107 | Russ Shafer-Landau on the Reality of Morality
What this covers
Russ Shafer-Landau, Professor of Philosophy at the University of Wisconsin-Madison, and Sean Carroll discuss the case for moral realism—the view that objective moral truths exist independently of human attitudes and beliefs. Shafer-Landau defends non-naturalism, arguing that while moral facts are grounded in and supervene on natural features, they cannot be reduced to or explained by physics and chemistry alone. He frames the discussion around what he calls the "data of ethics"—strong intuitions such as that genocide is wrong, that some moral claims are true and others false, and that morality makes genuine, inescapable demands on us independent of self-interest. Shafer-Landau contends that moral realism best accounts for these data, whereas competing positions—nihilism, which denies moral reality altogether, and constructivism, which grounds morality in human attitudes—must explain them away or face internal problems.
The conversation moves through a range of challenges and responses. Shafer-Landau addresses the causal closure objection: if physics completely describes all events in the universe, what causal work can non-physical moral facts do? He also tackles the epistemological problem—how we could know objective moral truths if moral facts do not causally interact with us or submit to scientific testing. On constructivism, he argues that grounding morality in human consensus or construction drains it of normative authority, making moral knowledge implausibly easy to obtain and licensing repugnant conclusions. Carroll raises the question of how non-naturalism avoids dualism, and Shafer-Landau responds by invoking weak emergence—the idea that moral and mental features are legitimate ways of describing certain physical arrangements without being fundamental. The discussion also clarifies that moral realism stays silent on which normative ethics—deontology, consequentialism, virtue ethics—is correct, and that it does not entail theism. Throughout, Shafer-Landau emphasizes the role of reflective equilibrium and fallibility in moral knowledge, and defends the self-refutation test against scientism itself.
Shafer-Landau argues that moral realism—specifically non-naturalist moral realism, the view that objective moral truths exist independently of human attitudes and are not reducible to scientific facts—best accounts for the data of ethics, even though it cannot offer an algorithm for resolving moral disagreement.
- Nihilism and constructivism conflict with strong moral intuitions (e.g., that genocide is wrong) and either deny moral reality or make moral knowledge too easy and authority-less.
- Morality carries genuine non-self-interested normative authority that naturalist and constructivist views struggle to ground.
- Moral facts supervene on but are not reducible to natural/physical facts, and scientism cannot justify itself by its own standard.
Moral realism superior explains why moral claims are true or false and why morality binds us.
- Moral realism does a superior job of accounting for the data of ethics—such as that some moral claims are true and others false, that some people are morally wiser than others, that moral judgments are beliefs that can be correct or incorrect, that there are reasons to do what is required, and that morality's demands are inescapable and enjoy priority over self-interest—whereas rival views must explain these data away.
“when it comes to the data of ethics, moral realism does a fantastic job of accounting for them”
- Morality possesses 'normative authority'—genuine, non-self-interested practical reasons to comply with its requirements regardless of one's preferences or interests—but constructivist views (e.g., cultural relativism) sap morality of this authority, since an iconoclast would have no non-prudential reason to obey norms that are merely a culture's uninformed prejudices.
“if you have a view according to which morality is created by us and for us, then those reasons may be either strongly diminished or in fact non-existent”
- The main case against nihilism is that it requires denying clear moral intuitions—such as that perpetrating genocide or gratuitously smacking a vulnerable child is wrong—and these intuitions are as probative as any evidence available in philosophy; abandoning them as a source of knowledge would create insuperable difficulties for gaining philosophical knowledge generally.
“some of our moral intuitions about cases like the two I've just described are as clear and as probative, actually, as any kind of evidence we can have in philosophy. So if you take a nihilistic kind of view, you've got to give up on that source of knowledge”
- On moral realism, one's meta-ethical position does not entail any particular normative ethical system (deontology, consequentialism, virtue ethics)—realists deliberately stay neutral about content and focus on the status of morality—whereas on constructivism the meta-ethic does let one read off moral content (e.g., from a culture's agreed norms).
“Most moral realists have tried deliberately to be as neutral as possible about the so-called normative ethical issues... we realists are not in the business of trying to answer that question, but rather talking about... the status of that system”
- Holding moral realism is not arrogant or dogmatic, because one must distinguish affirming that an objective morality exists from affirming that one knows all its contents; realists do the former but not the latter.
“You gotta distinguish between, on the one hand, affirming the existence of an objective morality, and then, on the other, affirming that you've got all of its contents. So, we do the first, but not the latter.”
Constructivism is self-defeating because human consensus cannot make moral claims true.
