Hume's law and Kant's law

Hume’s law is this: One cannot derive an ‘ought’ from an ‘is’. From the fact that war IS common it doesn’t follow that war OUGHT to be conducted etc.

Kant’s law is this: Ought implies can. “I ought to rescue the man in the burning house” implies “I can rescue the man in the burning house”.

Now to me both of these “laws” seem correct. It has, however, been claimed that they exclude each other. The reason is this. “Can” is an “is-word”; it deals with facts. Thus one can derive an “is” from an “ought”. But then, through contraposition, one can derive “not-ought” from “not-is”. And since “not-ought” qualify as “ought-word”, and “not-is” as “is-word”, one has derived an “ought” from an “is”. For example, from the fact that I cannot rescue the man in the burning house it follows that I am under no obligation to rescue the man in the burning house.

Any ideas of a solution?

ought implies nothing and nothing implies ought

can implies could

is implies am doing

sam implies I am

even if you could, you are under no obligation to save the man from the burning building

-Imp

Most interesting. I have never come across this “paradox” (you might call it) before. (Have you any references?) I suppose that one could twist a little and talk about the obligation being “voided” when it is impossible for you to perform the obligation rather than not having the obligation, but I don’t see that as of much help.

for hume, there is no selfless action- and there is no deducing moral statements from statements of fact.

iep.utm.edu/h/humemora.htm

plato.stanford.edu/entries/hume-moral/#io

for kant morality (categorical imperative) is a duty that is demanded by his idea of reasoning.

iep.utm.edu/k/kantmeta.htm#SH8d

rep.routledge.com/article/DB047SECT10

en.wikiquote.org/wiki/Immanuel_Kant

they do exclude each other…

solution? let kantian metaphysics about the idea of god (which actually agree with hume) hold sway…

plato.stanford.edu/entries/kant-metaphysics/#5

-Imp

I read about it in a doctoral dissertation on Hume’s law. The author (HÃ¥kan Salwén) doesn’t give any references, but writes: “Kant’s law is accepted by many and by some taken to entail that Hume’s law is false.” He doesn’t indicate who “some” is.

I have been thinking in similar lines: that “You ought to do x” should be analyzed into a factual and a normative part: “You can do x” and “If you can do x, then you should do x”. If this analysis is correct, it is clear that “ought” implies “can”. It is also clear that “not-can” implies “not-ought”, but only due to the factual component of “ought”. Thus the normative part of “ought” is, if this analysis is correct, not involved in derivations featuring “can”, and thus the “is-ought”-distinction could be upheld.

Seems to me that there are some pretty big assumptions: 1) “Can” is an “is-word”; it deals with facts. and 2) And since “not-ought” qualify as “ought-word”, and “not-is” as “is-word”, one has derived an “ought” from an “is”.

Can is not an is-word, one is a potential the other is an actual. Ie. I can lift that rock, I am lifting that rock.

I would say a “not-ought” is the exact opposite of an “ought-word”. Likewise with “not-is” and “is-word”.

But, how is “should do X” any different from “ought to do X”?

How is that other forum doing?

ken)

“Should” and “ought” aren’t different at all. I just needed another word. My idea is that “ought” contains a normative and a factual component, and that is thanks to this fact that factual words like “can” can imply “ought”. When this is done, the normative component isn’t touched, so to speak.

The other forum; well, it’s not very different from what it used to be. Mostly crap, but I guess it’s the same in most fora.
There are some posters worth reading though, like muxol.

kolibri)

1/ “Can” deals with potentiality, that’s true. But it is an “is-word” in the sense that it deals with physical reality. If it’s true that I can lift that rock, it’s true due to some physical facts. If it’s true that I ought not to kill, then it’s not due to some physical facts (at least not alone).

2/ What are your arguments?

Stefan,

Inherent to Hume’s Empiricism is the idea that perception is value-free, is this your position?

Dunamis

I don’t think that Hume’s law entails radical empiricism.

However, personally I think that perception sure is “theory-laden”, contrary to Hume and the empiricists. I also think that one needs to avoid a psychologistic trap. Surely many of our feelings, thoughts and perhaps even perceptions have both normative and factual components. But this doesn’t destroy the fact-value-distinction, any more than faulty reasoning destroy the logical laws. Just like logic, the division between facts and values isn’t on a psychological, but at a metaphysical, non-psychological level. (By the way, this isn’t very Humean, I’m afraid).

Stefan,

Well, if you permit value to enter into perception, Hume’s law does seem to have some problems, for the “is” would gain its meaning from a holism of beliefs to which its identity is connected. How would you handle the very simple:

There is a steak on the table.

You are starving.

You ought to eat it.

Dunamis

p.s. I think that the “derive” in Hume’s Law is specific to his empiricism

I don’t think you can call “can” a “is-word” if it does not have exactly the same characteristics as “is”.

The set “ought-words” contain all words with the same characteristics as ought, eg. “ought” and “should”. Not ought is the opposite of ought, thus it cannot be in the “ought-word” set.

Kolibri

1/ I have of course never said that “is” is synonymous with “can”. What I mean is, of course, that “can”, according to one classification (but of course not to every classification, for example the potential-actual classification) belongs to the same class as “is”, whereas ought belongs to another class. “Is”-words give you information about the world, which ought-words doesn’t.

2/ I give you virutally the same answer here. “Not-ought” is of course not synonymous with “ought”, but it belongs to the same class, namely the class of words that deals with morals.

