E.G.

In this post I’ll offer a few reasons to be pessimistic about the possibilities for experimental philosophy.  I should admit at the outset that I am not very well familiar with experimental philosophy.  Corrections of any wildly false claims made below would be much appreciated.

Consider the following scenario.  Suppose claims 1, 2 and 3 each seem to be intuitively appealing for independent reasons, yet the conjunction "1 and 2 and 3" yields a contradiction.  At least one of these claims has got to go.  Suppose philosophers have been talking for a long time about the question: Which of these claims can be discarded with the fewest associated philosophic and intuitive costs?  Schools of thought on this question have emerged.  "Anti-1ism" is the view that claim 1 has the least intuitive appeal; "anti-2ism" is the view that claim 2 has the least intuitive appeal; etc.  No consensus seems to be forthcoming; there are respected and prolific anti-1ists, anti-2ists, and anti-3ists.  Anti-1ism, anti-2ism and anti-3ism have thus each become "entrenched in the literature," and each view has developed its own philosophical tradition.

In an environment like this, people with extensive backgrounds in philosophy will be virtually certain to be very familiar with the anti-1ism/anti-2ism/anti-3ism debate.  In that case, it’s likely that their intuitions will have become "polluted."  They will no longer be in a position to know which way their intuitions are telling them to go.  It’s possible, in fact, that what they claim to be "intuitively plausible" will seem downright absurd to an ordinary non-philosopher.  This, I take it, is where experimental philosophy is supposed to come in.  To approach a solution to the "1 2 3 problem," an experimental philosopher might present versions of claims 1, 2 and 3 to a group of ordinary people and ask each subject whether she finds any of the claims difficult to reject.  It might turn out that although ordinary people find it very difficult to reject claims 1 and 3, claim 2 doesn’t have any intuitive appeal at all to them.  If this happens, then the case for anti-2ism will become quite strong.  Progress will have been made on a problem which philosophers have long been unable to solve.

I don’t have any principled objections to the basic approach just described.  The problem I have is somewhat more practical.  To get at the problem I have in mind, suppose that around 75% of philosophers are anti-2ists, 15% are anti-1ists, and 10% are anti-3ists.  If the views are distributed in this way, has the "1 2 3 problem" been solved?  In many disciplines (such as, say, physics), problems which have been "solved" command universal assent.  If only three quarters of all physicists believe X, for instance, then X is fairly controversial, and the "problem of X" should not be regarded as solved.  Perhaps the bar should be lowered for philosophy, though I don’t think so.  Let’s assume, at any rate, that in the scenario above, where the majority view commands the assent of only 75% of all philosophers, the "1 2 3 problem" is not solved; some number higher than 75% would be necessary.

Now suppose that experimental philosophers go out and find a similar distribution of views among educated laypeople.  That is, suppose they find that 75% of laypeople want to reject 2, 15% want to reject 1, and 10% want to reject 3.  What would these experimental philosophers conclude from this finding?  I suspect that they may conclude that rejecting 1 and 3 would come with a very high intuitive cost, and that the numbers strongly favor rejecting 2.  That is, they would take their results to strongly favor anti-2ism.  But this, I claim, wouldn’t be "fair."  It would mean that experimental philosophy would be thought to have "succeeded" without having outperformed the purportedly "failed" traditional philosophical approach.  Why should ordinary philosophy be deemed a failure at 75% if experimental philosophy can succeed with the same number?

As just one example chosen at random, consider this quotation from Nahmias at Experimental Philosophy:

We ran variations with positive action (saving a child from a burning building) and neutral action (going jogging). But in all three cases, a significant majority of subjects (68-79%) judged that Jeremy acted of his own free will or that he was praiseworthy for saving kid (89%) or blameworthy for robbing bank (83%).

For the background on this quotation, I encourage you to follow the link provided above.  The point I want to make is independent of the context in which these results are reported.  My point is this: If 68-79% of all physicists believed that if I let go of this ball, then it might not fall, I’d say there’s no consensus in physics about the laws obeyed by gravity.  Similarly, if only 68-79% of all philosophers would say that in some imagined scenario, Jeremy acted of his own free will, then I’d guess that there’s significant controversy among philosophers about free will.  I claim that the same standards should be applied to the population of laypersons whose intuitions are studied by experimental philosophers. 

