Sunday, July 05, 2009

Is it really against reason?

Is it really against reason to say, "I will do what is good for me, and what happens other people does not matter. Kant thinks I have to be willing to universalize my principles. Why? I'm not other people. Of course, let other fools be ethical. I will do what I have to do for me.


IlĂ­on said...

It isn't a sensible claim to say that "Thus-and-Such is against Such-and-Such" unless Such-and-Such contains an oughtness component.

Does 'reason' subsume 'morality?' Does 'morality' subsume 'reason?' Or, are they two different things, one of which ought to inform the use of the other?

a helmet said...

I think Kant is in line with the golden rule: Everything you want others do to you, you shall do to others. For that's the law and the prophets.

Steve Lovell said...

I've been looking at Gewirth's version of Kantian rationalism on and off for 10 years or so, and while I'm really not persuaded, his champion Derek Beyleveld is formidable. I think this is the best modern attempt to rehabilitate a thoroughgoing ethical rationalism. There are other weak "rationalisms" too, but they really aren't at all Kantian.

Does anyone have a take on Gewirth?

I can make a long post summarising the Gewirth/Beyleveld line of thought if it's of interest use to anyone.

Steve Lovell

JSA said...

@Steve - eah, I'd be interested to read a post; do you have a blog?

Laszlo Mero's famous game theory book figured Kantian ethics prominently throughout, and IIRC he contrasted Kant subtly with "Golden Rule", and ended up not being very favorable toward the golden rule. I'm not exactly recommending the book, since he got a little weird at the end (it's excellent intro to game theory, though). But I found the discussion of Kant vis-a-vis game theory thought-provoking.

Steve Lovell said...


No, I don't have a blog. I was just going to post here in this thread. I'll see what I can work up. It'll be a few days before I have it done.

If Vic sees fit he can turn it into a thread of its own.

Steve Lovell

Steve Lovell said...

I have to split this up as it's too long for a single comment

A summary of Gewirth’s / Beyleveld’s Argument

Basically Gewirth attempts to show that a fairly substantial moral principle is dialectically necessary. That is to say that one can only deny this principle at the cost of denying/contradicting that one is purposive agent. Thus this principle is internally connected with “the will of a rational being as such.” The substantial moral principle that the argument concludes with is known as the Principle of Generic Consistency (The PGC) and is stated thus: All prospective purposive agents have a (claim) right to freedom and well-being.
The argument comes in three stages. The first attempts to establish that an agent must regard their own freedom and well-being as necessary goods. The second then attempts to establish that on this basis an agent must claim they have a “prudential” right to their freedom and well-being. The third stage attempts to universalise from this result to the conclusion that an agent must be committed to the PGC.


A Prospective Purposive Agent (PPA) has purposes. (Def.)

Any PPA will be committed to

(1) I am a PPA

In theory (1) should not only be true for a PPA it should also be dialectically necessary. Perhaps this is established like this: Suppose a PPA denies (1). It seems that they must think this a good thing to do, they do it for a reason. But this would be enough to show that the denial is false, for only PPAs do things for reasons. Whether or not this is right is not really what I’m interested in. Rather I’m more concerned about the derivation from (1) to the PGC, The Principle of Generic Consistency.

Given the definition of a PPA, anyone accepting (1) must accept

(2) I aim to achieve end E

since one must conceive of ones own ends as good, anyone accepting (1) and (2) must also accept

(3) E is good

Where the agent in question also realises that

(2a) X is a necessary means to E

they must be committed to
(4) X is good

But the generic features of (successful) action are necessary means to any ends that I may have. To have any chance of acting (successfully) I must have these things. It seems (at least) plausible to think that this means that my freedom and well-being are such necessary means. I must think I have a reasonable chance of achieving my ends in order to bother attempting to. This seems to require some minimal sense of Well-Being/Self-Esteem. I certainly must be free of anything which would prevent action. This clearly means having some minimal kind of Freedom. So an agent who accepts (1) now seems to be committed to

(5) My Freedom and Well-Being (F&WB) are necessary goods

Part2 follows

Steve Lovell said...


