. 6
( 7)


behaviour to justify the bene¬ts which freedom of choice for
individuals confers.
In this example it may be more accurate to speak of God
allowing rather than using the uncertainties but the end effect
is much the same.
behaviour of collections of particles and human populations has been
noticed by other physicists. Social scientists have been somewhat lukewarm
about this and many would share Steve Fuller™s opinion that ˜Social Physics
can never explain complex decision-making processes such as choosing
whom to marry™ (New Scientist, 4 June 2005). Few physicists would claim
that it could!
c h a pt e r 1 2
The challenge to chance

The fact that God could do things through the medium of chance does
not mean, of course, that he actually does so. The view expounded
in the last chapter is challenged by those who take a strong view
of the sovereignty of God. One of the clearest attacks comes from
John Byl, who claims that to introduce ontological chance is scien-
ti¬cally unwarranted, philosophically objectionable and theologically
inconsistent with the sovereignty of God. This argument is met by
attempting to overturn each contention, and in particular, by arguing
that a sovereign God does not need to be directly involved in those
matters to whose outcomes he is indifferent.

bac k to go d ™ s s ov e r e i g n t y
It is one thing to argue that God could have created the world
in a manner which allows chance a real and important role. It
is another matter entirely to argue that he did actually do it
in that way. There are many things that we can do but there
are some which, our friends would argue, we would not have
done on the grounds that it would simply be ˜out of character™
for us to behave in that way. Maybe it would also be out
of character for God to act as I proposed in the last chapter.
After all, there is nothing in what we observe to show that what
looks like chance must be chance. In spite of the complexities
involved in mimicking chance, it is not impossible to do so. The
objection to what I have been arguing comes from theology,
The challenge to chance 197
not science. Some, such as Sproul and Overman, have left us
in no doubt about their view of the matter. A more closely
reasoned objection is provided by John Byl (2003) and I shall
make his article the point of departure for a defence of the
proposition that chance has its origin in God. There are two
stages to the argument. First, I take the objections which Byl
raises to the idea that chance has a place in God™s scheme of
things. Secondly, in answer to Byl, I aim to strengthen the case
by noting that there are positive theological gains in the added
¬‚exibility that an uncertain world allows.

b y l ™ s o b j e c t i on s
The central thesis of this book is that chance plays a posi-
tive role in the world and that it does not undermine God™s
sovereignty. John Byl has provided a sustained critique of
this position and his article provides a convenient basis for
examining the implication of his position in detail.
At the outset a possible source of misunderstanding, about
what Byl calls ontological chance, should be removed. This is
essentially the same as what I have been calling pure chance.
The term refers to happenings1 for which no cause can, even
in principle, be found. That is, there is nothing we can observe
which has any predictive value whatsoever. Radioactive emis-
sions are the archetypal example. It should be made clear that
it is not a necessary part of my thesis that ontological, or pure,
chance is a necessary ingredient of the created order. Rather,
that chance is present and plays a role which is indistinguish-
able from pure chance. Whether or not there are actually any
events which are uncaused can be left as an open question,

Byl™s point, of course is that there are no such happenings “ see the following

note. Here we are merely clarifying how such happenings are de¬ned.
God, Chance and Purpose
but I certainly do not to wish to rule out the possibility of
ontological chance at the beginning. The proposition I wish
to defend is that the world may be treated as if the uncertain-
ties are due to the operation of pure chance. How God might
engineer chance happenings is a separate question. But, for
the moment, I shall engage with Byl on his own ground.
Byl™s argument is that ontological indeterminism ˜is scien-
ti¬cally unwarranted, philosophically objectionable and theo-
logically inconsistent with a strong view of divine sovereignty
and providence™. I shall consider each of these objections in
First, the claim that ontological indeterminism is scienti¬-
cally unwarranted. In Byl™s view this is because there can be
no empirical evidence to settle the question of whether onto-
logical indeterminism exists. Even if, for example, radioactive
disintegration has all the marks of pure indeterminacy, we do
not have to believe, Byl argues, that the radioactive events we
observe are actually uncaused. He claims that to make such
an assertion we would have to establish that there was noth-
ing whatsoever in the observable world having any predictive
value. To observe the present or past state of the world in the
detail needed to establish this is clearly impossible, so we can
never be sure that we have not overlooked some feature which
plays a causative role. Essentially the argument is therefore
about the lack of empirical evidence for asserting that there
are uncaused events. Once this is accepted, Byl has a logi-
cally secure position but it does not follow that this makes the
opposing position ˜scienti¬cally unwarranted™. His view has
become increasingly exposed to the growing empirical evi-
dence that ontological indeterminism provides the simplest, if
not the only explanation of much that we observe. It is true that
the question can never be settled with absolute certainty but
there can be an accumulation of empirical evidence all point-
ing in one direction. A hypothesis which accounts for the
The challenge to chance 199
indirect empirical evidence has to be taken seriously, even if it
can never be supported by direct evidence. In summary, Byl™s
argument is not, in my view, suf¬cient to justify the claim that
the hypothesis of pure chance is scienti¬cally unwarranted.
Secondly, the philosophical objection comes from the
notion of lack of cause. Byl believes that the principle of suf-
¬cient reason2 requires that there always has to be a cause. If
there is no physical cause, then some other agent, human or
divine, must be responsible. This leads him to the position
that a sovereign God must be in absolute control of every-
thing that happens and, in particular, of determining events
at the quantum level. Byl is aware of Bell™s inequality and the
experimental evidence which it provides for the pure random-
ness of quantum events. Bell™s3 work places severe limitations
on the so-called ˜hidden variable™ explanations of quantum
behaviour which would allow the deterministic interpretation
which Byl requires. In particular, hidden-variable theories
violate the property of classical physical theories which for-
bid caused effects at a distance. However, Byl is prepared to
entertain such counterintuitive possibilities in order to retain
causality. Again his position can be maintained, but only as the
ground on which he stands is gradually eroded around him.

The principle of suf¬cient reason says that anything that happens does so for

a de¬nite reason. It is commonly attributed to Leibniz and was elaborated
by Schopenhauer into four forms. To invoke the principle in the present case
is to do little more than say that there must be a cause for every quantum
event. If that is the case, then it is obviously philosophically objectionable
to suppose otherwise.
John Bell (1928/90) was a Northern Irish physicist who proved a theorem

relevant to the interpretation of quantum mechanics. This result has been
described as the most profound discovery of science. Bell™s inequality is
derived from the theorem and this makes it possible to make an empirical
test of whether deterministic theories of quantum behaviour are possible.
A considerable body of empirical evidence has accumulated to show that
they are not.
God, Chance and Purpose
Thirdly, and this is the theological objection, the strong
view of the sovereignty of God requires that God knows
everything and is in detailed control of every happening. This
is in opposition to the view expressed by Peacocke, Ward and
others,4 including the present author, that God might willingly
give up some detailed control in order for a greater good to
be realised. I have argued elsewhere in this book (chapter 1,
p. 15 and chapter 11) that the strong view of God™s sovereignty
is not as well-de¬ned as this simple account suggests. Since
randomness at one level implies order at another level then, if
one expects the order at this second level to express purpose,
one cannot have it without randomness at the lower level. We
cannot have ˜simultaneous sovereignty™, if we can so describe
it, at both levels. This makes it dif¬cult, if not impossible,
to say exactly what the strong form of sovereignty actually
means. That, in turn, makes it very dif¬cult to know what the
doctrine of God™s sovereignty asserts about the world as we
now know it to be.
The most interesting aspect of Byl™s critique arises not in
relation to quantum events, but in relation to human decision
making. Human freedom seems to require a degree of indeter-
minism in the created order without which it becomes a mirage.
Hence, others have argued, the development of human free-
dom requires there to be suf¬cient space for that freedom to be
exercised. Chance seems to provide just the ¬‚exibility required
and therefore to be a precondition of free will. Byl argues that
this is not necessary and he opts for a compatabilist5 view

