14

scepticism
about other minds
• The problem of other minds
• The argument from analogy
• A problem for the argument from analogy
• Two versions of the problem of other minds
• Perceiving someone else’s mind

THE PROBLEM OF OTHER MINDS

We take it for granted in our everyday lives that we are not alone in the universe; that
there are other people who inhabit this place with us. As we will see, however, once
one starts to reflect on the matter, it isn’t entirely obvious what entitles us to this belief.
Why are we so sure that there are other people out there, people who have minds like
our own?
The problem confronting our knowledge of other minds is that, on the face of it at
least, we don’t actually observe other minds in the way that we observe objects like
trees and cars. After all, one’s mind seems to be something that underlies one’s body
and one’s bodily behaviour such that, although one’s behaviour manifests one’s mind,
simply observing an agent’s behaviour is not the same as observing their mind.
Accordingly, the thought runs, in order to know that someone is minded we have to
do more than merely observe their behaviour; we also have to infer that there is
something underlying that behaviour and giving rise to it – namely, a mind.
If this picture of how we come to know that there are other minds is correct, then
scepticism about the existence of such minds is just around the corner (i.e. the view
that knowledge that other minds exist is impossible). After all, if we have to infer the
existence of other minds from observed behaviour, then the question naturally arises
as to whether that observed behaviour could be manifested even though there is no
mind underlying the behaviour. Perhaps the ‘people’ that one interacts with on a daily
162 • do we know anything at all?

basis are nothing more than unminded automata or zombies who have no thoughts
and feelings at all. How would we tell the difference? (This is particularly troubling
in the case of zombies, where there is no obvious underlying physical difference.) This
difficulty concerning how we know that there are other people who have minds like
we do is called the problem of other minds.

THE ARGUMENT FROM ANALOGY

So how might one respond to the problem of other minds? Perhaps the most famous
line of response – a version of which is usually credited to John Stuart Mill (1806–73)
– makes use of a form of inductive reasoning known as an argument from analogy.
Essentially, the idea behind this approach to the problem of other minds is to maintain
that we can come to know that there are other minds by observing how the behaviour
of others mirrors that of our own (where we know that we are minded). The thought
is that since we know that we have minds, it follows that the behaviour of others
which is similar to our own shows that these others have minds too.

John Stuart Mill (1806–73)

If all mankind minus one were of one opinion, and only one person were
of the contrary opinion, mankind would be no more justified in silencing
that one person, than he, if he had the power, would be justified in silencing
mankind.
Mill, On Liberty

The English philosopher and economist, John Stuart Mill, was one of the most
influential men of his day. Like his father, James Mill (1773–1836), Mill was a
prominent liberal reformer committed to utilitarianism – the view that actions
are morally right to the extent to which they promote the greatest happiness in
the greatest number of people. He was a member of the British parliament and
forcefully argued for the rights of women.

The starting point for this argument is our knowledge both of the existence and nature
of our own minds. After all, we cannot seriously doubt that we have a mind, since who
then would be doing the doubting? (This is the point of Descartes’ ‘cogito’: ‘I think,
therefore I am’.) Moreover, it is also held that there cannot be any troubling sceptical
argument concerning our access to what is going on in our own minds because this
access is privileged. That is, we have immediate non-inferential access to what is going
on in our own minds – what we are thinking and feeling – and this means that our
scepticism about other minds • 163
knowledge in this regard is entirely secure (at least if any knowledge is). It follows that
we can put our knowledge of our own minds together with our knowledge of how
we, as minded creatures, behave, to determine what sort of behaviour a minded
creature should have.
For example, we might notice that when we are in pain, as when we accidentally burn
ourselves on a match, we respond in certain ways (e.g. by calling out). Suppose we
notice a number of these correlations between external stimuli (e.g. the burning of a
match, the tickle of a feather), external response (e.g. calling out, giggling), and the
associated mental state (e.g. pain, pleasure). Suppose further that we observe other
apparently minded people behaving in the same ways in response to the same stimuli
(i.e.they call out when burnt by matches and giggle when tickled with feathers).
Wouldn’t we then be entitled to inductively infer that there are other minds just like
our own?
Here is the form of the inductive argument that is being used here:
A1 There are patterns in my behaviour in response to external stimuli which
reveal that I am having mental states of a certain sort (e.g. my crying out in
response to being burnt by a match indicates that I am in pain).
A2 This same behaviour in response to external stimuli is exhibited by others.
AC These others experience the same mental states that I do, and so are
minded, just like me.
On the face of it, this looks like a good way of responding to the problem of other
minds.

