Is-Ought Problem

Frank Walton writes: “Evolutionist Dr. Massimo Pigliucci writes, ‘It has been pretty obvious since Darwin that we, indeed, are nothing but machines.’ Obviously, then, there wouldn’t be a problem if one machine ‘kills’ another machine. When an automobile slams and crashes into another automobile do we say that the cars murdered one another?”

Walton makes a very common mistake here, one that was first recognized by David Hume and which philosophers call the “is-ought problem” after Hume’s comment in Book 3 of his Treatise:

“In every system of morality… the author proceeds for some time in the ordinary way of reasoning… [and then] instead of the usual copulations of propositions, is, and is not, I meet with no proposition that is not connected with an ought, or an ought not.”

Hume goes on to say that an ought or ought not can never be deduced from an is or is not. In other words, a statement about a fact (or what is the case) does not entail a moral conclusion of what should be done or how we ought to react to the fact. In this case, Pigliucci’s ontological remark that human beings are machine-like does not lead to the moral conclusion that murder is justifiable. Walton inappropriately derives an ought from an is. This faulty thinking is another reason Kahlo was asked 17 times “If you don’t believe in God, why care about anything?”

Rape them Atheists!
What if you Saw a Miracle?
Apologetics Infographic #1: Atheism and Nothingness
Religious Experience – Recognizing God
  • Frank Walton

    Hi James,

    Thank you for opening this discussion. I don’t think I’m making the “ought from an is” fallacy because I deny that there is an “ought” in the case of machines. Thus, “[w]hen an automobile slams and crashes into another automobile do we say that the cars murdered one another?” The obvious answer to that rhetorical question is no. Morality is totally irrelevant if man is a mere machine (in the Darwinian evolutionary sense) thus there is no “ought.” Anyway, that’s how I see it.

    Hope that helps,

    Frank Walton

  • Frank Walton

    James: In this case, Pigliucci’s ontological remark that human beings are machine-like does not lead to the moral conclusion that murder is justifiable.

    Frank: *Shrugs* Nor would it lead to the moral conclusion that murder is unjustifiable.

  • Steven Carr

    Is Frank saying that he has no objection to anybody trashing his car?

  • Frank Walton

    Hiyah Stevie,

    What I’m saying is a car (ie a machine) isn’t a moral agent. I thought that much was obvious to you. *Shrugs* Oh, well.


  • Steven Carr

    Sorry. I thought you were saying that machines had no value.

  • Frank Walton

    Yes, actually, that’s what I was saying. I wrote, “Morality is totally irrelevant if man is a mere machine (in the Darwinian evolutionary sense) thus there is no ‘ought.’” So, what prompted you to think that I thought man as a machine had value?

    Oh, wait! You think if I said that if a machine had no value that would give any person a justifiable reason to trash my car which has no intrinsic value, corrrect? Gee, Stevie, what have we forgotten here: I didn’t agree with Pigliucci’s premise that man is a mere machine. Thus I don’t equate men to automobiles! Your question if anything is a red herring. Let me make it more easier for you, my friend. If an avalanche of rocks fell from a cliff and pummelled a boulder in two – would that mean the avalanche murdered the boulder? Obviously not. I’ll leave it to you, though, Stevie, every time you see a boulder you have every right to “murder” it if you like.

  • Alonzo Fyfe

    Frank Walton

    I have long considered this to be a mistaken interpretation of Hume.

    To say that ‘ought’ cannot be derived from ‘is’ means that ‘ought’ exists in a supernatural realm, a realm independent of the natural world. Yet, somehow, ‘ought’ is supposed to be able to interact with the natural world such that certain states cannot be adequately explained without mention, somewhere, of somebody realizing that X ‘ought’ to be the case and doing it for that reason.

    In other words, either (a) we have to be able to reduce ‘ought’ to some set of natural properties, (b) ‘ought’ is a supernatural property, or (c) ‘ought’ statements have no relevance in the real world.

    Of these options, clearly Hume thinks that ‘ought’ has relevance in the real world. Furthermore, Hume tends not to believe in supernatural properties. This leaves (a), ‘ought’ can somehow be derived from ‘is’.

    What Hume actually said is not that ‘ought’ cannot be derived from ‘is’, but that ‘ought’ cannot be derived from ‘is’ without mentioning the passions. The passions are no less real than other natural properties. However, the passions are internal, not external. They are not ‘perceived by reason’ but, rather, are written (by nature or by nurture) into the human emotions.

    Ultimately, Hume describes for ‘is’ relationships that are necessary in deriving ‘ought.’

    These are:

    (1) Is pleasing to self.
    (2) Is useful to self.
    (3) Is pleasing to others.
    (4) Is useful to others.

    ‘Useful’ simply means ‘can be used to bring about some other state which is pleasing to self or others.’ So, everything ultimately leads, directly or indirectly, to that which is pleasing to self or others.

    However, there are no supernatural properties involved. Nor are we forced to abandon ‘ought’ statements as something irrelevant in the real world.

    This, by the way, is why damage to a machine has no moral property in itself. Damage to a machine is only morally significant insofar as it harms the owner. This is because nothing can please or displease a machine. There is no “pleasing to others” for us to consider when a machine is involved.

