People intuit very strong feelings about right and wrong and good and bad. If we see a stranger help an old woman across the street, we think to ourselves: “that was the right thing to do.” Or if we see a child reach out and touch a hot iron and recoil in pain, we think: “that was a bad thing for that child to experience.” These statements of normativity and value are useful shorthands in colloquial dialogue, but it is a mistake to believe we are making any claims about reality. In this paper, I will argue that such claims are mistakes; that is, I will defend anti-realism about morality, the position that there are no facts about right and wrong (normativity) and good and bad (value). I will first review the two strains of realism, reductive and non-reductive realism, explaining why the problems they face are strong reason to not believe in them, and then defend the anti-realist position. My argument for anti-realism follows the structure:
P1. Reductive realism is false (history of failures, just-too-different intuition)
P2. Non-reductive realism is false (access argument, argument from queerness)
P3. Realism is either reductive or non-reductive
C. Realism is false (equivalently, anti-realism is true)
Be careful not to conflate my definitions of realism and anti-realism with the standard ones, which typically invoke the notion of mind-independent or objective facts about reality. I call theories which posit mind-dependent facts about reality realist theories. What I mean when I say anti-realism is simply that moral facts do not exist, mind-dependent or otherwise. Note that under these definitions, I consider relativist and ethical subjectivist positions to be realist, while in most philosophical discussions, they are considered anti-realist because they describe moral facts as mind-dependent or non-objective.
First, I will explain reductive realism and its problems. Reductive realism posits that moral facts exist and can be reduced to science. By ‘reduced’, I mean that these moral truths can be identically redescribed in scientifically acceptable terms. A simple form of utilitarianism is often interpreted as a reductive realist theory: the right action produces the greatest amount of happiness for the greatest number of people. In the naïve interpretation of this theory, happiness can be reduced to a brain state, so whether an action is right or wrong can be scientifically measured based on the neural information of the population.
Reductive realism is an attractive position to many because it preserves the idea that scientific facts are all that exist (i.e., that the scientific image captures all of reality), while also making room for the intuition that moral facts indeed exist.
With that gloss of reductive realism, I will now discuss two problems with it: the history of failures with reductive realist theories and the just-too-different intuition (Enoch, 2011).
I argue that reductive realist theories consistently expose themselves to counterexamples and that despite thousands of years of philosophers attempting to formulate robust reductive theories, counterexamples keep cropping up. Because of this, we should cast doubt on the effectiveness of reductive theories at capturing our notions of morality. To illustrate this point, I will now go through some reductive theories and their pertinent counterexamples.
Act utilitarianism: Let’s first return to the simple utilitarian theory I offered above where right actions maximize pleasure for the greatest number of people. Here, pleasure is reduced to measurable brain states, which are a part of the scientific image.
Problem: This theory faces the organ harvester problem. Suppose we have an array of five people who each have distinct failing organs (one person needs lungs, another a kidney, etc.) and will die unless they receive new organs. There is an organ transplant shortage, so these people will surely die soon. Under this theory, a doctor should kill one healthy person, harvest their organs, and use those organs to save the five dying patients because it maximizes happiness across five people’s lives in return for the loss of happiness of one person’s life. This conclusion strongly diverges from our moral intuitions, indicating that this theory’s reduction fails to capture the moral facts we intuit.
Subjective response theory (simple subjectivism): Under subjective response theory, the normativity of actions is determined by a tribe consensus about whether the action is right or not. This is a reductive theory because the normativity of a given action can be reduced to the observable reactions and attitudes of the community toward that action (either through a vote or through brain states, which can be located in the scientific image).
Problem: Imagine a community of psychopaths who all agree that killing unhoused people for fun is right. Say these psychopaths derive pleasure from the thrill murder gives them and as psychopaths, they don’t care at all about the pain they cause to those they attack. Even though there is a communal consensus about this normative statement, we outsiders look in and intuitively believe that despite the consensus, what they’re doing is still wrong. As such, the theory fails to locate our moral intuitions in something scientific.
Normative response-dependent theory: In this theory, you assign an action (or object) X to a property P (e.g., right or wrong) iff an idealized agent under idealized conditions would have response R to X. The agent’s response R explains the property P; it is the fact that the ideal agent under ideal conditions has a given response that constitutes the property X is assigned.
