In the current online issue of Skeptic magazine, Gary Whittenberger publishes “Meta Ethics: Toward a Universal Ethics — How Science & Reason Can Give Us Objective Moral Truths Without God”. In the course of his argument, he runs up against the “is” of science making the leap toward the “ought” of morality.

Whittenberger vaults the chasm! Does he make it? Here is the central quote (edited for formatting):

I view an “ought” statement as both a contingent prediction and an encouragement. If I say “You ought to shop at the grocery store today” then I am predicting that if you shop at the grocery store today, then you will have a good outcome for yourself and perhaps for others, and so I am now encouraging you to go there. This reduces the “mystery of the ought.”

If we think carefully about the first component, i.e., the contingent prediction, we may understand that it is really based on two “is” statements. It “is” a fact that in the past when you have shopped at the grocery store, you have almost always had a good outcome, and it “is” a fact that when under similar circumstances you have not shopped at the grocery store, you have almost always had a bad outcome.

The second component is also an “is” statement, i.e., it “is” a fact that I am right now encouraging you to shop at the grocery store today. Thus, the original “ought” statement is derived from three “is” statements, two about the record of past events and one about encouragement. We can derive “ought” statements from “is” statements, but we must do it carefully by the use of reason.

If Whittenberger is correct, he has solved David Hume’s “is-ought” problem. This problem has puzzled a lot of good minds in the intervening 275 years or so.

Well, fellow Stack Exchangers, did Whittenberger do it? (I am skeptical) And is the problem of induction the next to fall?

  • 2
    It depends on whether he has justified his usage of the term "good outcome." But the is-ought problem is solvable, even if Whittenberger didn't solve it.
    – causative
    Oct 5, 2022 at 2:35
  • 2
    It is so trivial to find counter-examples to his analysis of ought, that I won't bother to write one down. I will, however, mention that there are many meanings of ought, and the one he is analyzing is not even the right one. Oct 5, 2022 at 4:34
  • 2
    @DavidGudeman If I was wrongly persuaded that Joe committed the murder, then I couldn't have been given sufficient evidence. For instance, I wasn't given the fact of Joe's presence in the bed of his mistress. Sufficient evidence would include, at minimum, all the base physical facts of the situation. More than that, sufficient evidence is an amount of evidence at which a reasonable person settles onto a conclusion, such that additional (correct, factual) evidence is not necessary and won't affect his conclusion.
    – causative
    Oct 5, 2022 at 5:00
  • 5
    No, and, really, Whittenberger doesn't even try. "It is a fact that you had a good outcome" does not yield that one ought to repeat it. For that, we need the major premise that one ought to do what is "good", which is itself an ought. But Hume, or anybody else, never questioned that one can produce oughts from other oughts by combining them with is-es. The real question is what makes that "good", and how it conjures up the imperative that it ought to be done. Whittenberger has nothing to say about that.
    – Conifold
    Oct 5, 2022 at 5:05
  • 2
    The gist of the famous is-ought problem is nothing but an objective definition of the form of good which is above, nobler and beyond all other forms (per Plato), that's why this is the hardest problem perhaps in philosophy. And if you deem your reference gives everyone a satisfactory definition of the good, then this problem could be said to be solved per humanity's standard... Oct 6, 2022 at 3:04

5 Answers 5


I have issues with the definition of "good" the author provides in the article.

He proposes "the fulfillment of the basic biological values of survival, reproduction, well-being, and advancement for all persons", but why should this be an objective moral metric? The mere fact that he has to propose it and argue for it tends to indicate it's not that "objective" at all.

Who decided it was "good" when I went to the shop last time? Because my well being improved thanks to it? But who said my well being ought to be the goal of my actions? The author is clearly talking from a consequentialist position, but who said we ought to be consequentialists?

I feel like many people would disagree with the conclusions the author would reach by the consistent application of his principles. For exemple: "I know a sizeable portion of people drafted in a war get killed, and I strongly feel getting killed is bad for me (*), so I ought to dodge the draft or desert".

Many would disagree and say I ought to serve my country, and I would have a really hard time proving them wrong with consequentialist arguments of the shape "if you want A, you ought to do B" because we strongly disagree about A to begin with. The way the author addresses this contradiction is not satisfying, because I can always say I don't care about "well-being, and advancement for all persons", I feel God is more important, or my country, or my own well being before every one else and he can't tell me why I ought to think otherwise.

