The Beginning of Infinity

DAVID DEUTSCH
The Beginning of Infinity

Explanations that Transform the World
VIKING
VIKING
Published by the Penguin Group
Contents
Penguin Group (USA) Inc., 375 Hudson Street,
New York, New York 10014, U.S.A. .
Penguin Group (Canada), 90 EgIinton Avenue East, Swa: 700,
Toronto Ontario, Canada MtP z y 3
(a division ~f Pearson Penguin Canada Inc.)
Penguin Books Ltd, 80'Strand, London WC2.R ?Rt, England
Penguin Ireland"5 St. Stephen's Green, Dublin a, Ireland
(a division of Penguin Books Ltd)
Penguin Books Australia Ltd, z So Camberwell Road, Camberwell,
Victoria 3 U4, Australia
(a division of Pearson Australia Group Pty Ltd)
P . Books India Pvt Ltd, II Community Centre, Panchsheel Park,
engum New Delhi - 110 017, India Acknowledgements VI
Penguin Group (NZj, 67 Apollo Drive, RosedaIe,Auckland 063Z,
New Zealand (a division of Pearson New Zealand Ltd) Introduction Vll
Penguin Books (South Africa) (Pty) Ltd, Z4 Sturdee Avenue,
Rosebank, Johannesburg ZI96, South Africa
Penguin Books Ltd, Registered Offices: I. The Reach of Explanations I

80 Strand, London WCzR oRt, England
2. Closer to Reality 34
First American edition
Published in ZOII by Viking Penguin, 3. The Spark 42
a member of Penguin Group (USA) Inc.
4. Creation . 78
1 • 3 4 S 6 7 8 9 10
Copyright C David Deutsch, zan

5. The Reality of Abstra~ons 1°7
All rights reserved 6. The Jump to Universality 125
Excerpts from The World of Parmenid .. by Karl Popper, edited by Arne F. P~ (RoutIecIae. 1998). Used by 7. Artificial Creativity. 148
permission of the University of KIagenfun, Karl Popper Ubrary.
Wustration on page 34: Starfidd image from the Digitized Sky Survey (0 AURA) counesy of the Palomar Obeervatory
8. A Window on Infinity 164
and Digitized Sky Survey, created by the Space Telescope Science Institute, operated by AURA, Inc. for NASA. 9. Optimism 196
Reprnduced with permission of AURAISTScl.
10. A Dream of Socrates 223
Wustration on page 4z6: 0 IlettmannlCorbis
I I. The Multiverse . 258
LIBRARY OF CONGRESS CATALOGING IN PUllUCATION DATA
12. A Physicist'S History of Bad Philosophy 3°5
Deutsch, David.
The besinnins of infinity : explanations that transform the world I David Deutsch.
13. Choices 326
p.an.
Inc1udes bibliographicaI refereoca and index.
ISBN 978-0-67C>-OU7S-S (hardback) 14. Why are Flowers Beautiful? 353
Explanation ••• Infinite. 3. !icienc:e--PhiIosophy.1. TttIe.
1.
QI7S·3z.E9'7D48 zan IS- The Evolution of Cultu~e. 369
SOI-dcu
Z.OIIOO4I10
16. The Evolution of Creativity 398
Printed in the United States of America 17. Unsustainable 418
~nhout ~ting the '!&hIS under copyright reserved above, no pan of this publkation may be reproduced, stored in or 18. The Beginning 443
mtrocl~ mto a ~ system, or ~mi~ in an~ ~ or by any means (electronic, mechanical, photocopying,
~ or otherwise), WIthout the pnor written pemusslon of both the COPYright owner and the above publisher of
Bibliography
The ~ uploading. and distribution of this book via the Internet or via any other means without the permisajon of
the publisher II iI1cpI.~ PDnishable "!
law. PIeaae pllIChase only authorized elearonic editions and do not panicipate in Index
or encouraae dearonic puaq of COPyrightable materials. Your support of the author's rish1l is appreciated.
,~
1
INTRODUCTION
happen, entailsa journey through virtually every fundamental field of

scienceand philosophy.From each such field we learn that, although I
progress has no necessaryend, it does have a necessary beginning: a
cause,or an event with which it starts, or a necessary condition for it The Reach of Explanations
~ota~e,offan.dto thrive. Each of these beginnings is 'the beginning of
infinity . as viewed from the perspective of that field. Many seem,
su~rficlally, to be unconnected. But they are all facets of a single
attribute of reality,which I call the beginning of infinity.
Behind it all is surely an idea so simple, so beautiful, that when

we grasp it - in a decade, a century, or a millennium - we will
all say to each other, how could it have been otherwise?
John Archibald Wheeler, Annals of the
New York Academy of Sciences, 480 (1986)
To unaided human eyes, the universe beyond our solar system looks
like a few thousand glowing dots in the night sky, plus the faint, hazy
streaks of the Milky Way. But if you ask an astronomer what is out
there in reality, you will be told not about dots or streaks, but about
stars: spheres of incandescent gas millions of kilometres in diameter
and light years away from us. You will be told that the sun is a typical
star, and looks different from the others only because we are much
closer to it - though still some 150 million kilometres away. Yet, even
at those unimaginable distances, we are confident that we know what
makes stars shine: you will be told that they are powered by the nuclear
energy released by transmutation - the conversion of one chemical
element into another (mainly hydrogen into helium).
Some types of transmutation happen spontaneously on Earth, in the
decay of radioactive elements. This was first demonstrated in 1901, by
the physicists Frederick Soddy and Ernest Rutherford, but the concept
of transmutation was ancient. Alchemists had dreamed for centuries
of transmuting 'base metals', such as iron or lead, into gold. They never
came close to understanding what it would take to achieve that, so
they never did so. But scientists in the twentieth century did. And so
do stars, when they explode as supernovae. Base metals can be
I
THE BEGINNING OF INFINITY THE REACH OF EXPLANATIONS
transmutedinto gold by stars, and by intelligentbeingswho understand the naked eye, they can outshine a supernova for millions of years at
the processesthat power stars, but by nothing else in the universe. a time. They are powered by massive black holes at the centres of
As for the MilkyWay,you will be told that, despite its insubstantial galaxies, into which entire stars are falling - up to several per day for
appearance,it is the most massiveobject that we can see with the naked a large quasar - shredded by tidal effects as they spiral in. Intense
eye:a galaxy that includesstars by the hundreds of billions, bound by magnetic fields channel some of the gravitational energy back out
their ~utual gravitation across tens of thousands of light years. We in the form of jets of high-energy particles, which illuminate the sur-
are seeing it from the inside, because we are part of it. You will be told rounding gas with the power of a trillion suns.
th~t,alth?ugh o~ ni~t sk! appears serene and largely changeless, the Conditions are still more extreme in the black hole's interior (within
u~l~erseISseething with violent activity. Even a typical star converts the surface of no return known as the 'event horizon'), where the very
mtllIo~sof tonnes of mass into energy every second, with each gram fabric of space and time may be being ripped apart. All this is happening
r~leasmgas muchenergyas an atom bomb. Youwill be told that within in a relentlessly expanding universe that began about fourteen billion
th e range of our best telescopes, which can see more galaxies than years ago with an all-encompassing explosion, the Big Bang, that
t ere are stars in our galaxy, there are several supernova explosions makes all the other phenomena I have described seem mild and
per second,each brieflybrighter than all the other stars in its galaxy inconsequential by comparison. And that whole universe is just a sliver
put together.We do not k h I' ,
all id now were ife and intelligence exist if at of an enormously larger entity, the multiverse, which includes vast
,0UtSI e our solar system, so we do not know h '
explosionsare horrendous tra edies B ow many of those numbers of such universes.
devastatesall the I h g . ut we,~o know that a supernova The physical world is not only much bigger and more violent than
that may existther~~::~U~i:t may, be o~bltmg ~t,wiping out all life it once seemed, it is also immensely richer in detail, diversity and
technologyfar suneri g any intelligenr bemgs, unless they have incident. Yet it all proceeds according to elegant laws of physics that
a human at a ra:g~eortfobr,tll~
ours'flkt~lneutrinoradiation alone would kill we understand in some depth. I do not know which is more awesome:
I Ions 0 uometrs if h ' the phenomena themselves or the fact that we know so much about
werefilledwith lead shield; v s, even I t at entire distance
mg. let we owe our ' them.
they are the Source,through ,eXIstence to supernovae:
of which our bodies a' d tra nSmutatlOn, of most of the elements How do we know? One of the most remarkable things about science
The h ,n our P anet, l are composed is the contrast between the enormous reach and power of our best
re are p enomena that outshine s .
X-ray telescopein Earth bi d upernovae. In March 2008 an theories and the precarious, local means by which we create them. No
or It erected an losi human has ever been at the surface of a star, let alone visited the core
as a, 'gamma-ray burst' billion ' exp OSIonof a type known
5
acrossthe known univ:r:· It I lion light years away. That is halfway where the transmutation happens and the energy is produced. Yet we
form a black hole _ an b~' whasprobably a single star collapsing to see those cold dots in our sky and know that we are looking at
ligh 0 ject w ose gravity' ,
I t can escape from its i ISso Intense that not even the white-hot surfaces of distant nuclear furnaces. Physically, that
igh
bn ter than a million s I S InterIOr The losi
. exp osion was intrinsically experience consists of nothing other than our brains responding to
h upernovae and ld h
t e naked eye from Earth h ' wou ave been visible with electrical impulses from our eyes, And eyes can detect only light that
- tough 1 fai 1
seconds,so it is unlikelyth t on Y amt y and for only a few is inside them at the time. The fact that the light was emitted very far
ty' II a anyone here'S
pica y fading on a timescale of m saw I,t. upernovae last longer, away and long ago, and that much more was happening there than
to see a few in our galaxy even bef onths, ,:hlch allowed astronomers just the emission of light - those are not things that we see. We know
Another class of cosmic mo ore the ~nvention of telescopes. them only from theory.
known as quasars, ate in a dif£er::~;~ the Inten~ly luminous objects Scientific theories are explanations: assertions about what is out
ague.Too dlStant to be seen with there and how it behaves. Where do ,these theories come from? For
3
most of the history of science, it was mistakenly believed that we obtaining new knowledge, in contrast with the medieval fatalism that
'derive'them from the evidenceof our senses- a philosophical doctrine had expected everything important to be known already. Thus, despite
known as empiricism: being quite wrong about where scientific knowledge comes from,
empiricism was a great step forward in both the philosophy and the
history of science. Nevertheless, the question that sceptics (friendly and
Sensory experiences
'Derivation' unfriendly) raised from the outset always remained: how can knowledge
(such as 'Extrapolation', ~ Theories / knowledge of reality
'----- __ --J 'Generalization' or 'Induction') of what has not been experienced possibly be 'derived' from what has?
What sort of thinking could possibly constitute a valid derivation of
Empiricism the one from the other? No one would expect to deduce the geography
of Mars from a map of Earth, so why should we expect to be able
~o~ex~m~le,the philosopher John Locke wrote in 1689 that the mind to learn about physics on Mars from experiments done on Earth?
IShk~ white paper' on to which sensory experience writes, and that Evidently, logical deduction alone would not do, because there is a
that ISwhere all our knowledge of the physical world comes from logical gap: no amount of deduction applied to statements describing
An~therempiricist~etapho.r was that one could read knowledge fro~ a set of experiences can reach a conclusion about anything other than
t~e Bookof N~t~re by~akmg observations.Either way, the discoverer those experiences.
o kno,:ledge ISIts passive recipient, not its creator. The conventional wisdom was that the key is repetition: if one
But, m reality, scientifictheories are not 'derived' from anything repeatedly has similar experiences under similar circumstances, then
~: do not read them in nature, nor does nature write them into us: one is supposed to 'extrapolate' or 'generalize' that pattern and predict
ey are guesses- bold conjectures. Human mind h b that it will continue. For instance, why do we expect the sun to rise
rearran . b' '. s create t em y
. . gmgf"com I~mg,altering and adding to existing ideas with the tomorrow morning? Because in the past (so the argument goes) we
mtentlon 0 tmprovmg h Wi
at birth b t . h i b upon tern. e do not begin with 'white paper' have seen it do so whenever we have looked at the morning sky. From
, u Wit m om expect t" do,
abilityt 0 a Ions an mtentIOns and an innate this we supposedly 'derive' the theory that under similar circumstances
o Improveupon them using tho gh d .
isindeedessentialto sci b u t ~n experience, Experience
0 0
we shall always have that experience, or that we probably shall. On
ce
by empiricism It is notetnh ,ut ItsfroleISdifferent from that supposed each occasion when that prediction comes true, and provided that it
. e Source rom who h h
never fails, the probability that it will always come true is supposed to
0 0
main useis to choosebetw theori IC t eones are denved. Its

