Partial moral relativism?

Maybe morality doesn't converge everywhere.

Moral relativism states that moral truths are culture- or agent-dependent. Moral realism, on the other hand, states that moral truths are objective facts, similar to the physical laws of the Universe. Moral relativism is not the same as moral nihilism, which states that there are no moral truths whatsoever, and which I believe to be contradictory (if there are no moral truths, then two contradictory moral propositions would be equally false, independently of context, which, by virtue of the law of excluded middle, is impossible). However, it is a form of moral antirealism, in the sense that it rests on the idea that moral truths are subjective constructs and not objective facts about the world. They could be the result of a social contract, emerge from environmental and cultural constraints, be the work of a group that coerces others, or any combination of these.

I am not a moral antirealist. I believe that moral facts exist independently of any mind, but that they require minds to be instantiated. In this sense, it’s the same as saying that colors exist as a property of the universe, but require a perceiver to be instantiated: if there is no eye and no brain, then all that “exists” are the various wavelenghts of light. Colors only emerge when there is a combination of wavelenghts, eyes, and mind.

I used to think that this implied that all moral agents would converge towards a single set of moral truths. However, I recently realized that this may only be partly true. Let’s see how.

Intuition pump #1: The Sentient Rocks.

Suppose a species of intelligent aliens derives great pleasure from losing agency. Their goal is to become as similar to rocks as possible, in fact going as far as digesting their own brains like tunicates once they reach reproductive age, in order to live as almost non-sentient entities, fixed to some cliff, only experiencing a state of permanent pleasure with no intentional object. Suppose, moreover, that this strategy is evolutionarily advantageous for them — for instance, they reproduce like mushrooms, via sporogenesis, after they become “rocks”. Suppose this process produces lots of offspring that will live short, but relatively happy lives, until they reach reproductive age and become non-agentic too, thus completing the cycle. Finally, suppose that there are no threats to them in their corner of the universe, so this process can be sustained as long as their sun burns.

Now, humans value agency, and they perceive being deprived of it as a grave attempt to their integrity and values. In fact, depriving a human of agency without their consent (and, in most cases, even with their consent), is perceived as a great moral wrong, and rightfully so: since we are a species whose survival and reward systems literally depend on being agentic. This seems to imply, in most moral systems, that depriving a human of their agency is morally wrong. We perceive it as morally wrong even when it creates great pleasure, as can be attested by the fact that Brave New World is classified as a dystopia despite the fact that almost every one of the inhabitants of that world is living a happy life, free of disease, free of the fear of death, and filled with simple, unsophisticated, easy to obtain pleasures. The reason why we find it abhorrent is that they don’t have freedom, and depriving someone of their freedom is immoral.

The question, however, is: would it be morally wrong to take one of the aliens described above and force their transition into a “rock”? Remember, the consequence for them would be that their fundamental desire to become one would be fulfilled, that they would be able to reproduce, and that lots of blissful offspring would be created. Would it be immoral if a surgeon opened their skull and removed their brain? What if the surgeon were one of them? It seems highly unlikely.

“Ah, but you are trying to fool us! You thought we wouldn’t notice that you are still relying on a pain/pleasure scale and a hedonistic Utilitarian framework to evaluate the surgeon’s actions! This is just classical Utilitarianism in disguise!” you would exclaim. Indeed. Let’s see if I can wriggle my way out of this objection.

Intuition pump #2: The artsy Vulcans.

Suppose a species of Vulcans, as defined by David Chalmers in Reality+ and in this article. Vulcans are beings that are conscious, but have no hedonic states at all. They don’t perceive pain or pleasure, they don’t feel sad or happy, and they don’t value their own life. However, they can still reason, act, and value things without ascribing any hedonic value to them: they simply see them as “important”, in the same way a computer may prioritize or deprioritize the execution of a program. Suppose, like in the first intuition pump, that they are spatially isolated from the rest of the Universe, so that they have no influence over any sentient beings, who could indirectly instantiate a pain/pleasure scale.

Suppose that a group of these Vulcans started valuing, somewhat arbitrarily, an extreme form of art, which involved ending their own lives in various ways. Note that this is quite unlike what human daredevils do, which consists in defying Death by performing heroic actions (like, say, climbing a skyscraper on a live show), thus affirming the sublimity of the “human will”, etc. In the case of the Vulcans, it would be an entirely pointless act, with no other justification than “deciding to do it”. I cannot stress this enough: there is absolutely no external justification for the Vulcans, it’s a purely arbitrary act. If a human decided to do this “for no reason”, without being depressed and without facing an otherwise unbearable condition, we would see them as mentally deranged, obscene, and probably also abhorrent. Deontologists would scream “thou shallt not treat thyself as a means to an end!” Virtue ethicists would seethe: “how dare you forsake all measure in pursuit of such a silly thing!” Utilitarians would weep, thinking about all the expected value that is being squandered. Negative Utilitarians… wouldn’t care much, admittedly, as long as no suffering is involved, but they would probably warn against second-order effects.

However, none of these objections would apply, because these Vulcans, unlike humans, do not value their own life and don’t find anything at all pleasant or unpleasant, just valuable in some abstract sense, like a computer might switch the state of a bit from 0 to 1. Concepts like “expected value”, “measure”, and “suffering”, all depend on our ability to value things on the hedonic scale.

