© 2010 Blackwell Publishing Ltd

Ratio (new series) XXIII 1 March 2010 0034–0006

SCEPTICISM ABOUT PHILOSOPHY rati_447 1..16

Jason Brennan

Abstract
Suppose a person who is agnostic about most philosophical issues
wishes to have true philosophical beliefs but equally wishes to avoid
false philosophical beliefs. I argue that this truth-seeking, error-
avoiding agnostic would not have good grounds for pursuing
philosophy. Widespread disagreement shows that pursuing phi-
losophy is not a reliable method of discovering true answers to
philosophical questions. More likely than not, pursuing philosophy
leads to false belief. Many attempts to rebut this sceptical argument
fail.

1. Insider and Outsider Scepticism about Philosophy

Philosophers disagree immensely in significant ways. Our best

philosophers disagree over the doctrines, methods, and even the
aims of philosophy. Experts in all fields disagree, but disagree-
ment is more pervasive in philosophy than in most other fields. As
Thomas Kelly says, ‘Philosophy is notable for the extent to which
disagreements with respect to even those most basic questions
persist among its most able practitioners, despite the fact that the
arguments thought relevant to the disputed questions are typically
well-known to all parties to the dispute.’1
A sceptic might claim that radical dissensus shows that pursuing
philosophy is not a good means for discovering true answers
for philosophical questions. Dissensus shows that philosophical
methods are unreliable instruments of truth. Suppose an uncom-
mitted person comes to philosophy hoping to get true answers to
her philosophical questions. She wants to know what that nature
of causation is, what justification is, what rightness consists in,
what justice is, and so on. She notices that philosophers have
extensive disagreement about the answers to these questions and

1
Thomas Kelly, ‘The Epistemic Significance of Disagreement’, in John Hawthorne and
Tamar Gendler, eds., Oxford Studies in Epistemology (Oxford: Oxford University Press, 2006),
p. 173.
2 JASON BRENNAN

thus concludes that the probability of her getting the true answer
by pursuing philosophy is low. So, she becomes a sceptic about the
field of philosophy and walks away with her questions unan-
swered. Is she making a mistake?
In this paper, I consider scepticism of the sort that holds that
there are true answers to philosophical questions, but none of us
are in a good position to know these answers. This type of scepti-
cism admits of two sub-types. 1) An insider sceptic holds that even
the best philosophers lack good reasons to hold their views. So,
the insider sceptic thinks that philosophers who are not agnostic
about philosophical issues should become agnostic. 2) A person
who is merely an outsider sceptic, on the other hand, might accept
that many philosophers are justified in holding their views,
despite widespread disagreement. The outsider sceptic need not
hold that philosophers should change their beliefs or become
agnostic. However, the outsider sceptic also holds that people not
already committed to one philosophical position or another
should stay uncommitted. So, the outsider sceptic holds that even
if most philosophers are justified in accepting their different
views, a person who lacks philosophical beliefs ought to refrain
from using philosophical methodology and instead should
remain agnostic.
Suppose an uncommitted person, one who is currently agnostic
about basic philosophical questions, wishes to discover the true
answers to these philosophical questions. She is also equally con-
cerned to avoid false answers. She is thus willing to stop being
agnostic and come to believe a doctrine provided she does so via
a reliable method. For her, a reliable method is one that is at least
more likely than not to give her true beliefs. If these are her goals,
it is difficult to show that philosophy as we do it would be worth
doing. She might as well remain agnostic. This is not to say that we
philosophers must give up our doctrines and become agnostics
ourselves, but merely that a truth-seeking, error-avoiding agnostic
does not have good reason to pursue philosophy in the attempt
to discover the truth about philosophical questions. This paper
argues that the presence of widespread dissensus makes it difficult
to defend philosophy from outsider scepticism, if not insider
scepticism.
There are many reasons why philosophy is worth doing. Yet, it
would be disturbing if we cannot show the agnostic that philoso-
phy gets her the right type of value – true answers to philosophical
questions.

