The Big Picture
Page 7
25
quantum state,” also known as “the wave function,” which we can use to
26
calculate the outcomes of experiments that observe the system.
27
Quantum mechanics has supplanted classical mechanics as the best way
28
we know to talk about the universe at a deep level. Unfortunately, and to
29
the chagrin of physicists everywhere, we don’t fully understand what the
30
theory actually is. We know that the quantum state of a system, left alone,
31
evolves in a perfectly deterministic fashion, free even of the rare but annoy-
32
ing examples of non- determinism that we can find in classical mechanics.
33
But when we observe a system, it seems to behave randomly, rather than
34
deterministically. The wave function “collapses,” and we can state with very
S35
N36
35
Big Picture - UK final proofs.indd 35
20/07/2016 10:02:39
T H E B IG PIC T U R E
01
high precision the relative probability of observing different outcomes, but
02
never know precisely which one it will be.
03
There are several competing approaches as to how to best understand
04
the measurement problem in quantum mechanics. Some involve true ran-
05
domness, while others (such as my favorite, the Everett or Many- Worlds
06
formulation) retain complete determinism. We’ll talk about the alterna-
07
tives in chapter 21. All of the popular versions of quantum mechanics, how-
08
ever, maintain the underlying philosophy of Laplace’s analysis, even if they
09
do away with perfect predictability: what matters, in predicting what will
10
happen next, is the current state of the universe. Not a goal in the future,
11
nor any memory of where the system has been. As far as our best current
12
physics is concerned, each moment in the progression of time follows from
13
the previous moment according to clear, impersonal, quantitative rules.
14
•
15
16
There is a bit of a mismatch between Laplace’s notion of determinism and
17
what most people think of when they hear “the future is determined.” The
18
latter phrase conjures up images of destiny or fate— the idea that what will 19
eventually happen has “already been decided,” with the implication that it’s
20
been decided by someone, or something.
21
The physical notion of determinism is different from destiny or fate in
22
a subtle but crucial way: because Laplace’s Demon doesn’t actually exist, the
23
future may be determined by the present, but literally nobody knows what
24
it will be. When we think of destiny, we think of something like the Three
25
Fates of Greek mythology or the Weird Sisters of Shakespeare’s Macbeth,
26
wizened oracles who will use riddles to indicate our future path, which we
27
will try to escape from and fail. The real universe is nothing like that. It’s
28
more like an annoying child who likes to approach people and say, “I know
29
what’s going to happen to you next!” Then, when you ask what will happen,
30
the child says, “I can’t tell you.” And after it happens, they say, “See? I knew
31
that was going to happen!” That’s the universe for you.
32
The momentary or Laplacian nature of physical evolution doesn’t have
33
much relevance for the choices we face in our everyday lives. For poetic
34
naturalism, the situation is clear. There is one way of talking about the uni-
35S
verse that describes it as elementary particles or quantum states, in which
36N
Laplace holds sway and what happens next depends only on the state of the
36
Big Picture - UK final proofs.indd 36
20/07/2016 10:02:39
W h At dE t E R M I n E S W h A t W I l l h A PPE n ?
system right now. There is also another way of talking about it, where we
01
zoom out a bit and introduce categories like “people” and “choices.” Unlike
02
our best theory of planets or pendulums, our best theories of human behav-
03
ior are not deterministic. We don’t know any way to predict what a person
04
will do based on what we can readily observe about their current state.
05
Whether we think of human behavior as determined depends on what
06
we know.
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
S35
N36
37
Big Picture - UK final proofs.indd 37
20/07/2016 10:02:39
01
02
03
5
04
05
Reasons Why
06
07
08
09
10
11
12
13
14
15
16
17
In November 2003, Dutch pediatric nurse Lucia de Berk was sentenced
to life imprisonment without parole, for the murder of four children
under her care and the attempted murder of three others. Her case
became a media sensation for an unusual reason: it involved the misuse of
18
statistical reasoning.