The only things that might have some weight in Vulcan morality are “ends”, and if Chalmers is right in stating that consciousness is what matters, then they would be acting wrongly by ending it in themselves. The issue, here, is that they don’t value themselves! The whole species is unable, by construction, to value the continued existence of its members. The closest comparable species on Earth would be something like ants, but even ant colonies value survival at the colony’s level. So, why would Vulcans care about what Chalmers, a sentient human, has to say about whether their hobbies are moral or immoral? In fact, it is highly unlikely that anyone in Vulcan society would care even if the “artists” started promoting their lifestyle to others. After all, none of the other Vulcans values their own life either, so as long as there is no direct interference with their present goals, they would fail to see any issue with someone trying to convince them to put an end to it. For this reason, I believe that the intuition that consciousness is what matters is simply a result of human bias, because our consciousness is attached to hedonic states that we find intrinsically valuable.

This doesn’t mean that the Vulcans wouldn’t have “moral values”. In fact, if they were a cooperative species, they would certainly have a minimal set of moral values that enabled cooperation. For instance, the same Vulcans who would see no problem with ending their own lives would see a problem with non-consensually ending the life of another Vulcan, because it would deprive them of pursuing their goals. In the same manner, they would think it is morally wrong to deprive another Vulcan of their agency, since the ability to act is consubstantial to their species’ way of being.

In other words, this species would develop a set of moral rules that are entirely alien to ours. In fact, it would be so alien to ours that we would fail to evaluate their actions in terms of human morality.

Why I’m still not an antirealist.

After reading this, the antirealist may exult: “Finally! You saw the light! Morality is entirely subjective!”

Well, no. If it were entirely subjective, I would have to concede that these same Vulcans could also simply decide that they valued making sentient beings suffer, for no other reason that it switches a 0 to a 1 in their internal value system, and that this would pose no moral problem. However, since Chalmers also supposes that they are rational beings that can understand the world as well as we do, I think that this is wrong. These Vulcans would, in virtue of their being endowed with reason, be able to understand that sentient beings perceive an extra set of properties, namely, affective (or hedonic) states, that make certain experiences inherently attractive or repulsive to them. Even if they weren’t able to understand it in full, they could hear it direcly from sufficiently intelligent sentient beings, or observe the behavior of sentient beings that are subjected to painful stimuli and deduce that it harms them. Since they do not have hedonic states, they might also be less prone than humans to rationalize this suffering whenever it serves their interests. Finally, since these Vulcans do, by construction, have an understanding of the concept of “harm”, and a set of moral rules that enable cooperation, which requires the ability to accomodate the needs of others, they could, in principle, deduce that it is wrong to inflict suffering on sentient beings.

Interestingly, this moral relation is not reciprocal. Since Vulcans do not value their own existence, it is not possible for a sentient being to conclude that not harming a Vulcan is “wrong” in the same sense that it is “wrong” to harm a sentient being, and the Vulcans themselves would agree with this: after all, they wouldn’t even see why it would be wrong to convince another Vulcan to inflict severe harm on themselves in pursuit of trivial ends. For this reason, if they are endowed with a sense of morality, and if they are applying the “no harm” rule consistently, they should conclude that sentient beings matter in a way that is incomparable to them, thus rediscovering the indexical scale of value that I talked about in this article, where sentience is at the root of everything valuable. Perhaps the Vulcans would discover that, without sentience, they do not “actually” value anything, and are instead simply “reward seekers” in a mechanistic sense.

The implication is that this is not a form of moral constructivism, because it entails that there is still an objective way to measure moral worth: first, it is indexically dependent on sentience, and second, it also binds non-sentient agents who have alien moralities, even when it goes against their own interests. In other words, the Vulcans should conclude that they should have two sets of moral values: one relative to how to treat sentient beings, and one relative to how they treat other Vulcans, with one being more stringent and including more rules than the other: a somewhat counterintuitive consequence of the way their experience of the world is constructed.

Note that these Vulcans would be reaching the same conclusion as Utilitarians who think alleviating insect suffering matters immensely, or Kantians who believe shrimp can matter more than humans under certain conditions. I believe this is another strong indicator that sentience is what matters: if even non-sentient, rational agents zero in on sentience being what matters morally, then so shall we.

Conclusion.

This article outlines a form of “light moral relativism”, where several different moral systems could emerge, depending on the minds in which they are instantiated, with one overarching priority still constituting the “core” of all these systems as long as the ability to understand other minds is enabled. This idea is still in its infancy, and I am not fully certain of how it may evolve in the future as I think about it more carefully. In fact, I am still not certain that it is possible to construct a “value scale” without relying on hedonic valence. Moreover, this article rests on the idea that empathy (or at least, cognitive empathy) is a necessary condition for accessing moral truths. Would it be possible to reason one’s way to morality without having empathy? I do not have an answer, albeit I suspect that it would not. For this reason, even if moral realism is true, some beings might fail to access moral truth forever, due to them lacking the necessary “sensors”. It is similar to humans not having access to the what-it’s-likeness of ultraviolet light, despite the fact that we know, via the existence of birds, that it is possible to access it.

I am experimenting with a less academic writing style by linking the references directly in the article instead of listing them at the end. I believe this makes it more entertaining and comprehensible by allowing the reader to check them in real time. Feedback appreciated!