I use Humes idea about the 'is..ought to gap' (or my spin on it) mainly to make sure I have good reasons for what I do -and what I advise others to do- and to make sure actions are not based purely on cultural pressures or recklessly induced policies. I don't use it much to judge whether the actions are morally right or wrong but whether we rationally have good reasons for doing or not doing something based on our present knowledge and feelings.

Since the introduction of neuroscience I have seen many attempts to bridge the 'is..ought to gap' with facts; especially with facts about what makes us feel better or worse. By seeing more and more respected people claiming that they bridged the is..ought to gab and that the is..ought to gap is a myth I naturally have started to question whether I am wrong to use it THE WAY I USE IT.

The argument goes something like: if we can confirm that a specific action consistently lead to a specific chemical reaction in the test subject's brain and if this chemical always makes the test subjects feel good, then we can conclude that this is what we 'ought to' do, regardless of what anyone value.

Another version is similar but is based on survival i.e. that if it helps us survive or live longer then that is what we ought to do.

Both of these arguments claim to produce logical 'ought to' conclusions without using values. I welcome these studies and think we can learn a lot from them i.e. they can help us get ever closer to objective values (values we all agree from our subjective views are good or bad). But I don't think they bridge the 'is..ought to gap' for at least two reasons: Firstly, because (in the first case) you are implicitly saying that you value feeling good in the particular way the chemicals make you feel, and in the other case you are implicitly saying that you value to live longer. Secondly, because there are plenty of times when you value some other chemical even more (even if it might be true that you as a human always value feeling good in the particular way that chemical make you feel), and there are plenty of times when you value doing something dangerous or unhealthy over living longer. Therefore you can not use even what is objectively valuable (assuming for the sake of argument there are such values) alone to decide what you 'ought to' do. You still have the evaluate each unique situation; you have to 'look inside' and out of all the alternative consequences you can imagine you have to choose what you value most right now before you can decide what you 'ought to' do.

Does anyone know a strong case against the fact that 'in order to rationally justify an ought to (i.e. what you should or should not do) I have to present a value'? Or Should I embrace the 'is..ought to gap' if I value having rational reasons for the rules and guidelines I put up for myself and others?

ps. I'm not interested in the argument that 'I have a goal' (i.e. a fact) therefore 'I ought to do x'. Since having a goal means (to me) that I have a value i.e. it might be a fact that I have a goal (or even other types of values) but this goal is then a implicitly saying that 'I value...'. Therefore this argument does not bridge the 'is..ought to gap' without a value.

Definition background (how I define/understand value):

I use both value and desires to mean 'what is compelling to me'. I often separate them in that my desires are my subconscious (irrational) and may values are my conscious (rational) compulsions. But this separation is often useless since few people use it. So to me it make sense to say that I can potentially prove what someone value/desire by seeing what they do but I have only proven what they subconsciously value/desire right now; your rational values/desires have to come from a conscious evaluation in your mind. So you can value one thing but do another and therefore states of affairs can teach you that your subconscious desires are not in accordance with your conscious desires, but ultimately it is up to you whether you decide to leave them like that, reprogram you subconscious desires or change you conscious values.

  • Isn't it a logical problem? – Annotations Apr 9 '13 at 15:13
  • @RicardoBevilaqua Nah, I don't think it is. The is-ought gap; or the fact-value distinction is squarely within ethics. Typically naturalists try to close the gap and deny the distinction. – Dennis Apr 9 '13 at 17:01
  • I re-tagged to what I think is a more appropriate tagging of this question. I don't think the others are really on target here, but feel free to disagree. I just think that, for instance, the logic tag gets applied FAR too broadly. Like I said above, this is an issue about ethics and closely tied to naturalist ethical theories. – Dennis Apr 9 '13 at 17:11
  • I really don't mean this to be an issue only about ethics. Unless you stretch ethics to concern all actions regardless of whether they are about moral judgments. I use the 'is..ought to gap' mainly to make sure what I do and what I advise others to do is not purely cultural pressures or based on policies recklessly induced from other policies. Not to judge whether the actions are morally right or wrong but whether you rationally have good reasons for doing or not doing something based on what you know right now. So I'm curious about this question more as an issue of whether it is logical. – Kriss Apr 10 '13 at 0:43
  • @Dennis According to the dominant interpretation, Hume says that no ought-judgment may be correctly inferred from a set of premises expressed only in terms of ‘is,’ and the vulgar systems of morality commit this logical fallacy (SEP). – Annotations Apr 10 '13 at 3:34

This issue is not particularly well addressed IMO, mostly because people who do address it (e.g. the Churchlands) seem to end up talking past the people who see a problem.