Problem: This theory pushes back the ought rather than giving a good explanation for it. The further fact about where the normative property P comes from can be traced back to the idealized agent’s response. However, the theory lacks a justification relating the fact about the agent’s response to the property P i.e., what makes the agent’s response normatively authoritative.
There are many more reductive realist theories which each have their respective problems. These cases are not meant to decisively knock down reductive realism. However, the fact that there are many, many such theories and each of them has issues that highlight misalignments with our conception of morality is reason to disbelieve the idea that we can reduce discussions about morality to scientifically acceptable terms.
A more decisive argument against reductive realism is the just-too-different (JTD) intuition, which I will cover now.
A helpful way to understand the JTD intuition is to start with Hume’s is-ought gap. Hume argued that you cannot derive a moral conclusion (an “ought”) from a scientific description (an “is”). He noticed that to make the leap from descriptions about what is the case to what ought to be the case, you have to include an extra premise. For example:
P1. Burning a cat for fun causes pain to the cat
P2. You ought not to cause pain for fun (missing premise)
C. You ought not burn a cat for fun
Notice how making the jump from P1 to C without including a justification for P2 feels wrong since P2 seems like a categorically different fact than P1. This is useful mainly as a set-up, since defenders of reductive realism have responses to their exclusion of P2, such as the idea that P2 isn’t expressing something new but is instead just a redescription of P1. Remember that under reductive realism, the claim is that moral facts can be reduced to scientific ones, meaning that both the moral and scientific facts are equivalent redescriptions.
I will now explain the real critique, which was proposed by David Enoch.
The JTD intuition rests on the idea that there are some features about normative properties that can’t be captured in naturalistic descriptions. Take some normative property, like being right, and some natural property, like being pleasure-maximizing; the realist hopes to reduce the former to the latter. It is not simply enough for the anti-reductionist to say prima facie that normative and natural properties are different. The reductionist can respond with the example of water. It’s not obvious prima facie that H₂O and water are identical, though it is in fact the case that water reduces to H₂O. However, there is a special quality about moral facts that makes them different from other reductions. If someone were to ask, “is water H₂O?,” the answer is, very obviously, yes. Once this reduction has been found, it becomes immediately clear that water is the colloquial description of H₂O, the chemical description of the same thing. However, if someone were to ask “is it right to do things that are pleasure-maximizing?,” the answer to that question is not so immediately obvious. The conceptual distinction between something being right and it being pleasure-maximizing is so divisive that it seems categorically wrong to try to reduce the concept of rightness to something that pertains to the physical world. When tracking the idea of rightness, it seems that we will not be able to find some physical redescription such that there is this immediate, obvious consensus where we all agree that the moral quality of rightness reduces to some physical property.
Moral qualities also have this quality of pushy-ness, which seems just too different to assign to observable facts (Mackie, 1977). By pushy-ness, I mean that the fact guides you to do a certain action (i.e., you should do things that are right). This pushy-ness seems otherwise absent from the physical world. I discuss this pushy-ness in more detail with respect to non-reductive realism in Section 3, but a parallel version of the critique can be leveled against reductive realism:
P1. Moral facts have the property of being pushy
P2. No physical facts could have the property of being pushy
C. Moral facts are not physical facts
These failures make reductive realist theories much less attractive. But maybe when we make claims about right and wrong, we are really making claims using further facts about reality, beyond what science can give us. That is what the non-reductive realist hopes for.
I will now explain non-reductive realism (NRR) and the problems it faces. As a non-reductive theory, NRR posits some further facts about reality that the scientific image does not capture. In doing so, the non-reductive realist loses their clean picture of reality as filled with only scientific facts, but they retain the intuitive notion that moral facts do exist in a more convincing way than the reductive realist.
Under NRR, an action is wrong iff
The same form of these conditions can be applied for determining whether an action is right, good, or bad.
There are two approaches to defining NRR that are very similar: using a special form of ought or adding these additional clauses, as I’ve done, that capture the severity of wrongness. Let me motivate these additional clauses. Say your goal is to reach work as fast as possible. Suppose that despite this goal, you decide to avoid taking the highway to work, increasing your commute time by 30 minutes. If your goal is to get to work as fast as possible, you ought not to avoid taking the highway. However, this is a different type of ought than the ought we mean when discussing things related to morality. You ought not to avoid taking the highway is very different than saying you ought not to bully a child. This distinction between morally serious and unserious oughts is what those clauses account for.