Dunamis

Interesting line of thought. However, as I’ve said before, the difference between facts and values aren’t on a psychological or linguistic but on a metaphysical level. Searle famously tried to derive “ought” from “is”, and thought that he had done that through a five-step argument. Your argument is of the same kind. If such an argument indeed is valid, that only proves that either is a normative component contained in “is”, or a factual component contained in “ought” (as the conclusion in deductive arguments is always included in the premises).

(Perhaps it is misleading to talk about the “is”-“ought” distinction. In line with the above reasoning, “the fact-value-distinction” is a better name.)

Why do you think that the “derive” in Hume’s law is specific to his empiricism? As far as I know, many if not most philosophers believe in Hume’s law, and most of them are certainly not Humean empiricists. (They may of course be wrong about what their beliefs entail, though.)

I believe that the problem is the making two large groups of names. Kant did not say: All group of words associated with ought implies all groups of words associated with can.

I think one should only group words that are very similar together, and not if they can simply be put under a specific classification.

Stefan,

“However, as I’ve said before, the difference between facts and values aren’t on a psychological or linguistic but on a metaphysical level.”

Forgive me, but did you answer the question of the steak that “is”, the starvation that “is” and the eating that’s “ought”?

Further, there is no need to go to Searle. Quine destroyed the fact value distinction, when he undermined the verification of “facts”. A “fact” does not exist without its relationship to “beliefs” which orient and give it meaning. There is no value-free fact. “Facts” are metaphysical. There is no one who has ever presented a “fact” who did not do so without conceiving it within his or her value-belief dependent world picture.

“Why do you think that the “derive” in Hume’s law is specific to his empiricism?”

Because that is his point. The “ought” is a creation of habit and convention assembled out of empirical “is’s”. Kant is not “deriving” ought from is, but from transcendental categories. They simply are two distinct ways of looking at the world.

Dunamis

Dunamis/

No, because I said that it is not a question of linguistics or psychology but of metaphysics: fact and values are different. If an “ought” is derived from an “is”, that only shows that either does “ought” denote a fact (at least partly) or does “is” denote a value (at least partly).

I assume you’re speaking of facts as some kind of theory-free entities. I agree with that: such facts doesn’t exist, as all observation is theory-laden. But that hasn’t got anything to do with the present discussion: morals. Quine deals with the connection between theory and observation.

I don’t understand this. The issue isn’t why Hume thought one couldn’t derive “is” from “ought”, but why one would have to be a Humean empiricist in order to rationally believe in the validity of Hume’s law.

I dont’ see a contradiction. I understand “ought” to mean two things at the same time. If I “ought” to rescue someone from the buring house, it means that if I do it I will come closer to achieving my goal, whether that goal is happiness or being ethical. But when someone says “ought”, there is also an understood implication is that you actually can do it. “Ought” therefore has two meanings combined: one says that doing this will promote a given goal, and another says that it is actually possible to do this.

Hume’s statement makes sense because our observations of the exterior world, and our logical conclusions about them, cannot imply that we “ought” to do something. For it to be the case that we ought to do something, there needs to be a goal we are trying to work for, and this is not observable in the exterior world.

Given the defintion above, for it to be the case that someone “ought” to do something, you need both a goal and an action which you actually “can” do (which promotes that goal). Thus if you just have the “is”, which is the fact that you “can” do something, you only have half an “ought”. The other half requires a goal, which cannot be derived from the “is”.

Back to Kant, “can” is not derived from “ought”, so much as it is simply half of the meaning of “ought.”

When faced with problems like this, it is helpful to think of what the actual definitions of the words are. A good part of the time this provides a clear solution.

Stefan,

If an “ought” is derived from an “is”, that only shows that either does “ought” denote a fact (at least partly) or does “is” denote a value (at least partly).

If you simply mean that there is no qualifying difference between ought and is, then I agree with you. But then the apparent distinction that resides in Hume’s law simply does not exist. So what’s the problem? His distinction simply is invalid.

but why one would have to be a Humean empiricist in order to rationally believe in the validity of Hume’s law.

Because in referencing Hume’s law, you are referencing the use of terms as he intended, within his philosophical position. Otherwise it is meaningless to reference him, unless trying to authorize the “law” through his name alone. If “is” and “ought” is not being used as Hume intended it, simply reconstruct the “law” without reference to him and his empiricism.

It seems the whole “paradox”, is simply using “is” and “ought” outside of the restrictive senses in which they were meant by either philosopher.

Dunamis

Like I said, there is an interesting distinction between facts and values, more or less corresponding to the “is”-“ought”-distinction, but on a metaphysical, not linguistic level.

The only reason I mentioned Hume’s name is that this is perhaps the most well-known name of this distinction. Other names are “fact-value-disctinction”, “‘is’-‘ought’-distinction”, “‘is’-‘ought’-gap”, “Hume’s guilliotine”, etc. (They don’t denote exactly, but more or less, the same thing). I have no special interest in Hume but is interested in whether the fact-value-distinction holds.

Stefan,

“Like I said, there is an interesting distinction between facts and values, more or less corresponding to the “is”-“ought”-distinction, but on a metaphysical, not linguistic level.”

I’m still not understanding what you are saying. I am saying that the distinction is nonsense, and produces the apparent paradox. Drop the distinction and loose the paradox.

Dunamis