|
Mechanism, like all materialist philosophies,
breaks down on the problem of knowledge. If thought is the undesigned and
irrelevant product of cerebral motions, what reason have we to trust it? C.S.
Lewis1
The man who represents all
thought as an accident of environment is simply smashing and discrediting
all his own thoughts – including that one. G.K. Chesterton2
The subject to be explored in this chapter is a complex one, and one
on which much more could be written than I am qualified to write. The issue is the rational value of a
certain argument that Lewis presented against Naturalism. We shall refer to
this argument as the argument from reason.3
This argument became the centre of a fascinating debate between Lewis and
another philosopher, G.E.M.
Anscombe (1919-2001). Many Lewis scholars have had something to say
about this debate, but the majority have taken only a biographical interest
in it, and have had little to say about the debate’s philosophical content.
Indeed, it has been widely assumed, almost without discussion, that
Anscombe conclusively refuted Lewis’ argument.4
As we shall see, Lewis admitted that
Anscombe had shown the argument must be either reformulated or abandoned.
However, Lewis clearly held his argument to contain an important insight
and subsequently rewrote the offending chapter of Miracles.
While I am uncertain about the cogency of the argument from
reason, it is not so easily rebutted as Anscombe and others seem to have
supposed. The argument comes in a variety of forms, and each one highlights
the existence of philosophical issues of great complexity. In offering this
argument against naturalism, Lewis revealed that he “had a nose for”
genuine philosophical problems, or in the words of Victor Reppert, that he
had “outstanding philosophical instincts”.5
To give the reader some idea of what is coming, we begin with
some historical notes on the arena of the Lewis-Anscombe encounter, the
Oxford Socratic Club, and with some observations on a later “re-run” of
that debate. Following this, I offer a definition of naturalism and attempt
to locate the philosophical “problem” on which the argument from reason is
based. Before we get to the substance of Lewis’ argument we outline a
variety of forms that the argument could take and develop a few of them a
little further. We then elucidate Lewis’ argument as it appeared in the
first edition of Miracles, and consider the objections that Anscombe
offered at the Socratic. The argument from reason is then further developed
along lines suggested by the work of William Hasker and Victor Reppert. In
the final section, the chapter considers the relevance of evolutionary
theory to the issue at hand.
Setting
the Scene
Between 1942 and 1954 Lewis was the President of the Oxford Socratic
Club. This club had “come into existence to apply [the Socratic] principle
[“follow the argument wherever it leads,”] to one particular subject matter
– the pros and cons of the Christian religion.”6
Christopher W. Mitchell writes
As a University Club, the
Socratic was a phenomenon. Meetings routinely had standing room only.
During the years Lewis was president, the Socratic entertained some of the
most influential atheists of the day, along with the weighty arguments they
brought against Christianity. As the Socratic’s point man, Lewis was relied
upon to represent the Christian position and to argue its case against the
opposition.7,8
It was at the Socratic, on February 2nd
1948, that C.S. Lewis encountered Elizabeth Anscombe. At the time, Anscombe
was a research fellow at Somerville College, Oxford. She went on to become
a professor at Cambridge.9 Anscombe
rejected the argument of Miracles chapter three, which Lewis had
originally entitled “The Self-Contradiction of the Naturalist.” As a Roman
Catholic, she was no defender of naturalism, but was convinced that it
could not be refuted in the manner Lewis had proposed. Following the
debate, Lewis admitted his argument contained a “really serious hitch” and
that Chapter three of Miracles “ought to be rewritten”.10
Indeed, Lewis took the opportunity to
rewrite this section when a new edition of Miracles was published in
1960. The chapter was re-titled “The Cardinal Difficulty of Naturalism”.
Although Anscombe evidently approved of these changes, her comments
indicate that she still found the argument unpersuasive.11
Before we move from history to philosophy,12 I should point out that on February 2nd
1967, exactly nineteen years after the original debate, the Socratic was
the forum of a sequel to the Lewis-Anscombe encounter. In this re-run,
Anscombe defended her original position while philosopher John Lucas
undertook to uphold Lewis’ side of the debate. The debate was entitled “Is
Mechanism Self-Refuting?” This is of particular interest because Basil
Mitchell, who succeeded Lewis as president of the Socratic (and remained
president until its final meeting in 1972) records that
on
that occasion, I think it would be generally agreed, Lucas succeeded in
sustaining Lewis’ side of the argument. If one were to think in terms of
winners or losers, I think maybe that Lucas was the winner on points …
Elizabeth and John agreed as to what the original Lewis-Anscombe debate had
been about, and Lucas simply maintained that on the substantial issue Lewis
was right and that, for the sort of reasons Lewis had put forward, a
thoroughly naturalistic philosophy was logically incoherent. And the
outcome of that debate was to make it perfectly clear that, at the very least,
Lewis’ original thesis was an entirely arguable philosophical thesis and as
defensible as most philosophical theses are. So there was no warrant for
supposing that in the original debate Lewis had been shown to be just
hopelessly wrong.13
What
is Naturalism?14
As Lewis himself noted, Naturalism is a difficult doctrine to
define. I will not attempt to give a cut and dry definition of it, but hope
rather to give the reader a good intuitive feel for the kind of things to
which the naturalist is committed.
