The following strikes me as a pretty
persuasive argument against a thorough-going process reliabilism. Since I’m no
expert on the field, I don’t know how similar it is to existing arguments
against process reliabilism, which is to say that if this turns out to be a
boring repetition of familiar points, well at least it wasn’t intentional
plagiarism.
Process reliabilism says that the
justification of a belief is proportional to the reliability of the process
that generated the belief. This raises the generality problem, as stressed in
Conee and Feldman’s 1998 paper – what is the process by which the belief
is generated? Or, to put the point more obscurely, what are the individuation
conditions for process types being used in this formulation. At one level the
generality problem is the problem of making the basic claim of process
reliabilism contentful – if we are prepared to count gruesome enough types,
then every belief is the product of some very reliable processes, and some very
unreliable processes. But let’s assume that problem has been handled.
At another level, the generality problem
raises a tension that I think can’t be resolved for a full-blown process
reliabilist. On the one hand, we want processes to be instantiated more than
one time, or else we’ll be led to the crazy view that a belief is justified iff
it is true. So we don’t want the instantiation to be too fine-grained.
On the other hand, the definition of justification entails rather immediately
(so immediately that it might surprise you to learn how long it took me to
realise this) every belief generated by the same process is equally justified. To
the extent that justificatory status can be very sensitive to the particular
ways a belief is formed, that implies we want processes to be individuated
quite finely. I think, and I think I have an example that supports this, that
these two constraints can’t be satisfied at once. Onto the example…
DIAGNOSIS
Morgan is
displaying symptoms S. Dr Watson knows that symptoms S normally
imply that the patient has a liver disease. But he also knows that in some
cases, happily enough in all and only cases where the patient has genetic condition
C, a patient with symptoms S doesn’t have a liver disease, but in
fact has a kidney disease. Dr. Watson also knows that genetic condition C
is rare, only 1% of males and 7% of females are C. And he knows that
there’s no easy way to test for whether a patient has condition C, for
usually it has no readily observable effects. And he knows he has no other
relevant information about whether Morgan is has condition C. So Watson
concludes that Morgan has a liver disease.
How justified is Dr. Watson’s belief?
I think you don’t know enough to say yet,
because you don’t know whether Morgan is male or female. If Morgan is male,
then Watson’s belief is very well justified. If Morgan is female, then Watson’s
belief isn’t particularly well justified, for he should be taking more
seriously the possibility that Morgan has condition C. Even in that case, it isn’t a disastrous
belief, but not as well justified as in the case where Morgan is male. Since
the two possible beliefs are not equally well justified, we need to say that
they are the results of different processes.
That alone might not be a problem. Perhaps
we can find a different way of categorising beliefs such that the belief that a
male patient displaying S has a liver disease falls into a different
category than the belief that a female patient displaying S has a liver
disease, though I’m not entirely convinced that existing (pure) reliabilist
theories have the resources to do this.
The problem is that the example generalises.
If x and y are both relatively small numbers, and Watson knows
that x% of males have condition C and y% of females do,
then his conclusion that Morgan has a liver disease is more justified if Morgan
is male rather than female for any such x and y, even if they are very
close, say x = 4.5 and y = 5, or even, I’d guess, if x =
4.5 and y = 4.51.
That means that we’re going to have to posit
infinitely many different categories of belief-forming processes, just to
account for all the different possible processes via which Watson could form
the belief that Morgan has a liver disease. The problem is that when categories
belief-forming processes get so fine-grained, we will start to get some
lucky guesses counting as justified beliefs, because they are the only beliefs
ever formed by that process, and some unlucky reasoned judgments counting as unjustified
beliefs, again because of the small sample size. This I take it should be
intolerable.
One response to related problems raised in
the 1980s was to modalise the notion of reliability. Maybe I’ll come back to
that in later posts, but I think it should be pretty clear that won’t help. The
problem is that there’s too many darn worlds to possibly count successes and
failures of a process, and no other approach to summarising the data from
nearby possible worlds seems to be much use.
This is not a problem for theories of
justification that incorporate some aspects of process reliabilism, but also
build in some more traditional internalist evaluations of modes of reasoning.