See, for example, Peacocke (1993, pp. 152ff.), Ward (1996, especially

pp. 80ff.) and the author™s God of Chance (Bartholomew 1984).
Compatibilism is the philosophical doctrine which says that humans can be

free, in a real sense, in a deterministic world. In essence, individuals will
freely choose to do what it is determined that they shall do. In this way
humans can be held morally responsible for their actions.
The challenge to chance 201
of free will, which supposes that the choices which individ-
uals freely make are compatible with what God wills. All the
uncertainty is, therefore, on the human side.
At ¬rst sight, however, the alternative idea of genuine
human freedom makes things worse rather than better. If we
interpret it to mean that human beings act purely at random,
this is hardly what we would expect from responsible human
adults. The notion of free will is surely not the same thing as
˜blind™ will. Yet it is empirically true, as noted in chapter 8,
that many human activities do sometimes appear to be ran-
dom, at least in the sense that when observed in the aggregate,
behaviour appears to be exactly what one would have obtained
if decisions were made by some random process.
Before rejecting this idea out of hand, we must examine it
more carefully. It is here that my de¬nition of pure chance
comes to the rescue, because I de¬ned it simply in terms
of unpredictability. Is it really true that human choices are,
to some degree, completely unpredictable “ that is, there is
nothing we could observe about the individual or the world
which would be of the slightest use in explaining that person™s
I shall now argue that the position of a human choice maker
is fundamentally different from physical entities such as the
electron. A human decision can be rational and yet appear
to the outsider as if it were made at random. The reason is
that each individual has an internal mental world as well as an
external world. To appear random to the outsider, choices have
to be unpredictable by reference to any feature of the external
world because that is all they can observe. To appear rational
to the choice-maker they have to depend “ and may only
depend “ on features of the internal and external worlds. This
internal world consists of ideas, re¬‚ections, beliefs, attitudes,
perceptions, and so on, which may have been formed under the
God, Chance and Purpose
in¬‚uence of the external world but are not fully and publicly
observable in that world. If, of course, attitudes, beliefs and
opinions are elicited by someone in the external world, they
then become part of that world and could be used in predicting
behaviour. But how those internal factors are used by the
individual decision-maker is unobservable and yet may be
totally rational.
In summary, therefore, I am proposing that there is a
purely random element in human decision making, but only
in the sense that choices depend, in part at least, on inher-
ently unobservable factors known only to the choice maker.
This ensures that they will appear unpredictable to the exter-
nal observer because there is nothing available to the external
observer which has predictive value.
But what is the ˜self™ residing in our person that processes
this information, private and public, to arrive at a decision?
It seems essential to require that there is a self which depends
upon, but supervenes6 upon the body of experience which
each of us ¬nds in our internal world. It does not preclude
the possibility, of course, that the divine being should interact
with that self and thus be a party to our choices. However, this
interaction must be voluntarily restricted on the divine side to
avoid taking away the very freedom which the internal world
is identi¬ed to allow.
My conclusion, therefore, is that human choices are real and
they may have the appearance of pure randomness because the
causative factors behind them are part of the internal, private,
Byl goes considerably further than this in the matter of
human choice. I have recognised that individual decisions may
Supervenience is a term that is often used of mental states, where those

states depend upon the brain, say, but are more than can be accounted for
by those states.
The challenge to chance 203
be rational and yet, to the outside world, they may appear to
be random in some degree at least. This is because they are
based, according to my argument, on internal factors which
only the choice-maker knows about. This position allows that
an individual decision is rational, in that it is based on rea-
sons private to that individual. However, it does not exclude
the possibility that the autonomous core at the centre of the
person may be able to make an input to the decision which is
independent of all existing mental content. This would not be
perceptible to others because it would not be separable from
all the other unobservable variables, but it would allow the
individual person an individual identity with which free will
could be associated. Any such action would have to be either
the covert action of God “ who is the only other being to
whom the choice could be attributed “ or a decision of the
person made independently of God.
It is the latter that Byl will not allow. For him ˜creatures
cannot act independently of God™ (2003, p. 114, top). This is
consistent with his strong view of sovereignty which holds
that God is in ultimate and detailed control of everything “
including what we say and do. We might, mischievously per-
some freedom to his creatures!
This view of sovereignty is the classical “ and extreme “
view, the thread of which can be traced through the whole
of Christian history; this includes Augustine but is most fre-
quently associated with John Calvin. It is part of a systematic
theology which has been the source of profound, and some-
times bitter, debate over the last few centuries. Theologians
have wrestled, none convincingly I think, to reconcile real
human freedom with God™s sovereignty. It has been justly
remarked that the volume of literature of the debate through
Christian history is entirely out of proportion to the attention
God, Chance and Purpose
given to the subject in scripture “ and especially in the Gospels.
The good news, that the new life in Christ is open to all, and
the example of the Apostles in pursuing that belief cannot
be easily reconciled with the, admittedly brief, Pauline pro-
nouncements on predestination. In essence, one has to judge
between the literal truth of a few texts (taken out of context?)
and the overall thrust of the Gospel. In the present case, the
former seems to make nonsense of the latter. The total deter-
minism which Byl wishes us to accept is simply incredible,
partly because it ignores the ˜level™ of creation that we are
talking about. In this sphere one quickly runs into logical
contradictions. If Byl is correct, the opposition to his view,
so widespread in Christianity, must equally be an expres-
sion of the will of God “ a thought which bears prolonged
There is one obvious way in which the two views, set in
opposition here, might be reconciled. If, as I shall allow later,
all apparently chance happenings are really pseudo-chance
events, then the sovereignty of God is saved. Everything is
ultimately determined by God, yet what happens is no
different from what would have happened if chance had been
real! In principle, God could predict everything because he
knows the ˜formula™ which generates the random numbers.
This is true, but it also remains true that something indistin-
guishable from pure chance is being used, and pragmatically
this is no different from the position I have been advocating.
The real point here is that, although God chooses the for-
mula and knows what it is, he does not need to know. This
means that it is a matter of indifference to him which of the
many possible formulae he chooses to use. At this deeper
level therefore, there is still a basic uncertainty “ and the
prospect of an in¬nite regress if we pursue this line of thinking
The challenge to chance 205