First of all, naturalism, as we shall understand it, is a
doctrine about what kinds of thing exist, and the basic formula is that the
naturalist holds that only nature exists. This, of course, does not help us
very much as we are immediately led to ask what kinds of thing nature
includes. One popular way of defining naturalism is as the doctrine that
the only kinds of thing that exist are the kinds of thing whose existence
the sciences posit. While this may give us some help at an intuitive level,
it is not, unfortunately, an entirely satisfactory definition due to the
difficulty in distinguishing (at a philosophical level) science from
non-science and good science from bad.
We do a little better when we simply look for a common element
in those things whose reality the naturalist would deny. The naturalist
will, I think, wholly deny the existence of the things in list (a), and
will tend to struggle with the things in list (b).
(a) God, gods, spirits, ghosts, the soul (when
thought wholly distinct from the body)15,
the occurrence of miracles.
(b) Prophecy, ESP, answered prayer, telepathy, astrology.
There is plainly a common element to all these
things, but it is terribly difficult to say quite what it is. It seems to me,
however, that the problem that naturalists have which such things is that
they make persons, purposes and the mental too fundamental an element of
reality.16 In so far as a naturalist gives
any countenance to persons, purposes and the mental it is because they
believe they can be explained or understood in such a way as to make it
clear that non-purposeful realities are more fundamental.
There are various ways in which one kind
of thing could be “more fundamental” than another. Common to each of them is
the idea of dependence. The naturalist will want to say that things
ultimately possess the features that distinguish the personal, the
purposive, and the mental in virtue of possessing other features
that are not of these kinds. For example, in so far as the naturalist
accepts the reality of mental states at all, she will think each is really
just (constituted by) a certain kind of physical state. Naturalists will
also tend to reject the existence of anything that occupies neither space
nor time. This puts the naturalist is an odd position about mathematical
and logical truths, which do not seem to be made true by anything belonging
to space or time. The typical naturalist will think that such truths are
either true “by convention,”17 or are
really truths about the kind of thinking we would ourselves endorse, or …
well, you get the idea: in so far as the naturalist accepts such
non-spatio-temporal truths they think such truths are truths in virtue
of certain other truths that are not of that variety.
In short, the naturalist thinks that what we may call the fundamental
realities are impersonal, non-purposive, non-mental, temporal, and extended
in space. In so far as naturalists will countenance anything else they
believe such things are really nothing in addition to, and “exist” in
virtue of, the fundamental realities.
Naturalists are committed to saying that
none of the “non-fundamental” realities could exist in the absence of these
“fundamental” ones. Furthermore, any changes or differences in the “non-fundamental”
must be accompanied by (and in some sense due to) changes or
differences in the “fundamental”.18
For Lewis’ argument against naturalism,
the most important consequence of naturalism would be that one of the most
basic relations that can exist between two things is a causal
relation.19
Approaching the General Issue
The general issue that this chapter addresses is the relationship
between naturalism and our ability to think rationally. It would seem that
Lewis clearly believed the relationship to be one of some kind of
incompatibility, and produced various arguments to this affect. However, on
examining his writings the reader will be hard pressed to find a single
line of argument that encapsulates all his thinking about the relationship,
or lack thereof, between naturalism and our rational capacities. It is as
if Lewis saw as a single issue what is in fact a cluster of issues, and
that his writings are consequently somewhat confused on the point. After
carefully reading through as much of the relevant material as I can find, I
have discerned what seem to me several distinct lines of argument that
Lewis uses in the attempt to demonstrate some kind of incompatibility
between naturalism and (confidence in) our rational abilities. But before
we come to particular arguments, it is worth taking a step back in an
attempt to see the general picture that Lewis found so objectionable, and
which prompted his arguments.