© 2010 Blackwell Publishing Ltd

 SCEPTICISM ABOUT PHILOSOPHY 3
2. The Argument against Philosophy

Dissensus can be used in an argument against philosophy:

The Argument against Philosophy. The goal of philosophy is to

uncover certain truths. Radical dissensus shows that philosophi-
cal methods are imprecise and inaccurate. Philosophy continu-
ally leads experts with the highest degree of epistemic virtue,
doing the very best they can, to accept a wide array of incom-
patible doctrines. Therefore, philosophy is an unreliable instru-
ment for finding truth. A person who enters the field is highly
unlikely to arrive at true answers to philosophical questions.

This is a rough sketch of the argument. I will refine it as necessary

as the paper proceeds. Note that Argument against Philosophy
need not claim that no philosopher has found the truth. It is
possible that Kant got everything right. Yet, philosophy also has
arrived at hundreds of other incompatible doctrines. If philoso-
phy leads to the truth, it is only because it leads almost everywhere.
A person concerned only to get the truth would at the very least
try to believe some randomly chosen doctrine rather than be
agnostic, since there is at least some possibility that a random
doctrine happens to be correct. She might even pursue philo-
sophical methods if they increase the probability of being correct,
as it is possible they do. Perhaps philosophers are twice as likely
to have true answers to philosophical questions than non-
philosophers are. However, if this same person is equally con-
cerned to avoid false beliefs about philosophical issues, then she
would want to pursue philosophy only if it gave her a greater than
50% chance of getting a true belief. Given the degree of dissensus
in most fields of philosophy, it seems unlikely that philosophy
offers her this great a chance.
Here is an analogy. Suppose, thousands of people, each of
whom wants to go to São Paulo, randomly board all flights depart-
ing Dallas-Fort Worth. Suppose they fill all departing seats, but are
not told where they are going. Of these thousands, a few hundred
in fact will land in São Paulo. Most will arrive somewhere else.
Philosophy seems like this in many respects. It may bring some
people to the proper destination, but it dumps most somewhere
else. Actually, matters are worse than that. Travellers will know
whether they have arrived in São Paulo. In philosophy’s case,
some may indeed arrive at truth. However, they will not have

© 2010 Blackwell Publishing Ltd

4 JASON BRENNAN

discernibly better grounds for believing this than their mistaken

peers. They may believe themselves to have better grounds, and
their peers believe this about themselves as well. However, from
the outsider’s perspective, they look the same. They are smart
people doing the best they can, and they disagree. The outsider
has little reason to think one philosopher is closer to the truth
than the next, and little reason to think that if she became a
philosopher, she would do any better.
Here is another way of making the unreliability argument.
Suppose that there are 10 competing doctrines in the field of
philosophy of mind, each of which is accepted by 10 percent of
the members of the American Philosophical Association.
Suppose, optimistically, that on the nature of consciousness 10
percent of the members of the APA have the right theory.
Suppose also that we can regard all members of the APA as
epistemic peers, where two people are epistemic peers just in case
they are equals with respect to their degree of epistemic virtue
(thoughtfulness, freedom from bias, etc.) and their access to evi-
dence.2 An uncommitted person, looking at the field from the
outside, would worry that if she pursues philosophy, she will have
something like a 1 in 10 chance of getting the right answer to the
questions of the philosophy of mind. She sees that philosophical
methodology – studying arguments, making new arguments, cre-
ating new distinctions, reading texts, debating, etc. – generally
leads people to accept some theory or other of the nature of
consciousness. (Let us assume that everyone who studies the phi-
losophy of mind ends up accepting 1 of the 10 theories.) So, she
knows that philosophical methodology will result in her accepting
some theory, but from her standpoint, it is more likely than not
that it will be a false theory. The greater the degree of disagree-
ment among epistemic peers, the lower the probability that phi-
losophizing will get her to the truth.
This argument assumes than an agnostic outsider who ends
up pursuing philosophical methods will have either a random or
proportional chance of accepting any theory. I.e., I am working
on the assumption that she will either accept a theory at random
or with a probability proportional to the percentage of her
epistemic peers that accept any given theory. Real people prob-
ably do not have a random chance due to their background