een eones that h lr d b increase. Thus one supposedly obtains ever more reliable knowledge
That is what 'learning fro . ave a ea y een guessed.
m expenence' is of the future from the past, and of the general from the particular. That
However,that was not pro rl dersn .
centurywith the work of th pehilYun herstooduntil the mid twentieth alleged process was called 'inductive inference' or 'induction', and the
. e p osop er Karl P doctrine that scientific theories are obtained in that way is called
Itwas empiricismthat firstp .d d lausi opper. So historically
al scienceas we now k r?vIEe ~ ~ ~uslbledefencefor experiment- inductiuism. To bridge the logical gap, some inductivists imagine that
. now It. mpmclst hOI h
reJectedtraditional approach k p I osop ers criticized and there is a principle of nature - the 'principle of induction' - that makes
th . es to nowledge h d inductive inferences likely to be true. 'The future will resemble the past'
au onty of holy books d h . sue as eference to the
an ot er ancient o.
authonties such as priests and d . writings, as well as human is one popular version of this, and one could add 'the distant resembles
rulesof thumb and hearsay E ac~. enucs, and belief in traditional lore the near,' 'the unseen resembles the seen' and so on.
and . mpmClsmalso Contr d' d ' But no one has ever managed to formulate a 'principle of induction'
Surprlsmglypersistent idea tha th
0 •
a icte the opposing