While and undergraduate at University
College, Oxford, Lewis recorded in his diary that he was reading various
books and articles by the (in)famous Bertrand Russell. He makes particular
mention of Russell’s essay “The Free Man’s Worship”.20
Before we come to Lewis’ comments, it will be helpful to have some of
Russell’s remarks before us. Russell writes,
That man is the product of
causes which had no prevision of the end they were achieving; that his
origin, his growth, his hopes and fears, his loves and his beliefs, are but
the outcome of accidental collocations of atoms; that no fire, no heroism,
no intensity of thought and feeling, can preserve an individual life beyond
the grave; that all the labours of the ages, all the devotion, all the
inspiration, all the noonday brightness of human genius, are destined to
extinction in the vast heat death of the solar system, and that the whole
temple of Man’s achievement must inevitably be buried beneath the debris of
a universe in ruins – all these things, if not quite beyond dispute, are
yet so nearly certain that no philosophy which rejects them can hope to
stand. Only within the scaffolding of these truths, only on the firm
foundation of unyielding despair, can the soul’s habitation henceforth be
safely built.21
Russell clearly held that the right attitude in
the face of such a universe is stoicism: the universe may be uncaring,
without purpose, amoral and non-rational, but we need not be, indeed, we should
not be. We should, according to this early essay from Russell, endeavour to
live up to the ideals of virtue and reason.22
If we place along side these statements the following quote from another of
Russell’s papers, the grounds for Lewis’ own remarks are clearly revealed.
Man is a part of Nature, not
something contrasted with Nature. His thoughts and his bodily movements
follow the same laws that describe the motions of stars and atoms …
Undoubtedly we are part of nature, which has produced our desires, our
hopes and fears, in accordance with laws which the physicist is beginning
to discover. In this sense we are a part of nature, we are subordinated to nature,
the outcome of natural laws, and their victims in the long run.23
It
is with these thoughts in mind that Lewis records the following:
In his “Worship of a Free Man”
I found a very clear and noble statement of what I myself believed a few
years ago. But he does not face the real difficulty – that our ideals are
after all a natural product, facts with relation to all other facts, and
cannot survive the condemnation of the fact as a whole. The Promethean
attitude would be tenable only if we were really members of some other
whole outside the real whole: [which] we’re not.24,25
Lewis’s problem with Russell’s paper is that it attempts to hold
together the three thoughts that (a) the universe is fundamentally amoral
and non-rational, that (b) man is a creature capable of virtue and
rationality, (c) that virtue and rationality are the proper response to the
world, and that (d) man is a part of the amoral and non-rational universe.
These three thoughts, while perhaps not straight forwardly inconsistent,
certainly do not sit well together. But, as we saw in the previous section,
something like (a) and (d) are essential elements of naturalism, so it
would appear that naturalism “does not sit well” with the belief that we
are capable of rationality. It was, I believe, this sense of the lack of
fit between such theses as (a), (b), (c) and (d) that led C.S. Lewis to
formulate his various arguments against naturalism.
The Argument from Reason
As with so many important arguments, the argument
from reason comes in a variety of forms. It is, therefore, slightly
misleading to speak of the argument from reason. What all of these
arguments have in common is that they each begin with an assertion of the
following form.
Unless Px, no belief can be held for (good)
reasons.
The truth of such a claim is held to yield anti-naturalist
conclusions in one of two ways. It may be claimed that the truth of Px
is inconsistent with the truth of naturalism, or it may be held that the
best explanation of the truth of Px involves assumptions
incompatible with naturalism. These two kinds of argument may be called self-defeat
arguments and best explanation arguments. Formally, they would run
as follows.
Self-Defeat
Arguments from Reason
(SD1) Unless Px, no belief can
be held for (good) reasons.
(SD2) If naturalism is true, then Px
is false.
(SD3) Therefore, if naturalism is true no belief
can be held for (good) reasons.
(SD4) Therefore, if naturalism is true,
naturalism is not believed for (good) reasons.
(SD5) Therefore, either naturalism is not
believed for (good) reasons or naturalism is false.
Best-Explanation
Arguments from Reason
(BE1) Unless Px, no belief can
be held for (good) reasons.
(BE2) If Px is true, this fact
requires explanation.
(BE3) The best explanation of the truth of Px
is that some non-naturalist view is correct.
(BE4) Therefore, either no belief is held for
good reasons, or naturalism is probably false.
(BE5)
Therefore, naturalism is either (a) not believed for good reasons, or (b)
probably false.
If either form of argument is a success, if
either (SD5) or (BE5) can be supported, naturalism will be a prime
candidate for rejection. But can either of these arguments be sustained? If
we are to answer this question, we must begin with a search for possible
candidates for Px. It seems to me that if the human
activity we call “drawing an inference” is to be possible and if some of
those inferences are to be rational, at least the following conditions must
be met.
P1) States of mind are capable of
truth and falsity, which itself requires that they are capable of being
“about” things.
P2) Logical laws exist.
P3) We are capable of apprehending
logical laws.
P4) The apprehension of logical laws plays
an explanatory role in the acceptance of the conclusion of the argument as
true.
P5) The state of accepting the truth
of a proposition plays a crucial explanatory role in the production of
other beliefs, and propositional content is relevant to the playing of this
role.