2
Kelly, ‘Epistemic Significance of Disagreement’, p. 10.

© 2010 Blackwell Publishing Ltd

 SCEPTICISM ABOUT PHILOSOPHY 5
starting beliefs. A person who comes to philosophy as a Christian
is probably more likely to end up being a moral realist and a
natural law theorist than his atheist counterpart. A graduate
student who studies ethics at Harvard University is probably more
likely than a student at Australian National University to become
a Kantian. People have dispositions towards one theory or
another, and (in certain respects) non-random factors such as the
people with whom they study philosophy affect the probability
they will adopt any particular theory. Suppose, however, that our
truth-seeking, error-avoiding agnostic has no such dispositions
and manages to have even exposure to all competing doctrines.
Will pursuing philosophy assign her to a set of beliefs randomly or
in a probability proportional to the positions of her epistemic
peers? Perhaps the agnostic will remain agnostic since she has no
dispositions. Without a good empirical account of the mecha-
nisms of belief formation, I cannot be sure whether she has a
random chance of adopting any particular theory, adopting a
theory with a probability proportionate to the percentage of com-
parably virtuous philosophers accepting that theory, or has some
different probability altogether. From her viewpoint, the process
will seem random in some way. So, I use randomness here as a
hopefully good-enough substitute for the actual mechanism that
assigns beliefs.3
Under some circumstances, it could be reasonable for the out-
sider to think she has a better chance than others do of getting
things right. For example, consider an exceptional person with an
IQ many times greater than that of the average philosopher, with
an exceptional memory, who lived long enough to read every
philosophy book ever written, and who exhibited the epistemic
virtues far better than Kant or Hume did. This person could
legitimately conclude that she might do better than other philoso-
phers have. However, no real agnostic will be this exceptional. A

3
Thus, one possible way to defeat outsider scepticism would be to show that a truth-
seeking, error-avoiding agnostic is actually more likely than the rest of us to arrive at true
doctrines. Perhaps her lack of prior commitments makes philosophical methodology
reliable for her, if not for us. Whether this counter-argument will work depends on
empirical points about the mechanism of belief formation as well. Note also that this type
of response attributes our disagreements to bias. However, suppose it can be show that the
true agnostic has a good chance of getting the truth. There will still be a sort of leftover
outsider scepticism. The outsider who is not an agnostic might still regard philosophy
as unreliable, as having too great a tendency to allow people to rationalize their prior
beliefs, etc.

© 2010 Blackwell Publishing Ltd

6 JASON BRENNAN

good response to the sceptical worry should provide reason to

pursue philosophy for a truth-seeking, error-avoiding agnostic
with epistemic virtue comparable to a typical philosopher’s.
Philosophers do seem to aim for truth. Philosophy’s state of
dissensus may show us that philosophy is not worth doing if truth
is our goal. Pursuing philosophy is not a reliable method of
finding the truth about philosophical issues.

3. What Type of Defence Do We Want?

Suppose that accepting a philosophical theory cures tooth decay.

This would make the theory a good thing, but not a good theory.
Similarly, a good hammer is good at driving nails. If it is bad at
driving nails but using it cures tooth decay, this makes the
hammer a good thing, but not a good hammer.
So, what is the nail to which philosophy is the hammer? We
have sets of questions we want theories to answer. We want phi-
losophy to get us the truth. We want it to answer our questions
or to show us that the questions were mistakes (because they
represent pseudoproblems).
There are a number of types of defences of philosophy:

1. Epistemic: Philosophy is good because it gets us to the truth,

or something reasonably truth-like (such as understanding).
2. Intrinsic: Philosophy is good as an end in itself.
3. Instrumental: Philosophy is good for getting some values
other than truth.
4. Aretaic: Philosophy is good for fostering wisdom, good char-
acter, or various intellectual virtues.

Aretaic defences could be considered a subset of instrumental

defences. When academic philosophers defend philosophy, e.g.,
by explaining its value on the ‘Why Study Philosophy?’ webpages
many departments post for prospective majors, they often list
defences of types 2, 3, and 4. Each of these are good defences, and
conjoined they might be excellent reasons to study philosophy or
to pursue a philosophical career. They might be excellent non-
epistemic reasons for becoming an insider rather than an out-
sider, or for coming to accept some philosophical doctrines rather
than remaining agnostic. Yet, ultimately we want a defence of type
1. If we do not get that, there is something disappointing about
the philosophical enterprise.