SQurcesof error to. be ignored And ~ e senses are little more than that is usable in practice for obtaining scientific theories from ex-
• It was 0 ti • . .
P lffilStlCbeing
t all about periences. Historically, criticism of inductivism has focused on that
.. s
failure, and on the logical gap that cannot be bridged. But that lets same underlying laws of nature. For that is an empty statement: any
induotivism off far too lightly. For it concedes inductivism's two most purported law of nature - true or false - about the future and the past
serious misconceptions. is a claim that they 'resemble' each other by both conforming to that
First, inductivism purports to explain how science obtains predictions law. So that version of the 'principle of induction' could not be used
about experiences. But most of our theoretical knowledge simply does to derive any theory or prediction from experience or anything else.
not. take that form. Scientific explanations are about reality, most of Even in everyday life we are well aware that the future is unlike the
w~lch ~oes not consist of anyone's experiences. Astrophysics is not past, and are selective about which aspects of our experience we expect
primarily about us ~what we shall see if we look at the sky), but about to be repeated. Before the year 2000, I had experienced thousands of
what stars are: their composition and what makes them shine, and times that if a calendar was properly maintained (and used the standard
how they formed, and the universal laws of physics under which Gregorian system), then it displayed a year number beginning with
that ~appened. Most of that has never been observed: no one has '19'. Yet at midnight on 31 December 1999 I expected to have the
experienced a bi!lion years, or a light year; no one could have been experience of seeing a '20' on every such calendar. Ialso expected that
present. at the Big Bang; no one will ever touch a law of physics _ there would be a gap of 17,000 years before anyone experienced a'19'
e~~ept I~ their minds, through theory. All our predictions of how under those conditions again. Neither I nor anyone else had ever
t 1O~s :,~llo?~are deduced from such explanations of how things observed such a '20', nor such a gap, but our explanatory theories told
are. 0 10 ~ctlvlsm fails even to address how we can know about stars us to expect them, and expect them we did.
and the universe, as distinct from just dots in the sk As the ancient philosopher Heraclitus remarked, 'No man ever steps
The second fundamental' '" y.
in the same river twice, for it is not the same river and he is not the
scientific theories predict that ~sconcePtlo~ in mductivism is that
that 'th t e future will resemble the past' and same man.' So, when we remember seeing sunrise 'repeatedly' under
e unseen resembles th 'd ' 'the same' circumstances, we are tacitly relying on explanatory theories
will.) But in reality the futu ' e inlik an so on. (Or that it 'probably'
from the seen S . ft re ISun, I e the past, the unseen very different to tell us which combinations of variables in our experience we should
. clence 0 en predicts db' interpret as being 'repeated' phenomena in the underlying reality, and
spectacularly different fro hi - an rings about - phenomena
For millennia people dream adnytb109 that has been experienced before. which are local or irrelevant. For instance, theories about geometry
. me a out flying b h . and optics tell us not to expect to see a sunrise on a cloudy day, even
falhng. Then they disc d d ' ut t ey expenenced only
overe goo expl heori if a sunrise is really happening in the unobserved world behind the
and then they flew _ in th d anatory t eones about flying,
at or er, Before 19 h clouds. Only from those explanatory theories do we know that failing
ever observed a nuclear-fi ' ( , 45, no uman being had
ssion atomic-bomb) I' to see the sun on such days does not amount to an experience of its
never have been one in th hi exp OSlOn; there may
, e IStOry of the . not rising. Similarly, theory tells us that if we see sunrise reflected in a
explOSion, and the condit'. d UDlverse. Yet the first such
Ions un er whi h . mirror, or in a video or a virtual-reality game, that does not count as
accurately predicted but f c It would occur, had been
- not rom th . seeing it twice. Thus the very idea that an experience has been repeated
would be like the past E ' e assumption that the future
. d " . yen sunnse - th t f ' is not itself a sensory experience, but a theory.
10 UetlVISts- is not always b d a avounte example of
, d fr 0 serve every tw So much for inductivism. And since inductivism is false, empiricism
VleWe om orbit it may h enty-four hours' when
And h appen every ni.' . must be as well. For if one cannot derive predictions from experience,
t at was known from th I nety mmutes, or not at all.
the E:arth. eory ong before anyone had ever orbited one certainly cannot derive explanations. Discovering a new explanation
It ISno defence of ind ., is inherently an act of creativity. To interpret dots in the sky as white-
the future still UetlVlsm to point out th ' hot, million-kilometre spheres, one must first have thought of the idea
does 'resemble th at In all those cases
e past' in the of such spheres. And then one must explain why they look small and
sense that it obeys the
6
7
THEBEGINNINGOF INFINITY THE REACH OF EXPLANATIONS
cold and seem to move in lockstep around us and do not fall down authorities: sensory experience and whatever fictitious process of
Suc,hideas do not create themselves, nor can they be mechanically 'derivation', such as induction, one imagines is used to extract theories
derlv~~~romanything:they have to be guessed - after which they can from experience.
be Ctltl~IZe~and tested. To the extent that experiencing dots 'writes' The misconception that knowledge needs authority to be genuine
some,thIngInto our brains,it does not write explanations but only dots. or reliable dates back to antiquity, and it still prevails. To this day,
~or, ISnature a ~oo~: one could try to 'read' the dots in the sky for a most courses in the philosophy of knowledge teach that knowledge
IIfeltllme
- many lifetimes- without learning anything about what they is some form of justified, true belief, where 'justified' means designated
rea yare.
as true (or at least 'probable') by reference to some authoritative source
Historically,that is exactly what happened. For mill . or touchstone of knowledge. Thus 'how do we know ... ?' is trans-
careful b f h I enrua, most
in h ~ servers0, t ~skybelievedthat the stars were lights embedded formed into 'by what authority do we claim ... ?' The latter question
a 0 ow, ro~atIng celestial sphere' centred on the Earth (or that is a chimera that may well have wasted more philosophers' time and
~~: werehole~In the sphere,through which the light of heaven shone) effort than any other idea. It converts the quest for truth into a quest
ISgeocentrIC - Earth-centred - h f' .
havebeendirectlyde' d f t ~ory a the universe seemed to for certainty (a feeling) or for endorsement (a social status). This
rive rom expenence d dl misconception is called justificationism.
anyone who I k d ,an repeate y confirmed'
00 e up could 'directly b ' h .
and the stars maintainin hei I' a se~e. t e celestial sphere, The opposing position - namely the recognition that there are no
held up just as the th g t die re atlv~ POSitions on it and being authoritative sources of knowledge, nor any reliable means of justifying
'
h eltocentric _ centred eory pre icts. Yet 10 reality hi' ideas as being true or probable - is called fallibilism. To believers in
hi, t e so ar system IS
on t e sun not th E h the justified-true-belief theory of knowledge, this recognition is the
at rest but in complexmotion. Altho» e art - ~nd the ~arth is not
by observingstars it ' gh we first noticed a dally rotation occasion for despair or cynicism, because to them it means that know-
'
Earth, and of the obse ISnot a property of th ledge is unattainable. But to those of us for whom creating knowledge
h e stars at all, but of the
f rvers w 0 rotate with it I' I . means understanding better what is really there, and how it really
o the deceptivenessof th I. t ISa c assic example
it is at rest beneath 0 J:esenses: the Earth looks and feels as though behaves and why, fallibilism is part of the very means by which this is
ur reer, even though" II
the celestialsphere d ' bei .. It ISrea y rotating. As for achieved. Fallibilists expect even their best and most fundamental
. d ' espite emg VISiblein b d d
It oes not exist at all. roa aylight (as the sky), explanations to contain misconceptions in addition to truth, and so
The deceptivenessof the senseswas alwa s they are predisposed to try to change them for the better. In contrast,
- and thereby,it seemed t. y a problem for empiricism the logic of justificationism is to seek (and typically, to believe that one
th h ' or SCienceThe emp] " ,b
at t e senses cannot be d' " Inclsts est defence was has found) ways of securing ideas against change. Moreover, the logic
are 0 I th f ecepttve 10 themselv Wh of fallibilism is that one not only seeks to correct the misconceptions
, , n y e alse interpretations that es. at misleads us
ISIndeed true _ but only be we place on appearances. That of the past, but hopes in the future to find and change mistaken ideas
anyth'Ing cause Our sense th I that no one today questions or finds problematic. So it is fallibilism,
fallibl .Only Our interpretations of themsd emse ves do not say
, e.,But the real key to science is h 0, and those are very not mere rejection of authority, that is essential for the initiation of
whl~h Include those interpret f t at our explanatory theories _ unlimited knowledge growth - the beginning of infinity.
conJect~~e~criticism and testing~ Ions - can be improved, through The quest for authority led empiricists to downplay and even stig-
EmpIrICismnever did hi . matize conjecture, the real source of all our theories. For if the senses
auth· y . ac ieve ItS ai f lib
w ort · It denied the legitimacy of tr ~. ~ I erating science from were the only source of knowledge, then error (or at least avoidable
as sa utlry. But unfortunately it did th: ~tIonal.authOrities, and that error) could be caused only by adding to, subtracting from or mis-
y settIng up two other false interpreting what that source is saying. Thus empiricists came to believe
8
9
that, in addition to rejectingancient authority and tradition, scientists I am not asking what authority scientific knowledge is derived from,
should suppressor ignore any new ideas they might have, except those or rests on. I mean, literally, by what process do ever truer and more
that had been properly 'derived' from experience. As Arthur Conan detailed explanations about the world come to be represented physically
Doyle's fictional detective Sherlock Holmes put it in the short story in our brains? How do we come to know about the interactions of
'A Scandalin Bohemia','It is a capital mistake to theorize before one subatomic particles during transmutation at the centre of a distant star,
has data.'
when even the tiny trickle of light that reaches our instruments from
, Butth~t w,asitselfa capital mistake.We never know any data before the star was emitted by glowing gas at the star's surface, a million
mterprenng It through theories.All observations are, as Popper put it, kilometres above where the transmutation is happening? Or about
theory-laden,It and hence fallible, as all our theories are. Consider the conditions in the fireball during the first few seconds after the Big Bang,
nerv~ ~ignal.sreaching our brains from our sense organs. Far from which would instantly have destroyed any sentient being or scientific
providing direct or untainted access to reality, even they themselves instrument? Or about the future, which we have no way of measuring
are n~verex~e,riencedfor what they really are - namely crackles of at all? How is it that we can predict, with some non-negligible degree
electrIcalacnvitv Nor. for th d .
, '1', e most part, 0 we expenence them as of confidence, whether a new design of microchip will work, or whether
~mg whe~ethey really ar~- inside our brains. Instead, we place them a new drug will cure a particular disease, even though they have never
I~ the ;eahty beyond. We do not just see blue: we see a blue sky up existed before?
t ere, ar away.We do not just feel pain: we experience a headache For most of human history, we did not know how to do any of this.
or a stomach ache Th brai h '
, h . e ram attac es those interpretations - 'head' People were not designing microchips or medications or even the wheel.
stomac 'and 'u th ' ,
it If 0 P ere - to events that are in fact within the brain For thousands of generations, our ancestors looked up at the night sky
I se. ur senseorgans th I
, 1 emse ves, and all the interpretations that we and wondered what stars are - what they are made of, what makes