P6) Our reasoning processes provide us
with a systematically reliable way of understanding the world around us.26
Intentionality and the Apprehension of Logical Laws: First Thoughts
Our discussion of the argument from reason will focus upon P4
to P6, but it will be worth saying a little about P1
to P3 before we move on. We begin at the beginning, then, with P1.
Certain mental states have what philosophers call
“intentionality”. That is, certain mental states are about, represent or refer
to other things. In other words, certain mental states have content.
Thoughts and beliefs are such states. Graham might have a belief (or
thought) about how tall someone is, say the belief that his best
friend is 5'9". The content of Graham’s belief (as with any other)
is specified in the “that clause” of this last sentence. Graham’s belief
will be true if, and only if, his best friend is indeed 5'9" in
height. If mental states were not capable of intentionality, not capable of
being about anything, then no mental state would be rightly thought
of as “true”. But that some mental states are true is clearly an essential
part of the concept of rationality and rational inference. How does this
feed into the argument from reason? C.S. Lewis clearly thought that the
phenomenon of intentionality causes problems for naturalism. He writes that
the admission that our thinking can be rational
rules
out any materialistic account of thinking. We are compelled to admit
between [our thoughts and the world] that particular relation we call
truth. But this relation has no meaning at all if we try to make it hold
between the matter in [our brains and that in the world]. The brain may be
in all sorts of relations to [the world] no doubt: it is in a spatial relation,
and a time relation, and a quantitative relation. But to talk of one bit of
matter being true about another bit seems to me to be nonsense.27
No model yet devised has made a satisfactory
unity between our actual experience of sensation or thought or emotion and
any available account of the corporeal processes which they are held to
involve. We experience, say, a chain of reason; thoughts, which are ‘about’
or ‘refer to’ something other than themselves, … but physical events, as
such, cannot in any intelligible sense be said to be ‘about’ or to ‘refer
to’ anything.28
Lewis is not alone here. Today’s philosophers are no less puzzled by
the phenomenon of intentionality than he was. Indeed, it is far from clear
that mental states exhibiting intentionality could possibly arise in a
world whose fundamental constituents are non-intentional. The main attempts
to understand intentionality within the naturalistic scheme endeavour to
reduce the intentional to functional, causal, or computational
interactions. But it is far from clear that such accounts can succeed. It
appears that whatever functional, causal or computational interactions are
present they will never be sufficient to generate intentionality, or at
least not sufficient to determine the content of any mental state that may
be present. I do not expect these brief comments to persuade the reader
that naturalism is inconsistent with P1; after all, I have only
sketched the outlines of the argument. The interested reader is encouraged
to turn to better-qualified authorities.29
With that, we move to a brief consideration of P2,
according to which the existence of logical laws is a pre-requisite for the
existence of rational inference. Firstly, we must ask what we mean by
“logical laws”. Although the standards by which our thinking is rightly
evaluated include canons of inductive as well as deductive reasoning, and
although both are relevant to the argument from reason, we shall be
focussing on the standards by which deductive reasoning is evaluated. It is
to these standards that the term “logical laws” refers. If no such
standards exist, then there is no way in which we can evaluate our various
inferences, and no one inferences can any better than any other. The
question then arises as to what these standards are and where they come
from.
One particular feature of these standards that seems to cause
problems for naturalism is that the standards have a modal status. That is
to say, some of these standards couldn’t have been other than they are;
they are necessarily true. Take the law of non-contradiction, for example.
The law doesn’t just say that no two contradictory beliefs actually are
both true, but that no two contradictory beliefs could both be true.
The law of non-contradiction could be stated thus:
(LNC) For any p, it is not the case that both p and not-p.
Our point then, is that (LNC) is not merely true, it is necessarily
true; it could not have been false. To see how this could cause problems
for naturalism consider the fact that any physical state of affairs could
have failed to obtain, the fact that all physical states of affairs are
contingent. Take any true statement you like, if that statement only refers
to things that the naturalist allows onto the “ground floor” of reality,
then that statement could have been false. For this reason, it would appear
that the naturalist cannot accommodate such necessary truths as (LNC).
After all, the truth (LNC) will be ultimately dependent upon ground floor
realities, and surely if those ground floor realities are contingent, then
so is anything that depends upon them – including the truth of (LNC). But
this is a conclusion we know to be false.
Even if this reasoning is flawed, we must ask in what way the
contingent realities generate necessary truths like (LNC). The particular
arrangement of the contingent realities certainly cannot logically imply
the necessity of our logical laws, for logical implication is itself a
modal notion. However, it is very hard to see how a necessary truth could
be a necessary truth if generated in any other fashion. If this is right,
the modal cannot be explained in non-modal terms, and since naturalism is
committed to thinking otherwise, naturalism cannot accommodate P2,
which seems to imply the existence of such modal realities.