- The core objection to constructivism is that no human being—individually or collectively, even the best moral philosophers—has the authority for their say-so to make moral claims true, because all humans are fallible, biased, and limited in processing information; this argument extends even to divine command theory, since if God's say-so cannot make something right, then human say-so certainly cannot.
“I don't think that my say-so makes it so when it comes to morality. I don't have that kind of authority... if even God's say-so doesn't make it so, then surely, we, imperfect, puny, little, fallible human beings' say-so isn't going to make it so, either.”
- A constructivist who judges another culture's morality as inferior does so by reference to their own society's standards, so the claim that those standards are morally superior is question-begging; the realist avoids this by appealing to an objective standard, though the realist incurs other costs.
“are the standards by reference to which you're rendering your judgements, are those standards in any way morally superior to the standards of my own society, and that's something that's I think going to be very hard for you to answer... that looks pretty question begging”
- On a relativist form of constructivism where morality is grounded in cultural consensus, moral knowledge becomes objectionably easy to obtain—one need only know one's society's mores and the relevant empirical facts—which both yields repugnant implications (e.g., that subordinating women would be morally right if a culture endorsed it) and makes moral knowledge too cheap.
“if you're a constructivist, it makes, perhaps paradoxically, it makes moral knowledge too easy to get. Most people don't think that that could be a problem with the moral view, but I do think it's a problem”
- The objection from moral disagreement is fundamentally epistemological—the worry that there is no way even in principle to adjudicate fundamental moral disagreements—but even in the worst case where such disagreements are intractable, this would not by itself impugn moral realism, which is in the first instance a metaphysical claim about what exists, not about whether everyone can be convinced.
“even if there were no way to do that, it's not clear to me why that would impugn realism, 'cause realism is in the first instance a claim, a metaphysical claim. It's a claim about what there is, namely moral reality, not a form of creation.”
Non-naturalist moral realism is metaphysically coherent and distinct from theism or scientism.
- Moral features supervene on natural/non-moral features—two situations identical in all non-moral respects must be morally identical—and a thing has a moral feature in virtue of its natural features, yet morality is an additional, irreducible layer that sits on top of the natural and cannot be reduced to it.
“the so-called supervenience of the moral on the natural or the non-moral is just what you said, namely that if two situations are identical in all non-moral ways, then morally speaking, they have to be identical... but I also think on top of that... there's morality that can't be reduced to any of that. It does depend on it.”
- Morality's status resembles chemistry: one can be a realist about chemical kinds (we discovered, not invented, hydrogen) while holding chemical facts are realized by, and in principle traceable to, subatomic arrangements; similarly, every particular moral fact is physically realized, yet what it is to be morally admirable or required is not itself a physical feature, marking an ontological division between moral and natural features.
“the so-called supervenience of the moral on the natural or the non-moral is just what you said, namely that if two situations are identical in all non-moral ways, then morally speaking, they have to be identical... but I also think on top of that... there's morality that can't be reduced to any of that. It does depend on it.”
- Non-naturalist moral realism is distinct from theism; one can hold that objective moral truths exist independently of human construction without grounding them in divine command or divine agency, contrary to a common assumption among non-philosophers.
“There's a very common thought among non-philosophers that if you're going to take up with a view like mine, you've got to ground morality in some kind of divine command or some kind of divine agency, but that's just a mistake”
- Within moral realism, non-naturalism holds that an objective moral reality exists but is discontinuous from scientific reality; moral truths cannot be discovered by laboratory methods or controlled experiments, even though scientific investigation can aid moral inquiry.
“morality is not really like the sciences. No matter how much scientific investigation we do, we're not going to be... We can be certainly aided in discovering what's right and wrong by reference to that, but the truths aren't scientific ones”
- Scientism—the view that all truths are in principle given by some science—is self-undermining, because the core scientistic claim that all truths can be explained by science is not itself a truth confirmable by science; if confirmable at all it would be confirmed by philosophy, so the assumption that everything can be explained by physics is itself in question.
“the core scientistic claim, namely that all the truths there are can be explained by science is not itself a truth that can be explained by science... It's one that, if it's confirmable... is to be confirmed by philosophy”
Moral knowledge proceeds through intuitive anchors, coherence-building, and fallible discovery methods.
- The proper moral epistemology starts with the intuitions about which we are most confident (e.g., genocide is immoral), builds a coherent network of beliefs from them, checks intuitions for distortion by prejudice or ignorance, and maintains a background openness to error; one can never be certain, but that uncertainty does not distinguish moral knowledge from knowledge in any other sphere.