conSCIOUS y and unconsciousl h hei
fallI'ble . h y attac to t eir outputs, are notoriously them shine, what their relationship is with each other and with us -
- as Witnesst e eel ti I h h
illusionand conjuring tri ~s ~a-sp ere t .eory,as ~ell as every optical which was exactly the right thing to wonder about. And they were
is. It is all theoretical i tiC . 0 ~e perc~lve nothmg as what it really using eyes and brains anatomically indistinguishable from those of
C n erpretatlon: conjecture modern astronomers. But they discovered nothing about it, Much the
onan Doyle came much closer h .
BoscombeValleyMystery' h' h d H to t e truth when, during 'The same was true in every other field of knowledge. It was not for lack of
evidence'(evidenceab ,e ~ olmes remark that 'circumstantial trying, nor for lack of thinking. People observed the world. They
out unwitnessed eve t ). , , tried to understand it - but almost entirely in vain. Occasionally they
, , . It may seem to POlo't . n s IS a very tricky thing
very straIght to hin b ' recognized simple patterns in the appearances. But when they tried to
your own point of view a littl one t g, ut If you shift
I e you may find . '. . find out what was really there behind those appearances, they failed
uncompromisingrna' It pomtmg m an equally
nner to somethi . 1 .
nothing more deceptiv th n~ entire y different ... There is almost completely.
. i6 e an an obVIOUSfa t' Th I expect that, like today, most people wondered about such things
SCient c discovery.And that a' . c. e same holds for
know? If all our theories "gam raises the question: how do we only occasionally - during breaks from addressing their more parochial
. d ongmate locally:
mm s, and can be tested 0 I Ill' as guesswork in our own concerns. But their parochial concerns also involved yearning to
, . nYocaybye'
Contamsuch extensiveand ,xpenence, how is it that they know - and not only out of pure curiosity. They wished they knew
h accurate knowl d bo how to safeguard their food supply; how they could rest when tired
we ave never experienced? e ge a ut the reality that
without risking starvation; how they could be warmer, cooler, safer, in
"'The te less pain - in every aspect of their lives, they wished they knew how
nnwucoined by the pha_ft.._"- ..
~ Norwood B,.U$SeUUA _
to make progress. But, on the timescale of individual lifetimes, they
&&lUUIOO.
10
: .. II
~ , :~ .... ..
'
almost never made any. Discoveries such as fire, clothing, stone important that was knowable had already been discovered, and
tools, bronze, and so on, happened so rarely that from an individual's was enshrined in authoritative sources such as ancient writings and
poi~t of v~ew the world never improved. Sometimes people even traditional assumptions. Some of those sources did contain some
~ealIzed,(with somewhat miraculous prescience) that making progress genuine knowledge, but it was entrenched in the form of dogmas along
In practical ,:ays would depend on progress in understanding puzzling with many falsehoods. So the situation was that all the sources from
phenomena In the sky. They even conjectured links between the two which it was generally believed knowledge came actually knew very
s,uchas myths, which they found compelling enough to dominate the~ little, and were mistaken about most of the things that they claimed
lives - yet which still bore no resemblance to the truth. In short, they to know. And therefore progress depended on learning how to reject
wanted to create knowledge, in order to make progress, but they did their authority. This is why the Royal Society (one of the earliest
not know how.
scientific academies, founded in London in 1660) took as its motto
hiS
Td was th~ ~i~a~ion from our species' earliest prehistory, through 'Nullius in verba', which means something like 'Take no one's word
h
t e awn of clVIlIzatlon and thr hi , ,
, histi ' oug Its imperceptibly slow increase for it.'
In sop isncanon - with m '
any reverses - until a few centuries ago Then However, rebellion against authority cannot by itself be what made
a powerful new mod f di .
late r becarne kn ow e 0 . iscovery and explanation emerged , which
the difference. Authorities have been rejected many times in history, and
revolut' b n.as SCIence.Its emergence is known as the scientific only rarely has any lasting good come of it. The usual sequel has merely
ton, ecause It succeeded aim . di , been that new authorities replaced the old. What was needed for the
ledge at . bl ost imme lately in creating know-
Wh a ndotlcea e rate, which has increased ever since sustained, rapid growth of knowledge was a tradition of criticism. Before
at ha changed? What mad' ffecri . the Enlightenment, that was a very rare sort of tradition: usually the
the physical world h II e,science e ectrve at understanding
people now doing fW etn afi pre.vlOus ways had failed? What were whole point of a tradition was to keep things the same.
, r o erst h time that d h diff ' Thus the Enlightenment was a revolution in how people sought
question began to be k d ,rna e tel terence? ThIS
as e as soon as , b knowledge: by trying not to rely on authority. That is the context in
and there have been rna fli SCIence egan to be successful,
ny con lctmg answe ' , which empiricism - purporting to rely solely on the senses for knowledge
But none, in my view: h h d rs, some contammg truth.
, as reac e the heart f h - played such a salutary historical role, despite being fundamentally
my own answer, I have to' I' lot e matter. To explain
The . ifi grve a Itt e context first false and even authoritative in its conception of how science works.
SClentl c revolution was art of ' , .
the Enlightenment which I b P a WIder mtellectual revolution, One consequence of this tradition of criticism was the emergence of
, a so roughr p , a methodological rule that a scientific theory must be testable (though
moral and political phil h ~ogress rn other fields, especially
U-L' osop y, and m the i , , this was not made explicit at first). That is to say, the theory must make
1uOrtlmately,the term 'the E I' e mstltutlons of society.
hil n Ightenme t' ' d predictions which, if the theory were false, could be contradicted by
P osophers to denote a' n IS use by historians and
violently opposed to ea h vthafletyof different trends, some of them the outcome of some possible observation. Thus, although scientific
COer. What I mea bv i 'II
we go atong. It is one of se 1 n y It WI emerge here as theories are not derived from experience, they can be tested by ex-
ra
and is a theme of this bookveB aspect~ of 'the beginning of infinity' perience - by observation or experiment. For example, before the
Enligh . ut one thmg th 11 ' discovery of radioactivity, chemists had believed (and had verified in
tenment agree on is th ' at a conceptions of the
rebellion against authority in at It ;as a rebellion, and specifically a countless experiments) that transmutation is impossible. Rutherford
Rejecting authority , regar to knowledge. and Soddy boldly conjectured that uranium spontaneously transmutes
f b mregardtok w1d into other elements. Then, by demonstrating the creation of the element
~: stracr analysis. It Was a necessaryno ;, ~e was not just a matter
ore the Enlightenment, it wasgen:;:Ultl~~for progress, because, radium in a sealed container of uranium, they refuted the prevailing
y eyed that everything theory and science progressed. They were able to do that because that
13
earlier theory was testable: it was possible to test for the presence of will later appear on stage unharmed. Those are testable predictions. I
radiu~. I~ contrast, the ancient theory that all matter is composed of may experience many conjuring shows and see my predictions vindi-
combinanon, of the elements earth, air, fire and water was untestable, cated every time. But that does not even address, let alone solve, the
because it did not include any way of testing for the presence of those problem of how the trick works. Solving it requires an explanation: a
components. So it could never be refuted by experiment. Hence it could statement of the reality that accounts for the appearance.
nev~r be - and never was - improved upon through experiment. The Some people may enjoy conjuring tricks without ever wanting to
Enhghtenment was at root a philosophical change. know how they work. Similarly, during the twentieth century, most
, The physicist Galileo Galilei was perhaps the first to understand the philosophers, and many scientists, took the view that science is incapable
:m~ortance of experimental tests (which he called cimenti meaning of discovering anything about reality. Starting from empiricism, they
trials b~ ordea~') as distinct from other forms of experi~ent and drew the inevitable conclusion (which would nevertheless have horrified
obsekrvatlOn,
which can more easily be mistaken for 'reading from the the early empiricists) that science cannot validly do more than predict
B00 of Nature' 't b T '
h ,,' esta I rty ISnow generally accepted as the defining the outcomes of observations, and that it should never purport to
c aractenstlc of the scientific h d P
of d " met 0 • opper called it the 'criterion describe the reality that brings those outcomes about. This is known as
emarcatlon between science and non-science. '
instrumentalism. It denies that what I have been calling 'explanation'
Nevertheless, testability c h b ' ,
ientif '. annor ave een the deCISIve factor in the can exist at all. It is still very influential. In some fields (such as statistical
scienn c revolution either C '
predictions had al b' o~trary to what IS often said, testable analysis) the very word 'explanation' has come to mean prediction, so
n that a mathematical formula is said to 'explain' a set of experimental
thumb for making aWfla!s beled quue common. Every traditional rule of
mt a e or a camp fire i bl data. By 'reality' is meant merely the observed data that the formula is
prophet who lai h re IStesta e. Every would-be
calms t at the sun will
theory. So does every g bl hI go out next Tuesday has a testable supposed to approximate. That leaves no term for assertions about
night - I can feel it' S am e~ WhO~as a hunch that 'this is my lucky reality itself, except perhaps 'useful fiction'.
• 0 w hat ISt e VItal ' Instrumentalism is one of many ways of denying realism, the common-
that is present in scien b b ' progress-enablIng ingredient
ce, ut a sent from th bl h . sense, and true, doctrine that the physical world really exists, and is
prophet and the gambl ~ e testa e t eories of the
er.
accessible to rational inquiry. Once one has denied this, the logical
The reason that testability is not eno '
and cannot be the purp f' ugh IS that prediction is not, implication is that all claims about reality are equivalent to myths,
, ose 0 SCIence C'd ' none of them being better than the others in any objective sense. That
a conjuring trick. The p bl f" onsi er an audience watching
, ro em acing them h h is relativism, the doctrine that statements in a given field cannot be
as a sCientificproblem Alth gh i as muc the same logic
t d ' . ou m nature there i , objectively true or false: at most they can be judged so relative to some
o ecelve us intention 11 ere ISno conjurer trying
, a y, we can be m ifi d ' cultural or other arbitrary standard.
essentIally the same r ysn e In both cases for
th eason: appearances
Instrumentalism, even aside from the philosophical enormity of
e explanation of a co", are not self-explanatory. If
th nJurmg tnck were .d " reducing science to a collection of statements about human experiences,
ere w?uld be no trick. If the ex I ,evI ent m,Its appearance,
were eVIdent in their appearance P a~~tl?nS of phYSIcal phenomena does not make sense in its own terms. For there is no such thing as a
would be no need for science ' emknP1fICIS~ would be true and there purely predictive, explanationless theory. One cannot make even the
The blem : as we ow It simplest prediction without invoking quite a sophisticated explanatory
pro em IS not to p di .
, re ICt the t . k'
lOs~ance,predict that if a con' nc s appearance. I may: for framework. For example, those predictions about conjuring tricks
VarIous th jurer seems to pi' , apply specifically to conjuring tricks. That is explanatory information,
, cups, ose cups will later ace vanous balls under
predict that if the conjurer a . appear to be empty; and I may and it tells me, among other things, not to 'extrapolate' the predictions
ppears to saw sOmeone in half th to another type of situation, however successful they are at predicting
, at person
IS
conjuring tricks. So I know not to predict that saws in general are observation, so conflicting ideas in a broader sense are the occasion
harmless to humans; and I continue to predict that if I were to place for all rational thought and inquiry. For example, if we are simply