One popular line among naturalists is to say that logical
laws arise out of the relations between our ideas or concepts. Our concepts
batchelor and unmarried are by themselves sufficient to
generate the necessity of all batchelors are unmarried, and
something similar is supposed to go for all other necessary truths.
However, once more we must ask how our concepts generate such necessary
truths. To put it another way, if logical laws arise out of the relations
between our concepts, we can ask: just what kind of relations does the
naturalist have in mind? If the answer is “logical relations”, it is clear
that he cannot have explained the existence of all laws of logic
naturalistically. But if any other answer is given, the explanation will
fail to explain.
Perhaps the naturalist could jettison the idea of necessity,
and hold that none of the standards by which we evaluate our beliefs are
necessarily true. This, however, will not help the naturalist to defend his
position. Take one line of argument that the naturalist often uses in
support of his view.
(N1) Using fewer and less problematic
assumptions, Naturalism is capable of explaining more features of the world
around us than are its competitors.
(N2) Any theory in this position ought to be
accepted.
(N3) Therefore, naturalism ought to be accepted.
The reader will have gathered by now that I question the truth of
(N1), but that is not the point at issue here. The point is that someone
who accepts the argument’s premises only need accept its conclusion if it
is impossible that the premises be true, while the conclusion false. That
is, the conclusion only need be accepted if one also accepts that (N4) is a
necessary truth.
(N4) If (N1) and (N2) are true, then (N3) is true.
But if we accept this as a necessary truth, then we have not taken
the naturalists advice in jettisoning necessity. In short, the naturalist
only has a right to think his position supported by valid arguments if he
allows the existence of necessary truths, but if he allows the existence of
necessary truths, he owes us some explanation of how these can be fitted
into his naturalistic scheme.
Again, I don’t expect the reader to be wholly convinced. The
issue is a technical one, on which I am no authority. Let it suffice to say
that the existence of mind-independent logical laws is not obviously
compatible with the truth of naturalism.
We now move to consider P3. According to this
principle, if we are capable of making rational inferences, we must be able
to apprehend the laws of logic. This is because if one is to make a
rational inference one must be able to apprehend the logical law with
reference to which that inference is rational. To rationally infer q
from the conjunction of p with if p then q, one must be aware
of the logical law according to which this conclusion follows from those
premises.30 Now, whatever account the
naturalist gives of the laws of logic, it cannot turn out to be (on
naturalistic assumptions) a complete mystery as to how we come to awareness
of those laws. It is evident that these laws are not confirmed (as logical
laws) by experience or experiment, so our knowledge of them seems to come
from some other source. But unless the laws of logic are mind-dependent, I
cannot see how (in a naturalistic scheme) we could possibly become aware of
them. On the other hand, if these laws are mind-dependent it is far from
clear that they can play the role we normally believe them to play. As
Lewis says, “Unless the measuring rod is independent of the things measured
we can do no measuring.”31 Unless the
standards by which our inferences are judged is independent of those
inferences, any “judging” that takes place will be a mere mockery. As with
P1 and P2, I don’t expect these comments to convince
the reader that P3 really is incompatible with naturalism. What
does seem evident is that all three represent prima facie problems
for the naturalistic world-view. Furthermore, these are not problems that
the naturalist can avoid simply by denouncing P1, P2
and/or P3, for in doing so she makes herself vulnerable to the
argument from reason.
C.S.
Lewis’ Initial Formulation of the Argument from Reason
We will approach our consideration of P4, P5
and P6 through C.S. Lewis’ various expositions of the argument
from reason. As explained above, Lewis offered the argument in a variety of
forms. We begin our thinking with the formulation of the argument that
Anscombe attacked. The following quotes should serve to give the reader a
good idea of how Lewis’ argument ran.
All
possible [inferred] knowledge … depends on the validity of reason. If the
feeling of certainty which we express by words like must be and therefore
and since is a real perception of how things are outside our own
minds really ‘must’ be, well and good. But if this certainty is merely a
feeling in our own minds and not a genuine insight into realities beyond
them – if it merely represents the way our minds happen to work – then we
can have no knowledge. Unless human reason is valid no science can be true.
It follows that no account of
the universe can be true unless that account leaves it possible for our
thinking to be a real insight. A theory which explained everything else in
the whole universe but which made it impossible to believe that our
thinking was valid, would be utterly out of court. For that theory would
itself have been reached by thinking, and if thinking is not valid that
theory would, of course, be itself demolished. It would have destroyed its
own credentials. …
We must believe in the
validity of rational thought, and we must not believe in anything
inconsistent with its validity. But we can believe in the validity of thought
only under certain conditions. Consider the following sentences. (1) ‘He
thinks that dog dangerous because he has often seen it muzzled and he has
noticed that messengers always try to avoid going to that house.’ (2) ‘He
thinks that dog dangerous because it is black and ever since he was bitten
by a black dog in childhood he has always been afraid of black dogs.’
