The Fabric of Reality David Deutch

particular heliocentric theory that Galileo was defending either. (So much,
then, for his having been convinced by accumulated observations!) But for
all that, the Church took no position in this controversy. The Inquisition did
not care where the planets 
appeared to be; what they cared about was
reality. They cared where the planets really were, and they wanted to
understand the planets through explanations, just as Galileo did.
Instrumentalists and positivists would say that since the Church was
perfectly willing to accept Galileo’s observational predictions, further
argument between them was pointless, and that his muttering ‘
eppur si
muove’ was strictly meaningless. But Galileo knew better, and so did the
Inquisition. When they denied the reliability of scientific knowledge, it was
precisely the explanatory part of that knowledge that they had in mind.
Their world-view was false, but it was not illogical. Admittedly they believed
in revelation and traditional authority as sources of reliable knowledge. But
they also had an independent reason for criticizing the reliability of
knowledge obtained by Galileo’s methods. They could simply point out that
no amount of observation or argument can ever prove that one explanation
of a physical phenomenon is true and another false. As they would put it,
God could produce the same observed effects in an infinity of different ways,
so it is pure vanity and arrogance to claim to possess a way of knowing,
merely through one’s own fallible observation and reason, which way He
chose.
To some extent they were merely arguing for modesty, for a recognition of
human fallibility. And if Galileo was claiming that the heliocentric theory was
somehow proven, or nearly so, in some inductive sense, they had a point. If
Galileo thought that his methods could confer on any theory an authority
comparable to that which the Church claimed for its doctrines, they were
right to criticize him as arrogant (or, as they would have put it,
blasphemous), though of course by the same standard they were much
more arrogant themselves.

---

So how can we defend Galileo against the Inquisition? What should Galileo’s
defence have been in the face of this charge of claiming too much when he
claimed that scientific theories contain reliable knowledge of reality? The
Popperian defence of science as a process of problem-solving and
explanation-seeking is not sufficient in itself. For the Church too was
primarily interested in explanations and not predictions, and it was quite
willing to let Galileo solve problems using any theory he chose. It was just
that they did not accept that Galileo’s solutions (which they would call mere
‘mathematical hypotheses’) had any bearing on external reality. Problem-
solving, after all, is a process that takes place entirely within human minds.
Galileo may have seen the world as a book in which the laws of nature are
written in mathematical symbols. But that is strictly a metaphor; there are no
explanations in orbit out there with the planets. The fact is that all our
problems and solutions are located within ourselves, having been created by
ourselves. When we solve problems in science we arrive through argument
at theories whose explanations seem best to us. So, without in any way
denying that it is right and proper, and useful, for us to solve problems, the
Inquisition and modern sceptics might legitimately ask what scientific
problem-solving has to do with reality. We may find our ‘best explanations’
psychologically satisfying. We may find them helpful in making predictions.
We certainly find them essential in every area of technological creativity. All
this does justify our continuing to seek them and to use them in those ways.
But why should we be obliged to take them as fact? The proposition that the
Inquisition forced Galileo to endorse was in effect this: that the Earth is in
fact at rest, with the Sun and planets in motion around it; but that the paths
on which these astronomical bodies travel are laid out in a complex way
which, when viewed from the vantage-point of the Earth, is also consistent
with the Sun being at rest and the Earth and planets being in motion. Let me
call that the ‘Inquisition’s theory’ of the solar system. If the Inquisition’s
theory were true, we should still expect the heliocentric theory to make
accurate predictions of the results of all Earth-based astronomical
observations, even though it would be factually false. It would therefore
seem that any observations that appear to support the heliocentric theory
lend equal support to the Inquisition’s theory.
One could extend the Inquisition’s theory to account for more detailed
observations that support the heliocentric theory, such as observations of
the phases of Venus, and of the small additional motions (called ‘proper
motions’) of some stars relative to the celestial sphere. To do this one would
have to postulate even more complex manoeuvrings in space, governed by
laws of physics very different from those that operate on our supposedly
stationary Earth. But they would be different in precisely such a way as to
remain observationally consistent with the Earth being in motion and the
laws being the same out there as they are here. Many such theories are
possible. Indeed, if making the right predictions were our only constraint, we
could invent theories which say that anything we please is going on in
space. For example, observations alone can never rule out the theory that
the Earth is enclosed in a giant planetarium showing us a simulation of a
heliocentric solar system; and that outside the planetarium there is anything
you like, or nothing at all. Admittedly, to account for present-day
observations the planetarium would also have to redirect our radar and laser
pulses, capture our space probes, and indeed astronauts, send back fake