I will address the common form of NRR which includes grounding, the idea that actions ground their normative element. So a bad action grounds an ought-not-to-do-it-ness. For example, the fact that you ought not to hit people would be grounded in the fact that hitting them causes them pain. Here, we are not deriving the ought from the fact about what is; they are distinct but also inseparable facts. There is a metaphysically necessary instantiation of the ought-not-to-do-it-ness if the bad action exists. A helpful way to think of this is like a bijection, or pairing, between a set of facts about what is the case and a set of facts about what ought to be the case — they are separate sets of facts but necessarily linked. This is unlike the reductionist theory, in which both facts are identical.
I will now discuss common problems levied at the non-reductive realist, specifically the access argument (inspired by Harman, 1998) and the argument from queerness (Mackie, 1977). Another strong argument is the supervenience challenge, which I leave for the reader to independently explore.
The access argument essentially argues that because moral truths are unobservable and don’t help us explain what we do observe, we should not believe in them. It goes like this:
P1. If NRR is true, there are facts about normativity and value.
P2. If so, we should be able to know such facts and reliably have beliefs about them.
P3. If we know these things, it can only be by observation or inference to the best explanation (IBE) from observation.
P4. We cannot know moral truths through observation or IBE.
C. NRR is false.
I will now defend the key premise P3. The heart of this argument is in epistemology: that the scientific image is complete and we can only gain knowledge of moral facts through empiricism. Our cognitive faculties evolved to know things through perception and inference and track things in the physical environment, and moral facts don’t seem to be trackable in that way. Harman’s claim is that moral properties are neither observable nor do they offer the best explanation for what we observe.
Be careful to note that Harman’s claim is not that we can only know things through observation alone; you can know things through IBE without observing them. Take the fact that we know negatively charged electrons exist. We can’t observe electrons, but we have a good reason to believe they exist because they explain other phenomena we do observe (e.g., the photoelectric effect, why cathode rays bend in a magnetic field, etc.).
Though we cannot observe them, moral facts also don’t provide the best explanation for what we do observe. Imagine the case of a child deliberately burning the neighbor’s cat for fun. We can observe that the child is laughing and also that the cat is yelping and running away. As such, we can use IBE to infer that the child is causing the cat pain for fun. However, we cannot infer that the child ought not cause the cat pain for fun because that adds no explanatory value to the scientific observations we make (e.g., about the kid’s action or people’s reactions, which can be explained with complete scientific knowledge without needing to invoke morality). The non-reductive realist’s picture of reality includes this other distinct ought-not-to-do-it-ness, but Harman argues that we should not accept this because we have no reason to believe such a property exists. Facts about normativity and value don’t relate to particles or observable reality — they’re something we try to attach onto reality after-the-fact.
As the argument goes, since IBE is the only possible way for us to learn of moral truths and we cannot do so (though we should be able to), moral truths must not exist.
I will now review two counterarguments to the access argument, the weaker of which I will briefly overview (Sturgeon) and the other of which I will spend more time investigating.
Sturgeon’s counterargument concedes the IBE model of epistemology but argues that we can in fact figure out moral truths using IBE, challenging P4. Take the case of the Holocaust. Realizing its tragedy, we observe that almost everyone condemns Hitler. Under Harman’s position, all we would be able to derive from this observation is that people condemn Hitler. However, this assertion may not entirely capture the notion of what we observe because it deflates Hitler’s actions and the mass hatred for Hitler into a fact about psychology. To account for the severity of peoples’ disgust and its universality, as well as the horrific things that Hitler did, the best explanation might instead be that Hitler was morally depraved. The moral fact is a better explanation for Hitler’s actions and the strength and universality of people’s reactions, since we might not expect to see such horrible things being done or the succeeding strength and universality of disgust if the fact was simply psychological. A standard reply to this argument is that if we dive into the details about Hitler’s upbringing and psychology and did the same specification for all the people who react the way they do, introducing the moral fact doesn’t add any explanatory value. It is instead simply an extra label of the natural facts that, under IBE, has no justification.