At this point, though, I think experimental philosophy faces a dilemma.  If experimental philosophy "lowers the bar," and claims (for instance) that 75% of laypeople is a conclusive result (a success), whereas 75% of philosophers is inconclusive (and a failure), then someone will need to explain why the bar is so much lower for experimental philosophy than it is for regular philosophy.  But if experimental philosophy doesn’t lower the bar, then it seems doubtful that experimental philosophers will be able to provide the high numbers they’ll need to outperform "regular" philosophy. 

Which, if any, of the following claims are true?

(1) Conservatives are able and willing to reasonably disagree with their opponents.  Liberals cannot stand to have others disagree with them; liberals will do anything to compel assent.

(2) Conservatives don’t really want to find consensus on issues; they just like to argue, and they engage in argument as an end in itself.  Liberals, on the other hand, sincerely care whether their arguments are convincing to their opponents.

(3) Liberals don’t really believe anything.  Liberals care only about consensus, and are willing to compromise their deepest values in order to acheive it.

(4) Conservatives are stubborn and pig-headed; they can see only one side of any given issue.  Conservatives are unwilling to concede anything, even when doing so would make their position stronger and closer to the truth.

There’s a theme running through all these claims.  In all four claims, liberals are associated with agreement, consensus, and compromise; conservatives are associated with disagreement, controversy, and stubbornness.  I’ve recently seen variations on all four of these claims in various comment threads around the internet.  Is there some common source?  For my part, I don’t see much truth in any of these four claims.  I assume that there are a fair share of stubborn liberals, and that there are also a fair share of compromise-happy conservatives.  Even as extremely rough generalizations, I’m not sure I see reason to think any of the above claims are true.  But it’s odd that these themes are popping up somewhat frequently in discussion threads. 

The new Philosophers’ Carnival is up at Studi Galileiani.  Looks like a lot of good stuff.  I’m amused by the summary of my post on Dreier’s Conjecture:

Staying with moral theory, Andrew Sullivan at E.G. considers Dreier’s Conjecture; namely, that non-consequentialist approaches can be reduced to consequentialism. Sullivan disagrees that the distinction between the two is empty and sets out a detailed argument to this effect, concentrating on the premise that for any non-consequentialist theory there is a "counterpart" consequentialist theory leading to the same verdict in identical circumstances (with the caveat that both have to be characterised as “plausible“).

This is a pretty fair summary of my post, but I have one minor quibble: I am not Andrew Sullivan.  I guess the person who wrote this summary noticed that I had written a post entitled "Andrew Sullivan" and assumed that I wouldn’t be writing about Andrew Sullivan unless I was Andrew Sullivan.  That is an incorrect, but forgivable, assumption.

UPDATE: As Hugo mentions in comments below, he has amended the post so that the first mention of Andrew Sullivan is corrected.  There still remains a second mention of Sullivan in the post, but that’s OK.  It’s not a bad thing to be confused for a talented and successful pundit like Andrew Sullivan.  I’m just glad the summary didn’t read something like this:

Staying with moral theory, Bill O’Reilly at E.G. considers Dreier’s Conjecture; namely, that…

UPDATE (five minutes later): Hugo’s fixed all the Sullivan-mentions in the summary now.  Thanks, Hugo! 🙂

From Ebert’s Glossary of Movie Terms:

We’re Alive! Let’s Kiss!

Inevitable conclusion to any scene in which hero and heroine take cover from gunfire by diving side-by-side into a ditch, and find themselves in each other’s arms, usually for the first time. Cf. HIGH ROAD TO CHINA.

There is usually something bad about concluding a scene in this way.  But what is bad about it, exactly?

(more…)

This site looks interesting, although I haven’t had time to really explore it yet.

Connard points out that Islamist terrorists have finally captured G.I. Joe.  This is a sad day for the war on terror/war on Cobra.

Here are some pictures of Cobra Commander.  Here is a picture of someone else. 