I think the real problem is that we don't know what "ought" means. We surround our oughts with their own little network of self-consistent terms, but what really is a bare "ought"? In contrast, the conditional ought ("if you want to avoid skin cancer, you should wear sunblock") is non-problematic. The naturalistic denial of the bare ought is not something I've seen stated in quite this way, but I've seen something that I interpret to be it from many naturalistic philosophers.

So a naturalist would presumably respond to the is/ought dichotomy by saying something like: anything that you simply say that you ought do is conditional, you just may not be aware of the conditions (they may be evolutionarily selected and not easily accessible to introspection); if you wonder why you feel that you ought do this and not that we need to consider the mechanistic basis for these feelings as well as the logical inference; and if you are uncomfortable with the results of this inquiry it is probably not that they are invalid but rather that the feeling of oughtness is an ad-hoc adaptation to social life (or the demands of having a brain composed of different computational modules) and thus may be intrinsically illogical.

You then end up with a philosophically weighty "is" but a rather anemic "ought" that lacks much of the universal compulsion sought by value systems. Given our genetic and behavioral similarities (and evolutionary constraints) you may yet recover something essentially universal for humans, but covering all rational thinking beings may be out of reach.

(As of when I wrote this, Wikipedia quotes a passage from Patricia Churchland that explains her take; I am not sure whether this will remain given that the article is heavily marked for revision.)

Edit: let me add a little bit more about what the is-ought distinction is (from my perspective anyway), and why the above is a sensible maneuver.

Suppose I have some measure of value--let's call it V--and you can make factual statements about how V will change in various scenarios. Now I can without any problem construct statements of the form "if you want V to increase by vX then you should do X" assuming that I have correctly concluded that doing X will in fact increase V by vX. However, if I provide you with this information, you might simply say, "Oh, that's interesting, but I'm not going to do X."

The question is whether there is a rational way for me to change your mind based on factual accounts alone. I could ask you what you value and use your metric U, but you could just say, "Well, I know that increases U, but I'm not doing it," which would not be irrational for you unless we knew that prediction of U is the only mechanism by which you select your actions and even then it's not clear that it would be wrong for you to decrease your U; rather we would only learn that it's impossible. So it's very hard to see a way out from stubbornness to rational compulsion (or, really, any impulse to act at all, if you think about it). If you envision defining value as to be that which is compelling to you, you've just shifted the problem: now value and compulsion are linked by definition, and you can read off value from states of affairs, but this still doesn't truly constrain your behavior because states of affairs are powerless to indicate to you that your values are wrong.

The naturalistic counter to this is that this sort of elaborate value-land (or realm of oughts) living over factual affairs is logically possible but not at all a good model of how our subjective sense of value actually happens. Rather, the bare ought, compelled by detached value, is arguably not even a sensible construct; this simply isn't how it works in real humans. If you want to understand how motivation and value and "oughtness" actually works--why we have the intuitions of these things that we do, and so on--then you need to approach the problem differently.

If you really want value in there, it's easy enough to inject it. Dopamine is pretty close to the universal currency of value in our brains. Restating oughts in terms of maximization of dopamine release (over evolutionary timescales, if that's what it takes) doesn't really lend any clarity, though; it just fulfills your wish to have value attached to ought.

Hopefully this clarifies what I am trying to convey.