Ernie Sosa’s virtue reliabilism is like this, and my theory, which is
reliabilist about observational beliefs and (sorta kinda) foundationalist about
non-observational beliefs isn’t either. But a theory that is all process
reliabilism all the time really looks like it has problems with DIAGNOSIS.
Posted by Brian Weatherson at 2:20 am
Comments Off
There’s been a rush on the vagueness
experiment in the last few hours, from where I have no idea. Anyway, as best I
can tell from looking through the counters (and taking into account comments
like Ehud’s that they hit some of the counter pages
because they were just looking around) the score is now Consistency 53 – Contextualism
12. I’m going to be away from the computer for a few hours – at this rate the
over/under for the combined score when I get back is about 100.
UPDATE: It turns out that the flood of
responses to the vagueness experiment are because of this rather kind link
by Matthew Yglesias, who runs one
of the best combined academic/political blogs around. Go read it, and if you
agree you can even vote
for him in Dwight Meredith’s Koufax
Awards. The Koufax
Awards are for the best lefty blogs around, and are allegedly named after the
best lefty pitcher ever. Though in that case why they aren’t named the Grove awards
is a bit of a mystery. Perhaps it’s because if the award were really for best
lefty pitcher they’d have to change their name to the Johnson
awards sometime between when Randy starts next year’s All-Star game and when he
wins next year’s Cy Young
award. Oh, in the experiment the score is now 71-18, so everyone who took the under on the bet I mentioned above wins.
Posted by Brian Weatherson at 3:06 pm
Comments Off
Via Martijn
Blaauw I got a notice of this graduate
student conference in epistemology to be held in Amsterdam next May. It
looks like fun, and not just because it’s in Amsterdam. Anyone who can get
funding for going to Amsterdam and participating in a fun philosophy conference
should pause and reflect on just how much good fortune they possess. Sometimes grad students have all the luck!
UPDATE: I didn’t read the fine print very closely. It seems the deadline for submitting papers to this conference has passed. I don’t know how strict they will be about enforcing things like deadline rules. (It’s at the University of Amsterdam, you’d think there wouldn’t be things like rules anyway.) But if they are strict this isn’t as appealing as it first looked. Thanks to Alyssa Ney for picking up this little detail that I missed.
Posted by Brian Weatherson at 10:11 pm
Comments Off
I’ve been emailing with Adam Elga about his Dr Evil paper
(and my objections
to it) and on at least one point I’ve been totally trounced. I said that predicaments
were only situations that were in some way unpleasant, but Adam used the term
to cover all sorts of situations, even ones involving the twins from the Coors Light commercials. Adam
replied that Michael
Jordan uses ‘predicament’ the way he does.
“We’ve got 26 wins
and we still have 35 games left,” Jordan said. “We’ve got a good chance of
putting ourselves in a good predicament, which all along I felt like we could.
In some ways you want to think greedy, but nut-cutting time is starting to
come.”
You know I don’t think I’ve ever lost an
argument so convincingly since the last time I tried disagreeing with Tim
Williamson.
Posted by Brian Weatherson at 10:06 pm
Comments Off
I’m constantly amazed at the speed things
move at in internet time. One would expect a priori that daily
readership would be a function of, inter alia, how often the site is
updated. If the site is updated rarely, then people might only check it once a
week or so, whereas if it’s updated hourly, it will be checked more often. And
to a fairly large extent this is true. When I was posting a lot in September
and early October, my daily hits were much higher than in November, when I was
posting much less.
One might have also thought a priori
that cause and effect here would take time to play out. Updating five times a
day can’t increase hit numbers until people actually check the site, which may
not be for a week if they think you’re updating weekly. But this seems to be
wrong. A few days of actually putting content back on the site, and yesterday
was the most hits I’d got in a single day. There’s a lesson or sorts I’m
sure.
Posted by Brian Weatherson at 11:47 pm
Comments Off
From the one man’s modus ponens is another
man’s modus tollens department, here’s a passage on time travel from Ned
Markosian’s entry on time
in the Stanford Encyclopaedia of Philosophy.