Also, the idea that wrongdoers end up suffering bad consequences of their actions appears as quite naive and simplistic. He gives the exemple of rape, but considering how many rape cases end up in "he says she says" situations where the alleged rapist is at most inconvenienced, depending on how much depraved pleasure they got out of their crime, it might not be considered a net loss (and that's ignoring all the rapists who never get caught). What is more it seems by this logic it's getting caught that has bad consequences, and therefore the natural conclusion is the very cynical motto that "the only thing forbidden is getting caught", not a very moral position if we go by the general opinion, I'm afraid.

(*) but is it true? Maybe there is a place in paradise for people who died at war. At least there is no scientific proof of the contrary, which shows the shortcoming of claiming to have moral values based on scientific facts.


Well, pleasant as this analysis is, it misses the core issue in the is/ought problem. Put simply, there is no necessary reason why the 'ought' statement must depend on factual historical experience. If I suggest you ought to go to a particular grocery store, it could be for any number of reasons:

  • I've had good experiences there, and bad experiences elsewhere, or...
  • I saw an advertisement that they were having a sale, or...
  • I drove past the store once and thought it 'looked interesting', or...
  • My sister-in-law owns the store, and I want to drive business her way, or...
  • The holy book ordains we will find salvation by doing commerce with others of our faith, such as the store owner...

Only the first implies any actual experience with the store that could be evaluated as a 'good' or 'bad' outcome. The rest are at best momentary impressions, and at worst actual manipulations against the interests of the listener. I mean, while it may in fact be the case that the road to salvation starts by shopping at Grocery X, that is clearly not a matter of personal historical experience.

Mr. Whittenberger's reasoning works fine for a limited class of prosaic worldly events, but it has no traction on more complex moral issues. Consider statements like:

  • You ought to get an abortion
  • You ought not to murder people who annoy you
  • You ought not to buy crypto
  • You ought to pray each day for forgiveness

Few (if any) people have sufficient experience on these kinds of events to evaluate them as factual 'is' statements. They instead have to rely on things like word-of-mouth, ideological positions, tenets of faith, philosophical analyses, beliefs, feelings, impressions, expertise, etc., none of which can be meaningfully reduced to 'is' statements in the way Whittenberger suggests.

  • 1
    I don't agree with Whittenberger either but is/ought relation in your examples depends on what one counts as part of "is". Eg if "not having a child at this given instant" is part of my "is", easily follows that i "ought" to get an abortion.
    – Nikos M.
    Oct 5, 2022 at 18:34
  • 1
    A further example is our needs at any given instant (eg being hungry, thirsty, etc..) Needs are the archetypal example of "is" that actually encodes an "ought"
    – Nikos M.
    Oct 5, 2022 at 18:36
  • 1
    @NikosM.: That's not strictly correct. A woman who is pregnant has two courses forward: one with the child and one without. We can say that she is with child — that's an empirical fact — but we cannot easily say which path she ought to pursue. It's unlikely she herself has much (if any) experience with abortions or having children; certainly not enough to draw an empirical inference (an is) about 'good' or 'bad'. Oct 5, 2022 at 23:45
  • 1
    @ScottRowe: And the fact that humans have agency is why 'ought' cannot be reduced to 'is'. Oct 6, 2022 at 1:55
  • 1
    @ScottRowe: But that's the sticking point. The real issue here is that people want to have agency, but want to discard responsibility. Moral assertions ('oughts') imply responsibility for actions and consequences (and thus dissonant emotions like shame, regret, or self-hatred). Reducing 'ought' to 'is' absolves all responsibility — we do what we do because that's what 'is' — and thus reduces cognitive dissonance. It's sadly nihilistic, but a powerful motivation to keep grasping after the fantasy. Oct 6, 2022 at 12:08

I like to imagine Hume's is-ought problem with just laying pipes vs. actually connecting them to a water reservoir.

We need a source of the magical "ought"-fluid! Without it, I can lay pipes all day long yet the whole system will remain dry.

But if we have finally found a source of "ought", we sure can try to channel it somewhere else. This just means deriving another "ought" from a given "ought" with a philosophical argument. Hume never denied that this was possible.

So for any purported solution of Hume's is-ought problem, carefully look if there isn't a hidden source of "ought" somewhere!

In Gary Whittenberger's argument, it lies in the tacit assumption that we ought to aim for a "good outcome" (for ourselves).

I don't deny that this comes easy for most people, but it is still an "ought" supplied from the outside. And this issue gets way more pressing if we consider other people. E. g. in a slaver society, you could replace "grocery store" with "slave market" and get a "sensible" suggestion:

"You ought to shop at the slave market today".