I will now address the stronger principles reply to Harman’s access problem, which takes issue with P3. Proponents of the principles argument posit that they can gain knowledge of moral truths a priori, meaning that we have good reason to believe such truths without observation of them. They provide evidence for this claim by giving examples where IBE fails and where rationality has to step in to give us knowledge. Take the example of determining whether a water bottle is “grue” (green or blue). We can’t observe “grue” when we look at the water bottle, but we can ascertain whether it is green or blue using principle-based rationality: if the bottle is green, then it is grue, or if the bottle is blue, then it is grue (logical disjunction from propositional logic). Another example is that when a car knocks over a stop sign, you don’t believe that it’s a car because that would be the best explanation for the stop sign falling over (the sign could just be old, or maybe there was a storm); rather, you deduce that the object is a car because it retains certain properties (e.g., having four wheels, having a steering wheel, etc.). In the same way, when you observe a child burning a cat for fun, it doesn’t have to be the case that the fact of that action’s wrongness comes from IBE. Instead, you could use a principles-based model:
P1. The child burning the cat for fun is a case of deliberately causing pain for fun
P2. If action X deliberately causes pain for fun, you ought not to do X (a priori fact that we can know)
C. The child ought not burn the cat for fun
Thus, we can determine whether an action X is wrong by reasoning about it from certain a priori moral truths we do know. However, it feels equally as implausible to believe that we somehow have a priori knowledge of these foundational moral principles. Our moral faculties developed as a result of evolution, culture, and social interaction; even though we can know things a priori, it would be strange to think that our minds are equipped to discover facts about morality through reflection alone. Given that we perceive the existence of moral facts, it is much more plausible that those intuitions are a product of evolutionary and social factors.
An analogy to math helps make belief in a priori moral facts more palatable for the realist. Most of us accept the axioms about math and take statements like 2 + 2 = 4 to be true. To believe that 2 + 2 = 4 is true, you must commit yourself to non-reductive realism about math since we are unable to observe in the world this abstract statement about addition. We might be able to observe the fact that moving two apples to a group of two apples produces a group of four apples, but that doesn’t prove the proposition about the relations of abstract numbers; regardless, we intuitively hold this abstract proposition to be true. If we already hold a priori facts about math to be true, holding a priori facts about morality seems less extraordinary.
This also gives trouble to the anti-realist: if they don’t accept a priori truths about ethics, then perhaps they shouldn’t accept a priori truths in math. One theory, called nominalism, does exactly this; nominalists argue that 2 + 2 = 4 is indeed not a fact about reality but instead a useful construction we developed to make sense of the world and achieve practical advances in math and science (Field, 1980). The idea that math is a useful fiction is perfectly plausible, especially considering that math is scoped within fundamental axioms that mathematicians take to be true without justification; however, it does leave you with the weird, flashy perspective that math does not give us facts about reality.
Yet, the anti-realist isn’t forced to give up math if they disbelieve in a priori moral facts, and a belief in math might not provide the realist as much comfort as they might have hoped; the analogy between ethics and math breaks down. Math has significantly more consensus about its axioms and conclusions than ethics does, so it is much more plausible to think that we have a priori knowledge about facts in math than about facts in ethics. Mathematicians across millennia have converged on ideas and results and points of disagreement can generally be resolved through proofs and rigor. Ethics feels categorically different: the range of things that were moral and acceptable in ancient civilizations (cannibalism, ius vitae necisque, incest, pederasty, etc.) that are not today is vast, and even modern civilization has undergone significant ethical revolutions (slavery, women’s rights, etc.). If we had a priori access to foundational moral facts, it seems unlikely that we’d see such dramatic disagreement and moral turmoil. So it seems that the anti-realist may not have to give up math and that the realist has the burden to explain how such profound disagreement exists given our ability to know a priori moral facts.
From the principles argument, we address the idea that gaining knowledge about normative statements is difficult and of a priori moral principles, likely impossible, if such statements exist at all.
I will now discuss Mackie’s argument from queerness, which is a more formal support to the intuition that this invisible moral quality does not exist.
Mackie delineates the queerness of moral facts into two kinds. The first kind is an epistemological queerness, which posits that it would be strange for our mental faculties to be able to track non-natural moral facts; I tangentially discussed this above and won’t cover it again to avoid redundancy. The other kind, which I will address now, is the metaphysical queerness, which posits that moral facts are quite different from the other facts about reality.