---

messages from them and return them with appropriate moonrock samples,
altered memories, and so on. It may be an absurd theory, but the point is
that it cannot be ruled out by experiment. Nor is it valid to rule out any theory
solely on the grounds that it is ‘absurd’: the Inquisition, together with most of
the human race in Galileo’s time, thought it the epitome of absurdity to claim
that the Earth is moving. After all, we cannot feel it moving, can we? When it
does move, as in an earthquake, we feel that unmistakably. It is said that
Galileo delayed publicly advocating the heliocentric theory for some years,
not for fear of the Inquisition but simply for fear of ridicule.
To us, the Inquisition’s theory looks hopelessly contrived. Why should we
accept such a complicated and 
ad hoc account of why the sky looks as it
does, when the unadorned heliocentric cosmology does the same job with
less fuss? We may cite the principle of Occam’s razor: ‘do not multiply
entities beyond necessity’ — or, as I prefer to put it, ‘do not complicate
explanations beyond necessity’, because if you do, the unnecessary
complications themselves remain unexplained. However, whether an
explanation is or is not ‘contrived’ or ‘unnecessarily complicated’ depends on
all the other ideas and explanations that make up one’s world-view. The
Inquisition would have argued that the idea of the Earth moving is an
unnecessary complication. It contradicts common sense; it contradicts
Scripture; and (they would have said) there is a perfectly good explanation
that does without it.
But is there? Does the Inquisition’s theory really provide alternative
explanations without having to introduce the counter-intuitive ‘complication’
of the heliocentric system? Let us take a closer look at how the Inquisition’s
theory explains things. It explains the apparent stationarity of the Earth by
saying that it 
is stationary. So far, so good. On the face of it that explanation
is better than Galileo’s, for he had to work very hard, and contradict some
common-sense notions of force and inertia, to explain why we do not feel the
Earth move. But how does the Inquisition’s theory cope with the more
difficult task of explaining planetary motions?
The heliocentric theory explains them by saying that the planets are seen to
move in complicated loops across the sky because they are really moving in
simple circles (or ellipses) in space, but the Earth is moving as well. The
Inquisition’s explanation is that the planets are seen to move in complicated
loops because they really are moving in complicated loops in space; 
but
(and here, according to the Inquisition’s theory, comes the essence of the
explanation) this complicated motion is governed by a simple underlying
principle: namely, that the planets move in such a way that, when viewed
from the Earth, they appear just as they would if they and the Earth were in
simple orbits round the Sun.
To understand planetary motions in terms of the Inquisition’s theory, it is
essential that one should understand this principle, for the constraints it
imposes are the basis of every detailed explanation that one can make
under the theory. For example, if one were asked why a planetary
conjunction occurred on such-and-such a date, or why a planet backtracked
across the sky in a loop of a particular shape, the answer would always be
‘because that is how it would look if the heliocentric theory were true’. So
here is a cosmology — the Inquisition’s cosmology — that can be
understood only in terms of a different cosmology, the heliocentric

---

cosmology that it contradicts but faithfully mimics.
If the Inquisition had seriously tried to understand the world in terms of the
theory they tried to force on Galileo, they would also have understood its
fatal weakness, namely that it fails to solve the problem it purports to solve. It
does 
not explain planetary motions ‘without having to introduce the
complication of the heliocentric system’. On the contrary, it unavoidably
incorporates that system as part of its own principle for explaining planetary
motions. One cannot understand the world through the Inquisition’s theory
unless one understands the heliocentric theory first.
Therefore we are right to regard the Inquisition’s theory as a convoluted
elaboration of the heliocentric theory, rather than vice versa. We have
arrived at this conclusion not by judging the Inquisition’s theory against
modern cosmology, which would have been a circular argument, but by
insisting on taking the Inquisition’s theory seriously, in its own terms, as an
explanation of the world. I have mentioned the grass-cure theory, which can
be ruled out without experimental testing because it contains no explanation.
Here we have a theory which can also be ruled out without experimental
testing, because it contains a bad explanation — an explanation which, in its
own terms, is worse than its rival.
As I have said, the Inquisition were realists. Yet their theory has this in
common with solipsism: both of them draw an arbitrary boundary beyond
which, they claim, human reason has no access — or at least, beyond which
problem-solving is no path to understanding. For solipsists, the boundary
tightly encloses their own brains, or perhaps just their abstract minds or
incorporeal souls. For the Inquisition, it enclosed the entire Earth. Some
present-day Creationists believe in a similar boundary, not in space but in
time, for they believe that the universe was created only six thousand years
ago, complete with misleading evidence of earlier events. 
Behaviourism is
the doctrine that it is not meaningful to explain human behaviour in terms of
inner mental processes. To behaviourists, the only legitimate psychology is
the study of people’s observable responses to external stimuli. Thus they
draw exactly the same boundary as solipsists, separating the human mind
from external reality; but while solipsists deny that it is meaningful to reason
about anything outside that boundary, behaviourists deny that it is
meaningful to reason about anything inside.
There is a large class of related theories here, but we can usefully regard
them all as variants of solipsism. They differ in where they draw the
boundary of reality (or the boundary of that part of reality which is
comprehensible through problem-solving), and they differ in whether, and
how, they seek knowledge outside that boundary. But they all consider
scientific rationality and other problem-solving to be inapplicable outside the
boundary — a mere game. They might concede that it can be a satisfying
and useful game, but it is nevertheless only a game from which no valid
conclusion can be drawn about the reality outside.
They are also alike in their basic objection to problem-solving as a means of
creating knowledge, which is that it does not deduce its conclusions from
any ultimate source of justification. Within the respective boundaries that
they choose, the adherents of all these theories do rely on the methodology
of problem-solving, confident that seeking the best available explanation is
also the way of finding the truest available theory. But for the truth of what