A PROBLEM FOR THE ARGUMENT FROM ANALOGY

Although initially persuasive, the argument from analogy runs into problems on closer
inspection. For one thing, the style of argument being employed here is not a good
one, even if we set aside the more general worries one might have about inductive
arguments that we looked at in Chapter 10. Compare the argument given above with
the following inductive argument:
1 Box A is brown and it contains a book.
2 Boxes B, C and D are brown.
C Boxes B, C and D contain a book.
Clearly, this is a very bad style of argument in that the mere fact that one brown box
contains a certain item does not give us any reason to believe that any other brown
box contains that sort of item. The problem with this argument is that it only considers
a particular instance of a brown box, an instance which we have no reason to think is
164 • do we know anything at all?

representative of brown boxes in general. As we noted in Chapter 10, however, good

inductive arguments are always ones that reason from representative premises to
conclusions. Accordingly, we cannot reason from this instance to a more general
conclusion that applies to any brown box we care to pick.
Notice that the following argument would be OK:
1* Lots of brown boxes have been observed over many years and in a wide range
of environments and they have all contained books.
2 Boxes B, C, and D are brown.
C Boxes B, C, and D contain a book.
If it is indeed true that we have observed a representative range of brown boxes and
found them all to have a book in them, then there is no problem in justifiably
concluding that any other brown box we find will also have a book in it. The trouble
is, however, that the argument from analogy is more akin to the first of these ‘brown
box’ arguments than the second. The reason for this is that it begins with the
observation of a correlation in a single case (between my behaviour and my
mindedness) and draws conclusions about the relationship between behaviour and
mindedness in general. But that is a very bad way of reasoning, as the first ‘brown box’
argument shows.
If we were entitled to suppose that our case is somehow representative of minds in
general, so that what holds for my mind would hold for others, then we could properly
use an argument from analogy to draw conclusions about the existence of other minds.
But how would we come by such a supposition without in the process simply
assuming that which is to be shown (i.e. that there are other minds out there which
are like my own)?
This is not the only problem facing the argument from analogy, but it is the most
decisive one. One cannot legitimately infer simply that there are other minds on the
basis of one’s own case.

TWO VERSIONS OF THE PROBLEM OF OTHER MINDS

As if the problem of other minds as it is presented above weren’t bad enough, there
is a second difficulty lurking here. This is that even if we could come to know that
there exist other minds, it isn’t at all clear how we could come to know that these
other minds are like our own. There are two problems here which can easily be run
together if one isn’t careful. The first is whether any other minds exist, regardless of
what those minds are like. The second is whether, given that other minds exist, those
minds are like our own.
Clearly, one could answer the first problem without having any answer to the second.
In order to see this point, take it for granted for a moment that there are indeed other
scepticism about other minds • 165
minds. Now ask yourself how you can be sure that other people’s minds are like your
own. A standard motif of science fiction movies, for example, is that of the alien taking
over someone’s mind. In such a case, we have someone who may well nearly always
behave as she used to, but who no longer thinks and feels like a human but like an
alien. How would we tell the difference if there was nothing in the alien’s appearance
or behaviour to give the game away?

Invasion of the Body Snatchers

The main premise of the 1956 film Invasion of the Body Snatchers is that people
are being quietly replaced by alien duplicates. In many ways, though not in all
ways, these aliens act just like the people they have replaced, which is what
makes it so difficult to tell the alien duplicates apart from the ‘real’ people.
Presumably, while these aliens look and act like real people, they do not
experience the world as we do. This raises the question, central to this section of
the chapter, of how we can be sure that others are minded in the specific way
that we suppose them to be. How do we know that they feel pain like we do,
for example? After all, the alien duplicates act just like we act, so it seems that
we cannot tell what their minds are like just by observing their behaviour. But
if we can’t do it in this way, then how can we do it?

Indeed, we don’t need to consider science fiction movies in order to get an example
of this sort of ‘deviant’ mindedness. After all, some people are colour-blind, for
example, and so see colours very differently to ‘normal’ people. Others have unusual
senses of taste and hearing, perhaps being unable to taste/hear things that others can
taste/hear, or tasting/hearing them differently. Often we can tell that this is happening
because it has an impact on someone’s behaviour. For example, if a certain fruit that
tastes sweet to others tastes very sour to them, then they will respond with disgust
upon tasting it. We can easily imagine cases, however, in which another person
experiences the world very differently and yet this difference does not manifest itself
in experience. For instance, suppose that someone sees red as blue and vice versa.
Accordingly, they would grow up calling what they experience as blue ‘red’, and vice
versa. Would this ever come to light? It might in that it might affect how they respond
to other colours on the spectrum, for example. Equally, however, it might not in that
this person might just go through life systematically mistaking red for blue and blue
for red. If this is possible, however, then it raises the question of how certain we can
be that we are all experiencing the world in the same way. Perhaps we have just
learned to categorise the world in a standard way, even though the subjective natures
of our experiences are in fact very different from case to case?
166 • do we know anything at all?