a ball under a cup, it really would go there and stay there. curious about something, it means that we believe that our existing
The concept of a conjuring trick, and of the distinction between it ideas do not adequately capture or explain it. So, we have some
and other situations, is familiar and unproblematic - so much so that it criterion that our best existing explanation fails to meet. The criterion
is easy to forget that it depends on substantive explanatory theories and the existing explanation are conflicting ideas. Ishall call a situation
about all sorts of things such as how our senses work how solid matter in which we experience conflicting ideas a problem,
and ,li~t behave, and also subtle cultural details. Kno~ledge that is both The example of a conjuring trick illustrates how observations provide
familiar and uncontroversial is background knowledge. A predictive problems for science - dependent, as always, on prior explanatory
theo~ whose explanatory content consists only of background know- theories. For a conjuring trick is a trick only if it makes us think that
ledge ISa rule of thumb. Because we usually take background knowledge something happened that cannot happen. Both halves of that proposition
fborgrant~d, rules of thumb may seem to be explanationless predictions depend on our bringing quite a rich set of explanatory theories to the
ut that ISalways an illusion. '
experience. That is why a trick that mystifies an adult may be uninterest-
uk ere is always an explanation, whether we know it or not for why ing to a young child who has not yet learned to have the expectations
a ru e of thumb works D . h '
I '. . enymg t at some regularity in nature has an on which the trick relies. Even those members of the audience who are
::y~:gna,tTlohnt~s effectiv~ly ~he ~ame as believing in the supernatural _ incurious about how the trick works can detect that it is a trick only
, a s not conJurmg It'S ct I ", because of the explanatory theories that they brought with them into
an explanation hi' a ua magIc. Also, there is always
n the auditorium. Solving a problem means creating an explanation that
always parochia;t: : ~~ e ~f ~humb fails, for rules of thumb are
Y
stances. So if an e 0 "I' on y in a narrow range of familiar circum- does not have the conflict.
, un fami tar f eature ' d Similarly, no one would have wondered what stars are if there had
and-balls trick the r I f h b' were mtro uced into a cups-
, u e 0 t urn I st t d izh . not been existing expectations - explanations - that unsupported things
prediction. For instance I co a e mig t easily make a false
it would be possl'ble t' fUld not tell from the rule of thumb whether fall, and that lights need fuel, which runs out, and so on, which con-
o per orm the t ick ' h I' flicted with interpretations (which are also explanations) of what was
of balls. If I had an expl ion of h ric Wit Ighted candles instead
ananon 0 ow th tri k k seen, such as that the stars shine constantly and do not fall. In this case
Explanations are also essential "e IC wor ed, I could tell.
first place: Icould not h for arnvmg at a rule of thumb in the it was those interpretations that were false: stars are indeed in free fall
, ave guessed those p d' , and do need fuel. But it took a great deal of conjecture, criticism and
tncks without havin d re IctlOns about conjuring
g a great eal of expl ' testing to discover how that can be.
- even before any specific I ' anatory Information in mind
. , , exp anatlon of h h ' k A problem can also arise purely theoretically, without any obser-
Instance, It IS only in the li t ow t e tnc works. For
abstracted the concept f gh of explanations that I could have vations. For instance, there is a problem when a theory makes a
.k 0 cups and balls f
prediction that we did not expect. Expectations are theories too.
tnc ,rather than sav red d bl rom my experience of the
'" an ue e if ' Similarly, it is a problem when the way things are (according to our
cups were red and the ball bl . ' ven I It so happened that the
'
h a d Witnessed. s ue In every ,
Instance of the trick that I best explanation) is not the way they should be - that is, according to
The essence f " our current criterion of how they should be. This covers the whole
o expenmental testin '
apparently viable theories k bog IS that there are at least two range of ordinary meanings of the word 'problem', from unpleasant,
fl' , nown a utth' ,
~on ICtlng predictions that can be di " e ~ssue In question, making as when the Apollo 13 mission reported, 'Houston, we've had a problem
ust as cOnflicting predictions hlstIngUlshed by the experiment here,' to pleasant, as when Popper wrote:
areteoc'r. .
. caslon lor experiment and
16
Ithinkthat thereisonlyonewayto science- or to philosophy,for that pre-existing knowledge of how objects work and how audiences work,
matter:to meeta problem,to seeits beautyand fall in love with it; to as well as how existing tricks work. So where did the earliest conjuring
getmarriedto it andto livewithit happily,till death do ye part - unless tricks come from? They must have been modifications of ideas that were
youshouldmeetanotherand evenmorefascinatingproblem or unless, not originally conjuring tricks - for instance, ideas for hiding objects
indeed,youshouldobtaina solution.Butevenifyoudo obtain a solution, in earnest. Similarly, where did the first scientific ideas come from?
youmaythendiscover,to your delight,the existenceof a whole family Before there was science there were rules of thumb, and explanatory
of enchanting,thoughperhapsdifficult,problemchildren . , . assumptions, and myths. So there was plenty of raw material for
Realism and the Aim of Science (1983) criticism, conjecture and experiment to work with. Before that, there
Experimental testing involves many prior explanations in addition were our inborn assumptions and expectations: we are born with
to the onesbeingtested,such as theories of how measuring instruments ideas, and with the ability to make progress by changing them. And
work. The refutation of a scientifictheory has, from the point of view there were patterns of cultural behaviour - about which I shall say more
o~someonewho expected it to be true, the same logic as a conjuring in Chapter 15.
trick- the only differencebeingthat a conjurer does not normally have But even testable, explanatory theories cannot be the crucial ingre-
acc~ssto unknown laws of nature to make a trick work. dient that made the difference between no-progress and progress. For
they, too, have always been common. Consider, for example, the ancient
"Sincetheories
, can contra diIC t each other, b ut there are no contra-
dictions in realitv every bl ' 1 h Greek myth for explaining the annual onset of winter. Long ago, Hades,
fla ,'" pro ern signa stat our knowledge must be
wed or Ina~equate. Our misconception could be about the reality god of the underworld, kidnapped and raped Persephone, goddess of
we are observmg or abo t h . spring. Then Persephone's mother, Demeter, goddess of the earth and
"
both. For Instance u ow our perceptIons are related to it or
a " ,k ' agriculture, negotiated a contract for her daughter's release, which
b ' . conJuring tnc presents us with a problem only
ecause we have mlsconcept' b h specified that Persephone would marry Hades and eat a magic seed
whICh Imp
o . I' h Ions a out w at 'must' be happening -
ies t at the kn owe1 d h that would compel her to visit him once a year thereafter. Whenever
were seeingis d f ' 1i ge t at we used to interpret what we
e ectrve, 0 an expert st d. " Persephone was away fulfilling this obligation, Demeter became sad
be obVIOUS . hat i
w at IShappening
eepe In conJurIng lore it may
.f h ' and would command the world to become cold and bleak so that
trick at all but m 1 h d - ~venI t e expert did not observe the
ere year a mlsleadi f' nothing could grow.
who was fooled b ' Thi . ng account 0 It from a person That myth, though comprehensively false, does constitute an ex-
y It. IS IS another 1f b
explanation'if one h' genera act a out scientific planation of seasons: it is a claim about the reality that brings about
. as a mlsconceptio b .
one's expectations m ( n, 0 servatlons that conflict with
ay or may not) s ' our experience of winter. It is also eminently testable: if the cause of
conjectures but no am f b . pur one Into making further winter is Demeter's periodic sadness, then winter must happen every-
,,' ount 0 0 servingwill he mi .
until after one has tho gh f b . correct t e misconception where on Earth at the same time. Therefore, if the ancient Greeks had
'gh , u to a etter Idea' . ,
n t Ideaone can explain th h ' In contrast, If one has the known that a warm growing season occurs in Australia at the very
. th
m e data. Again th
e p enomenon if h
even I t ere are large errors
. , e very term 'data' (" ,. moment when, as they believed, Demeter is at her saddest, they could
mgthe 'data', or reJ'ecti givens) ISmisleading. Amend-
' , ng some as erroneo . f have inferred that there was something wrong with their explanation
f
o SClehtlficdiscoverv a d h ' us, ISa requenr concomitant
'I <r» n t e crucial 'd t ' of seasons.
unn theory tells us what t 1 k 1: a a cannot even be obtained
A 0 00 ror and h d Yet, when myths were altered or superseded by other myths over the
new conjuring trick . ow an why.
Lik . ISnever totall 1 '" course of centuries, the new ones were almost never any closer to
I e a new SCientificth ". y unre ated to exrsnng tricks.
re' eory, It IS formed b . the truth. Why? Consider the role that the specific elements of the
arrangmg and combining th ide fr y creatively modifying, Persephone myth play in the explanation. For example, the gods
e I eas om " .
eXisting tricks. It requires
.:118
19
provide the power to affect a large-scale phenomenon (Demeter to author could not have denied that the role of all the details could be
command the weather, and Hades and his magic seed to command played equally well by countless other things.
Persephone and hence to affect Demeter). But why those gods and not The Persephone and Freyr myths assert radically incompatible things
others?
f In Nordic myth 0Iogy, seasons are caused by the changing about what is happening in reality to cause seasons. Yet no one, Iguess,
~rtu~s o~ Freyr, the god of spring, in his eternal war with the forces has ever adopted either myth as a result of comparing it on its merits
o hCOh a~ lda.rkn~ss: Whenever Freyr is winning, the Earth is warm; with the other, because there is no way of distinguishing between them.
w en e IS osmg, It IScold. If we ignore all the parts of both myths whose role could be easily
That myth accounts for the se b replaced, we are left with the same core explanation in both cases: the
myth. It is sli htl b ~s~ns a out as well as the Persephone
gods did it. Although Freyr is a very different god of spring from
worse at exprain! ettherat elxp~ammg the randomness of weather, but
mg t e regu anty of se asons, b ecause real wars do not Persephone, and his battles very different events from her conjugal
ebb d fl
an ow so regularly (except . f h visits, none of those differing attributes has any function in the myths'
themselves) In the P h mso ar as t at is due to seasons
. ersep one myth the I f h respective accounts of why seasons happen. Hence none of them
and the maei d. . ,ro e 0 t e marriage contract
gic see ISto explain that r I' B " . provides any reason for choosing one explanation over the other.
a magic seed and not h ki egu anty, ut why IS It specifically
. . any ot er ind of rna . ;l Wh . . . The reason those myths are so easily variable is that their details are
VISitscontract and not h gIC. Y IS It a conjugal- barely connected to the details of the phenomena. Nothing in the
. some ot er reason f
action annually! For' here i or someone to repeat an problem of why winter happens is addressed by postulating specifically
. mstance ere IS a' I
the facts just as well. Pe hoi vananr exp anation that fits
. . rsep one was not rid h a marriage contract or a magic seed, or the gods Persephone, Hades
year in spring when h e ease - s e escaped. Each
, er powers are at th . h igh and Demeter - or Freyr. Whenever a wide range of variant theories can
on Hades by raiding the d I elf ei t, she takes revenge
account equally well for the phenomenon they are trying to explain,
spring air. The hot air t~n ed~orl d and cooling all the caverns with
. us ISp aced rise' h there is no reason to prefer one of them over the others, so advocating
causmg summer: D s mto t e human world
. . emeter celebrates Ph' a particular one in preference to the others is irrational.
annIversary of her esca b ersep one's revenge and the
th E h pe y commanding I That freedom to make drastic changes in those mythical explanations
e art. This myth ace f p ants to grow and adorn
ad' . ounts or the same b . of seasons is the fundamental flaw in them. It is the reason that myth-