And for another
thing, as I mentioned at the beginning of this section, we often think about
time travel stories; but it is very plausible to think that a story cannot
depict things that are downright impossible. For example, it is natural to
think that there could not be a story in which two plus two are five, or in
which there is a sphere that both is and is not red all over. (This seems
especially true if the story is told pictorially, as in the case of a movie.)
Hence, if time travel is impossible, then we should not even be able to
consider any story in which time travel occurs. And yet we do so all the time!
One task facing the philosopher who claims that time travel is impossible,
then, is to explain the existence of a huge number of well-known stories that
appear to be specifically about time travel.
Hmm. I seem to remember drawing exactly the
opposite conclusion from this. It was sort of a crucial point in my imaginative
resistance paper that since we can represent impossible time travel
situations in fictions, fictional representability did not entail possibility.
(This argument was lifted in its entirety from some almost parenthetical
remarks in Tamar Gendler’s paper on imaginative resistance.) So who’s drawing
the right inferences here?
Well, I think I am. (No? Really?!) Ned’s
looking for an argument that time travel is possible. But this argument
overgeneralises, for if it worked it would be an argument that many kinds of
time travel are possible, including changing the past Back to the Future style
time travel that most everyone agrees is impossible. Since we can represent
that kind of time travel, the fact that we can also represent the more seamless
kinds of time travel where the past and future all ‘fit together’ hardly shows
that that kind of time travel is possible.
Posted by Brian Weatherson at 11:32 pm
Comments Off
I was talking to Juan Comesaña about practical
rationality yesterday, and a family of puzzle cases came up about which we
couldn’t figure out what to say. So I thought I’d share them with you and see
if I could let you all figure out what my intuitions are, because right now I’m
not sure I can.
(By the way, much of the talking was done in
a mock interview, in which Juan looked pretty impressive. Anyone out there who’s
looking to hire a talented young epistemologist/ethicist should have his agent
on their speed dial. (Alert: strained American football analogy forthcoming.) Being
an epistemologist and an ethicist should be a natural combination, like being a
wide receiver and a cornerback. But for some reason it’s not only rare, the two
fields are put into two different groups, so departments usually have to fill
two lines to cover both needs. When roster space is tight, multi-skilled philosophers
should be more highly valued than they actually are. I don’t really mean to
compare Juan to Deion Sanders, but that’s just a limitation of my talent at
drawing analogies.)
Anyway, the examples. To get the framework
in place, imagine that you’re more or less a Humean about practical value. In
particular, you think that being practically rational means, most of the time, acting
so as to satisfy your preferences. But, you think, actions that you only
believe will lead to preference satisfaction because you hold irrational
beliefs are not practically rational. So in GOOD BEER I am practically
rational, in BAD BEER I am not.
GOOD BEER
I’m watching football, and I realise that I want a beer. I believe there is
beer in the fridge because I put some there not long ago. So I stroll out onto
the porch and get myself a beer.
BAD BEER
I’m watching football, and I realise that I want a beer. I believe there is
beer in the fridge, but only because I hope there’s beer in the fridge, and I
always believe the world is the way that I hope it is. So I stroll out onto the
porch to get myself a beer.
If you don’t agree with us about
those cases then either you disagree so deeply about practical rationality that
you’re not going to care much about the following examples, or you’ve got a
different concept in mind to the one we have. But I hope you can at least
imagine agreeing with us about those cases. If you do agree with us, then you’ll
probably agree that there are potentially difficult cases when a decision is
partially based on an irrational belief and partially on perfectly rational
beliefs. For instance, the following families of cases are troubling. (The
cases will be underdescribed because I’m interested in how intuitions vary as
we vary some of the parameters in them.)
BEER AND
SANDWICHES
I’m watching football, and I realise that I want a beer and a sandwich. I
believe that both are in the fridge, because I remember putting sandwiches in
there, and that there is beer in the fridge, but that’s only because I hope
there’s beer in the fridge, and I always believe the world is the way that I
hope it is. So I stroll out onto the porch to get myself a beer and a sandwich.