Both
sentences explain why the man thinks as he does. But the one
explanation substantiates the value of his thought, the other wholly discredits
it. Why is it that to discover the cause of a thought sometimes damages its
credit and sometimes reinforces it? … The real difference is that in the
first instance the man’s belief is caused by something rational (by
argument from observed facts) while in the other it is caused by something
irrational (association of ideas).
We may
state it as a rule that no thought is valid if it can be fully explained
as the result of irrational causes. … Obviously, then, the whole
process of human thought, what we call reason is … valueless if it is the
result of irrational causes. Hence every theory of the universe which makes
the human mind a result of irrational causes is inadmissible, for it would
be a proof that there are no such things as proofs. Which is nonsense.
But Naturalism, as commonly
held, is precisely a theory of this sort. The mind, like every other
particular thing or event, is supposed to be simply a product of [physical
causes]. … And [physical causes are] not supposed to be rational. All
thoughts whatever are therefore the results of irrational causes.32
It
would be impossible to accept naturalism itself if we really and
consistently believed naturalism. For naturalism is a system of thought.
But for naturalism all thoughts are mere events with [non-rational,
physical] causes. It is, to me at any rate, impossible to regard the
thoughts which make up naturalism that way and, at the same time, to regard
them as a real insight into external reality. …
Every particular thought … is
always and by all men discounted the moment they believe that it can be
explained, without remainder, as the result of irrational causes. Whenever
you know what the other man is saying is wholly due to his complexes or to
a bit of bone pressing on his brain, you cease to attach any importance to
it. But if naturalism were true, then all thoughts whatever would be wholly
the result of irrational causes. Therefore, all thought would be equally
worthless. Therefore, naturalism is worthless.33
It seems clear that in these passages Lewis is endorsing some kind
of Self-Defeat Argument from Reason. In outline, the argument appears to
run like this:
(1) Naturalism is a system of thought.
(2) If naturalism is true all thoughts are
ultimately the result of certain irrational causes.
(3) No thought (and so no system of thought) can
be reasonable if it results from irrational causes.
(4) Therefore, if naturalism is true, the thought
that it is true is unreasonable.
(5) Therefore, naturalism is either untrue or
unreasonable.
(6) So, we ought to reject naturalism.
Anscombe’s Criticisms
In her “A Reply to Mr C.S. Lewis’s Argument that “Naturalism” is
Self-Refuting”, Anscombe offers at least three different criticisms of
Lewis’ argument. Firstly, Anscombe contends that Lewis’ argument trades on
a confusion of “irrational” with “non-rational” causes. Secondly, she
raises issues surrounding the scepticism (about our reasoning) that Lewis
thinks the naturalist cannot evade. Thirdly, Anscombe distinguishes various
senses of the word “because”, and claims that when different senses of the
word are in play the proffered explanations are not competitors.
Irrational
vs. Non-rational Causes
According to Anscombe, Lewis’ argument trades on a confusion between
irrational and non-rational causes. When a person’s entertaining a thought
is understood as a physical event, naturalism is of course committed to
saying that the event has causes like any other. He is not, however,
committed to saying that those causes are “irrational”. Antony Flew, in his
own response to Lewis, puts the point with characteristic clarity.
Lewis is too carefree in his
talk of “rational” and “irrational.” Why must atoms, or systems of neurons,
or whatever may be the terms of the scientific explanation of my mental
processes, be either rational or irrational? Can they not be just
non-rational – things to which the rational/irrational distinction does not
apply? Lewis would surely not say that atoms were immoral. But then, must
they be moral? Of course not. Lewis would say that the distinction does not
apply to the sort of things in terms of which “naturalists” would give
their causal explanations of mental processes. But since atoms are neither
rational nor irrational, the argument breaks down, for the causes by which
the “naturalist” explains his own thinking are no longer irrational and the
“naturalist” thesis no longer refutes itself.34
Lewis granted the irrational/non-rational
distinction, as is evident from the changes he made to the revised edition
of Miracles. We shall return to the issue shortly, for now we simply
reformulate the argument from reason as follows.
(1) Naturalism is a system of thought.
(2') If naturalism is true all thoughts are
ultimately the result of certain non-rational causes.
(3') No thought (and so no system of thought) can
be reasonable if it results from non-rational causes.
(4) Therefore, if naturalism is true, the thought
that it is true is unreasonable.
(5) Therefore, naturalism is either untrue or
unreasonable.
(6) So, we ought to reject naturalism.