19
Some direct evidence was brought against de Berk, but it was flimsy. In
20
one case, for example, the victim (“baby Amber”) was alleged to have been
21
poisoned by the drug digoxin, but doctors pointed out that similar chemi-
22
cal signals could have arisen naturally. The crucial part of the case against
23
de Berk wasn’t any incontrovertible evidence of individual murders, but
24
rather the supposed statistical unlikelihood of so many deaths occur-
25
ring while a single nurse was on duty. One expert testified that there was
26
less than 1 chance in 342 million of such a coincidence. The prosecution
27
argued, successfully, that the improbability implied by this calculation
28
meant that a lower burden of proof should be used when evaluating the
29
30
single incident.
31
The problem was that the calculation was entirely bogus. It was plagued
32
by elementary mistakes, from multiplying probabilities that weren’t inde-
33
pendent to “fishing” for seeming coincidences in large numbers of events.
34
After the conviction, other experts put forward alternative calculations,
35S
ranging from 1 in 1 million to 1 in 25, depending on precisely how the ques-
36N
tions were asked. Further investigation showed that the infant mortality
38
Big Picture - UK final proofs.indd 38
20/07/2016 10:02:39
R E A S O n S W h y
rate at the hospital had been higher in the years before de Berk had been
01
hired than it became once she started working there, not really the effect
02
one would expect the presence of a serial killer to have. Ultimately, doubts
03
about both the statistical arguments and the direct evidence led to a retrial.
04
In 2010, de Berk was fully acquitted of all charges.
05
But math mistakes alone are not sufficient to account for Lucia de Berk’s
06
wrongful conviction. What started the ball rolling was a psychological con-
07
viction: the idea that something as horrible as these infant deaths couldn’t
08
just be random; someone must be to blame. There must be a reason why it
09
happened. As horrible as the death of a child necessarily is, it becomes more
10
sensible to us if it can somehow be explained as the result of someone’s ac-
11
tions, rather than simply random chance.
12
Looking for causes and reasons is a deeply ingrained human impulse.
13
We are pattern- recognizing creatures, quick to see faces in craters on Mars
14
or connections between the location of Venus in the sky and the state of our
15
love life. Not only do we seek order and causation, but we favor fairness as
16
well. In the 1960s, psychologist Melvin Lerner proposed the “Just World
17
Fallacy” after noticing people’s tendency to blame victims of misfortune
18
when something went wrong. To test his idea, he and his collaborator Car-
19
olyn Simmons conducted experiments in which subjects were shown other
20
people apparently suffering the effects of electrical shocks. Afterward, many
21
of the subjects— who knew nothing about the people supposedly being
22
shocked— passed harsh judgments against them, berating their character.
23
The more violent the shocks appeared to be, the harder the subjects were on
24
the victims.
25
26
•
27
Searching for reasons why things happen is by no means an irrational pur-
28
suit. In many familiar contexts, things don’t “just happen.” If you are sitting
29
in your living room and a baseball suddenly crashes through your window,
30
it makes sense to look outside and expect to see some kids at play. Giant
31
whales do not spontaneously come into existence several miles in the air.
32
Our familiar intuitions concerning cause and effect have developed over
33
evolutionary time because they provide useful guides for understanding
34
how the world really works.
S35
The mistake is to elevate this expectation to an unbreakable principle.
N36
39
Big Picture - UK final proofs.indd 39
20/07/2016 10:02:39
T H E B IG PIC T U R E
01
We see things happen, and we attribute reasons to them. Not only with
02
events at home and people’s personal fates but all the way down to the ba-
03
sics of ontology. If the world consists of certain things and behaves in cer-
04
tain ways, we think, there must be a reason why it is so.