n It IStestable (and' f 0 servatl0ns as the original
wh . m act refuted) by th .' making in general is not an effective way to understand the world. And
at It asserts about reality] e same observatIons Yet
that is so whether the myths are testable or not, for whenever it is easy
it is the opposite of _ th 1 ~s~rkedly different from - in many ways
E h e original myth to vary an explanation without changing its predictions, one could just
very ot er detail of th . as easily vary it to make different predictions if they were needed. For
wint h e story apart fro . b
er appens once a year,' . ' m ItS are prediction that
example, if the ancient Greeks had discovered that the seasons in the
myth was created to explain :~~:t as ea~il! variable. So, although the
northern and southern hemispheres are out of phase, they would have
to that purpose Wh' asons, It ISonly superficiall d d
m k . en ItS author was deri ya apte had a choice of countless slight variants of the myth that would be
I a e a goddess do something won ermg what could possibly
consistent with that observation. One would be that when Demeter is
t must have been a marriage once a year, he did not shout 'Eureka!
made that h . contract enfor db' sad she banishes warmth from her vicinity, and it has to go elsewhere
c oice - and all hi ce y a magic seed' He
cultural and '. IS substantiv hoi . _ into the southern hemisphere. Similarly, slight variants of the Perse-
artIstICreasons and be e c oices as author - for
at a II. He rna I h ,not cause of th phone explanation could account just as well for seasons that were
ya so ave been try' e attributes of winter
metaphorically _ but he I mg to explain aspects of hu marked by green rainbows, or seasons that happened once a week, or
capacity re am COnce ed . man nature sporadically, or not at all. Likewise for the superstitious gambler or
as an explanation of seas . rn w~th the myth only in its
the end-of-the-world prophet: when their theory is refuted by experience,
OtIS, and in that respect even its
:u
they do indeed switch to a new one; but, because their underlying Somewhat less trivially, it leads to the rejection of authority, because
e~planationsa~ebad, they can easily accommodate the new experience if we adopt a theory on authority, that means that we would also have
without changing the substance of the explanation. Without a good accepted a range of different theories on authority. And hence it also
explanatory theory,they can simply reinterpret the omens, pick a new implies the need for a tradition of criticism. It also implies a meth-
dat~,and make essentiallythe same prediction. In such cases, testing odological rule - a criterion for reality - namely that we should conclude
one s theory and abandoning it when it is refuted constitutes no that a particular thing is real if and only if it figures in our best
pr~gress towards understanding the world. If an explanation could explanation of something,
easily explain anything' th ' fi ld '
hi In e given e , then It actually explains Although the pioneers of the Enlightenment and of the scientific
not mg.
revolution did not put it this way, seeking good explanations was (and
In ~beeneral,when theories are easily variable in the sense I have remains) the spirit of the age. This is how they began to think. It is
descn d, experimental t ti ' 1m
es ing ISa ost useless for correcting their what they began to do, systematically for the first time. It is what made
errors. I ca11such theories b d l '
, a exp anations. Being proved wrong by that momentous difference to the rate of progress of all kinds.
experiment,and changingth theori h
h ' e eones to ot er bad explanations does Long before the Enlightenment, there were individuals who sought
not get t elf holders one jot closer to the truth. ' good explanations. Indeed, my discussion here suggests that all progress
Becauseexplanation pla S hi
testabilityis of littl 'Yh t IScentral role in science, and because then, as now, was due to such people. But in most ages they lacked
e use m t e case of bad I ' contact with a tradition of criticism in which others could carryon their
to callmyths sup " exp ananons, I myself prefer
, erstltlons and similar theori , ideas, and so created little that left any trace for us to detect. We do
theymaketestablep d' , , eones unscientific even when
re icnons But It do know of sporadic traditions of good-explanation-seeking in narrowly
you use so long as it d' es not matter what terminology
, I oes not lead y I ' defined fields, such as geometry, and even short-lived traditions of
somethingwonhwh'l b ou to cone ude that there IS
I e a out the Perseph h criticism - mini-enlightenments - which were tragically snuffed out, as
apocalyptictheory or th bl one myt ,or the prophet's
N ' e gam er's delusi iust b ' , I shall describe in Chapter 9. But the sea change in the values and
or ISa person capabl f ki on, Just ecause ISIt testable.
willingto drop a th e 0 hen ,In~ progress merely by virtue of being patterns of thinking of a whole community of thinkers, which brought
eory w en It ISrefur d. about a sustained and accelerating creation of knowledge, happened
a better explanation of th Ie. one must also be seeking
frame of mind. e re evant phenomena. That is the scientific only once in history, with the Enlightenment and its scientific revolution.
An entire political, moral, economic and intellectual culture - roughly
As the phYsicistRichard Fe
learnedabout how to k f ynman said, 'Science is what we have what is now called 'the West' - grew around the values entailed by the
, bl eep rom fooling I quest for good explanations, such as tolerance of dissent, openness to
vana e explanations th bl ourse ves,' By adopting easily
will be bl ' e gam er and pro h change, distrust of dogmatism and authority, and the aspiration to
a e to continue fooHn th p et are ensuring that they
!ust as thoroughly as if the ha~ emselves no matter what happens. progress both by individuals and for the culture as a whole. And the
msulating themselves fro~ faci:dop~ed untestable theories, they are progress made by that multifaceted culture, in turn, promoted those
about what is really there in th ~ e~ldence that they are mistaken values - though, as I shall explairt in Chapter 15, they are nowhere close
The quest for go d I ~ P YSlcalWorld. to being fully implemented.
pr' 'I 0 exp anatlons is I b li
Now consider the true explanation of seasons. It is that the Earth's
th tncip e not only of science,but of t' e!eve, the basic regulating
~ feature that distinguishesth he EnlIghtenment generally. It is axis of rotation is tilted relative to the plane of its orbit around the
~h ers, ~nd it impliesall those o~~:appro~c~es to knowledge from all sun. Hence for half of each year the northern hemisphere is tilted
avediscussed:It triviallyimpl' rhcondltlons for scientific progress towards the sun while the southern hemisphere is tilted away, and for
lest atp di , the other half it is the other way around. Whenever the sun's rays are
re Ictlon alone is insufficient.
l:
l
;"
::
(
~ing vertically in one hemisphere (thus providing more heat per theory would have been refuted, just as, in the event, the Persephone
uOlt~~a of the surface) they are falling obliquely in the other (thus and Freyr myths were refuted by the opposite observation. But the
proVIdingless). difference is, if the axis-tilt theory had been refuted, its defenders would
have had nowhere to go. No easily implemented change could make
tilted axes cause the same seasons all over the planet. Fundamentally
new ideas would have been needed. That is what makes good ex-
planations essential to science: it is only when a theory is a good
explanation - hard to vary - that it even matters whether it is testable.
Bad explanations are equally useless whether they are testable or not.
Most accounts of the differences between myth and science make
too much of the issue of testability - as if the ancient Greeks' great
mistake was that they did not send expeditions to the southern hemi-
sphere to observe the seasons. But in fact they could never have guessed
The .trueexpl
. ' 0f seasons (not to scalel)
anatlon that such an expedition might provide evidence about seasons unless
they had already guessed that seasons would be out of phase in the
! j That is a good explanation _ hard two hemispheres - and if that guess was hard to vary, which it could
a functionalrole F .' to vary, because all its details play
I1 . or Instance we kn d have been only if it had been part of a good explanation. If their guess
of our experience f '. ow- an can test independently
o seasons - that surf til d was easy to vary, they might just as well have saved themselves the
1 heat acebeated lessth.. h . aces te away from radiant
I .sphereins~ e,t\n
anw en they a faci .
in.aa constant
. -. . dir . re. 'acing It, and that a spinning
boat fare, stayed at home, and tested the easily testable theory that
• ·~·t"",~m
winter can be staved off by yodelling.
In terDlaof~ies .. £ .......•.•.•.• '. •.•. ectlon.And we can explain why,
tilt appears..inoQt.e;lageO~. ~£~ heat. and mechanics. Also, the same So long as they had no better explanation than the Persephone myth,
hOI'lZo . '.' .,~t". natIon 0 .where th there should have been no need for testing. Had they been seeking good
.. nat.different """'Aft f . . .'. .'. e sun appears relative to the
the..co ldness'Of.theW~.' ~-oyear,
'..1.d.." '. d..' .
In the Per h "
. sep one myth, 10 contrast,
explanations, they would immediately have tried to improve upon the
d 1.1U IS cause by D ' myth, without testing it. That is what we do today. We do not test every
• o DOtgenerallycool th . . emeters sadness - but people
ha Veno way ollmo. .". eu.:h'surround' h
.... lOgs..W en they are sad and we
testable theory, but only the few that we find are good explanations.
th •'.' wIDgt at Derner .'d ' Science would be impossible if it were not for the fact that the over-
the. ewodd'c>ther ·.~.·th ..
'•.. .. ..'. ~~m$etoWInt~itse1f 0 f er ".sa ,or th.
a t she ever cools
. whelming majority of false theories can be rejected out of hand without
.: ·mo.on'~~~~,tlle~s ..tilt . ••... ." . necould not substitute
JIlOGnInthe ijqrdc>esnbtr ' '." 8tQry,becaU$e the position of the any experiment, simply for being bad explanations.
sun's rays hea,tlng tl1eEatthepe~tltselfOJ1ce a Year,and because the Good explanations are often strikingly simple or elegant - as Ishall
one casU' . -'. . .. ~eIDtegral to th I . discuss in Chapter 14. Also, a common way in which an explanation
all thiY lDCO~rate any Stories ab<>uthoe exp anatIon. Nor could
s, because If the true e I ".. w the sun god feels about can be bad is by containing superfluous features or arbitrariness, and
the Ea _1. - xp anatIon of w'. '. sometimes removing those yields a good explanation. This has given
. &UI-lJun·rnotionthen h lOter IS In the geometry of
and If the "" . ow anyone feel b " rise to a misconception known as 'Occam's razor' (named after the
ho re were some flaw in that e '. .'. s a out It IS irrelevant,
fourteenth-century philosopher William of Occam, but dating back
....
_....
::..a_.n
...y~ne.fel.•.tWoul.·d.'.•..,.u. tit ".nl.. .'.iXplanatlQn~ then no story about
' .•.• 4"~
' ....•... ~.. '.'
t~Y"'1~n flaut.
.t.1-...-.. ' '...•....•....
;__ ... _~ to antiquity), namely that one should always seek the 'simplest ex-
'Ml·.... twoh.o....,.~..... .' '.'". ·····~ ~~"J;lH:fhe .... planation'. One statement of it is 'Do not multiply assumptions beyond
" .....
··~_~ifi~l\~ ....•..
G.. "... "
~IS~be.OUt Qf phase
necessity.' However, there are plenty of very simple explanations that
"~~~Qbe'lAPhase, the
are neverthelesseasily variable (such as 'Demeter did it'). And, while might be, say, Aristarchus of Samos, who gave the earliest known
assumptions'beyond necessity' make a theory bad by definition, there arguments for the heliocentric theory in the third century BeE.
have been many mistaken ideas of what is 'necessary' in a theory. Although you know that the Earth is a sphere, you possess no
Instrumentalism,for instance,considers explanation itself unnecessary, evidence about any location on Earth south of Ethiopia or north of
and so do many other bad philosophies of science as I shall discuss the Shetland Islands. You do not know that there is an Atlantic or a
in Chapter 12. '
Pacific ocean; to you, the known world consists of Europe, North Africa
When a formerly good explanation has been falsified by new and parts of Asia, and the coastal waters nearby. Nevertheless, from
observations, it is no longer a good explanation, because the problem the axis-tilt theory of seasons, you can make predictions about the
h~s e~panded to include those observations. Thus the standard weather in the literally unheard-of places beyond your known world.
sCient~ficmethodology of dropping theories when refuted by experi- Some of these predictions are mundane and could be mistaken for