If there were facts about morality, they would be of a strange sort different than all the other facts about reality. Specifically, Mackie argues that moral facts have this pushy-ness unlike any other fact about the world. The properties of being blue, being negatively charged, or being water aren’t pushy properties; they don’t invoke a certain kind of action as a result of them being true. Moral facts are categorically different: if an action has the property of being wrong, you are by-definition being pushed to not do it. When thinking about what facts exist and don’t exist, it seems odd to have this special category of pushy facts when everything else we know to be true is not pushy. Since moral facts would be unlike any other facts we’ve ever encountered in that they have this action-guiding property, it would be rational to doubt their existence and preserve a harmonious ontology (a clean theory of what exists).
Mackie’s argument operates in the context of a broader commitment to a naturalist ontology, meaning that his picture of reality only contains natural facts and entities. Under naturalism, there is no possibility for non-natural, ‘queer’ moral facts (given that we can’t turn to reductive realism). I discuss reasons to stick to a naturalist ontology at the end of Section 4.
As I outlined at the onset, because the problems with realist views are too great, we are left with anti-realism, the view that there are no facts about right and wrong, or good and bad. Remember that the definition of anti-realism we are dealing with does not contain ‘objective’ so I’m grouping mind-dependent facts about reality under realist worldviews. Under this definition, the two primary strains of anti-realism are noncognitivism and error theory.
Under noncognitivism, moral statements equate to emotional expressions. So moral statements like “causing pain for fun is wrong” is a linguistic shorthand for something like: “causing pain for fun!” with a disapproving tone. This circumvents the need to have moral facts by instead introducing the idea that moral statements aren’t really about truth but about expressions of emotion. This nicely retains our use of moral language and provides an explanation for the pushy-ness of moral language. Though it is outside the scope of this paper, see the Frege-Geach problem for a challenge to this view.
Under error theory, all moral statements are systematically false since they refer to non-exist properties (like wrongness, goodness, etc.). Since wrongness doesn’t exist, the error theorist argues that the statement “causing pain for fun is wrong” is systematically false; this is typically argued for using Russell’s theory of descriptions, which I will briefly explain now.
Take the sentence: “the present king of France is bald.” There is no present king of France, so under Russell’s theory, the sentence is false. Russell posits that a statement like the “the present king of France is bald” is really a conjunction of three different propositions:
If any of those three propositions are false, then the entire statement “the present king of France is bald” is false as well. It’s not exactly the same structure, but the same logic can be applied to moral statements. If you say “causing pain for fun is wrong,” that presupposes the existence of wrongness, and since that presupposition is false, the entire moral statement must be false as well.
It is not obvious that we would develop such strong intuitions about morality if there were no facts about it, so anti-realists often provide an account for where these mistaken intuitions about normativity and value come from. One theory is that we evolved moral talk to help us cooperate with each other and maximize our survival. With moral talk, we can maintain a cohesive social order and control the actions of each other without actual physical control or supervision.
The inconsistency of our moral intuitions might be best explained by evolution rather than facts about reality. Take the famous argument put forth by Peter Singer in his “Famine, Affluence, and Morality.” We have a strong moral intuition to inconvenience ourselves to save a drowning child but not the same burning intuition to inconvenience ourselves to donate to UNICEF and give starving children food. Our moral intuitions don’t persist for people we are detached from, physically or otherwise. This might be explained by group selection in evolutionary theory; groups survive, so we care about others inside our groups but not those outside of the group. This might be reason to cast doubt on our moral intuitions or at the very least, provides a solid explanation for why they exist in the first place (rather than the alternative explanation that they are tracking facts about reality).
Now I will discuss two main arguments against anti-realism: the permissibility problem, and the companions of the guilt argument (with an emphasis on the epistemological problem). These problems can be levied at both noncognitivism and error theory.
First, I will discuss the permissibility problem. This is unintuitive byproduct of anti-realism. It goes like this:
P1. If an action X is not wrong, then X is permissible. (If X is wrong, then X is impermissible.)
P2. Under anti-realism, X is never wrong (since the wrongness doesn’t exist).
C. Under anti-realism, X is permissible.
This is a formalization of the main gripe people have with anti-realism. It is true that under anti-realism, you cannot say that the Holocaust was wrong and in fact you must say that it was permissible, which obviously provokes a visceral reaction of disgust in people. It seems that you should be able to label certain terrible actions as wrong.
One response is to be okay with simply expressing disgust or anger towards things you would have otherwise described as wrong. That might be satisfactory: you can’t say that the Holocaust was wrong, but you can condemn and decry it and express your disgust towards it.