PERCEIVING SOMEONE ELSE’S MIND

One way in which one might respond to the problem of other minds – in both its
forms – is to question its guiding premise that our knowledge of other minds is by its
nature inferential. After all, common sense would seem to suggest otherwise on this
score. Suppose I see someone writhing in agony on the ground before me. Do I really
need to make an inference in order to know that he is in pain? Can’t I just see, directly,
that he is in pain?
The thought is thus that perhaps, at least when it comes to some very clear-cut cases,
I could know that someone is having a certain experience – of being in pain, say –
simply by looking at them. And if I can know what kind of experience someone is
having in this direct way, then presumably I can also come to know that this person is
a creature with a mind that is capable of experiences in the first place. That is, I can
come to know, without inference, both that there is someone else with a mind and
that the experiences that this person has are at least in certain respects like mine. If
this is right, then the worry that the inference involved in the argument from analogy
is unsound does not get a grip, at least not on these select cases of direct knowledge
of other people’s minds.
At first pass, this proposal might look like mere dogmatism, but notice that this sort
of view is structurally very similar to the direct realism as regards perceptual
knowledge that we looked at in Chapter 7. One of the key motivations for direct
realism was the thought that we should resist the inference from the fact that our
perceptual experience could be undetectably misleading to the claim that what we
are directly aware of in perceptual experience is only the way the world seems to us
rather than the way the world is. Although it is true that in deceived cases, such as the
scenario in which I am visually presented with a mirage of an oasis, I am not directly
aware of the world but only with the way the world appears, this should not be
thought, says the direct realist, to entail that in non-deceived cases, such as that in
which I am actually looking at an oasis in the distance, I am not directly acquainted
with objects in the world.
One might apply the same line of reasoning here. There clearly are cases in which one
might make a judgement about what someone is experiencing and be wrong.
Moreover, we can certainly conceive of cases in which one makes a judgement that
something has a mind – a robot, say – when in fact it doesn’t. Conceding this much,
however, doesn’t by itself ensure that you can never know what someone else is
experiencing – or, indeed, that they have a mind – just by observing them. Why should
the cases in which one’s judgements go wrong dictate whether one has knowledge in
cases where one’s judgements go right? Of course, such knowledge, if it is possessed,
is bound to be fallible – we could be wrong. But then we are usually happy to grant
knowledge in the absence of infallibility, so why not here?
scepticism about other minds • 167
CHAPTER SUMMARY
• The problem of other minds concerns the fact that it seems that we are unable to
observe another person’s mind in the same way that we can observe physical
objects like tables and chairs. So how, then, do we know that there are other minds
in the first place?
• One way to try to resolve this problem is to make use of the argument from
analogy, which notes correlations between our behaviour and our mental states, and
thereby inductively draws conclusions about the mental states of others who
behave in ways that are similar to how we behave.
• The style of reasoning employed in the argument from analogy is defective,
however, since one cannot legitimately reason from a correlation that holds in a
single (and apparently unrepresentative) case to a general conclusion that applies
to many cases.
• We then distinguished two closely related problems that are involved in the
problem of other minds. The first (noted above) is whether other minds exist. The
second is whether, given that other minds exist, these minds are like our own. As
we noted, it could be that we are able to know that there are other minds, but are
nevertheless unable to know that these minds are like our own. This is because it
seems possible that other people might experience the world very differently from
how I experience it, but in such a way that these differences in subjective
experience are undetectable to others.
• Finally, we looked at one way in which one might respond to the problem of other
minds (in both its forms), which is to hold that we can, at least sometimes, have
direct knowledge of another person’s mind. For example, if I see someone writhing
around on the ground before me, I could come to know, without needing to make
any inference, that this person is in pain. We noted that such a view is very
controversial.

STUDY QUESTIONS
1 Why might it be thought problematic to suppose that one can know that there are
other minds? What is it about our beliefs in the existence of other minds that makes
them suspect?
2 What is the argument from analogy, and how is it supposed to resolve the problem
of other minds? What difficulties does this argument face? Does this argument
succeed in showing that we can have knowledge of other minds?
3 Explain, in your own words, why there is a difference between doubt about the
existence of other minds, and doubt that others have minds like one’s own. What
special reasons might there be to doubt the latter?
4 Is it plausible to suppose that one can directly observe someone else’s pain, and
thereby come to know, without inference, that they are in pain? If one could, then
how would this help us resolve the problem of other minds?