© 2010 Blackwell Publishing Ltd

 SCEPTICISM ABOUT PHILOSOPHY 7
It is not enough that philosophy leads to some truths; it needs
to lead to truths about philosophical issues. If philosophical theo-
ries helped us learn the truth about physics, that would not quite
be the target value. There are some distinctly philosophical ques-
tions we want philosophy to answer.4 Ultimately, we need 1*.

1*. Proper Epistemic: Philosophy is good because it gets us to the

truth (or something reasonably truth-like) about philo-
sophical issues.

Below, I will consider a number of defences of philosophy. In the

next section, I consider common defences and explain why they
are inadequate. Many of them fail because they do not provide a
proper epistemic defence of philosophy, but simply show it to be of
instrumental or aretaic value. In the section following the next, I
consider more pressing objections that hold that there is reason-
able disagreement among philosophers who are epistemic peers. If
reasonable disagreement is possible, this implies that at least some
philosophers can justifiedly say to themselves, ‘Even though my
epistemic peers disagree with the theory I believe, my theory is true
and I am justified in believing that it is true.’ I will argue that even
this sort of defence is not enough to satisfy the truth-seeking,
error-avoiding agnostic. This defence at best explains why insider
scepticism is unwarranted, but does not explain why outsider
scepticism is unwarranted. That is, the possibility of rational dis-
agreement can explain why we philosophers who have views are not
required to give them up in light of disagreement, but it does not
explain why a truth-seeking, error-avoiding agnostic should pursue
philosophy and come to adopt any views.

4. Some Inadequate Defences

In this section, I discuss some anti-sceptical defences of philo-

sophy I have encountered. The defences are individually and
collectively unsatisfactory. Each defence captures something
important, and collectively they may justify pursuing philosophy.

4
Questions that were once thought to be philosophical have a tendency to become
questions for the social or natural sciences. The border between philosophical and non-
philosophical questions is fuzzy. However, without saying how to make the distinction, I will
assume there is something like a core of questions that we reasonably can expect to remain
part of philosophy.

© 2010 Blackwell Publishing Ltd

8 JASON BRENNAN

However, they do not show that we can regard philosophy as

producing the right sort of value – true answers to philosophical
questions. It is not necessary to go into much depth with these
defences, because it can be shown rather quickly that they are not
the right type of defence.
Recall that the agnostic about philosophical issues is consider-
ing pursuing philosophy with the goal of getting true answers to
philosophical questions. This outsider sees the degree of dissen-
sus and thinks to herself, ‘At most one of these theories for any
given issues is correct. It’s possible that if I study philosophy, I will
produce a new theory that competes with these others. Each of
these philosophers thinks her own theory is more likely to be true
than her competitors’ theories. I realize that if I study philosophy,
I will come to think that way about whatever theories I come to
accept as well. However, from my standpoint now, I have to regard
each of the competing theories as something like equally likely to
be true, or perhaps likely to be true in proportion to how many
good philosophers accept the theory. It’s possible none of them
are true. If so, then much more likely than not, I will end up
accepting a false theory. So, I should remain an outsider and an
agnostic.’ The general problem with the defences listed below is
that even if they give this outsider good reasons to study philoso-
phy and to accept doctrines rather than be agnostic, these
defences do not give the right kind of reason. I.e., they do not give
proper epistemic defences of philosophy. Some of the other defences
fail because they rest on bad arguments, even if they are attempts
at proper epistemic defences.

A. The Argument Undermines Itself. There is a facile defence: The

Argument against Philosophy undermines itself. The general posi-
tion that philosophy is irrational fails to pass self-inspection. ‘Phi-
losophy is irrational’ is a philosophical position. If philosophy
is irrational, so is the view that philosophy is irrational. If philoso-
phical argumentation never establishes any position, then the
anti-philosophy position cannot be justified by philosophical
argumentation. The Argument against Philosophy refutes the
Argument against Philosophy. Even if this defence works, it is
embarrassing if this is the best defence philosophy has. Yet, it is
not obvious that the defence succeeds. It may just be that all
philosophy is unreliable except anti-philosophy philosophy.
The outsider sceptic’s position is that philosophical methodol-
ogy is unlikely to bring her to the truth about philosophical

© 2010 Blackwell Publishing Ltd

 SCEPTICISM ABOUT PHILOSOPHY 9
questions. One might argue that the sceptic used philosophical
reasoning to arrive at this conclusion, and so the sceptic cannot
consistently be a sceptic. However, it may just be that a small set
of philosophical issues is answered and that philosophical meth-
odology works reliably on a small set of issues, i.e., just in the
areas needed to make the sceptic’s argument. For instance,
perhaps the sceptic needs probability, an account of the notion
of an epistemic peer, some notion of reliability, and not much
else.