“The place to start is with our very strong moral convictions and then, trying to build from them a coherent network of moral beliefs, all the while being open to the possibility that we're mistaken”
- Progress in a philosophical domain comes from first identifying the data of that domain—claims inquirers collectively have good reason to believe at the outset of inquiry, which are defeasible but require equally strong justification to overturn—and then constructing theories that meet criteria of adequacy in accounting for that data.
“the way to make progress in gaining understanding of a given area in philosophy is by first trying to identify the data within that domain, and then constructing theories that have to meet various criteria of adequacy in the attempt to account for that data”
- There is no determinate method or algorithm for making progress in moral disagreements—just as there is arguably no genuine 'scientific method' beyond vague generalities—so realists can only offer anodyne guidance (keep an open mind, learn the relevant facts, avoid judgment-clouding emotions) rather than any algorithm for moving from current intuitions to settled high-confidence views.
“I don't think that there is a determinant method by which we can always make progress in our discussions with one another about what's morally right or wrong... I don't have anything remotely algorithm-like to offer”
- Whether constructivism or realism is correct, the method for engaging those who don't share our views is the same: appeal to shared bases the discussion partner is open to or already accepts, since finding common ground makes consensus more likely.
“The only way you can try to make progress there, I think, is to appeal to things that your discussion partner, some things that they're already going to be, at least, open to hearing and ideally will accept.”
- Intellectual virtues of humility, modesty, and recognition of fallibility sit best with moral realism, because—as with a physicist facing a cosmos not of their making—appropriate humility follows from believing there is a truth one is trying to discover rather than make up; this makes it harder for a constructivist to consistently profess fallibility.
“the explanation of why those are the appropriate attitudes to take is because you don't get to make up the truth. You're there to discover a truth not of your own making. I think the same is the case when it comes to morality.”
The epistemological objection to non-naturalism—that we cannot access objective moral truths—is realism's hardest challenge.
- A sharper objection to non-naturalist realism is that if it is true there must be a plausible epistemology explaining how we access objective moral truths, yet there appears to be no such epistemology—since on non-naturalism scientific means are insufficient and there is no moral object that impinges on our senses to correct us, unlike walking into a table to refute external-world skepticism.
“When it comes to the moral world, there's no... Nothing that's going to impinge on your senses directly... that's going to give you the sort of check, you get to indicate that your moral intuitions or your moral view, more generally, is mistaken.”
- Causal-closure or causal-exclusion arguments risk excluding too much: if everything is ultimately explained by fermions and bosons, then beliefs and desires also lose explanatory room unless they are nothing but arrangements of fundamental particles—a view few in philosophy of mind hold—so jettisoning moral features on these grounds may force jettisoning much one is unwilling to give up.
“I'm a little worried about causal closure arguments or causal exclusion arguments, because I think they're going to, in the end, exclude too much... that leaves little room for beliefs and desires”
- Some people simply will not see reality—moral or otherwise—and cannot be argued out of their errors, due to forces like self-interest and ideological indoctrination, so the failure to persuade someone (even after exposure to films, memoirs, or narratives) does not show there is no moral reality they are missing.
“It's possible for someone to just be blind to the moral equality of a woman or an African-American and there's nothing you can say... they may bring their blinders... to the reading or to the film watching, and they may be unconvinced in the end.”
Nihilism is incoherent because nice nihilists' preferences contradict denying objective morality.
- A nihilist can coherently be a 'nice nihilist'—preferring that people not commit genocide or harm children and acting decently—provided they refrain only from claiming that their preferences register genuine morality or immorality; the inconsistency charge against nice nihilism fails.
“I myself actually don't think there's any inconsistency in being a nice nihilist... All they're forbidden from saying, so to speak, so as to maintain consistency is to say that their preferences register some kind of genuine immorality or morality.”
- Meta-ethics is a branch of moral philosophy that, rather than answering what is actually right or wrong, steps back to ask about the nature of moral reality, how we can know it, and the nature of moral judgments (e.g., whether they are beliefs or desire-like states).
“meta-ethics is not trying to... It is a branch of moral philosophy, but it's not trying to answer questions about what actually is right or wrong... rather it's trying to take a step back and ask about the nature of moral reality, if there is one”
- Meta-ethical positions can be divided into three big families: nihilism (there is no moral reality), constructivism (moral reality exists but is constructed by human attitudes individually or collectively), and moral realism (a moral reality exists objectively, awaiting discovery, not made by humans).
“there are three really big families of views. One's a view that we can call nihilism... Another view is a view according to which there is a moral reality, but the reality is a constructed one. These folks are called constructivists... And then lastly, there is the family of views known as moral realism”