before we began to accurately predict weather patterns (outside of looking at the last 100 years of records and setting a percentage chance of rain based literally on how many of those 100 prior calendar days had rain). Two things advanced meteorology more than anything else: satellites in space, and computers capable of running complex predictive models from massive repositories of historical data in near-real-time. What a space-level view of morality would mean is a bit beyond me, and modelling morality still falls back to the first objection I described: those models would have to be designed with judgments built into the mathematics, and run for an ultimate purpose of social-moral manipulation.

  • To be clear, I think that our moral response is largely innate and unconscious… part of our brain structure. I think our moral strictures are cultural inventions.

  • Carstonio

    Rudimentary moral behavior has been observed in other animal species, so it’s likely that the moral sense is an evolutionary adaptation that aids in species survival. We probably perceive the sense in the same way we see faces in inanimate objects like cars. (I detest the “evil jeep” aftermarket grill covers, which seem to embody the same spirit as the Punisher logo stickers.) As much respect as Francis Collins deserves for his work with DNA, he errs in his assumption that natural selection is inadequate to explain the origin of the moral sense – he’s using the argument from incredulity.

    But the question of what constitutes right and wrong is probably not empirically discoverable, since our concept of morality is shaped to some degree by social conditioning. I don’t know how one would prove that right and wrong exist as Platonic ideals. They might be more like a debate over who is the best rock guitarist of all time – a somewhat subjective question that still relies on shared principles.

    Looking at Haidt’s moral foundations theory, the three that conservatives throw in (loyalty, authority and sanctity) all have to do with hierarchy, which is how conservatism tends to define society. This may be related to research on brain structures that suggest that conservatives have a stronger sense of fear. (Which could explain why many of them confuse a visceral reaction of disgust, such as I might have to sauerkraut, with actual moral revulsion.) My stance is, if one’s argument for a particular concept of right and wrong is based on a concept of what is best for society, one is already conceding the liberal moral foundations of harm/care and fairness/proportionality, since the whole point of a society is mutual benefit.

  • The authors rightly point out that calling animal behavior moral or altruistic is problematic. Sure, there are behaviors that appear to benefit another organism at the expense of the acting organism; in terms of evolution, though, this requires some sort of corresponding benefit to the organism being inconvenienced. This is the problem with hypothesizing natural selection as the source of morality. We can propose all the selectionist just-so stories we want, but morality is a lot more than trade-offs involving resources and reproductive success. Some awareness of the ethical context of our behavior is essential in defining the behavior as moral.

    Saying that our concept of morality is shaped to some degree by social conditioning is a real understatement. As you say, the many hierarchies established in human societies—in families, communities, and societies—and one’s sense of place within them are staggeringly complex and yet they influence our moral sense in various ways for good and bad. And if we can’t even agree on what society is for, then any consensus about what constitutes morality is impossible. I can appreciate the need to find some basis for morality in reliable methods of inquiry, but I’m just as skeptical as the authors that it’s a realistic goal.

  • Carstonio

    Sure, there are behaviors that appear to benefit another organism at the expense of the acting organism; in terms of evolution, though, this requires some sort of corresponding benefit to the organism being inconvenienced.

    That’s overly simplistic. Natural selection is less about the survival of the individual organism and more about the survival of the species, and the two aren’t necessarily mutually exclusive. The benefit from the behaviors isn’t an absolute dichotomy either. A dog that would try to take all the kill for itself risks its long-term survival, because of the protection that the pack affords all its members, whereas a dog that takes only its share helps promote that protection.

    Very true that awareness of ethical context is essential. We can’t assume that species that exhibit rudimentary moral behavior lack any amount of that awareness.

    I’m really challenging Collins’ “God did it” assumptions regarding the origins of morality.

  • PD

    Hi Shem, I just wrote a long comment (what else is new) which got caught in the spam filter. Would you please release it from cyberlimbo? Thanks.

  • PD

    A few thoughts on the fact/value dichotomy you discuss..

    The commonly held view (at least since Hume) that there is a hard and fast dichotomy separating description/facts and prescription/values has been called into question by several interesting philosophers, including John Dewey and Hilary Putnam in the US. I think it’s reasonable to talk about a fact/value *distinction* in which the 2 can overlap or become, in Putnam’s words, “entangled.” (see his book The Collapse of the Fact-Value Dichotomy).