B. Disunity of Science. One could argue that science is less unified

than commonly thought. Thomas Kuhn claims that the appear-
ance of unity is largely a myth propagated by ahistorical science
textbooks.5 It may also be that philosophy only appears to have
more disagreement to us philosophers because we are most famil-
iar with philosophy. If we were better informed, we would realize
that there is just as extensive disagreement in biology and physics
over fundamental issues as there is in philosophy. This approach
may deflate science, making philosophy seem less inferior in com-
parison, but it does not show us that philosophy is truth-tracking.
Our truth-seeking outsider is not impressed. Also, deflating
science also improves the comparative position of astrology, phre-
nology, and creationism.

C. Lists of Accomplishments. Another type of defence is that offered

by Wilbur Urban, former president of the American Philosophical
Association. In 1925, Urban attempted to validate the rationality
and progressiveness of philosophy by listing its recent accomplish-
ments.6 Urban’s list looks strange. Much of it is hard to under-
stand, so it is unclear whether the claims of progress are worth
much. The clearer items are problematic. For one, he claims that
philosophy has made progress because there is no movement
back toward Kant. However, eighty years later, we see numerous
defenders of forms of transcendental idealism, Kantian construc-
tivism, and the like. He also claims that philosophers have shown
that value cannot be reduced to something else and that evolution
cannot fully explain values. However, eighty years of neo-

5
T. S. Kuhn, The Structure of Scientific Revolutions (Chicago: University of Chicago Press,
1996), pp. 136–138.
6
Wilbur M. Urban, ‘Progress in Philosophy in the Last Quarter Century’, The Philo-
sophical Review 35:2 (1926), pp. 93–123.

© 2010 Blackwell Publishing Ltd

10 JASON BRENNAN

naturalist metaethics and sociobiology shows that this claim is not

obviously true. Though I agree with Urban, I have many epistemic
peers who disagree. Third, he cites the growth of logic as a formal
discipline. This is one of philosophy’s major accomplishments,
but it is not clear that this helps. Formal logic may have less
disagreement than other fields, but it is also the place where
philosophy comes closest to being mathematics.
Any list will be contentious. Probably, if I were to make a list of
philosophy’s recent accomplishments, it would seem esoteric,
strange, irrelevant, wrong, and/or silly to philosophers eighty
years from now.
The outsider remains unimpressed. She can look at such lists
and ask, do we yet know what right action is, what justification is,
what knowledge is, what justice is, and so on? There remains
extensive disagreement over these fundamental issues, and she
remains worried that philosophy is unlikely to deliver her the
truth.

D. Progress as Destruction.7 Some philosophers defend philosophy

by saying that our work at least shows what theories are false. For
instance, Gettier demolished the justified true belief analysis of
knowledge. Quine, Putnam, and others eradicated logical positiv-
ism. Gödel showed us that Principia Mathematica did not axioma-
tise arithmetic. If this is progress toward truth, it must be progress
by elimination.
Refuting inadequate past theories clears the path for good
answers, but does not thereby give us good answers. (Even nega-
tive ‘progress’ tends to be reversed, as once dead doctrines, such
as Ross’ moral theory, are resurrected, albeit in better forms.8)
Often, there are potentially infinite numbers of possible theories
in any sub-field. So, even if over the past 2500 years of philosophy,
we have managed to show that a few thousand theories are inad-
equate, that does not show us we are any closer to the truth. On
the other hand, suppose there are a finite number of theories. If
so, permanently refuting a theory increases the probability one
will accept the correct theory. In this case, the agnostic might have
reason to pursue philosophy, but only if enough theories had

7
This phrase comes from Toni Vogel Carey, ‘Is Philosophy Progressive’, Philosophy
Now 59 (2007), accessed online (3/15/07) at http://www.philosophynow.org/issue59/
59carey.html
8
E.g., Robert Audi, The Good in the Right, (Princeton: Princeton University Press, 2004).