t h e a dvan tag e s o f c h an c e
In the second stage of this chapter I move on from the negative
to the positive side of my reaction to Byl™s case. It is logically
possible that everything in the world is just like those who
take Byl™s part say it is. The movement of every single particle
in the universe could be controlled precisely by God with the
whole creation being a wonderfully orchestrated (technically
speaking) symphony to the glory of God.7 If things have hap-
pened in this particular way, they clearly could have happened
that way and God could have chosen to do it that way. Why
then should he have contemplated any other way, much less
committed himself to implementing it? Because, I shall argue,
it gives greater freedom for God to express his true nature and
purpose. In short, it is a more ˜Godlike™ way to do it.
Let us begin by returning to the creation of planets bearing
intelligent life. I pointed out that one could be virtually cer-
tain that these would arise ˜by chance™ if the universe were
suf¬ciently large, and furthermore, that there were likely
to be a ˜few™ planets rather than just one. Is this view of
creation theologically credible? At ¬rst sight the notion of
God™s ˜casually creating™ a few planets here and there might
seem an uncharacteristic way of carrying out such a momen-
tous undertaking, but what are the alternatives? Most obvi-
ously, and at the other extreme, one might contemplate the
fabrication of the present earth in a specially selected corner
of the universe by the direct action of the Creator “ whatever
that might mean. But what then would be the point of the
vast universe beyond? Surely not just to impress the inhabi-
tants of earth with the prodigal power of the Creator? Would
A useful account of the debate on the question of whether the universe is

deterministic is given by Taede Smedes in ˜Is our universe deterministic?
Some philosophical and theological re¬‚ections on an elusive topic™ (2003).
God, Chance and Purpose
it not be more God-like to suppose that a universe on the
observed scale was in some way necessary for the existence
of this planet? If we opt for the ˜chance hypothesis™, it was
necessary for the universe to be large enough for humanity to
arise with near certainty “ but does not this relegate mankind
to the edges of creation as a small and insigni¬cant by-product
of something much bigger? Not if we can get out of the habit
of judging importance by sheer physical size. The vast amount
of energy and matter necessary to start the human enterprise
is suf¬cient evidence of the signi¬cance of what was being
undertaken. Having absorbed all of this, the reader might still
be uneasy with the seeming haphazardness of it all. Surely
God would care where, exactly, the habitable parts of the uni-
verse occurred? Probably not and the reason lies in the hidden
assumptions which we so easily transport from our everyday
affairs into the uncharted territory of the cosmos. Location
may be the prime issue in the domestic housing market but
it has no signi¬cance in space. The universe looks much the
same in whichever direction we view it and from whatever
location. So it makes no difference to God, we may suppose,
just where these things happen. It is the local conditions that
matter, not the cosmic setting. The signi¬cance and meaning
of a Mozart opera does not depend on whether it is performed
in Melbourne or Montreal. It is the character of the creation
itself that matters.
These conclusions embody an important principle which
underlies much of our thinking about God™s action in the
world. Roughly stated it says that God does not need to be
directly involved in any events whose outcomes have no rel-
evant consequences for his wider purposes. This is not an
abrogation of his responsibilities but the assumption of his
proper place in the scheme of things. Analogies are always
incomplete and can be misleading but the development of the
The challenge to chance 207
analogy of a management hierarchy, which has already been
used in chapter 9 (p. 152), may point us in the right direction.
Large and complex organisations such as companies or
even nation states may have a single person at the helm “ a
chief executive or president, let us say. That person has overall
responsibility to the shareholders or electorate for whatever
happens. Typically their role is to provide a sense of direction
and purpose, vision, ideas, judgement and all those things
that we sum up in the term ˜leadership™. Such a person is not
expected to check every invoice or interview candidates for
of¬ce jobs. Those things are delegated down the line to an
appropriate level. The person at the top is directly concerned
at all levels and their resignation may be called for if their
subordinates fail. They are responsible for creating a system
which does the job, not for doing the job themselves.
Something similar may be true of God. He is responsible for
every single thing but he does not need (nor should he be)
involved in executing it down to the last detail. That can safely
be left to the autonomous system which he himself has created
for that very purpose. We would justly criticise anyone in an
organisation who continually interfered with the work of their
subordinates. Similarly we might question those who expect
God to concern himself, unnecessarily, with those parts of the
creation which have been designed to look after themselves.

l i f e on e a rt h
Next we turn again to the arena where the manner of God™s
involvement has been most hotly disputed, namely the appear-
ance of human life on earth. Is life the natural consequence of
the interplay of chance and necessity or was it specially cre-
ated by God ˜with his own hands™? Unfortunately, the issue
God, Chance and Purpose
has become clouded by the way in which the battle lines have
been drawn up. This has more to do with questions of how
the book of Genesis is to be interpreted than with the deeper
question of how, or whether, God is at work. I am not about to
rehearse the debate of creationism versus evolution but to look
at how God might have acted in the evolutionary process.8
I start with three accounts which might be given of evo-
lutionary history. One is that it is the product of a purely
deterministic process. On that account any appearance of ran-
domness in the historical record is just that “ appearance. Any
use we make of probability theory to describe what we observe
is just an empirical description of our ignorance, not of any
ontological uncertainty. Theistic determinism says that this
is the way God intended it to be and his hand is to be seen
in everything that happens, however trivial. It covers both
what one might call the short-term development that cre-
ationism requires and the more long-term view of atheistic
The second view is the one where most things which happen
are actually due to chance, as the modern scienti¬c account
suggests, but that God intervenes occasionally to give things
a nudge in the desired direction. At particularly signi¬cant
junctures it may need a particular mutation or other micro
event to keep evolution on track. As I have noted several times
already, there is no way we can distinguish this model from
complete indeterminism because interventions are suf¬ciently
rare to be lost in the greater mass of genuine randomness.
Also, as I have noted, it is not clear whether such occasional
interventions would be suf¬cient to do their job. In other
words, the method might not work.

See note 7 of chapter 11.
The challenge to chance 209
The third account is that provided by so-called natural-
istic science “ where the chance events are what the critics
call ˜blind chance™ as though the adjective quali¬ed the noun
in some signi¬cant way. It is this third view which creation-
ists and supporters of Intelligent Design have had in their
sights as the main threat to theology. It appears to banish God
entirely and replace him by the ogre called blind chance. This
was never a fully defensible view and it is even less so now
than it once was. It was always possible to argue that this was
another example of God™s acting with the aid of chance. Thus,
if life had emerged from the seemingly chaotic goings on in
nature then it certainly could do so, and given enough oppor-
tunities, it certainly would have done so. The main dif¬culty
arose from the view, championed by Gould, that there were so
many paths for evolution to follow that it was incredible that
our particular path, the one leading uniquely perhaps to us,
should have materialised. It was at this point that many apol-
ogists have felt the need to invoke the intervention of God
to make sure that the right path was chosen. The ubiquity of
convergence, which was discussed in the last chapter, radi-
cally alters the need for such interference. For if the paths of
evolution converge it becomes much more likely that all (or
many, at least) roads will ˜lead to Rome™. In addition, if the
main sphere of God™s activity turns out to be in interacting
with human minds, the greater part of evolutionary history
may be seen as a preparatory stage which was designed to
deliver what was necessary with the minimum of interfer-
ence from outside. Theologically speaking, this third alterna-
tive, of God™s creating by chance, may prove to be the most
This last point brings us to the threshold of a major issue
which needs a chapter to itself. Its origins go back to the third
God, Chance and Purpose
example given in the last chapter, concerned with how God
might use uncertainty in a constructive way in the realm of
human affairs. For those like John Byl who, as I have already
noted, believe that there can be no human action apart from
God, this question does not arise. But for the rest of us it opens
up a whole new vista.
c h a pt e r 13
Choice and chance