The
Threat of Scepticism
The second line of response to Lewis focuses on Lewis’ claim that if
naturalism is true then no human reasoning is “valid”. Anscombe suggested
that this latter claim is without meaning. One can meaningfully assert that
a particular piece of reasoning is valid or invalid but, Anscombe
contended, one cannot meaningfully make either claim about human reasoning
in general. This is because we only acquire of concepts of validity and
invalidity though experience of particular instances of reasoning of both
kinds.
If this is correct, we cannot have the concept of invalidity
without having first encountered at least one instance of valid reasoning
(and one instance of invalid reasoning). But if we have encountered one
instance of valid reasoning, then it cannot be true that all human
reasoning is invalid. From these considerations, Anscombe concluded that
Lewis’ argument cannot succeed unless we have reason to think that we
cannot even have the concept of validity in the naturalist’s world. The
defender of the argument from reason might at this point suggest that
naturalism does indeed entail that we cannot have the concept of validity.
But Anscombe has a ready response: all talk about “not having the concept
of validity” must be incoherent, for if we did not have the concept we
would not know what someone was talking about when they make that
statement. Such talk is either unintelligible or simply false.
This, however, does not settle the matter. Putting this
objection alongside the argument from reason, to which of the premise(s) is
Anscombe objecting here? It would appear that she is objecting to an
unstated entailment of (2') and (3'): if naturalism is true then none of
our thinking is reasonable (or “valid”). She might, therefore, be
interpreted as claiming that the conjunction of (2') and (3') is itself
unintelligible. This, however, seems wildly implausible. In any case, we
needn’t worry too much about this criticism of Lewis’ argument, for the most
that Anscombe’s argument shows is that complete scepticism about our
reasoning abilities is incoherent. Our argument can be restated so as to
avoid these problems by reformulating the argument so as to be arguing that
naturalism itself cannot be meaningfully asserted, because if it could it
would itself entail something that cannot: that none of our thinking is
reasonable (or valid).
There is, however, another related issue that has bothered
some commentators on the argument from reason. One way of presenting a
“self-defeat” argument might run as follows.
(ST1) Unless we have a convincing response to philosophical
scepticism, we cannot know anything.
(ST2) If naturalism is true, we have no such response to scepticism.
(ST3) Therefore,
if naturalism is true, its truth can never be known.
For obvious reasons, Victor Reppert labels arguments like this
“Sceptical Threat Arguments”. This kind of argument would proceed by
raising “sceptical doubts about the validity of reasoning, and then [go on]
to argue that such doubts can be resolved only if naturalism is denied.” As
Reppert observes, such arguments will be objected to “by many people in
contemporary philosophy on the grounds that no absolute security against
such doubts is available from any quarter, and that even if it were it is
not needed.” He goes on to point out that neither Theism nor any other
kinds of anti-naturalism provides any more security against these doubts
than does naturalism.35
But not all arguments from reason (or even all Self-Defeat
arguments from Reason) are sceptical threat arguments. On the contrary, the
argument from reason may begin by “assuming that validity is an established
fact” and asking “whether, in a naturalistic world, one can account
for the fact that it is valid.”36
Non-Competing
Explanations
Anscombe nowhere clearly expresses her third objection, though it
should be obvious to anyone reading her paper just what that objection is.
She reconstructs Lewis’ argument in the following passage.
You argue that the naturalist
hypothesis about human thinking implies that no human thinking is rational
… For if a man produces what purports to be the conclusion of an argument,
in order that what he says should be rational he must say it because
he has reasoned; but the naturalist hypothesis say that he says it because
of certain natural causes; and if these causes fully explain his
utterance, if the chain of causes is complete, there is no room for the
operation of such a cause as the man’s own reasoning.37
What Anscombe objects to here is the idea that there being a “full”
explanation of an event in causal terms entails that there is “no room” for
any other kind of explanation. She argues that there are many kinds of
explanation: causal, historical, logical, and psychological to name just
four. According to Anscombe, if two explanations are to compete with one
another (so that if one applies there is “no room” for the other), they
must belong to the same kind. In particular, a causal explanation does not compete
with a logical or psychological one. Flew makes the same point.
Lewis and others who produce
similar arguments are snared by the chronic ambiguities of words like
“cause,” “reason,” “because.” If asked “What is the reason why you think
this is true?” I may reasonably answer either “It was thrashed into me at
school,” or “It follows from such and such true premises.” Both these
answers simultaneously may be sound, for they are answers to what are
really quite different questions. I shall call the senses of “reason,”
“cause,” etc., which ask for the first type of answer the historical
senses … , and shall call the senses which ask for the second type of
answer the logical senses … If the reason (historical) why I think
my mental processes are determined by neurone changes is itself something
to do with neurone changes, this has no necessary bearing on the questions
whether there are, or whether I have, any logical reasons, any good
arguments, for thinking this thought about the causation of my mental processes.38
This, I think, represents the most significant of the challenges
that Anscombe (and Flew) put to Lewis’ argument against naturalism. From
the revisions that Lewis made to his argument, it seems obvious that he
would have agreed.