:ent is i~plied by the requirement for good explanations. The best induction: you predict that due east or west, however far you travel,
1 tan~tions ~rethe ones that are most constrained by existing know- you will experience seasons at about the same time of year (though the
e
e gh - mdudmg other good explanations as well as other knowledge timings of sunrise and sunset will gradually shift with longitude). But
f h
o t e p enomena to be e lai d Th '
th h ,xp ained, at is why testable explanations you will also make some counter-intuitive predictions: if you travel
at ave passed stnngent test b only a little further north than the Shetlands, you will reach a frozen
whi h i , s ecome extremely good explanations
rc is m tum why th im of '
of kid ' , e maXim 0 testability promotes the growth region where each day and each night last six months; if you travel
now e ge m SCience.
further south than Ethiopia, you will first reach a place where there
Conjecturesare the products f " ,
with imao1nation' that i 0 creattvetmagInation. But the problem
are no seasons, and then, still further south, you will reach a place
e- is at It can create fi ' where there are seasons, but they are perfectly out of phase with those
truth, As I have sugge t d hi' cnon much more easily than
explain expen'e ,s e, stoncally, virtually all human attempts to everywhere in your known world. You have never travelled more than
, nce m terms ofa id li ' a few hundred kilometres from your home island in the Mediterranean.
m the form of myths d Wi er rea Ity have indeed been fiction,
, ogma and mistak You have never experienced any seasons other than Mediterranean
rule of testability is an ' ffici en common sense - and the
msut cienr check h mi ones. You have never read, nor heard tell, of seasons that were out of
quest for good explanati'o d he i on sue mistakes. But the
ns oes t e Job. ' inaf phase with the ones you have experienced. But you know about them.
and therefore they a . mventmg alsehoods is easy,
, re easy to vary 0 f d ' What if you'd rather not know? You may not like these predictions.
explanations is hard but th h d nee oun ; dIscovering good
, e ar er they fi Your friends and colleagues may ridicule them. You may try to modify
are to vary once found Th id I are to nd, the harder they
, , . e I ea that ex I ' the explanation so that it will not make them, without spoiling its
is rucelydescribedby th ' P anatory SCiencestrives for
hi h e quotatIon from Wh 1 ' agreement with observations and with other ideas for which you have
t is c apter: 'Behindit all' 1 ee er With which I began
iS Sure y an id '
when we grasp it _ in add i ea so Simple,so beautiful that no good alternatives. You will fail. That is what a good explanation
II eca e a centu ' ' will do for you: it makes it harder for you to fool yourself.
a say to each other, how ld' ry, or a millennium - we will
N cou it h b For instance, it may occur to you to modify your theory as follows:
ow we shall see how thi 1 a,ve een otherwise? [my italics].'
answe th s exp anatton-based . 'In the known world, the seasons happen at the times of year predicted
rs e question that I ask d b conCeption of science
about unfamili~r aspeets of real~ ; ove: how do we know so much by the axis-tilt theory; everywhere else on Earth, they also happen at
Put YOurselfin the place f ty , those times of year.' This theory correctly predicts all evidence known
the axis-tilt explanation of se an anCient astronomer thinking about to you. And it is just as testable as your real theory. But now, in order
assume th. asons. For the k f' to deny what the axis-tilt theory predicts in the faraway places, you
at you have also adopted the helisa eo. Simplicity, let us
ocentnc theory. So you have had to deny what it says about reality, everywhere. The modified
theory is no longer an explanation of seasons, just a (purported) rule creators is precisely that it does not have to be extrapolated. By its
of thumb. So denyingthat the original explanation describes the true nature as an explanation, when its creators first thought of it, it already
cause of seasons in the places about which you have no evidence applied in our planet's other hemisphere, and throughout the solar
has forced you to deny that it describes the true cause even on your system, and in other solar systems, and at other times.
home island. Thus the reach of an explanation is neither an additional assumption
Supposefor the sake of argument that you thought of the axis-tilt nor a detachable one. It is determined by the content of the explanation
theory y~~self. It is your conjecture, your own original creation. Yet itself. The better an explanation is, the more rigidly its reach is determined
because
, It ISa good explana t'Ion - h ar d to vary - It
' IS
, not yours to - because the harder it is to vary an explanation, the harder it is in
modify:It ~~s an autonomous meaning and an autonomous domain particular to construct a variant with a different reach, whether larger
o~ap~hcabdIty.Youcannot confine its predictions to a region of your or smaller, that is still an explanation. We expect the law of gravity to
~ °th°skn1Og.Whetheryou like it or not, it makes predictions about places be the same on Mars as on Earth because only one viable explanation
o own to you and unkn di , of gravity is known - Einstein's general theory of relativity - and that
tho ghr f d own to you, pre tenons that you have
"I u 0bian ones that you h ave not t h ou gh't of. TIlted planets m
SImIar or Its 10 other sol
, is a universal theory; but we do not expect the map of Mars to resemble
c li I ' ar systems must have seasonal heating and the map of Earth, because our theories about how Earth looks, despite
the being excellent explanations, have no reach to the appearance of any
n~:e~~e:~e::::: l:h most distant galaxies, and planets that we shall
have yet to form rZ w~redestroyed aeons ago, and also planets that other astronomical object. Always, it is explanatory theories that tell
us which (usually few) aspects of one situation can be 'extrapolated'
originsinsideone'brai: :h:~~ reaches out, as it were, from its finite
evidencefrom a small art f as been ~ffectedonly by scraps of patchy to others.
This reach of expla p , 0 ~ne hemisphere of one planet - to infinity. It also makes sense to speak of the reach of non-explanatory forms
nanon, IS anothe ' f of knowledge - rules of thumb, and also knowledge that is implicit in
of infinity'.It is the abili f r meanmg 0 'the beginning
those that they were cr ltyt°d someIof them to solve problems beyond the genes for biological adaptations. So, as I said, my rule of thumb
ea e to so ve about cups-and-balls tricks has reach to a certain class of tricks; but I
The axis-tilt theory is an exam I'. ' ,,
explain the changes' th ' p e. It was ongmally proposed to could not know what that class is without the explanation for why the
'
ComblOedwith 10 e sun s angle of I' , rule works.
a littl kn e evanon during each year.
' e owledge of heat d ' , Old ways of thought, which did not seek good explanations, permitted
expIamed seasons And 'h an spmnmg bodies, it then
' " WIt out any furth di no process such as science for correcting errors and misconceptions.
expIamed why seasons er mo ification it also
h' are out of phase' h " Improvements happened so rarely that most people never experienced
w Y tropical regionsdo not h th 10 t e two heffilspheres, and
at midnight in polar reoion ave em, and why the summer sun shines one. Ideas were static for long periods. Being bad explanations, even
rna II h been unaw s - t hree phenom ena 0f w hiich Its
y we ave
l)& , creators the best of them typically had little reach and were therefore brittle and
are.
Thereac h of an I ' unreliable beyond, and often within, their traditional applications. When
so ethin exp anatlon is not a' rin 'I ' ideas did change, it was seldom for the better, and when it did happen
, m, g that the creator of the e I P, Clpe of mduction'; it is not
It. It ISnot Part of th ' xp anatlon can use to obr ' , ify to be for the better, that seldom increased their reach. The emergence
a fter we h th
e creatIve proc
ess
II
at a WI fi d
am or JUStl
ave e explanation " e n out about it only of science, and more broadly what I am calling the Enlightenment, was
to do with' - SOmetImeslong ft S ' the beginning of the end of such static, parochial systems of ideas. It
eXtrapolation' or" duen a er, 0 It has nothing
anyoth II ,10 Uetlon' or with 'd .. initiated the present era in human history, unique for its sustained, rapid
er a eged way.It is exactly th oth I envmg' a theory in
the explanati e . er wa d creation of knowledge with ever-increasing reach. Many have wondered
on of seasons reach £ .. y toun : the reason that
es ar Outside the experience of its how long this can continue. Is it inherently bounded? Or is this the
beginning of infinity - that is to say, do these methods have unlimited more often a theory is confirmed by observation the more likely it
potential to create further knowledge? It may seem paradoxical to claim becomes.
anything so grand (even if only potentially) on behalf of a project that Induction The non-existent process of 'obtaining' referred to above,
has swept away all the ancient myths that used to assign human beings Principle of induction The idea that 'the future will resemble the past',
a special sign~cance in the scheme of things. For if the power of the combined with the misconception that this asserts anything about
~uman faculties of reason and creativity, which have driven the En- the future.
h~te~ent, were indeed unlimited, would humans not have just such Realism The idea that the physical world exists in reality, and that
a significance? knowledge of it can exist too.
And yet, as I mentioned at the beginning of this chapter, gold can Relativism The misconception that statements cannot be objectively
be created only by stars and by intelligent beings. If you find a nugget true or false, but can be judged only relative to some cultural or
of gold anywhere in th ' other arbitrary standard.
' e umverse, you can be sure that in its history
th ere was either a supernov . II' be' .
' a or an mte igent mg With an explanation. Instrumentalism The misconception that science cannot describe
And If you find an explanaf i the universe you know
Ion anyw hereere in reality, only predict outcomes of observations.
t h at th ere must have be 'II" ' Justificationism The misconception that knowledge can be genuine
ld en an mte igent bemg. A supernova alone
wou not suffice.
or reliable only if it is justified by some source or criterion.
But - so what? Gold is i . Fallibilism The recognition that there are no authoritative sources of
of things ir h I' I ' , mportant to. us, but m the cosmic scheme
as Itt e Significance E I ' knowledge, nor any reliable means of justifying knowledge as true
we n d h . . xp anations are important to us:
ee t em to survive But' h hi or probable.
scheme of thing bout ISt ere anyt mg significant, in the cosmic
s, a out explanatio th . Background knowledge Familiar and currently uncontroversial
process that happ insid b ' n, at apparently puny physical
ens mSI e rains) I h II dd knowledge.
Chapter 3, after so flecn . s a a ress that question in
me re ecnons about appearance and reality. Rule of thumb 'Purely predictive theory' (theory whose explanatory
content is all background knowledge).
Problem A problem exists when a conflict between ideas is ex-
TERMINOLOGY
perienced.
Explanation Statement about h ' h
Good/bad explanation An explanation that is hard/easy to vary while
and why. w at IS t ere, what it does, and how
still accounting for what it purports to account for.
Reach The ability of some ex I '
The Enlightenment (The beginning of) a way of pursuing knowledge
those that they were crear d p analnons to solve problems beyond
Creativity Th ' e to so ve. with a tradition of criticism and seeking good explanations instead
e capacity to crear of reliance on authority.
Empiricism The' ,e new explanations.
fr misconception that w 'd ' , Mini-enlightenment A short-lived tradition of criticism.
om sensory experience. e enve all our knowledge
Theory-laden Th ' Rational Attempting to solve problems by seeking good explanations;
, ere IS no such thin ' , actively pursuing error-correction by creating criticisms of both
experIence of the world g as raw experience. All our
un' comes through I f existing ideas and new proposals.
conSCIOUS interpretatio ayers 0 conscious and
Inductivism Th' n. The West The political, moral, economic and intellectual culture that
b " e misconception that scie ifi ' has been growing around the Enlightenment values of science, reason
y generalizing or extra I ' nn c theones are obtained
po attng repeated experiences, and that the and freedom.
THE BEGINNING OF INFINITY THE REA CH OF EXPLANATIONS
MEANINGS OF 'THE BEGINNING OF INFINITY