© 2010 Blackwell Publishing Ltd

 SCEPTICISM ABOUT PHILOSOPHY 11
been or could refuted that she were more likely than not going to
accept the true one. But this is not the case.
Additionally, this defence does not explain philosophers’
actual behaviour. Suppose philosophy is progressive because it
can show, at least, which theories are false, and the point is to
arrive at the truth though elimination. This would justify con-
structing, debating, examining, and attacking theories, but not
accepting a theory. It would not give the agnostic reason to believe
anything.

E. Consensus Just Around the Corner. One could concede that

current dissensus shows that philosophical methods are ineffec-
tive, but then assert that philosophy could become effective in the
future. Philosophers use the wrong methods. We need to con-
tinue working until we discover the right methods. Then agree-
ment will follow. Indeed, we could even take agreement as a sign
that we have discovered the right methods.
The natural sciences began making progress when a change in
methods was adopted. Scientists dropped the Aristotelian para-
digm; i.e., they began doing extensive ‘artificial’ experiments
rather than just making observations. Also, they accepted math-
ematics as a tool for modelling nature. Could there be similar
methodological revolutions for philosophy?
Philosophers have made this claim before and tried to intro-
duce new methods. Hobbes argued that progress could be made
and agreement would be possible if philosophers would just start
with clearly stated, sensible definitions. David Hume called the
Treatise an attempt to introduce empirical methods into philoso-
phy. Kant’s Copernican Revolution meant to resolve the
rationalist-empiricist debates by exposing an unnoticed, mistaken
common assumption. Thus, seeking consensus by finding the
right methods has been tried and has not yet worked. After twenty
five hundred years, the claim that consensus is going to appear
once we get the right methods is implausible.
We are more inclined to think disagreement is a permanent
fixture. In fact, it seems that widespread philosophical consensus
is more likely to come from irrationality and intellectual corrup-
tion than from honest inquiry. The very best philosophers
throughout history have produced radically different doctrines.
(Part of what makes philosophers great is that they do an excellent
job defending novel doctrines.) Thus, it seems that we should not
expect convergence as philosophers become more rational. On

© 2010 Blackwell Publishing Ltd

12 JASON BRENNAN

the contrary, our best philosophers tend to diverge rather than

converge.

F. Philosophy as Maieutic. Philosophy gives birth to new fields.

Philosophers invented economics, political science, sociology,
physics, biology, etc. If we take a realist view of theories in these
fields, then philosophy is indirectly truth-tracking, as it produces
other fields that find the truth.
However, there are two worries with this sort of defence. First,
even if it is an epistemic defence of philosophy, it is not a proper
epistemic defence. We want philosophy to find answers to philo-
sophical questions, such as whether God exists, what the nature of
knowledge is, what is right and wrong, and so on. There is a view
that philosophy is the field of residual speculation, and perhaps
over time philosophy will self-destruct as it gives birth to special
sciences capable of answering its questions. However, arguably,
there is a common core of questions that cannot be made non-
philosophical. (This point is, of course, subject to contention.)
Though, looking backward, we can see how some questions were
mistakenly treated as philosophical, this does not give us good
reason to think that all questions will one day be turned over to
other fields. So, insofar as we legitimately believe that there will
always be philosophical questions, the maieutic defence of phi-
losophy is not enough.
Worse, the birth rate appears to be dropping. Philosophy is not
founding new fields as often as it used to. At least when viewed in
isolation, the maieutic defence suggests that pretty soon we
should stop practicing philosophy, because the expected utility
(in terms of founding new fields) is too low.

G. Developing Critical Thinking Skills. Another unsatisfactory

defence of philosophy is the claim that it develops critical think-
ing skills and various intellectual virtues. No doubt philosophy
does foster such virtues, but the defence is still unsatisfactory
because it is an aretaic rather than a proper epistemic defence.
That philosophy develops such skills is an excellent reason for
undergraduates planning to work in other fields to major in it.
Still, the outsider sceptic is not impressed, as this defence does not
explain how applying philosophical skills to philosophical ques-
tions reliably generates true answers. In addition, this aretaic
defence is somewhat embarrassing, in that it does not do much to
differentiate philosophy from playing logic games or Sudoku.