My last post on this topic left some threads hanging.  In this post I’ll try to pick up some of those threads.  Ideally you will have read the previous post before reading this one, but if you’re already familiar with this topic you might not really need to do that.

Recall that, according to Dreier’s Conjecture, there is, for every non-consequentialist theory, a conceivable "counterpart" consequentialist theory which provides the same verdicts in the same circumstances as the "original" non-consequentialist theory.  In the previous post, we saw that even if Dreier’s Conjecture is true, this is not enough to motivate the view that the consequentialism/non-consequentialism distinction is unimportant, insignificant or otherwise "empty."  Among other things, that result requires another thesis, which I will call the Verdict Thesis:

Verdict Thesis: A moral theory M is not importantly distinguishable from another moral theory N if M and N yield the same verdicts in the same circumstances.

The Verdict Thesis is probably false.  It is probably false because, as was noted in the previous post, moral theories do all sorts of important things other than yield verdicts.  But it is probably false for another reason, as well. 

Let’s say that a theory is remotely plausible if it is such that at least one minimally rational, consistent, well-informed person, who is not utterly morally depraved, could accept it.  Now consider a non-consequentialist theory N1.  Suppose Dreier’s Conjecture is true.  Then there exists a consequentialist counterpart to N1; let’s call it M1.  But suppose N1 is remotely plausible, while M1 is not.  For instance: suppose that you’d have to believe that George Bush has six heads in order to consistently accept M1, whereas to accept N1 you do not need to believe any such wildly false claims.  This fact alone, I would think, permits us to "importantly distinguish" between M1 and N1.  So even though M1 and N1 yield the same verdicts in the same circumstances, we can importantly distinguish between them.  Thus the Verdict Thesis is false.

In the example just discussed, the existence of M1 shows that Dreier’s Conjecture is true, at least for the case of N1.  But I doubt anyone would think, even for a minute, that this means that the consequentialism/non-consequentialism distinction is "empty."  Why would no one think this?  Here’s my suggestion.  For Dreier’s Conjecture to be significantly true, it has to be the case that if you start out with a remotely plausible non-consequentialist theory N, you’d better be able to find a remotely plausible consequentialist "counterpart" theory M.  If all the possible counterparts to N are not even remotely plausible, then even if Dreier’s Conjecture is true, it’s not very interesting that it is true.

So, what is needed is to supplement Dreier’s Conjecture so as to accommodate this "remote plausibility of the counterpart" requirement.  To meet this need, I suggest we consider "P-Dreier’s Conjecture": 

P-Dreier’s Conjecture: For any remotely plausible non-consequentialist theory N, there exists a remotely plausible "counterpart" consequentialist theory M which yields the same verdicts in the same circumstances as the original.

If we can show that P-Dreier’s Conjecture is true, then at least we’ll have the beginnings of a case for the view that the consequentialism/non-consequentialism distinction is empty. 

Unfortunately for those (such as myself) who want the consequentialism/non-consequentialism distinction not to be empty, it appears that P-Dreier’s Conjecture is true.  In Dreier’s original paper on this topic you can find evidence for that view, although as I recall Dreier himself does not seem to notice that some notion like that of plausibility should figure into his case.  I won’t provide a case for P-Dreier’s Conjecture here.  But consider this: Most remotely plausible non-consequentialist theories take account of the "intrinsic" features of a given action as reasons for prohibiting or prescribing that action.  As Dreier and others have shown, it is usually not hard to incorporate those features into a coherent theory of value, and thereby to produce a "counterpart" consequentialist theory.  Moreover, it seems usually to be the case that when those features are so incorporated, they yield a theory of value which is intuitively plausible for the same reasons that the original, non-consequentialist theory was plausible.  This should mean that if you start out with a remotely plausible non-consequentialist theory N, you can use it as a guide to produce a remotely plausible theory of value, and thereby to produce a remotely plausible consequentialist theory M.   So, I tentatively suggest that P-Dreier’s Conjecture is true.

