According to moral subjectivism, nothing is innately moral or immoral. To give an example; "you should not steal" would be no more valid than "you should steal". Both would be opinions. Subjective, emotional positions. Not logical positions. We assign the negative or positive value to something like theft. Theft itself has no innate negative or positive value. And there is no reason to do or not do it. A conversation on morality would be as pointless as a conversation on favorite video games, ice cream flavors, etc ...

Are there arguments against that?

  • Are there cogent arguments for moral subjectivism? Although people might conflate their own desires with the supposed will of God, still, defining morality relative to God at least means defining it intersubjectively. Beyond that, Mackie has described the very concept of moral facts/properties as "objectively prescriptive": he argues that such things are too "weird" to exist in our world, but then he is an error-theorist, not a subjectivist. And many other definitions of morality as objective/intersubjective are available. Mar 15 at 1:41
  • Moreover, suppose we take deontic logic at face value. Then it is "objectively true" that if something is forbidden, it is obligatory for that thing not to be done; or if only two things are permitted in some context, then there is an obligation disjoined over the two permissions. While the elementary inputs into the deontic operator "grid" might be subjective, derivations from those inputs would not be (in the same way, anyway), would they? Mar 15 at 1:43
  • @KristianBerry what about secular arguments against moral subjectivism? Are there any?
    – ActualCry
    Mar 15 at 2:21
  • AFAIK, Mackie is not a theist. Then we have Plato (maybe a transtheist, though), or G. E. Moore, W. D. Ross, Prichard, Rawls, Korsgaard, Huemer, Parfit, Scanlon, Susan Neiman, Onora O'Neill, Allen Wood, John Stuart Mill, Bentham, Peter Singer, Nick Bostrom (e.g. his question of infinitary ethics), Bernard Williams, etc. and so forth and so on. Mar 15 at 2:47
  • Not sure about Bernard Williams, now that I think of it. Maybe he's outside the subjectivism/objectivism debate, IDK. I also forgot to mention Gilbert Harman, who is a relativist, but relativism is not the same as subjectivism anymore than objectivism is the same as absolutism. Mar 15 at 3:48

2 Answers 2


Ultimately morals originate in values, which originate in instinct and or nature. There is naturally a lot of reasoning, inference, and trial-and-error along the path in coming up with said morals. The differences you see between persons and groups result from differences in both nature and nurture, though arguably mostly the latter since most humans have a lot of nature in common.

Even a simple negative feedback mechanism, like a thermostat, can be said to have values. For example, in the cold of winter, opening a window could be deemed immoral to the instinct or nature of the heater since it "wants" to maintain a certain minimum temperature.

For absolute moral subjectivism to be rational, there would have to be little if any shared desires or values between the beings in question. We know this not to be the case when looking not only at human society, but at nature in general. Humans have plenty of needs and wants in common, and this applies to most if not all other known life.

On top of that, even from the perspective of passing on one's genes, humans and many other lifeforms here have plenty of genes in common; so being at least rudimentarily courteous to our neighbours has plenty of benefit if our genes are to be valued.

Moreover, the complex ecosystem around us has a lot of interdependence, where any significant interruptions to other beings can come back to us. Hence, we have inherent reasons to care about others, including those seemingly quite distinct in form and function.

One thing about morality is the more you think about it, the more you see the vast interconnectedness of our actions and their effects on one another, including back unto ourselves in some, often indirect path. Maximum viability of the species and biosphere depends on rational regard for these interconnections. Denying our interdependence and personal role in the greater picture is collective neglect, perhaps suicide. The more powerful and pervasive a creature, the truer and faster this holds. With strength comes responsibility.


Here's a very simple argument. Question the premises if you will, but acknowledge its informal validity, anyway.

  1. Moral subjectivism indelibly tends to degenerate into moral non-cognitivism. {Reasoning: there is too little difference between an honest report and a sincere expression, at least (or especially) in this case, for the degeneration not to go through.}
  1. If moral non-cognitivism were true, then "ought" couldn't imply "can." Emotivist talk like, "Hooray for saving the world!" does not consist in assertoric functions, wherefore it cannot entail an assertoric function like, "'Ought' implies 'can.'" The same goes (with some peculiar qualifications) for prescriptivist talk like, "Go save the world."
  1. But even if "ought" doesn't imply "can," it can imply that.
  1. Therefore (by modus tollens), moral non-cognitivism is false.
  1. Therefore, moral subjectivism is false too. QED
  • I don’t see 2 here. How does “boo, murder!” imply it is impossible to murder? It may not imply it is possible to murder, but that doesn’t man it implies it is impossible to murder. Mar 15 at 4:35
  • That's not what I said at all. I said that non-cognitivist representations are syntactically incapable of implying cognitivist ones; ought-imples-can is a cognitive representation; ergo... Mar 15 at 10:21
  • "Boo murder!" implies neither, "Murder is possible," nor, "Murder is impossible," then. ::: I guess I could also try to rephrase the point directly in terms of subjectivism, though (a subjective report about one's moral emotions doesn't potentially entail objective free will, either). Mar 15 at 10:38
  • Interesting, but I’m not sure how central moral sentences not having assertoric functions is to non-cognitivism. The hallmark of noncognitivism is the idea that moral sentences have no truth value. The former is not necessarily implied by this. If not “syntactically”, then pragmatically, non-cognitivism does permit “ought-implies-can”. For instance, “Boo, murder!” does pragmatically imply murder is possible, otherwise the speaker wouldn’t have any feelings about it. Yet, the sentence is still not truth-apt. Mar 18 at 5:31
  • I have a pretty naive view of truth-aptitude, I think. I even think, "X is beautiful," is truth-apt. I'm a strong believer in excluded-middle so that's part of it. Mar 18 at 9:51

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .