Think: A Compelling Introduction to Philosophy (7 page)

If Descartes's project is to use reason to fend off universal doubt
about the truthfulness of reason, then it has to fail.

Hume's challenge seems convincing. It looks as though Descartes
was doomed to failure. So what should be the outcome? General
scepticism, meaning pessimism about whether there is any har-
nmony at all between the way we believe things to be and the way
they are? Or something else? Other possibilities need introduction.

One way of thinking-Hume's own-accepts the view that our
system of belief needs some kind of foundation. However, it denies
that that foundation could have the kind of rational status that
Descartes wanted. The veracity (truthfulness) of our senses and
reasonings is itself part of the foundation. It cannot itself be
demonstrated by standing on some other `original principle. For
all of us, outside the philosophical study, it comes naturally to trust
our common experience. We grow up doing so, and as we grow up
we become good at recognizing danger areas (illusions, mirages)
against the background of natural beliefs we all form. The selfcorrective nature of our systems of belief, mentioned above, is all
we need. We could call this approach non-rational or natural foun-
dationalism. (Not of course implying that there is anything irrational about it. It is just that the things in the foundation do not
have the demon-proof way of `standing to reason' that Descartes
had hoped for.) Hume himself gave a number of arguments for
side-lining any appeal to rationality, and we visit some of them in
due course.

The emphasis on natural ways of forming belief chimes in with
another strand in Hume and other British philosophers of the
seventeenth and eighteenth centuries, which is their distrust of the
power of unaided reason. For these philosophers, the best contact
between mind and the world is not the point at which a mathematical proof crystallizes, but the point at which you see and touch
a familiar object. Their paradigm was knowledge by sense experience rather than by reason. Because of this, they are labelled empiricists, whereas Descartes is a card-carrying rationalist. The
labels, however, conceal a lot of important detail. For example, at
some points when he gets under pressure, Descartes himself
appears to say that the really good thing about clear and distinct
ideas is that you can't doubt them when you have them. This is
not really a certification by reason, so much as the very same kind
of natural potency that Hume himself attaches to basic empirical
beliefs. And soon we visit an area where the champion of British
empiricism, John Locke (1632-1704), is as rationalist as the best of
them. Great philosophers have a disturbing habit of resisting labelling.

On this view, I)escartes's problem was that he relied too much
on the powers of reason. Instead, we can appeal to nature, here
meaning our natural propensities to form beliefs and to correct
them. And what of the Evil Demon? On this story, the true moral of
I)escartes's struggles is that if we raise the question whether our
experience and reasoning (en bloc) accords with the way the
world is (en bloc), it will take an act of faith to settle it.'God' simply labels whatever it is that ensures this harmony between belief
and the world. But, as Hume says in the passage just quoted, we do not find a need to raise this question in normal life. The hyperbolic
doubt, and the answer to it, is in this sense unreal.

This may sound sensible, or it may just sound complacent. But
to blunt the charge of complacency, we can at least notice this. Regarding the doubt as unreal does not have to mean that we simply
turn our backs on the problem of harmony between appearance
and reality: how we think and how things are. We can approach it
from within our normal framework of beliefs. In fact, when Hume
himself approached it in this way, he became overwhelmed by difficulties in our ordinary ways of thinking about things: difficulties
strong enough to reintroduce scepticism about our ability to know
anything about the world. This is the topic of Chapter 7.

However, one piece of optimism is available to us, two centuries
later. We might thus suppose that evolution, which is presumably
responsible for the fact that we have our senses and our reasoning
capacities, would not have selected for them (in the shape in which
we have them) had they not worked. If our eyesight, for example,
did not inform us of predators, food, or mates just when predators,
food, and mates are about, it would be of no use to us. So it is built
to get these things right. The harmony between our minds and the
world is due to the fact that the world is responsible for our minds.
Their function is to represent it so that we can meet our needs; if
they were built to represent it in any way other than the true way,
we could not survive. This is not an argument designed to do away
with the Evil Demon. It is an argument that appeals to things we
take ourselves to know about the world. Unfortunately, we have to
visit in time the area of Hume's doubts, where things we take ourselves to know about the world also serve to make that knowledge seem doubtful.

A rather different response shrugs off the need for any kind of
'foundations', whether certified by reason, as Descartes hoped, or
merely natural, as in Hume. This approach goes hack to emphasizing instead the coherent structure of our everyday system of beliefs:
the way they hang together, whereas the sporadic experiences or
beliefs we get in dreams are fragmentary and incoherent. It then
points out an interesting feature of coherent structures, namely
that they do not need foundations. A ship or a web may he made up
of a tissue of interconnecting parts, and it derives its strength from
just those interconnections. It does not need a 'base' or a 'starting
point' or 'foundation'. A structure of this kind can have each bit
supported by other bits without there being any bit that supports
all the others without support itself. Similarly, if any one belief is
challenged, others can support it, unless, of course, it turns out that
nothing else supports it, in which case it should be dropped. The
Austrian philosopher Otto Neurath (1882-1945) used this lovely
metaphor for our body of knowledge:

We are like sailors who on the open sea must reconstruct their
ship but are never able to start afresh from the bottom.

Any part can he replaced, provided there is enough of the rest on
which to stand. But the whole structure cannot be challenged en
bloc, and if we try to do so, we find ourselves on l)escartes's lonely
rock.

This approach is usually called 'cohere ntism'. Its motto is that
while every argument needs premises, there is nothing that is the premise of every argument. There is no foundation on which everything rests. Coherentism is nice in one way, but dissatisfying in
another. It is nice in what it does away with, namely the elusive
foundations. It is, however, not clear that it offers us enough to replace them. This is because we seemed able to understand the possibility represented by the Evil Demon-that our system of belief
should be extensive and coherent and interlocking, but all completely wrong. As I said in the introduction to this chapter, even as
children we fall naturally into wondering whether all experience
might be a dream. We might sympathize with Descartes's thought
that if the options are coherentism or scepticism, the more honest
option would be scepticism.

It is good, then, to remember four options in epistemology
(the theory of knowledge). There is rational foundationalism, as
attempted by Descartes. There is natural foundationalism, as attempted in Hume. There is coherentism. And brooding over all of
them, there is scepticism, or the view that there is no knowledge.
Each of these has had distinguished defenders. Whichever the
reader prefers, he or she will find good philosophical company.
One might think that Descartes got almost everything right, or
that he got almost everything wrong. The baffling thing is to defend whichever answer commends itself.

LOCAL SCEPTICISMS

Scepticism can he raised in particular areas, as well as in the global
fashion of Descartes. Someone might be convinced that we have, say, scientific knowledge, but be very doubtful about knowledge in
ethics or politics or literary criticism. We find particular areas
shortly where it does not take hyperbolic doubt, only a bit of caution, for us to become insecure. However, there are other nice examples of highly general areas where scepticism is baffling. The
philosopher Bertrand Russell (1872-1970) considered the example
of time. How do I know that the world did not come into existence
a very few moments ago, but complete with delusive traces of a
much greater age? Those traces would include, of course, the modifications of the brain that give us what we take to be memories.
They would also include all the other things that we interpret as
signs of great age. In fact, Victorian thinkers struggling to reconcile
the biblical account of the history of the world with the fossil
record had already suggested much the same thing about geology.
On this account, around 4,000 years ago God laid down all the misleading evidence that the earth is about 4,000 million years old
(and, we can now add, misleading signs that the universe is about
13,000 million years old). This was never a popular move, probably
because if you are sceptical about time, you quickly become sceptical about everything, or maybe because it presents God as something like a large-scale practical joker. Russell's possibility sounds
almost as far-fetched as Descartes's Evil Demon.

However, there is one highly intriguing thing about Russell's
scenario. This is that it can actually be argued to be scientifically
more probable than the alternative we all believe in! This is because
science tells us that `low-entropy' or, in other words, highly ordered
systems are more improbable. In addition, as physical systems like
the cosmos evolve, entropy or disorder increases. The smoke never returns into the cigarette; the toothpaste never goes back into the
tube. The extraordinary thing is that there was ever enough order in
things for the smoke to be in the cigarette or the toothpaste to be in
the tube in the first place. So, one might argue, it is `easier' for a
moderately disordered world, such as the world is now, to come
into existence, than it is for any lower-entropy, more orderly ancestor. Intuitively, it is as if there are more ways this can happen, just as
there are more ways you can get four-letter or five-letter words in
an initial hand of seven letters in Scrabble, than there are in which
you can get a seven-letter word. It is much more probable that you
get a four-letter word than a seven-letter word. Similarly, the argument goes, it is as if God or Nature had less to do, to make the world
as it is today out of nothing, than to make the lower-entropy world
as it is supposed to have been some thirteen billion years ago out of
nothing. Therefore, it is more probable that it happened like that.
In a straight competition for probability between Russell's outlandish hypothesis and common sense, Russell wins. I leave this for
the reader to ponder.

THE MORAL

How then should we regard knowledge? Knowledge implies authority. the people who know are the people to whom we should
listen. It implies reliability: the people who know are those who are
reliable at registering the truth, like good instruments. To claim
knowledge implies claiming a sense of our own reliability. And to
accord authority to someone or some method involves seeing it as reliable. The unsettling scenarios of a Descartes or a Russell unseat
our sense of our own reliability. Once we have raised the outlandish possibilities, our sense of a reliable connection between the
way things are and the ways we take them to be goes dim. We could
regain it, if we could argue that the scenarios are either impossible,
or at least have no real chance of being the way things are.The difficulty is that it is hard to show them to be impossible, and in these
abstract realms we have no very good sense of probabilities or
chances. So it is difficult to argue that they have no chance of being
true without relying on the very opinions that they query. Hence,
scepticism permanently beckons, or threatens, us. We may be
tracking the world reliably, but we may not. To revert to the engineering analogy I used in the Introduction, the structure of our
thought seems to span large gaps: here, the gap between how things
appear and how they might be. We hand ourselves the right to cross
those gaps. But if we (1o this trailing no very good sense of our own
reliability or harmony with the truth, then that right seems illfounded. And this is what the sceptic insists upon. Any confidence
in a harmony between the way we take things to be, and the way
they are, will seem to be a pure act of faith.