05
This mistake has a name: the Principle of Sufficient Reason. The term was
06
coined by German philosopher and mathematician Gottfried Leibniz, but
07
the essential idea had been anticipated by many earlier thinkers, most no-
08
tably by Baruch Spinoza in the seventeenth century. One way of stating it
09
would be:
10
11
Principle of Sufficient Reason:
12
For any true fact, there is a reason why it is so, and why some-
13
thing else is not so instead.
14
15
Leibniz once formulated it simply as “Nothing happens without a rea-
16
son,” which is remarkably close to the maxim “Everything happens for a
17
reason,” which you can buy on T-shirts and bumper stickers today. (Alter-
18
natively, designer and cancer survivor Emily McDowell sells empathy cards
19
reading “Please let me be the first to punch the next person who tells you
20
everything happens for a reason.”) Leibniz did grant that sometimes the
21
reasons would be knowable only by God.
22
Why would anybody believe not only that we can usually attribute rea-
23
sons to things that happen but that every single fact about the universe is
24
associated with a particular reason? There is an obvious alternative, after all:
25
that some facts have reasons behind them, but that there are also “brute”
26
facts— things that are simply true, with no further explanation possible.
27
How are we to judge whether brute facts are part of the basic ontology of
28
the world?
29
•
30
31
Whenever we are confronted with questions about belief, we can employ
32
the technique called abduction, or “inference to the best explanation.” Ab-
33
duction is a type of reasoning that can be contrasted with deduction and
34
induction. With deduction, we start with some axioms whose truth we do
35S
not question, and derive rigorously necessary conclusions from them. With
36N
40
Big Picture - UK final proofs.indd 40
20/07/2016 10:02:39
R E A S O n S W h y
induction, we start with some examples we know about, and generalize to
01
a wider context— rigorously, if we have some reason for believing that such
02
a generalization is always correct, but often we don’t quite have that guar-
03
antee. With abduction, by co
ntrast, we take all of our background knowl-
04
edge about how the world works, and perhaps some preference for simple
05
explanations over complex ones (Occam’s razor), and decide what possible
06
explanation provides the best account of all the facts we have. In chapters
07
9 and 10 we will explore this method of inference more fully under the
08
topic of Bayesian reasoning.
09
In the case of the Principle of Sufficient Reason (PSR), for simplicity
10
let’s divide the possibilities into two competing claims: that every fact has
11
a reason that explains it (the PSR is true), or that some facts do not (the PSR
12
is false). To each claim we assign some prior credence— the degree of belief
13
we start out with. Then we gather evidence, by looking at how the world
14
works, and update our credences appropriately.
15
The usual strategy of defenders of the Principle of Sufficient Reason is
16
not to gather evidence but to proclaim that what we have is a “bedrock
17
metaphysical principle.” That is to say, it’s the kind of thing we can’t even
18
imagine not being true. Accordingly, they assign a prior credence of unity
19
to every fact having a reason, and a prior credence of zero to the existence
20
of brute facts. Given that choice, no evidence is going to have any effect on
21
your credences thereafter; you will always believe that every fact is associ-
22
ated with a sufficient reason.
23
Our standards for promoting a commonsensical observation to a “meta-
24
physical principle” should be very high indeed. As Scottish philosopher
25
David Hume— who, if anyone, deserves to be called the father of poetic
26
naturalism, perhaps with his Roman predecessor Lucretius as the
27
grandfather— pointed out, the Principle of Sufficient Reason doesn’t seem
28
to rise to that level. Hume noted that conceiving of effects without causes
29
might seem unusual, but it does not lead to any inherent contradiction or
30
logical impossibility.
31
When pressed as to why we can’t live without the Principle of Sufficient
32
Reason, its defenders generally fall back on one of two angles. They may try
33
to defend it by appealing to some other bedrock metaphysical principle.
34
Leibniz, for example, had something he called the Principle of the Best,
S35
N36
4 1
Big Picture - UK final proofs.indd 41
20/07/2016 10:02:39
T H E B IG PIC T U R E
01