Another response is to argue, unlike the error theorist, that normative statements are neither right nor wrong. Proponents of this view take on a different linguistic theory than Russell’s theory of descriptions, which is outside the scope of this paper, and accept the notion that moral statements are neither true nor false (i.e., such statements have a truth-value gap). They avoid the permissibility problem by claiming that P2 is neither true nor false which stymies the conclusion.
Now I will discuss what I consider to be the strongest argument against anti-realism: the ‘companions in the guilt’ argument. I will specifically focus on how eliminating oughts in ethics might force anti-realists to eliminate oughts in epistemology. The basic argument is that normative facts in epistemology have the same features as normative facts about morality, so arguments we can levy against non-reductive realism about morality can also be used to undermine non-reductive realism about epistemology, specifically arguments like Mackie’s argument from queerness. When I say normative facts in epistemology, I mean things like “you ought to hold true beliefs” or funnily, “you ought to believe in anti-realism.” If we find normative facts in morality metaphysically weird, then we should find these same categorical oughts in epistemology just as weird; they are pushy, irreducible to science, and ‘queer.’ Because normative statements in epistemology are companions in the guilt and face the same problems moral facts do, we should leave them behind and say that such facts don’t exist either.
There are a few responses to this critique. One comes from Quine’s tradition of naturalized epistemology, in which he reframes normative statements as conditional or hypothetical rather than categorical. So rather than saying something like “you ought to hold beliefs that line up with the evidence,” you might say something like “if you want to predict the world accurately, you ought to hold beliefs that line up with the evidence.” As such, you can do away with the categorical oughts that feel metaphysically special and reframe the oughts in a much weaker sense that is purely instrumental and constrained by our desires.
Another response is the quietist one, in which you simply leave out all use of oughts in language. Instead, you simply state whether beliefs are reasonable or good predictors of the future, and you hold beliefs that fall into those categories without prescribing that others do the same.
All of these responses try to cling onto a naturalist ontology that in the spirit of Mackie and others, try to eliminate any facts that don’t neatly fit into science. When faced with a set of facts, ontological naturalists try to reduce those facts to science and if impossible, throw them away. Anti-realists have to decide whether in trying to cling to this ontology they are losing too much: moral normativity, epistemological normativity, maybe math, and more! For some this may be a cost too great and an unnecessary commitment to science.
I argue though that this tradition of naturalism is worth keeping around. A full defense of this naturalist ontology is outside the scope of the paper, but I will provide a very brief defense here since it is why I lean towards anti-realism.
Every physical fact has a sufficient physical cause (causal closure hypothesis), so if non-natural facts were to exist, they should either have causal powers and intervene in the natural world, which we have no evidence for (physics explains all), or they have no causal powers at all. In the latter case, we have strong priors to believe that we would not be able to access such supernatural facts since our mental faculties evolved to gain knowledge from experience and IBE, and the a priori existence of moral facts about reality feels both anthropocentric and unlikely given moral disagreement. Naturalism cleanly explains reality, and we should defer to simpler explanations of what is. It is more probable that our moral inclinations are evolutionary and social than that they are queer facts about reality out of distribution from our understanding of everything else.
So what now? The anti-realist position is certainly unintuitive. We use moral language every single day, so does a belief in anti-realism mean we have to give all of that up? I argue not. We would be mistaken to believe that we are talking about things that exist, but regardless, we must use moral language; it is deeply ingrained into society. Anti-realists can think of colloquial use of moral language as a linguistic shorthand for emotional expressions, like in the non-cognitivist tradition, or as a pragmatic fiction that helps us communicate, which is more in line with the error theorist’s approach. Though this is an unintuitive position to hold, it captures the sentiment that moral language doesn’t actually have any bearing on reality.
You can reinterpret your day to day moral judgments with renewed skepticism — what you may have previously thought to be undeniable rights and wrongs are suddenly less dogmatic. The force of moral claims, like that stepping on a dog’s tail for fun is wrong, suddenly loses much of its gravity. This is a bitter pill to swallow, but it in no way tilts you toward doing things that cause people harm. With the intuitive goals of human prosperity and preserving human dignity, you can retain your oughts (as conditional ones) and live your life in pursuit of such goals. But while doing so, you must realize that those goals are not baked into reality or what everyone should necessarily be striving for. It’s humbling and existential, but a step towards truth and away from being blindly driven by biology and culture.