    One example of overlap is our use of certain *thick* concepts like rape or murder. Let’s say you’re reading a news report and come across a sentence like, “Jones was convicted on multiple charges of rape, one of which was perpetrated against an innocent 17 year old girl.” This is descriptive, of course, but the word ‘rape’ is simultaneously descriptive and evaluative. It’s a *thick term/concept.* Nor should this be terribly surprising. After all, we internalize norms and values in social contexts, in situations that can be described. Some of these situation (such as cheating on an exam) are described not just as factual events (“So and so cheated again!”) but as events deemed undesirable or in your words “proscribed”– where that also implies the prescription of honesty and independent effort as desiderata. We can easily observe as children that the teacher accusing a kid of ‘cheating’ is none too pleased. That is– cheating is discouraged or as you say, proscribed.

    We can distinguish factual statements like “There is a house on 1st and Broadway” and evaluative ones like “Jones is a bad person” largely because the latter is decontextualized. The reification of “goodness” and ”
    badness” as metaphysical properties in-themselves removes them from more practical and observable contexts. Ask what makes Jones so bad and you may end up hearing lots of descriptions of his behavior. Thick terms that catalogue his misdeeds might include descriptions of him or her cheating, stealing, committing fraud, vandalizing, spousal abuse and so on. These are descriptive terms which happen to encode both information about specific behavior and judgments *of* those behaviors. Sure, a subjectivist can argue that the behavior is “intrinsically neither good nor bad.” But the cultural relativist (like myself) has no such luxury.

    Re: radical subjectivism: We learn to function in society, I think, by internalizing norms, values, roles and the like. We don’t first learn a bunch of facts and then ‘subjectively’ come to evaluative conclusions for ourselves in isolation (I would argue); but rather we learn how to do things the *right* way relative to social context, as with Wittgenstein’s idea of Forms of Life/Language Games. We learn about rules that reflect the strongly held values, ideals and attitudes of those who raise us (family, teachers, friends, the babysitter, media etc.) We may later engage in practical reasoning within the languages we’ve learned and come to specific conclusions (whether to do x or y in a moral dilemma) more or less independently. But even this skill of practical reasoning is, as Aristotle says in his Nich. Ethics, the result of habituation until it becomes “2nd nature.”

    By pressuring the fact/value dichotomy (not dissolving it but relaxing it) we see that the whole topic is so complex that it’s really preposterous to suppose that empirical science can provide that “space-level view” of morality, as you rightly suggest. It seems that the best view is not from outer space but from *within cultures and subcultures* that we learn to understand rationally, emotionally and behaviorally at the same time so that we can get on with people and live hopefully productive lives. Over time, members of culture do things like reevaluate the laws, mores, customs, values and beliefs that constitute the normative sphere so that these things are not static and there is some kind of agency ( another complex topic). But the main point is that when we stop looking for morality in some Platonic realm of “ideas” or in essential definitions of “The Good” or perhaps some alleged “morality module” in the brain, and instead look at the concrete practices that cause moral upset, conflict (abortion, foreign policy, how to define a marriage, etc.), I think we’ll notice that values and norms are nested in situations that can’t be described without these thick terms, especially those that are important enough to be codified into laws (e.g.’murder’).

  • Bridging the gap

    Social science certainly can help us understand how morals evolve. Science can clarify how we are biologically prepped for certain ethics. Even that strong that some geneticists claim that our political preferences are partly predisposed. Frans van der Waal and other scientist have shown morals are not uniquely human. Therefore it is. Ery unlikely that morals need religion or God as their origin.

    Science itself is biased and is not unfallable. Social darwinism abused Darwins theory of evolution to justify European imperialism and later Nazism. For along time Anthropologists and genetecists supported those ideas as if they had a scientific justification. Even if they had to adopt data to make it fit the political theory.
    The same happened with Marxism. Marx original observations were bent into a killing doctrine.

  • Jim Jones

    We’re a very long way indee from understanding the human brain, however we don’t KNOW as yet that this is impossible. If it isn’t, it can be solved.

  • Paul Nedelisky

    Hi Shem, thanks for reviewing James Hunter’s and my book. Just read your response to Peterson’s comment and thought I’d address your concern about our dismissiveness towards moral pragmatism. Basically, this is because while a course of action can be called “good” or “bad” relative to achieving some goal, this doesn’t get us to what we understand as genuine goodness or badness. I think the example we use in the book is that an action can be “good” relative to the goal of robbing a bank, i.e., disabling the alarm system. But of course that doesn’t make it or the goal genuinely good. Put even more starkly, concentration camps may have been “good” relative to Hitler’s goal of exterminating Jews, in the sense that they helped him towards his goal, but they certainly weren’t genuinely good, in an ethical sense. Too often, the alleged moral guidance science offers is merely of this instrumental/relative-to-a-goal sort. So while we’re fine with speaking of actions as “good” or “bad” relative to their goals, we want to be clear that this isn’t genuine ethical goodness and badness, and hence is no substitute for addressing whether the goals in question really are good or bad. Hope that makes sense!

    All best,
    Paul