The discussion of free will and order at the end of chapter 12 brings
us up against the central question of whether human choices which
appear random can be really random. We must, therefore, attempt
to unravel the intimate relationship of choice and chance. The ¬rst
question here concerns in what sense human choices can be said to be
random. The idea linking the two is that of unpredictability, which is
the key characteristic of both free choice and chance. This leads on to
a consideration of how God can ˜create™ chance and to the observation
that total unpredictability, at every level, is impossible.

a pa r a d ox i c a l s i t uat i on
Choice and chance stand in a paradoxical relationship to one
another. If choices are free then, presumably, they are not
entirely predictable; for if they were predictable they would
be determined, in part at least, and hence not free. If chance
implies lack of predictability, then chance events could hardly
be the result of deliberate choice. Yet rational choices, when
viewed collectively, do often appear as if they were random. If
all happenings, which to us appear to be completely random,
were to be attributed to God, then they would have to be his
deliberate choices, rationally made, so it is dif¬cult to describe
them as totally random.
The purpose of this chapter is to unravel the confusions
which lie hidden in such assertions. These issues have surfaced
God, Chance and Purpose
several times in earlier chapters but the time has now come to
make them the focus of our attention. This will involve going
over some old ground again but will serve to emphasise its
importance in the present context. I begin with the matter of
human choice, leaving God out of the picture for the moment.

a r e h u m an c h o i c e s r e a l ly r an d o m ?
This appears to be a very silly question. It is of the essence
of our human nature at its best that we make informed and
rational choices; how then can we speak of them as random?
To some scientists, however, there is nothing silly about the
idea. Marvin Minsky (1987, p. 306),1 for example, has put the
matter categorically.
We each believe that we possess an Ego, Self or Final Center of Control,
from which we choose what we shall do at every fork in the road of
time. To be sure we sometimes have the sense of being dragged along
despite ourselves, by internal processes which, though they come from
within our minds, nevertheless seem to work against our wishes. But
on the whole we still feel that we can choose what we shall do.
He goes on,
According to the modern scienti¬c view, there is simply no room at
all for ˜freedom of the human will™. Everything that happens in our
universe is either completely determined by what™s already happened
in the past or else depends, in part, on random chance. Everything,
including that which happens in our brains, depends on these and only

A set of ¬xed, deterministic laws. A purely random set of accidents.

Minsky™s remark was made twenty years ago towards the end of a distin-

guished career. It may be that his view would not be widely held today, or
be expressed in such categorical terms, but the quotation serves to de¬ne an
extreme position.
Choice and chance 213
There can be no room on either side for any third alternative. What-
ever actions we may ˜choose™, they cannot make the slightest change
in what might otherwise have been because those rigid, natural laws
already caused the states of mind that caused us to decide that way. And
if that choice was made by chance “ it still leaves nothing for us to

To digress for a moment, any statement like that raises
profound questions about the meaning which can be attached
to it. For if Minsky is speaking of all minds, then presumably
his own is included. If what comes out of his mind is solely
the result of deterministic laws and random chance, it is not
clear why such an automaton™s utterances should correspond
to anything in the real world or, for that matter, why similar
automatons like ourselves should believe them. The same
problem arises whenever a human mind attempts to make
statements about the class of human minds to which it, itself,
belongs. This chapter is no exception and rational discourse
can only continue if we reject Minsky™s view. It is not open
to us to set ourselves above the debate by supposing that we
alone have a privileged position which exempts us from the
necessity of applying our conclusions to ourselves!
That being said, it is an empirical fact, already noted, that
human choices, when viewed collectively, are often virtually
indistinguishable from random choices.2 How then do we
reconcile this collective character with what we believe to be
the purposefulness of our own individual choices?

It must be emphasised that randomness is not a property of a single event

but of a collection. To say that human choices appear to be random is,
therefore, a statement about a set of actions and not about any single act.
To justify such a statement we have to observe a collection of choices made
under similar conditions and to ¬nd that the relevant observed frequency
distributions were what they would have been if they had been generated
by a known random process.
God, Chance and Purpose
The clue to the resolution of the paradox lies in the way
that I have de¬ned chance: the essence of a chance or random
event, remember, is that it cannot be predicted, even partially,
from anything we can observe about the state of the world past
or present. In the case of human choice, this would certainly
be true if everything a person thought or did was, as Minsky
supposed, the result of random ¬rings in the brain (any partial
dependence on any other ¬xed aspect of the world would
reduce, but not eliminate, the uncertainty so does not affect
the point being made). The obvious way to reconcile choice
and chance is thus to go along with Minsky, and those who
think like him, and to say that they are the same thing.
But there is another way to do it which does not rule out at
the start the notion of a rational-thinking human being who
is affected by factors not detectable by external observers but
known only to the subject. In that case the choice making could
be perfectly rational yet the rationale would be hidden from
anyone external to the individual. To clarify this statement
I refer back to the distinction between a person™s external
world and their internal world which I made in chapter 12. The
internal world may be partly constructed from what happens in
the outside world but will also have an input from the ˜self™. The
individual™s choices, as observed from outside, are therefore
partly unpredictable and, in the aggregate, will display the
same characteristics as a process in which there is a truly
random element.
In practice, of course, the inner and outer worlds will
become mixed up because of the two-way traf¬c between
them. Something of the internal world will be revealed in pat-
terns in past choices, and there will be a modi¬cation of the
inner world as information from outside is absorbed to become
Choice and chance 215
part of the inner world. It may also happen that choices appear
random to us only because we have not noticed causal fac-
tors which are operating. Indeed, one may suppose that this
is normally the case since our knowledge is so limited.
None of this precludes factors in the external world from
having a causal role in what we decide. Genetic and environ-
mental factors immediately come to mind. Some people go as
far as to say that they are wholly responsible3 but, if this were
so, one might expect our choices to be much more predictable
than they actually are. But this fact certainly does not rule
out the possibility that there may still be a large component
of unpredictability in our behaviour, arising from the factors
which are exclusive to ourselves.
We are very familiar with what we regard as the mixture
of determinism and free will in much of what we do. Much
market research, for example, depends on the fact that there
are things which we can observe about people “ their income,
hobbies, and so on “ which are useful predictors of purchasing
behaviour. Voting behaviour is another example of something
which is partially predictable. Income, occupation and place
of residence all tell us something about the party for which
someone will vote “ especially at national elections. These pre-
dictors may have been better in the past than they are today and
their effect will vary from one culture to another. But, when
all of this has been taken into account, all voters, for example,
It is common for people to suppose that human decisions are explained

by a mixture of ˜nature™ and ˜nurture™. Most of the arguments which take
place are about the proportions of each in the mix. If this were literally
true, all such decisions would, in principle, be predictable. Once the genetic
and environmental factors were identi¬ed and exhaustively measured, the
decision would be a foregone conclusion. One would have expected there
to be a reasonable number of non-trivial examples where outcomes were
well predicted but such examples are not easy to ¬nd.
God, Chance and Purpose
believe that they have a real choice and that this is not totally
constrained by their economic and social circumstances. In
short, they will think of voting as an exercise of their free will.
Free will, in the libertarian4 sense, at least, requires that we
should be able to make choices that are not wholly determined
by factors which are, in principle, observable by others.
All that I have said so far can be accommodated within
a naturalistic framework. No appeal need be made to God
to explain the unpredictable choices which people make. The
essential point thus far is that the individual has a world of
private experience which permits rational decision making
and which is not accessible to outside observers.
However, this leaves out of account two matters which
are important if we wish to bring theological factors into the
discussion. The ¬rst adds another dimension to the naturalistic
interpretation already discussed above. If God acts through
individuals, and if his mind is hidden to us, the unpredictability
we observe in the actions will be, partly at least, a re¬‚ection
of God™s participation in those actions. The second case is
where apparently random things happen without any human
participation at all as, for example, when particles are emitted
from a radioactive atom. This takes us back to God™s action
in the physical world and it may seem rather remote from the
sphere of human action. Nevertheless it is convenient to see
it as a limiting case of the problem of human decision making
which arises in the case when we eliminate the human element
and thus leave God as the only remaining decision maker.
Can God also do unpredictable things “ unpredictable, even
by himself? I take the two cases in the order in which I have
introduced them.