It follows from (5) that

(6) I ought to protect/pursue my F&WB

The criterion for this ought is prudential. It is, if you like, a hypothetical imperative. However, it is not one which can be escaped by not having the relevant goals. All that is required is that one have SOME purposes, and since one’s F&WB is necessary for the achievement of those ANY goals there is an imperative to protect/pursue one’s F&WB.
Now using the same criteria for ‘ought’ statements, that is ones own interests (broadly) anyone committed to (6) is also committed to

(7) Other PPAs (PPAOs) ought not interfere with my F & WB

So anyone committed to (1) is now committed to (6) and (7) too!
Now, since “oughts” are correlative with rights anyone committed to these propositions will also be committed to

(8) I have a claim right to my F & WB

This is, however, not a moral claim at this point ... since it is only a claim based upon my own criteria of ‘oughtness’, not necessarily anyone else’s criteria.

Notice however, that each of the steps (should!) follow by pure logic. An agent denies that he/she is a PPA if they deny any of (1)-(8).


Now the logic of the argument so far seems to require that an agent committed to (1) believe that

(8a) My being a PPA is sufficient grounds for my having a claim right to F & WB

This is because accepting (8a) seems to be equivalent to seeing that from one’s own perspective (1) entails (8). This seems a bit dubious, but fortunately is expanded upon by the “Argument For/From the Sufficiency of Agency.” (The ASA)

(a) Suppose that my having a claim right to my F&WB is due, not to being a PPA but to something else, about me, G.
(b) But if G is not a feature necessary for a PPA to have (not internal to the idea of a PPA) then, it will not be dialectically necessary for me as PPA to claim a right to my F&WB. That is, if I didn’t have G, I would not have the right, so since it is not dialectically necessary to affirm that I have G, neither is it dialectically necessary for me to affirm that I have the right to my F&WB.
(c) But it is dialectically necessary for me to claim a right to my F&WB.
(d) So I must assert that the sufficient ground of this right is my being a PPA.

The next stage of the argument utilises the logical principle of universalisation (LPU).

(8b) If A’s possessing feature x is sufficient grounds for his possessing feature y, then B’s having x is also sufficient for B’s having y.

Now as the agent must accept LPU, he must accept that if being an agent is sufficient to ground his claim right ... then it must be sufficient to ground it for others too.
so the agent must accept that

(9) All PPAs have a right to their F&WB

Further anyone going through this chain of reasoning will realise that the same reasoning will apply to all PPAs. Which is to say that not only do I contradict/deny that I am a PPA if I reject (9), but so to does any other PPA. Thus all PPAs are necessarily committed to (9). Thus, allegedly, the PGC is established.

I'll leave any comment on this reasoning until later. Others can feel free to comment in the meantime.

Steve Lovell

Steve Lovell said...

No other commentators on this? Oh well. All the more for me.

I have several problems with the argument, but I think the big problem is somewhere between (7) and (9) and depending on how you finess the terms you can either fault the move to or from (8).

The basic issue to my mind is that the reference to the "oughts" being grounded in the reasoners desires has disappeared in Stage III. If we leave in this reference then it isn't clear whether the application of the LPU will deliver (9). If particular, suppose read (7) as

(7') Other PPAs (PPAOs) ought not interfere with my F & WB if I am to have my desires fulfilled.

And we universalise from this without translating in to the language of rights. To do this we need

(8a') My being a PPA is sufficient grounds for other-directed oughts which must be followed if I am to have my desires fulfilled.

Which universalises to

(9') Anyone being a PPA is sufficient grounds for other-directed oughts which must be followed if they are to have their desires fulfilled.

Now this may be a fair univeralisation, but it isn't what is stated by the PGC. In particular it doesn't seem to give us any reason to act in favour of other agents. The only reason to suppose it does is to think that other people's desires are as important as one's own. Now this may be true, but is hardly something the argument can assume.

I have very good reason for treating my desires differently that other people's: that they are mine. To take a parallel case, some of my behaviour towards my daughter is not only explained by the fact that she is a child, but more specifically by the fact that she is my child. To assume, without argument that all desires (including those of other people) should be given the same weight is like assuming without argument that I should show no favouritism to my child over other people's children.

Steve Lovell