, h . but mstea
. d set up a tradition of criticism.
justifications for t eones, reach. they explain more
. .d have enormous. .
ENCOUNTERED IN THIS CHAPTER Some of the resulting leas . d The reach of an explanation
.. 11y designe to. .
- The fact that some explanations have reach. than what they were origma . that we make about It
is an intrinsic attribute of it, not a~ assumption
- The universal reach of some explanations. d i d tivism claim. .
- The Enlightenment. as empiricism an m uc ranee and reality, explanation
Now I'll say some more about appea
- A tradition of criticism.
and infinity.
- ConJ'eeture: the origin of all knowledge.
- The discovery of how to make progress: science, ..th e scien of
. tifi.c r evo-
lution, seeking good explanations, and the political principles
the West.
- Fallibilism.
SUMMARY
Appearances are deceptive. Yet we have a great deal of knowledge

about the vast and unfamiliar reality that causes them, and of the
elegant, universal laws that govern that reality. This knowledge consists
of explanations: assertions about what is out there beyond the appear-
ances, and how it behaves. For most of the history of our species, we
had almost no success in creating such knowledge. Where does it come
from? Empiricism said that we derive it from sensory experience. This
is false. The real source of our theories is conjecture, and the real source
of Our knowledge is conjecture alternating with criticism. We create
theories by rearranging, combining, altering and adding to existing
ideas with the intention of improVing upon them. The role of experiment
and observation is to choose between existing theories, not to be the .
source of new ones. We interpret experiences through explanatory
theories, but true explanations are not obvious. Fallibilism entails not
looking to authorities hut iustead acknOWledging that we may always
be mistaken, and trying to COrrect errors. We do SO hy seeking good
expla~ations - explanations that are hard to vary in the sense that
cbang,ng the details would ruin the explanatiun. This, not experimental
testing, was the decisive factor in the SCientific revolution and also
in the ~qu e, rapid, sustained PlOgress in other fields that have partici-
pated 10 the Enlightenment. That Was a rebellion against authority
wlucb, unhke most such rebellions, tried nor to seek authoritative

The Beginning of Infinity - by David Deutsh

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Beginning of Infinity - by David Deutsh

Uploaded by

Copyright:

Available Formats

DAVID DEUTSCH

Penguin Books Ltd, Registered Offices: I. The Reach of Explanations I

Copyright C David Deutsch, zan

happen, entailsa journey through virtually every fundamental field of

Behind it all is surely an idea so simple, so beautiful, that when

main useis to choosebetw theori IC t eones are denved. Its

a icte the opposing

MEANINGS OF 'THE BEGINNING OF INFINITY

Appearances are deceptive. Yet we have a great deal of knowledge

You might also like