The philosophical doctrine that human choices are neither causally deter-

mined nor random. They are genuinely free.
Choice and chance 217
How God acts in the world is a major problem by any
standards and this problem was discussed in chapter 9. The
most natural way to accommodate God™s action in the scheme
of things is to suppose that he acts partly, if not exclusively,
through interaction with the human mind. According to this
view, God™s acts would appear under the guise of human acts
and thus be indistinguishable from them, in principle, at least.
John Byl, and those who think like him, believe that no one
can act apart from God and hence that there are no genuinely
free choices. According to them our apparent freedom arises
from the fact that our choices are those which God has made
for us, and which are compatible with our wishes “ for which
he is also responsible. Even if we take this view, it remains true
that the choices which God makes in us will not be predictable
and hence, in the aggregate they will appear as if they had
been made at random.
At ¬rst sight the situation in the physical world is much
simpler because the entities with which it deals do not have an
˜internal™ world. The fact that they appear random is simply a
re¬‚ection of the fact that we know of no other observables in
the world which would have any predictive value. However,
on closer inspection, the matter is not so straightforward. From
the theological perspective there is no other person involved,
so any decisions made have to have been made by God. This
brings us back to the question of how God might determine
the outcomes of physical processes, bearing in mind the fact
that, because there is no separate physical cause, they will be
totally unpredictable and hence apparently random. How then
can God purposefully behave as if he had no purpose? I can
link this question to another which arises out of my central
claim that God uses chance for his own purposes much as we
use it for ours. In fact, these are really the same question which
may be posed as: how can God create chance?
God, Chance and Purpose

h ow c an go d c r e at e c h an c e ?
We now turn to look at things from God™s side. If there is
chance in the world and if God is somehow responsible for
it then what happens is, in a certain sense, his choice. How
can what he chooses be random? To clarify this conundrum we
have to think again about how chance events come about.
At this stage we are thinking primarily of the natural world,
so I use again the familiar example of the emission of particles
from a radioactive substance. Here we are back to the quan-
tum world where, to the best of anyone™s knowledge, there
are no predictors available. Thus, if we suppose that emissions
occur at random we know that we can deduce the form of sev-
eral observable distributions, as reported when discussing the
same situation in chapter 3. For example, if we measure the
waiting time to the next emission it turns out that the propor-
tion of occasions on which we shall have to wait more than a
minute, say, is given by a simple exponential expression. Alter-
natively, if we count the number of emissions per minute, the
frequency distribution will follow Poisson™s law. There is an
important feature of random processes which has been noted
above and needs to be reiterated here. This is, that when we
observe randomness, there is independence between events.
This does not need to be explicitly mentioned because it is
covered by our requiring that the past history have no predic-
tive value. For example, the time of emission of a particle does
not depend in any way on the timing of other preceding emis-
sions, and that is what is meant by saying that it is independent
of them.
If God determines outcomes then his actions are part of
what physics studies. If pure chance is part of the created order
and is therefore a work of God, how does he do it? Put another
way, if God is the cause of everything, how can he cause
Choice and chance 219
something which, by de¬nition, appears to have no cause?
This problem was identi¬ed in God of Chance (Bartholomew
1984, p. 102) and commented on by Byl (in Byl 2003, p. 107)
and others, but its importance does not seem to have been
generally noticed. If, for example, quantum happenings have
all the characteristics of chance events with no discernible
cause, how does God, I repeat, bring them about?
It is always possible, of course, that we are deluding our-
selves and that there are no purely chance events. We can
produce pseudo-chance events, and if we can do so, then they
would surely not be beyond God. But let us suppose, for a
moment, that there are events which happen independently of
the state of the universe. (In passing, it is worth noting that
there is also a problem here for non-theists, but for them it
is part of the broader question of why there is anything at
all. This then becomes the point at which they come to an
impasse and have to admit that rational enquiry can take them
no further. Theists, on the other hand, have the no less dif¬cult
problem of explaining why, in God™s world, some things can
happen without a cause.)
We should make considerable progress on this question if
we could see how God might engineer randomness in some-
thing as simple as a sequence of random tosses of a coin. Coin
tossing is not random in the purest sense but is a very sub-
tle example of a pseudo-random series. Tossing a coin is a
dynamic matter where the impulse imparted to the coin, the
air currents and a multitude of other physical factors deter-
mine the outcome according to the laws of matter which are
well understood. Nevertheless, coin tossing is near enough
random to provide a convenient and familiar framework in
which to discuss the problem. The question now becomes:
how might God generate a sequence of heads or tails (or zeros
and ones) which was purely random?
God, Chance and Purpose
The following very tentative suggestion is based on the idea
that if causal forces must be present yet have no effect, their
individual effects must, in some sense, cancel out. We might
think of it as a kind of tug-of-war in which the two physical
propensities to produce heads or tails might just balance. Since
there has to be one outcome or the other, and since there is
nothing to decide between the two options, what actually
occurs is a matter of chance. Buridan™s ass5 provides a familiar
picture of the situation. Placed between two equally tempting
bales of hay, where each is equally accessible, the ass has no
reason to prefer one over the other. Which way the scales are
tipped cannot then be predicted from a knowledge of the two
forces involved moving in either direction because they are
equal but opposite.
A much simpler solution to the dilemma, touched on above,
is to suppose that God, like us, can produce pseudo-random
events, not genuinely random events. That is, behind every
happening for which we can see no causal factors, there lies,
deeply buried in the mind of God, a pseudo-random-number
generator. For those who demand that God be in total control,
this provides the determinism they are looking for. But in
interpreting this we must recall what I have repeatedly said
about ˜levels™. What is random at one level may be determined
at the next higher level. Chance and determinism are not
mutually incompatible alternatives but each may imply the
other “ but at a different level. This is tantamount to saying
that total randomness is impossible anyhow.