© 2010 Blackwell Publishing Ltd

 SCEPTICISM ABOUT PHILOSOPHY 13
5. Rational Disagreement

Here I consider at greater length the possibility of rational dis-

agreement as a response to the problem. Consider two or more
epistemic peers holding contrary views on the same issue. Can
they each justifiedly believe either A or B?

A. My epistemic peers disagree with me on this issue. I am

right, but they are each wrong. I am justified in holding my
view, but they are not justified in holding theirs.
B. My epistemic peers disagree with me on this issue. I am
right, but they are each wrong. I am justified in holding my
view, and they are also justified in holding theirs.

If rational disagreement is possible, then sometimes it is justifiable

for more than one member of a dispute to accept A, B, or some-
thing similar.
In current epistemology, there is disagreement about whether
rational disagreement is possible. E.g., Richard Feldman argues
that reasonable disagreement between peers is not possible under
common circumstances, because there is generally at most a
uniquely justified belief in light of a given set of evidence.9 Adam
Elga holds that when one discovers that one disagrees with an
epistemic peer, one should give the peers’ views equal weight as
one’s own.10 David Christenson argues that when on has disagree-
ment with peers, this typically should occasion belief-revision
towards the views of one’s peers and vice versa.11 In contrast,
Thomas Kelly holds that one often need not revise one’s views in
light of discovering disagreement with one’s peers because one
believes they have misjudged the evidence.12 Gideon Rosen holds
that rationality is permissive and that sometimes one is permitted to
choose among competing theories when given a set of evidence.13

9
See Richard Feldman, ‘Reasonable Religious Disagreements’, in Louise Antony. ed.,
Philosophers Without Gods, (Oxford: Oxford University Press, 2007); Richard Feldman,
‘Epistemological Puzzles about Disagreement’, in Stephen Hetherington (ed.), Epistemol-
ogy Futures (Oxford: Oxford University Press, 2006).
10
Adam Elga, ‘Reflection and Disagreement’, Noûs 41 (2007), pp. 478–502.
11
David Christensen, ‘Epistemology of Disagreement: the Good News’, Philosophical
Review 116 (2007), pp. 187–217.
12
Kelly, ‘Epistemic Significance of Disagreement’.
13
Gideon Rosen, ‘Nominalism, Naturalism, Philosophical Relativism’, Philosophical
Perspectives 15 (2001), pp. 69–91.

© 2010 Blackwell Publishing Ltd

14 JASON BRENNAN

Nicholas Rescher explicitly addresses the problem of philo-

sophical dissensus. He argues that philosophers choose to reject
different theses – and thus establish conflicting schools of
thought – because they accept different cognitive values or weigh
the cognitive values differently.14 Cognitive values are the epistemic
traits by which we assess a doctrine, e.g., coherence, plausibility,
generality, importance, informativeness, elegance, etc. A philoso-
pher who more strongly values plausibility and intuitiveness is likely
to accept different doctrines from those a philosopher who more
strongly values systematicity would accept. Rescher argues that
differences over the relative weights of cognitive values cannot fully
be resolved. According to Rescher, rational theory acceptance
means accepting a theory that does justice to one’s cognitive values.
Different theorists can reasonably accept different values to differ-
ent degrees. So, rational disagreement is possible.
It might be thought that the possibility of rational disagreement
will bear on whether outsider scepticism is warranted in light of
philosophical dissensus. For instance, Peter van Inwagen discusses
people who have heard philosophical debates but have remained
agnostic. He then says,

I think that any philosophy who does not wish to be a philo-

sophical sceptic . . . must agree with me that . . . it must be pos-
sible for one to be justified in accepting a philosophical thesis
when there are philosophers who, by all objective and external
criteria, are at least as equally well qualified to pronounce on
that thesis and who reject it.15

However, it is possible that whether the agnostic should become a

sceptic and whether the non-agnostic philosopher should become
an agnostic sceptic are distinct problems. Perhaps rational dis-
agreement is possible among peers, and this excuses non-agnostic
philosophers from having to become agnostic sceptics. However,
as I will argue in this section, even if this is so, this does not give
reason for the truth-seeking, error-avoiding agnostic to become a
non-agnostic. Rather, she should be a sceptic about philosophy.