In passing, it should be noted that we use this notion of "remote plausibility" to modify the Verdict Thesis, like so:

P-Verdict Thesis: A moral theory M is not importantly distinguishable from another moral theory N if (a) M and N yield the same verdicts in the same circumstances, and (b) If N is remotely plausible, then M is remotely plausible.

It might be thought that if we combine P-Dreier’s Conjecture with the P-Verdict Thesis, we will have a pretty good case for the view that the consequentialism/non-consequentialism distinction is empty.  But I want to show that that is not the case.

Recall the discussion in my previous post.  There, it was suggested that Dreier’s Conjecture only tells half the story.  Dreier’s Conjecture says that for any non-consequentialist theory there exists a "counterpart" consequentialist theory.  It does not say that for any consequentialist theory there exists a non-consequentialist "counterpart" theory.  Thus what is need, I claimed, is to consider a thesis which I called "Reverse Dreier’s Conjecture":

Reverse Dreier’s Conjecture: For any consequentialist theory M, there exists a "counterpart" non-consequentialist theory N which yields the same verdicts in the same circumstances as the original.

We are now in a position to introduce the notion of "remote plausibility" into Reverse Dreier’s Conjecture to yield

P-Reverse Dreier’s Conjecture: For any remotely plausible consequentialist theory M, there exists a remotely plausible "counterpart" non-consequentialist theory N which yields the same verdicts in the same circumstances as the original.

I claim that if we combine P-Dreier’s Conjecture, the P-Verdict Thesis, and P-Reverse Dreier’s Conjecture, we have a very good argument for the view that the non-consequentialism/consequentialism distinction is empty.  To put this point a little differently, lay your eyes on the following:

1. P-Dreier’s Conjecture is true.
2. The P-Verdict Thesis is true.
3. P-Reverse Dreier’s Conjecture is true.

I think that if 1-3 are accepted, then one can provide a very strong argument for the view that

4. The consequentialism/non-consequentialism distinction is empty.

One might take issue with premise 2.  I am not interested in doing that; as was mentioned previously, others (e.g. Doug Portmore) have given good reasons to reject 2.  I will here offer a case against 3.

To do this, what we need is to find a remotely plausible consequentialist theory for which no remotely plausible "counterpart" non-consequentialist theory exists.  This would provide a clear counterexample to P-Reverse Dreier’s Conjecture.  This, in turn, would show that premise 3 is false, and therefore that at least the argument we’re considering just now can’t be used to show that the consequentialism/non-consequentialism distinction is empty.

To get at the counterexample to P-Reverse Dreier’s Conjecture which I have in mind, consider the following scenario.  Suppose, back in 1937 (or whenever), Hitler is faced with a choice: (A) To carry out the holocaust, resulting in the deaths of millions of innocent people; or (B) not to carry out the holocaust, sparing millions of innocent people.  (Probably, Hitler never really was faced with a singular moment of choice like this; the holocaust was probably the result of thousands of comparatively smaller decisions on his part, as well as on the parts of others.  But for the purpose of this scenario, assume there was just one moment of choice.)  We should and would say: Hitler ought to have made choice (B) — and it was terribly wrong to make choice (A).

Now suppose that whether Hitler chooses (A) or (B) will have effects for the next thousand years.  That is, suppose that the world in the year 2937 would be different if Hitler chose (A) than it would be if Hitler chose (B).  And suppose that, by means of some extremely complicated causal interconnections, the amount of pleasure experienced between 1937 and 2937 would be slightly higher if Hitler chooses to do (A) than if Hitler chooses to do (B).

Given this information, most people will still say that Hitler ought to choose to do (B).  But I’m guessing you can see where I’m going with this.  Given this supposition about the long-term effects of choosing between (A) and (B), at least some versions of hedonic utilitarianism are committed to the view that Hitler ought to choose (A).  Such versions say, for instance, that the only thing of value is the quantity of pleasure; it does not matter who experiences the pleasure, or why, or when.  And our obligations are simply to choose those actions which, over time, maximize the total amount of that one thing of value.  Since choosing to carry out the holocaust would, over the course of one thousand years, result in the experience of a (very slightly) greater amount of pleasure than not choosing to do so, Hitler ought to have carried out the holocaust.