The dilemma of the ass which starves because of its inability to make

a decision often comes into philosophical discussion. It dates from the
fourteenth century but is said to have been discussed by Aristotle (384“
322 bc).
Choice and chance 221

tota l u n p r e d i c ta b i l i t y at a l l
l e v e ls i s i m po s s i b l e
I have repeatedly stated that lawfulness in the aggregate may
be the product of randomness at the individual level. What is
unpredictable at the latter level, therefore, necessarily becomes
predictable at the higher level. The converse may also be true
but we are not at liberty to have unpredictability at both levels
simultaneously. This fact places restrictions on a God who
operates at the individual level, and thus places a question
mark against the total freedom which we might require to be
possessed by someone who was truly God.
To make the point clearer, let us return yet again to the
emission of particles from a radioactive source. Nothing we
can observe about the process “ such as the length of time
from the last emission “ has any predictive value for the tim-
ing of the next emission. One consequence of this is that the
frequency distribution of the intervals between emissions has
a characteristic and predictable form known as the negative
exponential. It looks like the example given in ¬gure 3.1 (see
p. 35), which shows, among other things, that the frequency
declines as the length of interval increases.
Next, suppose that the actual timing of the emissions is
determined by God. In order for his choices to conform to
the empirical facts, it must not be possible for the human
observer to detect any departure from randomness. Detec-
tion would be possible if the emerging frequency distribution
departed from the negative exponential form. We can there-
fore imagine the Deity having to keep an eye on the overall
frequency distribution at the higher level, as well as making
sure that nothing in the detailed history of the process at the
lower level, including the recent inter-event intervals, reveals
God, Chance and Purpose
any non-independence. It would surely be much simpler and
just as effective for the Deity to buy or borrow a pseudo-
random-number generator and let that determine when emis-
sions should occur! In other words, it is not possible to do
better than a randomising device, so why should God not use
the best tool available? What would it tell us about his nature
if he chose the second best?
For all things that depend only on the aggregate behaviour
of random events there is no obvious advantage in actually
determining when each individual emission occurs. Where
individual happenings might matter is in circumstances where
they trigger some macro event, perhaps through the amplify-
ing effect of a chaos model. A tiny proportion of deliberately
engineered events would then pass unnoticed in the mass of
randomness and those could be inserted by the Deity at will,
with no possibility of detection. This puts God™s particular
actions beyond the reach of science, and it promises to dis-
engage the two ¬elds from potential con¬‚ict. However, this
raises the awkward question which I posed in connection with
the coin-tossing example in chapter 8. This whole manner
of working smacks of subterfuge, which is hardly a God-like
characteristic. But maybe there is no other way, given the
other requirements of running the whole creation. Further, it
is not clear why the transparency, so often demanded in human
decision making, should necessarily be part of the divine char-
acter. It does leave open, however, the question of why such
¬ne-tuning should be necessary except, perhaps, to counter
the unwanted consequences of human free will. But then we
must ask why a more direct intervention at the level of human
decision making might not be more effective.
c h a pt e r 14
God and risk

The central thesis of this book is that God uses chance. This appears
to carry the implication that God takes risks. This simple statement
has repercussions for most parts of theology but here we focus on the
central issue. What is required is a theology of risk and this is what
the chapter aims to provide. After some theological preliminaries, I
commend a view which it is proposed to call critical orthodoxy.

d o e s go d ta k e r i s k s ?
We live in a society obsessed with risk. Risk assessment and
risk management are part of everyday life in business and
industry. We are exposed to all manner of hazards, not only
to life and limb, but to our comfort and general welfare. The
insurance industry has long existed to alleviate the problem by
spreading risk but the threat of litigation and crippling dam-
ages has heightened awareness of the problem. It is hardly
surprising that there should be a ready market for Peter Bern-
stein™s book Against the Gods: the Remarkable Story of Risk
(1998).1 But risk also raises questions for theology. Gregerson
(2003b) has taken steps towards a theology of risk and fol-
lowed this up (Gregerson 2006) with a study which, among

Bernstein™s book also contains much else that would provide useful back-

ground reading for this book, about sampling for example.

God, Chance and Purpose
other things, challenges the views of some contemporary soci-
All of this work looks at risk from the human side. There
is also a God-ward side on which we concentrate in this chap-
ter. The idea that God uses chance poses many problems for
theology, and the chief of these, perhaps, is the question mark
it places against the sovereignty of God. This challenge has
occupied us a great deal in the earlier part of the book, espe-
cially in chapter 12. It also lies behind many of the issues
discussed in this chapter but I shall keep it in the background
as we approach things from a different angle.
If my thesis that God uses chance is accepted, we cannot
avoid the consequence that he appears to be taking risks. How-
ever, it may not be immediately obvious why this is such a
problem. After all, the whole thrust of the argument has been
that chance can be used to achieve determinate ends, so we
might naturally wonder what risks are being run. To clarify
this we need to recall the differing levels of nature and society.
Near certainty at one level may be the product of uncertainty
lower down, and it is at that lower level that outcomes may
have unwanted effects. If it is true that the evolutionary pro-
gression to ourselves is the product of chance and necessity,
then it is painfully obvious that this is not achieved without
waste and suffering along the way. If that is so, God could not
avoid taking risks. To get the global picture right, it appears
that it may have been necessary to get many of the details
wrong. This is our starting point.
Two questions immediately pose themselves. First, is God
really a risk taker?, and secondly, why did he create a world
in which we are exposed to so many risks? The aim of
this chapter is to suggest answers to both questions but ¬rst
we need to be clearer about what they mean and why they
God and risk 225
To say that God is a risk taker means that he takes “ or has
taken “ actions where the outcome was intrinsically uncer-
tain and which might turn out contrary to his intentions. It
is immediately clear why this is a serious question for Chris-
tians. A God, in the classical mould, who is omnipotent and
omniscient, ought not to be at the mercy of his own creation
like this. Such a God ought not to need to take risks. Even if,
for a moment, we concede that things might occasionally get
out of hand, God should surely have the power to get things
back on track before any harm is done. Indeed, the language
of risk seems to have no place in describing the nature and
role of the God of the Bible and Christian orthodoxy.
Such a God does not fare much better if we judge him in
relation to the second question. Many of the risks we run lead
to immense suffering and damage. Even when we allow for
the fact that much of this is self-in¬‚icted, the residue is still
capable of supporting a serious charge. Either God deliber-
ately created a world in which there was bound to be much
hurt and suffering, or he did not have suf¬cient control over
things to stop them going awry. In the former case we have to
question his very nature, and in particular, the Christian claim
that his nature is love. In the latter it is his omnipotence that is
undermined. Furthermore, if he cannot fully control things,
it is dif¬cult to see how he could act effectively in the world
and so his providence has to be added to the list of orthodox
If we take all this into account, the credibility of the whole
edi¬ce of Christian orthodoxy begins to look decidedly shaky
and there is no lack of critics on hand to drive that fact home to
the Christian™s disadvantage. But perhaps there is no cause for
alarm after all, for these gloomy prognostications only follow
if we give a positive answer to the ¬rst question. If God is
a risk taker then we may indeed be in trouble, but surely
God, Chance and Purpose
centuries of Christian theology cannot be so lightly over-
turned? Should we not take our stand on the certain rock of
revelation and dismiss all of this as idle speculation? I think
not. And my reasons lie in the fact, which has repeatedly been
brought to our attention, that there is much in the scienti¬c
picture of the world that seems to be inconsistent with a theol-
ogy which sees God in detailed charge of every single thing. I
have touched upon these matters in earlier chapters but now I
tackle them in a more systematic way. Although they are usu-
ally dealt with in the context of God™s sovereignty, I wish to
discuss them from ¬rst principles under what I shall call na¨ve ±
orthodoxy. If we take the new knowledge about the role of
chance in the world seriously, we must either revise our view
of God or join those who have abandoned him altogether.
Earliergenerationswouldhave beenpuzzled by such a claim
because, from their perspective, the boot would have been on
the other foot. For them science seemed to involve replacing
to the Bible™s view of an omnipotent God. At the very core of
science was the lawfulness of nature typi¬ed by Newton™s laws
of motion. And have not Christians, including Newton him-
self, seen this as testifying to the faithfulness and reliability of
the God who created it all? In the almost mechanical necessity
of the physical world, there seemed to be no place for uncer-
tainty and risk taking. To be sure, this view was essentially
deism and posed other problems for the believer, and some
ingenuity was needed to see how room could be found for
special providences. But the general providence expressed in
the laws of nature seemed secure and no one could level a
charge of capriciousness at the God to whom science seemed
to point.
Times have changed and science has moved on. It is not
that the laws of science have turned out to be wrong. They are
God and risk 227
as valid as ever they were. But we now see them as a partial
description of an immensely complicated universe in which
chance plays as important a part as necessity. Before we can
begin to reconstruct a credible theology, which is consistent
with our new knowledge, we must recall the different kinds of
uncertainty which we encounter in the world. I have identi¬ed
three broad categories: the pure chance which we observe
primarily at the quantum level; accidents which include the
mutations of evolution; and, ¬nally, the uncertainties of human
decision making.

