So I've been reading on the is-ought distinction.

Let's say I have a supercomputer and am given the initial conditions of an isolated system which contains humans. Then I see the time evolution of the system and see aha! these are the moral value systems that will evolve in this system.

But I think I'm contradicting Hume because I've used science which tells us what is and not what ought to be?

Of course one can claim my machine only predicts the logical fallacies made by that civilization. But in that case the counter question becomes is Hume's law relevant to moral systems at all in the first place?

  • 1
    That some moral value systems do evolve tells us nothing conclusive as to what the moral values ought to be. How is this contradicting Hume? Oughts are relevant because we want to know what to strive for, Hume's point is that what happens to happen is of limited use in this regard. It can tell us something about means to an end, but not what the end ought to be.
    – Conifold
    Commented Oct 25, 2021 at 9:09

2 Answers 2


Your confusion comes from the fact that a prediction about whatever moral values a given society is going to adopt is not an ought statement, but an is. It is to say, you're not stating what the society ought to be, you are just stating what it is going to be.

It can be illustrated with the following thought experiment: let's say I have devised the perfect sociological model, akin to the psychohistorical model described in Foundation: whatever the current state of a given society, I can gather the relevant data, plug it into my equations and voila, I can tell you what this society is going to be in 10 or even 100 years.

One could argue it is not possible, because people have free will and are unpredictable or whatever, but it's not the point so bare with me.

Now lets say I am in Germany in 1920 and I use my model, and I predict the moral choices that will be made by the German population at large in the next decades are going to lead to the Nazi regime, the war, the holocaust. I of course disapprove of this evolution, this is my ought statement: I think that society ought to stay peaceful and democratic. But at the same time my equations are reliable, they tell me the regime is going to be fascist. This is my is statement.

You can see that they differ, and it is not because people have free will or because science is unreliable, but because what I say the world to be and what I think it should be are two entirely different things.

  • At some point though I'm sure you will agree our morals converge to 'how it ought to be' or more and more will remorse the tradegy of the holocaust. Does this idea of convergence enable one to say this is an ought statement? Note I am not making a logical deduction but a correlation. Commented Oct 25, 2021 at 12:52
  • Natural selection will end up with any set of moral values that only guarantees the survival of the considered society. Or not, and the population will go extinct. Is a set of moral values that just barely keep us alive what ought to be? Probably not. We probably can do better than just surviving. Also, what ought to be is subjective, many people have a different of what ought to be. But only one state of things will ever be the case.
    – armand
    Commented Oct 25, 2021 at 13:34

There is no ought in determinism.

A computer simulation is the closest thing to a deterministic system we can think of. We can assume that the outcome is completely determined by the initial conditions, if there are no random malfunctions or human intervention.

Your fallacy here is that you assume that the system contains humans. That is not possible. A sufficiently advanced simulation of a human would be considered as an artificial life form having instincts, preferences, agenda and morality of its own. That would be a human intervervention that renders the system nondeterministic.

An insufficiently advanced simulation would have only a simulation of morality designed by whoever programmed the initial conditions.

  • en.m.wikipedia.org/wiki/… Commented Oct 25, 2021 at 12:54
  • Please, explain the relevance of this link to my answer. Commented Oct 25, 2021 at 13:45
  • All the laws of physics are formulated as initial value problems. When you say "Your fallacy here is that you assume that the system contains humans" your indirectly saying the laws of physics do not apply to humans Commented Oct 25, 2021 at 13:48
  • A deterministic system cannot contain humans. Humans tend to decide things, like moral values. In a deterministic system all decisions have to be made outside of the system, before it is started. Nothing can be decided or created within the system. No runtime input is allowed. Commented Oct 26, 2021 at 6:30
  • How does one create a coherent view of this with science then? Commented Oct 26, 2021 at 7:02

You must log in to answer this question.

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