---

lies outside those boundaries, they look elsewhere, and what they all seek is
a source of ultimate justification. For religious people, divine revelation can
play that role. Solipsists trust only the direct experience of their own
thoughts, as expressed in Rene Descartes’s classic argument 
cogito ergo
sum (‘I think, therefore I exist’).
Despite Descartes’s desire to base his philosophy on this supposedly firm
foundation, he actually allowed himself many other assumptions, and he was
certainly no solipsist. Indeed, there can have been very few, if any, genuine
solipsists in history. Solipsism is usually defended only as a means of
attacking scientific reasoning, or as a stepping-stone to one of its many
variants. By the same token, a good way of defending science against a
variety of criticisms, and of understanding the true relationship between
reason and reality, is to consider the argument against solipsism.
There is a standard philosophical joke about a professor who gives a lecture
in defence of solipsism. So persuasive is the lecture that as soon as it ends,
several enthusiastic students hurry forward to shake the professor’s hand.
‘Wonderful. I agreed with every word,’ says one student earnestly. ‘So did I,’
says another. ‘I am very gratified to hear it,’ says the professor. ‘One so
seldom has the opportunity to meet fellow solipsists.’
Implicit in this joke there is a genuine argument against solipsism. One could
put it like this. What, exactly, was the theory that the students in the story
were agreeing with? Was it the professor’s theory, that they themselves do
not exist because only the professor exists? To believe that, they would first
have had to find some way round Descartes’s 
cogito ergo sum argument.
And if they managed that, they would not be solipsists, for the central thesis
of solipsism is that the solipsist exists. Or has each student been persuaded
of a theory 
contradicting the professor’s, the theory that that particular
student exists, but the professor and the other students do not? That would
indeed make them all solipsists, but none of the students would be agreeing
with the theory that the professor was defending. Therefore neither of these
two possibilities amounts to the students’ having been persuaded by the
professor’s defence of solipsism. If they adopt the professor’s opinion, they
will not be solipsists, and if they become solipsists, they will have become
convinced that the professor is mistaken.
This argument is trying to show that solipsism is literally indefensible,
because by accepting such a defence one is implicitly contradicting it. But
our solipsistic professor could try to evade that argument by saying
something like this: ‘I can and do consistently defend solipsism. Not against
other people, for there are no other people, but against opposing arguments.
These arguments come to my attention through dream-people, who behave
as if they were thinking beings whose ideas often oppose mine. My lecture
and the arguments it contains were not intended to persuade these dream-
people, but to persuade myself — to help me to clarify my ideas.’
However, if there are sources of ideas that behave as 
if they were
independent of oneself, then they necessarily 
are independent of oneself.
For if I define ‘myself as the conscious entity that has the thoughts and
feelings I am aware of having, then the ‘dream-people’ I seem to interact
with are by definition something other than that narrowly defined self, and so
I must concede that something other than myself exists. My only other
option, if I were a committed solipsist, would be to regard the dream-people

---

as creations of my unconscious mind, and therefore as part of ‘myself in a
looser sense. But then I should be forced to concede that ‘myself had a very
rich structure, most of which is independent of my conscious self. Within that
structure are entities — dream-people — who, despite being mere
constituents of the mind of a supposed solipsist, behave exactly as if they
were committed 
anti-solipsists. So I could not call myself wholly a solipsist,
for only my narrowly defined self would take that view. Many, apparently
most, of the opinions held within my mind as a whole would oppose
solipsism. I could study the ‘outer’ region of myself and find that it seems to
obey certain laws, the same laws as the dream-textbooks say apply to what
they call the physical universe. I would find that there is far more of the outer
region than the inner region. Aside from containing more ideas, it is also
more complex, more varied, and has more measurable variables, by a
literally astronomical factor, than the inner region.
Moreover, this outer region is amenable to scientific study, using the
methods of Galileo. Because I have now been forced to define that region as