    Now, pleasing to self may be determined. But its being determined makes it no less real.

    Alonzo Fyfe
    The Atheist Ethicist

  • Jim Lippard


    I think you’re right… I would just say that “This is because nothing can please or displease a machine” would not be correct for a machine that has internal passions.

    Frank seems to be arguing that there are no oughts if there is no God, but surely there is no necessary connection between the existence of God and the existence of the passions. When Frank says “Morality is totally irrelevant if man is a mere machine,” what does “mere” mean in this context? The evidence of human evolution is overwhelming, but how human beings got here doesn’t change the fact that they have desires, intentions, thoughts, and actions, or the fact that the human neurological system is immensely complex.

  • Alonzo Fyfe

    Jim Lippard

    You are correct — once we create machines that have internal passions, then we will have machines with moral rights and duties.

    After all, our neighbors are machines with passions and, it is in virtue of these passions that they have rights and duties.

  • James Still

    I was trying to keep it short so I didn’t describe Hume’s thinking to the level that Alonzo did so well. Certainly it would be wrong to say that an ought can never come from an is; we do it all the time as we calibrate our actions in accordance with our values. I only want to attribute to Hume the idea that an ought is not logically deduced from an is.

  • Frank Walton

    Thank you Alonzo and Jim for shedding some light on the issue. Surely, you can get an ought from an is in light of scientific law. For instance “What goes up eventually comes down” (that’s an “is”). We can conclude from a regularity of “ises” that “If you throw a ball up in the air, it should come back down” (that’s the “ought”). As my friend Trey said, “But this action, when it takes place, is mechanical, in a very real sense. There is no option about it. Even if it were possible (which it is not) that such a rock would go up and *NOT* come back down, one would not think to accuse the rock (or the person who threw it) of violating the law of gravity and be criminally answerable for such violation. However, to break a law of morality, we *DO* think in terms of being criminally answerable, and we know that it *IS* possible for a person to do something against an established law of morality, because they are not mechanical and invariable the way that, say, the law of gravity is. Not even people have a choice in whether they follow the laws of thermodynamics; they just do because that’s how reality works. Morality, on the other hand, is different: you can establish all the ‘ises’ you want to when it comes to morality, but that doesn’t ever imply an ought in this case, because moral ‘oughts’ are not predictive, they are prescriptive. It’s not saying what will likely happen based on past evidence, but what one *SHOULD* do based on what is right, regardless of what others have done in the past. Indeed, when such a rule or law is broken, we seek retribution, justice, punishment for that broken law. That’s not even possible in the category of laws of thermodynamics.”

  • Frank Walton

    I should also add that whatever pleasure or desire the machine has is dictated by the laws of physics; thus doing harm to ones pleasure (desire, etc.) as a means to human morality is making the “is-ought” fallacy. :o)

  • Jim Lippard


    I agree with you that in legal and moral rules, there is not the predictability of behavior of the governed that there is in most scientific laws–they are normative and prescriptive, not descriptive. Legal and moral rules are not scientific laws.

    Human beings are usually not so predictable. We don’t know whether, if we knew all the appropriate underlying causes, if human behavior would follow deterministically from preceding causes or if quantum effects are “magnified” in the brain and there really is no predicting, but either way, it does not preclude holding people to moral standards of reasonability.

    The macroscopic causes of human behavior are their beliefs, desires, and intentions, which are, along with their behavior, factors measured in holding them responsible. We say that people “could have done otherwise” based on seeing relevantly similar people in relevantly similar situations who do otherwise (or even seeing the person being evaluated do otherwise in other relevantly similar situations). The kind of cause of behavior that is considered exculpatory or reduces the blameworthiness of the offense is either an external cause (e.g., coercion) or a measurable defect (mental illness, mental retardation, the lack of the ability to distinguish right from wrong).

    Further, even if one holds a simplistic notion of determinism (and we know that, at the bottom, determinism is false), it *still* makes sense to hold people responsible, because the very acts of holding people responsible are still causal factors in their behavior (through their effects on beliefs, desires, and intentions).

    If you haven’t read Dennett’s _Elbow Room_ (or his more recent _Freedom Evolves_), I recommend them.

    It’s worth noting that the issues of free will vs. determinism are completely independent of the existence of God. Postulating a separate spirit or soul does nothing to eliminate this problem, it just moves it to the point of interaction or into soul-stuff (is soul-stuff deterministic or indeterministic, neither choice makes the philosophical issues go away). It actually creates a whole host of additional issues, since what we know about the brain shows that lots (I’d argue all–the gaps to put soul in are rapidly disappearing) of major functions of thought, perception, and consciousness occurs on the material side–it’s processing done by the brain itself.

  • Hume’s Ghost

    Alonzo explained it far better than I could, but I was going to say Hume did not mean that you can’t derive ought from is, but that you should be aware of the distinction.

    What Hume was saying was that because something is it does not necessarily mean it ought to be the case.

  • Frank Walton

    Hi Jim,

    After reading what you wrote, I’m apt at repeating myself but I won’t. Thank you though I liked the discussion.