  • Churchland is actually one person I heard say this and I think she is trying to put forward the case that we have a scientific base for moral judgements. She's trying to find objective values. But objective values do not help much when deciding what to do in day to day life because it doesn't help us prioritize between alternative actions in complex systems where only our gut judgments can show us what we value most. It doesn't help ME decide whether I should go to university or work at Mc Donalds; it only help me decide to rather greet people with a hi than to hit them in the face. :) – Kriss Apr 10 '13 at 1:09
  • I define ought to as "If you ought to do something, doing it is better than not doing it." (simple.wiktionary.org/wiki/ought) And I mean it as a rational exercise based on the beliefs you can express explicitely. It sound like the naturalist (according to your description) would accept an argument like "if I feel like I ought to do it then I ought to" (?). I understand this as a valid argument but are they not just leaving the value implicit because they do not yet know why (once they know it might change)? Would you say I don't understand what 'ought to' means? – Kriss Apr 10 '13 at 1:23
  • @Kriss - Er, well, you've just moved the problem from "ought" to "better". How do you judge better? How do you adjudicate between different opinions or definitions of better (even within yourself)? So I'm not yet sure you understand your "ought". Also, no self-respecting naturalist-philosopher would, I hope, accept the "if I feel like I ought to then I ought to" line you stated; I'd hope for some generalization so that you could understand when feelings are wrong--and just because "ought" is inspired by some feeling does not mean we cannot use the word for something more definitive. – Rex Kerr Apr 10 '13 at 2:04
  • I assume you would agree that I 'ought to' consider what I value more if I have two competing 'ought to'-value pairs? So are you saying that in these scenarios it is no longer a question of 'ought to' i.e. that it make sense to say that I 'ought to' to both, even if it is physically impossible (since ought to is not about prioritising)? – Kriss Apr 10 '13 at 2:30
  • @Kriss - Well, I'm actually not sure where you're going with this. If you are not careful you will package the is-ought disparity into the difference between better and ought and then no longer be able to find it. Hume's original criticism was against this sort of sleight-of-hand. That is, if you don't phrase it as "if you want things to be better, you ought to do X", you're left with the question: why should what is better have any potency to direct my actions? If you are positing a universal-better which is undeniably preferred, then you have to make sense of what "better" really is. – Rex Kerr Apr 10 '13 at 3:16

This is just an addition to the other answers from an analytic perspective, not an attempt to give an exhaustive reply.

The argument goes something like: if we can confirm that a specific action consistently lead to a specific chemical reaction in the test subject's brain and if this chemical always makes the test subjects feel good, then we can conclude that this is what we 'ought to' do, regardless of what anyone value.

In welfare economics, there is a a similar position (without the naturalist presumption) about the good as opposed to the ought called the preference satisfaction view, the view that if someone chooses A over B, then he is better off with A than with B. As John Broome remarks, "... the preference-satisfaction theory is obviously false, and no one really believes it." (Broome 1999, p. 4) Perhaps something similar could be said about the above passage?

Related to that, the view that preferences are revealed by choices is called revealed preferences theory in welfare economics and has also been challenged from various perspectives - e.g. based on a prior rejection of the preference satisfaction view, based on empirical phenomena (e.g. akrasia, Kahneman&Tversky's work, examples of intransitive preferences), or based on foundational value incomensurability and pluralism.

Turning to values in the narrow sense, in the analytic literature on values I know of (von Wright, Hansson, Chang, Gert, Rabinowicz, ...) , values are either tied to the ought by definition according to the fitting attitude analysis of values, hence cannot be naturalized anyway, or they are characterized by their logical structure. In the latter case, there is much agreement with your points, because in order to make sense of the "producing a good feeling" analysis, more has to be said - comparative values have to be explained. And once you lay out how comparative values work in naturalistic terms, you might end up with the preference models of this tradition, thus implicitly defining values. However, at the same time it must be pointed out that the logic of "ought to" and related notions (must, permitted to, may) is not exactly the same and not directly derivable from the logic of values (i.e. essentially peference logic), or at least I'm not aware of such work. For example Hansson (2001) lays out values based on preference logic in the first part of his book and norms based on deontic logic in the second part of the book. (And deontic logics have much more problems than the modelling of preferences and values, which is fairly well-understood.)

The bottomline is that there are many reasons why you are right and you can find a lot of literature on this topic within economics and philosophy, but of course at least for analytic philosophers the devil is always in the detail...


This seems to be Utilitarianism under a weird disguise. The question then remains should we accept the behaviour of what would make the National Socialist happy and what the child molester or the necrophiliac happy?

Or more in line with your brain science. Should we accept what leads to a specific chemical reaction in the National Socialist's brain as just. As long as it leads to a constant result.

Another version is similar but is based on survival i.e. that if it helps us survive or live longer then that is what we ought to do.

Then we really should never leave our houses. we can justify murder of anyone with a gun in his hand then also. It is not bad the kill that person with the gun in his hand. As long as there is a chance that I will live longer. The same would be true for just about anything that may possibly construed as a weapon.

It is clear to me that what makes you "live longer" or "feel good" is such a vague and ambiguous term that to have this as your moral compass would just lead you to be able to justify any action.

I personally have no faith in the virtue of humanity in general and if humanity is left to his own devices he will inevitably turn to his wicked nature. So that is why I would believe such vague ideas of good and bad are a recipe for disaster.

Not the answer you're looking for? Browse other questions tagged or ask your own question.