s o m e t h e o lo g i c a l i m p l i c at i on s o f t h e
way t h e wo r l d i s
Given that the world is such an uncertain place, we face some
serious questions about the apparently hit-and-miss nature of
some aspects of the creative process. Leaving aside the ques-
tion of the hazards attending the appearance of a habitable
planet, which I have already dealt with, let us focus on the
emergence of life. According to the likes of Stephen J. Gould,
Jacques Monod,2 Richard Dawkins and others, the appear-
ance of life was a chance happening, depending on accidental
copying errors in the reproductive process. If that was all there
was to it, Gould™s metaphor of the rerun ¬lm applies and our
presence is the almost incredible result of a random process.
On this reckoning, it was extremely unlikely that intelligent
human life would have been the outcome. That hardly sounds
like the strategy of a loving God intent on creating beings des-
tined for fellowship with himself for, if it was, it was almost
Jacques Monod™s book (1972 [1970]) is, perhaps, the original and clearest

exposition of the thesis that chance and necessity are both necessary and
suf¬cient to account for all living things. Dawkins and Gould are but two
of the most widely read authors who have propagated the idea.
God, Chance and Purpose
certain to fail. However, convergence comes to the rescue by
showing that the appearance of sentient beings, not so very
different from us, may have been almost certain.
Even if we take the view that the emergence of life on
earth was inevitable, there were still enormous risks to be run.
Collisions of the earth with meteors or asteroids could have
destroyed the ¬‚edgling life or so distorted the path of evolution
as to close the door to human development (something similar
seems to have happened to the dinosaurs). Volcanic eruptions,
radical climate changes or devastating diseases could likewise
have posed insuperable barriers for the progress of life towards
its intended culmination in human beings.3
Again, even if, against all the odds, human life did become
¬rmly established, the hazards of human history seem to put
the whole enterprise at risk. We know from that small segment
of society with which we are intimately acquainted, how much
often seems to depend on the merest whim. When magni¬ed
a millionfold onto a world scale with its recessions, wars and
conspiracies, it seems incredible that anyone, even God, could
control what was going on. On the face of it, at least, human
history does not look like a carefully orchestrated divine drama
moving inexorably to its intended end. The threat of global
disaster by nuclear war, famine or epidemic is still suf¬ciently
real to give us serious pause for thought.
The dilemma for the theologian is most acute when we
come to consider the place of the Incarnation in the scheme of
things. Think of all the mishaps which can occur in childhood
to nip a promising life in the bud. If, as is claimed, only 50
per cent of children at the time of Jesus reached their tenth

The possibility that a burst of gamma rays might be responsible for a mass

extinction 443 million years ago was raised by Mark Buchanan in New
Scientist, 30 July 2005.
God and risk 229
birthday, the risk of pinning the redemption of the world on
a single individual is apparent. Did God have to take these
risks? If he did not and Jesus was spared those risks, how can
we say that Jesus was truly human? For experiencing risk and
being changed by the experience is part of what it means to
be human. The doctrine of the full humanity of Jesus seems
to preclude the kind of special protection accorded to royal
personages and their likes today.
These considerations, and others like them, seem to take
us into a very different world from the one addressed by the
prophets and New Testament writers of the Bible. It is not
easy for those immersed in contemporary culture, in which
risk seems so deeply embedded, to take seriously the Christian
story as na¨ve orthodoxy presents it.

b e n e ¬ ts o f a r i s k y wo r l d
We have assumed, perhaps too readily, that risk is a bad thing
and that its ubiquity is an embarrassment to the believer. This is
far from the case. In evolution, for example, the almost prodi-
gal variety which nature throws up seems a terrible waste
from an economical human perspective, but it confers many
bene¬ts. It provides an insurance against the unexpected. For
example, by ensuring that there is a wide range of variation
among organisms of a particular type, it increases the chance
that there will be at least some survivors of any disaster capable
of coping with the new situation. This use of chance to coun-
teract the effects of chance is a subtle and surprising feature
of nature that has profound implications.
When we move on to human society there opens up a
whole new dimension of risk. Risk is not only something
which we seek to avoid or to insure ourselves against. It is
also something that we seek out, or even create, for ourselves.
God, Chance and Purpose
Mountaineering, caving or extreme physical sports are not
thrust upon us, and though they may cause death or serious
injury, there is no lack of people eager to take part, and prizes
are awarded to the successful. Likewise, no rational person
guided by economic criteria alone would gamble, because the
expected gain is never positive. Nevertheless, to many the
excitement and suspense when much is at stake seems to make
it worth while. The enduring appeal of many games depends
on just the right balance of skill and chance. In snooker, for
example, there would be little interest at the highest level
if chance played no part, and it is a prime function of rule
makers to create just that right balance. The rules must favour
the skilful but involve enough luck to leave the outcome in
doubt and keep interest alive.
Although games might be considered a rather trivial matter
to bring into this discussion, we have seen already how illu-
minating they can be. Games mirror life, and in some cases,
perhaps, provide a substitute for the thrills necessary to our
wellbeing of which we have been deprived by modern civil-
isation. The competitive instinct is deeply rooted and seems
to be essential for progress. It could be plausibly argued that
risk is a necessary ingredient for full human development. It
provides the richness and diversity of experience necessary
to develop our skills and personalities. This does not mean
that the risks are always welcome at the time. We can all look
back and identify occasions of great uncertainty which we
would have gladly avoided but which are seen, in retrospect,
to have contributed to our development. The notion of trial
and test in the religious sense, much of which involves facing
hazards of various kinds, is familiar enough to Christians. The
story of Job springs to mind, as does the temptation of Jesus
and his experiences leading to the Cross. The remarks in 1
Peter 1 about faith, like gold, being tested by ¬re, must have
God and risk 231
found many echoes in the experience of the early Church. It
is not easy to imagine a world without risk or to know what
its absence would imply. At the very least we should now be
aware that its abolition might not be wholly bene¬cial.