BEER AND COKE
I’m watching football, and I realise that I want either a beer or a Coke. For
some reason right now these are pretty good substitutes for each other for me
now. (If you can’t imagine this then substitute other products.) I believe that
both are in the fridge because I put some of each in there yesterday. So I
stroll out onto the porch and when I get there I decide I’d prefer a Coke so I
get one. But when I get back to the couch I’m a little upset, because I realise
I had just got myself a beer five minutes ago, and it isn’t finished yet, so I
could have had beer without moving too far from the couch.
I have my suspicions about what
my intuitions are in each (instance of each) case, but I’m sort of interested
in hearing what other people think before I post them, so maybe that will be
left until later today.
Posted by Brian Weatherson at 10:03 am
Comments Off
Wo thinks I’m wrong in
saying the world is gruesome. He’s got a point, though I think that the
underlying argument I wanted to make still goes through. Wo wanted to argue
that (1) was a better candidate to be a complete theory of everything than (2),
or presumably its cognates like (3).
(1) w
exists.
(2) Everything is F.
(3) There is a G.
In (3) I take G to be a predicate of
worlds that only this one satisfies. In (2) F is a predicate that any
thing satisfies in virtue of being a member of a world that is G. That
looks on the surface like a gruesome property, since it doesn’t make for
similarity among the things that satisfy it in any but a very artificial way. G
on the other hand has no such problem. Anything that satisfies G is just
like any (other) thing that satisfies G, probably because there’s only
one of them in the pluriverse. (Well, maybe there are more if there are
pluralities of duplicate worlds, a point that Lewis sometimes reveals
uncertainties about.)
So what I should have said is that (1) has
no virtues as a theory of the world over (3). I put this first by saying that
the world isn’t a particularly natural object. What I should have said is that
the world is exactly as natural as the property of being a G. Perhaps
the world is natural as objects go, though it looks to me like a fairly
scattered fusion of disparate elements. But if it is natural, then the property
of being just that way is also natural.
What also seems to be the case is that any
term that picks out w, or G, will have to be, in one sense or
other, magic. Because as natural as w might be, for any putative name
for w, say ‘@’, there will be several other eligible referents, all of
which will be just as natural as w, and all of which will satisfy our
uses of ‘@’ in all the respects that w does, perhaps save one. So for ‘@’
to refer to w rather than to w´ will take something that looks
like magic.
What about the one respect in which w
is not like w´, that we can make demonstrative reference to it. Well,
that’s an important difference, and it’s how we do manage to have names for
things like the actual world. But if you’re going to let demonstratives into
your theory of everything, then we could restate (3) as ‘Something is thus-ly’
demonstrating the way the world is. So I still don’t see a way in which
theories like (1) are better than theories like (3).
Posted by Brian Weatherson at 9:35 am
Comments Off
The vagueness
experiment is still running, though I guess most people reading this will
have taken it by now. The score right now is 26-3 in favour of sticking with
the first answer you give. I should stress that I don’t think this undermines
all the empirical claims made by contextualists about vagueness, or even any of
the claims made by all contextualists about vagueness. One different experiment
would have been to randomly start people at either end of a Sorites sequence
and walk them through it until they change their answer. Contextualists predict
that where people change their answer will be a function of which end they
start at, because there’s a bias to answer consecutive questions the same way.
That could be right, in fact I think it probably is, though it would be
interesting to test it.
UPDATE: It’s now 26-4. Is it fair to point
out that the last two votes in favour of changing one’s mind have come from
Cornell and St Andrews, two known locales of contextualists?
Posted by Brian Weatherson at 9:16 am
Comments Off
It’s not often I see an article in a major
web publication with my name at the
top of it. (It’s even rarer that I see this in a non-web publication
because I read so few of those.) Admittedly it didn’t have my name there as
such, but it sure looks like my name. In any case, the person Slate is
using my name as a nickname for doesn’t sound like the most pleasant
character ever. He did win a Rhodes scholarship, but from what I hear from
people around Oxford there’s little guarantee that that counts for much
character-wise.
Posted by Brian Weatherson at 9:12 am
Comments Off