So stealthily the pipes were connected to this "ought"-reservoir and the magical "ought"-fluid flows along. It didn't actually came from dry pipes as he suggested.

Disclaimer: I found the quoted argument rather crude compared to the usual attempts to derive "ought" from "is". So I didn't bother to read the whole essay.


In short, no. The essence of the is/ought problem is that certain types of statements cannot be justified without some degree of subjectivity. If I state that such-and-such soccer club is the best in the world, and you ask me to justify my claim, I can make progress in a logical way by citing various exceptional attributes and achievements of the club- perhaps they have scored more goals than any other, perhaps they have the largest average attendances at their matches, perhaps their team is the most valuable, and so on, but no matter how many facts I marshal to support my argument, you can still challenge it by asking why do those facts matter, and not other facts. Ultimately opinion has to creep in somewhere to justify the conclusion.

Indeed, the way to resolve the is/ought problem once and for all is to recognise that it is not a problem but a distinction.


The is-ought problem has long been solved, and not by Whittenberger. I recommend you read Viable Values for a discussion of the general method of bridging the is-ought gap using contingent predictions. This is a method that has been used by several philosophers to offer a moral philosophy that bridges the is-ought gap. Whittenberger is (very roughly) correct in his outline of what this method involves, but he is certainly not the first to use this method.

To deal with your specific question about the quoted section from Whittenberger, you should observe that this reasoning hinges on the idea that an outcome can be diagnosed as a "good outcome" for a person and others. This really just pushes things back a step. Any normative evaluation of a circumstance or outcome (e.g., that the outcome is "good") is part of the "ought" piece of the is-ought dichotomy. Remember that the "is" part is descriptive (you are pointing out that certain things exist and behave in certain ways) and the "ought" part is normative (you are saying that certain courses of action are good/bad and certain outcomes are good/bad). So this is not yet a solution to the is-ought problem, but it is pushing further back towards a more fundamental goal, presumably with the intention of having some primary contingency of this kind.

Despite this shortcoming, this is alluding to, and pushing you towards the correct solution, which lies in contingent prediction. The way that other philosophers have solved the is-ought problem is to extract some overarching normative goal that can be considered as a primary and then they say that the is-ought is bridged conditionally on the chosen objective to pursue that goal. In this formulation, you go from an is to an "ought" by saying:

If the facts of the world are X then doing A will most effectively achieve your goal of G. Thus, since your goal is G, you ought to do A.

The first part of this is the contingent prediction and the second part is the encouragement. Note that the contingent prediction requires you to have already established an overarching goal G that can be used as the contingency in the statement. Since this overarching goal is normative in nature, this is just pushing things back to an ultimate goal that still must bridge the is-ought gap somehow. The general method by which other philosophers have found an ultimate goal that bridges the is-ought gap is to look for a goal that is axiomatic --- i.e., it is implicit in all action and is cannot be denied without some kind of contradiction. If you read Viable Values you find a detailed argument that living and flourishing can be considered an axiomatic goal of this kind, since the occurrence of any type of action (including arguing against the argument I put here) involves living. I will not attempt to summarise this argument here, but I recommend you read it to see both the specific axiomatic argument and also the general technique of contingent prediction.

  • 2
    None of this answer crosses the is/ought barrier, because the whole point is to figure out if we can derive morality as a goal. The closest you come is waving at Viable Values in this quote: "you find a detailed argument that living and flourishing can be considered an axiomatic goal of this kind, since the occurrence of any type of action (including arguing against the argument I put here) involves living." Of course, one may live, and thrive, while being an evil SOB, so the hand waving does not actually point toward MORAL goals, all you imply is that we must have goals.
    – Dcleve
    Jan 1, 2023 at 23:25
  • It is odd to assert that direct reference to a philosophical treatise on the subject constitutes "hand-waving". Not every subject can be fit into a short SE post.
    – Ben
    Jan 1, 2023 at 23:37
  • You claimed a widely cited barrier in philosophy has been breached, but do not even summarize how this was done. And the hints you provided suggest this “solution” is not even relevant to the problem. An answer here should describe this supposed “solution”, and THEN provide the citations that allow interested parties to examine this accomplishment ourselves. Citing a consensus of other philosophers supporting this claimed achievement would also be highly relevant.
    – Dcleve
    Jan 2, 2023 at 2:48

You must log in to answer this question.

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