t h e o lo g i c a l p r e l i m i na r i e s
I indicated at the beginning of the chapter that chance and
uncertainty were widely seen as a threat to what I called na¨ve
orthodoxy. The question now to be considered is whether we
can begin to fashion a more critical orthodoxy that comes to
terms with the new without abandoning what is essential in
the old. I want to suggest that we should welcome the new
because its insights are almost wholly bene¬cial.
As a preliminary step it may help to make some remarks
according to which orthodoxy consists in assent to a long list
of propositions. From this perspective the ˜true™ faith is then
that which is held by those who assent to all items on the list.
Those of weaker faith place crosses or question marks against
some items and are seen as professing a somewhat watered
down version of the faith; sects are distinguished by the items
on the list to which they attach particular importance.
This is, of course, a caricature but something like it lies
behind much of the posturing in religious circles. It fol-
lows naturally from what we might term the mathematical
or deductive approach to knowledge.4 According to this,
certain truths are held to be given, such as the self-evident
axioms of geometry. Theology, like geometry, then consists
Foundationalism is a term often used to describe the equivalent way of

reasoning in theology. One starts from certain propositions which are taken
as ˜given™, in some sense, and which therefore provide the foundation for
the theology that is built on them.
God, Chance and Purpose
in working out their logical consequences. Propositions can
be established as de¬nitely true or false by the processes of
logic, and the aforementioned list then consists of those truths
which can be deduced from the axioms provided by revela-
tion. We do not need reminding that there is some diversity
of view about where the axioms are to be found and what they
To my mind, the process of scienti¬c inference provides
a better model for theology. Not so much in the process of
conjecture and refutation advocated by Popper5 “ though that
has its merits “ but in the inductive accumulation of knowl-
edge. I have advocated and illustrated this approach in my
book Uncertain Belief (Bartholomew 1996), arguing that all
knowledge is uncertain, in varying degrees.6 This is a bottom-
up approach. It yields nothing to its rivals in the importance it
attaches to the treasures of Bible and Church. The ˜faith once
delivered to the saints™ is expressed in the record of writings,
experience and practice of the Christian community. This is an
essential part of the evidence with which the theologian must
grapple in trying to construct a coherent account of reality
but it is not the whole of the evidence. It must be interpreted
Karl Popper started from the idea that, though one could never determine

with certainty what was true, it was sometimes possible to be certain about
what was not true. Science should thus proceed by a process of falsi¬cation.
For example, one could never be certain that all crows were black but the
proposition would be falsi¬ed if a single crow were to be observed which
was not black.
Richard Swinburne pioneered the inductive approach in theology using

Bayes™ theorem. The ¬rst edition of his book The Existence of God, published
in 1979, was a landmark publication in this ¬eld, though, for reasons set out
in Bartholomew (1996), I did not think the application of the method was
wholly convincing. The second edition, published in 2004, was described
by the publisher as ˜the de¬nitive version for posterity™. In the preface the
author acknowledges that his ˜critics are ˜many™ and that they ˜have provided
much help™ but it is not clear what form that help took.
God and risk 233
by reference to the knowledge from all sources including,
especially, science.
Pictures may help to make the distinction between the
deductive and inductive approaches to belief clearer. The
approach of deductive theology is like the house built on the
solid rock of certain truth. As long as the foundation remains
sound, the elaborate edi¬ce above stands ¬rm and secure. But
if the integrity of the foundation is compromised, cracks begin
to appear and although appearances can be maintained for a
while by papering them over, the pretence cannot be kept up
What about the inductive approach? Is that like the house
built on sand, as it will seem to many? In a way it is, but a better
analogy is the picture of a railway built across a bog which
was used on the cover of Uncertain Belief. When George
Stephenson built the Manchester and Liverpool Railway he
had to cross Chat Moss. The obvious way of building the
line would have been to sink piles onto solid rock beneath
and support the line in that way “ but no such foundation
could be found. So he ¬‚oated the line across on a bed of
heather lashed to hurdles. The weight was thereby distributed
so widely that even such a frail foundation as boggy ground
could support it. The strength, in this case, was derived from
the mutual support of the interlocking hurdles, none of which
was excessively loaded. I think it is better to regard the basis
of belief in these terms. Although no single element will bear
much weight if looked at in isolation “ as many critics have
shown only too well “ the combined strength is immense. It
is the cumulative and interdependent effect of all the frag-
ments of evidence which ultimately provides the securest
One important consequence is that the near certainties only
begin to emerge when we stand back and look at the whole
God, Chance and Purpose
picture. In many of the details we shall be mistaken without
always being able to tell which. It is in the common strands that
run through time and across diverse cultures that the essential
core is to be discerned. To take his words only slightly out of
context, John Wesley put it rather well when in his sermon on
the catholic spirit, he said:
No man can be assured that all his opinions taken together are true.
Nay every thinking man knows that they are not . . . He knows, in
the general, that he himself is mistaken; Although in what particulars
he mistakes, he does not, perhaps he cannot know.

We can take the analogy with scienti¬c method one step
further. It suggests how we may expect doctrine to develop
as knowledge advances. Science typically progresses not by
replacing old knowledge by new, but rather the old ¬nds
its place in the new as a special case or an approximation.
Newton™s laws of motion are no less true today than in the
seventeenth century. They are implicit within relativity the-
ory as an approximation valid in the normal range of human
perception. Although quantum theory was designed to deal
with the world of the very small, it is consistent with the
mechanics of the human-sized world. The new in each case
includes the old but has a wider range of validity, encompass-
ing the very large, the very small or the very fast. Similarly in
theology, the old orthodoxy will remain, but will now be seen
as part of a fuller truth valid over a wider range of time and
We have seen this happen already with the doctrine of
creation. Originally formulated in terms of the small world
of the Mediterranean basin and a primitive cosmology, it has
developed as our scienti¬c horizons have extended. It is now
an altogether greater thing and the God to whom it points is
thereby magni¬ed. The God whose space is measured in light
God and risk 235
years is immeasurably more magni¬cent than the tribal deity
of the early parts of the Old Testament.
Much the same could be said of evolution, though here
there are added complications arising from the resurgence of
creationism and the advent of the Intelligent Design move-
ment, whose claims were examined in chapter 7. The process
of understanding here is still growing and subject to lively
debate, especially in the United States of America.
Often new developments in human thought, such as the
evolution of life, have been seen as threatening to belief
because of what they appear to deny. Only as we become
accustomed to them and fully absorb their implications do we
begin to see that our understanding of God and his world has
been enlarged. The doctrine is not diminished but greatly
enriched. The new orthodoxy is richer, not poorer, as a

t h e t h e o lo g y o f r i s k
Can the threat to God™s sovereignty, apparently posed by
risk taking in all spheres, be treated in the same way as in
creation and evolution? Can a world in which chance seems so
threatening to our understanding of God really turn out to be
friendly to a more critical orthodoxy? If it turns out that we can
claim that the world of chance and theology are compatible,
can we go further and claim that our understanding of God
has been deepened and enriched by the new knowledge? In
the remainder of this chapter I hope to show, in outline at least,
that we can.
So we return to my opening question: ˜Is God a risk taker?™
In the light of my discussion this question has to be worded
more carefully because we have to distinguish between ulti-
mate goals and short-term deviations. We have seen that
God, Chance and Purpose
determinate ends may be achieved as the result of averaging
many random effects or by the interactions within the pro-
cess. This means that the end of the process may be virtually
certain, even though the path to that end is not determined. Is
it suf¬cient to preserve our understanding of God™s greatness
that he, as it were, gets there in the end “ or that he must never
put a foot wrong? It seems to me that it is the end that matters
and if deviations towards that end deliver side bene¬ts, the net
result may be gain.
To begin with creation, suppose someone claims “ as many


. 6
( 7)