Of course, most real hedonic utilitarians will probably try to get out of this result.  They might try to show, for instance, that the scenario I’m envisioning could never happen.  But at least some hedonic utilitarians will "bite the bullet" and say that, if we suppose, contrary to fact, that the holocaust would have produced a very slightly greater amount of value than otherwise, then carrying out the holocaust was the right thing to do.

Now, I don’t think this is a plausible verdict.  That is, I think that in the scenario just described, Hitler still ought not have carried out the holocaust.  I suppose most reasonable people would agree with me.  But some hedonic utilitarians, in fact, have endorsed the view that it is right to carry out actions like Hitler’s if such actions produce a slightly greater amount of pleasure.  I’ll go out on a limb and venture the guess that the Ethical Werewolf would accept this verdict; I seem to recall him once saying he would accept a similarly counterintuitive verdict.  (Apologies, of course, to the Werewolf if I am here misrepresenting his views.)  Since at least some smart, minimally well-informed, relatively self-consistent, non-depraved people have endorsed the view that pleasure is the only thing of value, and that one ought always to maximize value, even when doing so means committing atrocities, this view is not completely absurd.  In other words: I think this version of hedonic utilitarianism just described is remotely plausible, even though it is, for most people, a deeply counterintuitive position. 

So we have in hand a consequentialist theory which is remotely plausible and which yields the verdict that in the imagined scenario, Hitler ought to have carried out the holocaust.  We can easily produce a non-consequentialist theory which yields the same verdict.  But I claim we cannot produce a remotely plausible non-consequentialist theory which yields the same verdict.

My claim: In the imagined scenario, the only consideration which could remotely plausibly support the verdict that Hitler ought to have carried out the holocaust is the consideration to which the hedonic utilitarian points: the consideration, that is, that over a very long period of time, the holocaust maximizes pleasure.  I claim that no remotely plausible moral theory could support the hedonic utilitarian’s verdict without being based solely upon that singular consideration.  If I am right about this, then hedonic utilitiarianism is the only remotely plausible moral theory which can yield this particular verdict.  If so, then it follows that for at least one remotely plausible consequentialist theory, there does not exist any remotely plausible non-consequentialist counterpart.  And in that case, it follows that P-Reverse Dreier’s Conjecture is false.  Premise 3, therefore, fails.

To conclude, recall the argument described above:

1. P-Dreier’s Conjecture is true.
2. The P-Verdict Thesis is true.
3. P-Reverse Dreier’s Conjecture is true.
—
4. Therefore, the consequentialism/non-consequentialism distinction is empty.

I think this argument, if expanded, would work if one could gain assent to premises 1-3.  But others have given reason to think that premise 2 is false.  And the Hitler/holocaust case just described gives very good reasons to think premise 3 is false.  So the conclusion cannot be derived by means of the three premises above.  This is not to say that the conclusion cannot be derived by some other means.  But until those other means are proposed and defended, I think we should go on behaving as though the consequentialism/non-consequentialism distinction is non-empty.

Andrew Sullivan claims he’s going to put the "Dish as you’ve known it" on hiatus for a while in order to work on longer projects.  Sullivan is one of my favorite bloggers and I’ll miss his regular on-line presence, if he does indeed stick to his plan.  Sullivan’s is the first blog I ever read.  I was listening to some radio show, he was a guest, and he plugged his blog, so when I got home I looked it up.  For a long time thereafter, Sullivan’s was the only blog I read — I wasn’t really aware there were others.  Then one day I followed a link (I think it was to Crooked Timber) and discovered someone’s blogroll.

I greatly admired the hand-held camera-work in this movie, which in English translation is called "The Son."  I was brought so far into this movie’s world by some of the opening scenes that when I became bored with some later scenes, it did not even occur to me that I had the power to turn the movie off and find something more interesting to look at.