14
Nicholas Rescher, The Strife of Systems (Pittsburgh: University of Pittsburgh Press,
1985), pp. 95–115.
15
Peter van Inwagen, ‘It is Wrong, Everywhere, Always, and for Anyone, to Believe
Anything upon Insufficient Evidence’, in Eleonore Stump and Michael J. Murray, eds.,
Philosophy of Religion: The Big Questions (Oxford: Blackwell, 1996), p. 275.

© 2010 Blackwell Publishing Ltd

 SCEPTICISM ABOUT PHILOSOPHY 15
Note that if rational disagreement were impossible, this would
serve my thesis, as it would bolster the case for outsider scepticism.
If rational disagreement is impossible, then insiders (non-agnostic
philosophers) should become sceptics. Presumably this means
that outsiders (agnostics who have not studied philosophy) should
become sceptics as well, once they learn that all the insiders are
rationally obligated to become sceptics. However, I will assume for
the sake of argument that rational disagreement among epistemic
peers is possible. I will argue that even if it is possible, this will
not be enough to show the truth-seeking, error-avoiding agnostic
that she ought to pursue philosophy and adopt any views. The
possibility of rational disagreement does not defeat outsider
scepticism.
Since I am granting that rational disagreement is possible, I
need not consider Rosen’s, Kelly’s, or others’ arguments for ratio-
nal disagreement at length or with much precision. I need only
consider their conclusion: rational disagreement is possible. If so,
then it follows that when I recognize that my epistemic peers
disagree with me on some issue, sometimes I may still justifiedly
believe that my view is true. (I take it that believing that X and
believing that X is true are the same thing.) In addition, depend-
ing on one’s view of rational disagreement, this might mean I
am justified in believing I am justified, and perhaps even in believ-
ing that my peers are justified in having similar attitudes toward
themselves.
So, what rational disagreement arguments deliver us, at the
end, is something like B: ‘Even though my epistemic peers dis-
agree with me on this issue, I am right, they are all wrong, I am
justified in holding my view, and they are also justified in
holding theirs.’ If I am justified in holding B, then insider scep-
ticism is defeated. I am not required to become an agnostic and
a sceptic.
However, notice that B is not what truth-seeking, error-avoiding
agnostic wants. She comes to philosophy hoping to obtain true
answers to philosophical questions while avoiding error. If ratio-
nal disagreement is possible, then philosophical inquiry can get
her justified belief in various philosophical doctrines even in the
presence of disagreement, but that was not what she asked for. A
justified belief that one has the truth on some issue is a great thing
to have – I certainly would like to have that – but it is a poor
substitute for bona fide truth. The truth-seeking, error-avoiding
agnostic is not interested in this substitute.

© 2010 Blackwell Publishing Ltd

16 JASON BRENNAN

She might be impressed to learn (depending on what the stan-

dards are for rational disagreement) that many or perhaps all
philosophers are justified in their beliefs. With some good
fortune, we might discover that all actual philosophical disagree-
ments among members of the APA are reasonable ones, and so
no philosopher must do any belief revision or become a sceptic.
However, this still does not give the truth-seeking, error-avoiding
agnostic reason to become a believer. She wants a greater than
not chance of getting true beliefs about most philosophical
issues. Even a 100% chance of getting a justified belief that one has the
truth about philosophical issues or (more simply) a 100% chance of
getting justified beliefs about philosophical issues will not motivate her,
because there are not the same things as a true beliefs about most
philosophical issues. They are poor surrogates.
Rescher’s defence of philosophy is particularly clear in how it
fails to satisfy this sort of agnostic. (This is not to say his defence is
bad, but just that it is not what I called a proper epistemic
defence.) Rescher holds that it can instrumentally rational to
accept a theory based on one’s cognitive values. There is a plural-
ity of reasonable stances on the weights of these cognitive values.
So, for Rescher, rational disagreement rests precisely on these
cognitive values rather than on truth. But our agnostic is not
interested in these cognitive values unless they reliably get her to
the truth. Apparently, they do not, because ex hypothesi the
pursuit of theories by different people with different cognitive
values or weights for these values results in dissensus.
The agnostic asks us if we can get her the truth. In light of
dissensus, apparently we have to say no. All we can offer is justified
belief.

Brown University
Box 2005
Providence, RI 02912
[email protected]

© 2010 Blackwell Publishing Ltd