I was reminded of that experience, and more generally of how much I like to watch well-made movies shot with hand-held cameras, when I watched "Friday Night Lights" the other day.  I don’t know anything about movie-making, but I imagine it would take great skill to do a good job filming with a hand-held camera.  Your whole body, I guess, becomes the camera: If you jerk unexpectedly, the camera jerks unexpectedly, and the audience is immediately reminded that it is watching unreal, made-up events.  That’s why they invented the "steadicam," I guess.  But when done well, those little jerks can foster, rather than dispel, the illusion the movie aims to produce.  After all, in the real world, one’s eyes are continually darting here and there, and routine movements of one’s body are continually causing one’s field of vision to jerk this way and that, so we should expect to see the movie-world in the same way.  When it’s done well, then, I think hand-held cinematography allows us to see the world of the movie in a way very similar to the way in which that world would look, if we were really in it, watching the people and looking around.

Here’s an interesting post from Dworkin over at Left2Right.  In the course of the post, Dworkin considers the claim that "argument is only useful if there are common premises that both parties share."  (Obviously, the two "parties" here are the person giving the argument, and the person hearing the argument.)  Let’s call this the "principle of shared premises."  Dworkin says:

This [i.e., the "principle of shared premises"] might seem to be a truism; actually it is false. It is possible to argue ad hominem, i.e. showing an opponent that from premises he accepts, although I do not, one gets a contradiction. Thus he must give up at least one of his views. This, of course, does not show the view he abandons is false. Just that he has no grounds for holding it.

I think Dworkin means to refer to reductio-style arguments here, not ad hominem ones (though I’m not completely sure of this; see the usage note in the second link).  Reductio-style arguments do seem to be a counterexample to the principle of shared premises, but I’m not sure they really are.  Here I’ll show why, but first it will be worthwhile to say a few very basic things about deductive propositional logic. 

In deductive arguments, not all assumptions are created equal.  Let’s distinguish "regular assumptions" from "working assumptions."  Regular assumptions are those assumptions upon which my conclusion depends.  Consider this argument:

1. P
2. If P, then Q
3. If Q, then R
4. Q (from 1 and 2)
5. R (from 3 and 4)
6. Thus, if P then R

This argument is of a form sometimes called "conditional proof."  It should be obvious that my conclusion in 6 depends on the assumptions in lines 2 and 3, but not on the assumption in 1.  That is: As long as 2 and 3 are true, it does not matter whether 1 is true or false; 6 has to be true.  Thus 2 and 3 are regular assumptions, while 1 is a working assumption.

Now here’s a reductio-style argument:

1. P
2. If P, then Q
3. If P, then not Q
4. Q (from 1 and 2)
5. not Q (from 1 and 3)
6. Q and not Q (from 4 and 5)
7. Thus, not P

Again: To compel assent to my conclusion in 7, 2 and 3 must be true, but premise 1 obviously does not need to be true.  Thus 1 is a working assumption, while 2 and 3 are regular assumptions.

Now again consider the principle of shared premises, which says that "argument is only useful if there are common premises that both parties share."  "Premises" here presumably refers to what I’ve been calling "assumptions."  If the principle is supposed to say something about assumptions in general, then the principle is obviously false, as Dworkin says; we’ve already seen two argument-forms which are indisputably useful even though they do not require anyone, let alone "both parties," to believe all the assumptions involved.  But if the principle is supposed to be talking about just regular assumptions, not working assumptions, then I think the principle is defensible.  In that case, the principle would say that arguments are only useful when both parties assent to all the regular assumptions.  It’s not clear whether this is the case or not.  We know that arguments are not useful when the person hearing the argument doesn’t assent to all the regular assumptions; the question, then, is whether arguments can be useful when the person giving the argument doesn’t assent to all the regular assumptions, i.e. "begs the question against herself," as I have called it previously.

………………………..

After writing this post, I e-mailed Dworkin to ask whether he meant something other than "ad hominem."  He said:

No, I mean ad hominem.  Not in the sense of arguing about the caracter of the opponent, but in the sense of arguing that HE cannot accept all the views he holds.  This leaves open the question of the truth or falsity of what he gives up. 

Here is a source which uses the term as I intended to use it.

The second form of the ad hominem was identified by John Locke <link> in his Essay Concerning Human Understanding <link>, when he wrote that it was

to press a man with consequences drawn from his own principles or concessions.  This is already known under the name of argumentum ad hominem.