Continuing
the Debate
In fact, the beginnings of a response appear in a note from C.S.
Lewis that accompanied the Anscombe paper when it first appeared in The
Socratic Digest, No. 4 (1948)39. The
important part of that note runs as follows.
I admit … that the cause and effect
relation between events and the ground and consequent relations between
propositions are distinct. Since English uses because of both, let
us here use Because CE for the cause and effect relation (‘This doll
always falls on its feet because CE its feet are weighted’), and Because
GC for the ground and consequent relation (‘A equals C because GC
they both equal B’). But the sharper this distinction becomes the more my
difficulty increases. If an argument is to be verific the conclusion must
be related to the premises as consequent to a ground, i.e. the conclusion
is there because GC certain other propositions are true. On the
other hand, our thinking the conclusion is an event and must be related to
previous events as effect to cause, i.e. this act of thinking must occur because
CE previous events have occurred. It would seem, therefore, that we never
think the conclusion because GC it is the consequent of its grounds
but only because CE certain previous events have happened. If so, it
does not seem that the GC sequence makes us more likely to think the true
conclusion than not. And this is very much what I meant by the difficulty
in Naturalism.40
In the revision of Miracles, Lewis develops this line of
thinking in two ways. Firstly, he implies that – on the assumption of
naturalism – the fact that these two because relations are completely
distinct suggests that it would be a massive coincidence if the two systems
happened to reliably run together in our thinking. It is important to note
that unless such a coincidence is literally incredible, this line of
thinking will lead us to endorse a best-explanation form of the argument
from reason. Secondly, in a passage we shall refer to as the central
passage, the distinction in question leads Lewis to ask:
But even if grounds do exist,
what exactly have they got to do with the actual occurrence of the belief
as a psychological event? If it is an event it must be caused. It must in
fact be simply one link in a causal chain which stretches back to the
beginning and forward to the end of time. How could such a trifle as lack
of logical grounds prevent the belief’s occurrence or how could the
existence of grounds promote it?41
Lewis’ own suggestion on the naturalist’s behalf
is to say that one belief may cause another by being seen to be a ground
for it.42 Here we are in familiar
territory, for one thought can only be a logical ground for another if
logical laws exist, and one thought can only be seen to be a logical ground
for another if humans are capable of apprehending logical laws. In other
words, this response is only open to the naturalist if they can accommodate
both P2 and P3. If the naturalist wishes to make use
of Lewis’ suggestion, it would surely be incumbent upon them to explain how
this could be fitted into the naturalistic scheme.
We begin our discussion of these points from Lewis with a
brief consideration of the claim that if naturalism is to accommodate the
thought that our beliefs are regularly in accord with reason, he must posit
an incredible coincidence. In his response to Flew, Ernest Gellner puts
that problem well.
Now
from the proposition that there are always causes (which is what
[naturalism] amounts to) it does indeed not follow that there are no reasons
or no valid causes; this makes Flew think he has established his case. But
not at all, for some of those whom he is opposing are not arguing that from
the presence of causes it follows that there never are reasons, but merely
that if causes are present, their overlap with reasons … is entirely
fortuitous. This is by no means undermined by Flew’s distinction and
his insistence that there is no necessary connection (above all,
negatively) between causes and reason; on the contrary, Flew’s central premises
is the firmest support of the view he is attacking. …
The point can be put thus: if [naturalism] is
true, then it is always a mere coincidence that what we believe is also
true, that the reason we in fact follow also corresponds to valid modes of
reason.43
The response from Flew was inevitable
[A]ll other things being equal
and in the long run and with many dramatic exceptions, true beliefs about
our environment tend to have some survival value. So it looks as if
evolutionary biology and human history could provide some reasons for
saying that it need no be a mere coincidence if a significant proportion of
men’s beliefs about their environment are in face true. Simply because if
that were not so they could not have survived long in that environment. As
an analysis of the meaning of ‘truth’ the pragmatist idea that a true
belief is one which is somehow advantageous to have will not do at all. Yet
there is at least some contingent and non-coincidental connection between
true beliefs, on the one hand, and the advantage, if it be an advantage, of
survival, on the other.44
Lewis evidently saw the possibility of such a response, but thought
that it begged the question. He claimed that it amounted to the naturalist
arguing for the reliability of his cognitive faculties, and that if those
faculties really are in doubt such an argument should not persuade us.45 But this seems like a poor response to Flew.
Flew is not trying to remove doubts about our cognitive faculties, he is
attempting to stop those doubts from arising in the first place.46 The question, then, is whether Flew’s doubt
preventing strategy is successful. In this context, it will be successful
if evolution can explain why a person who has good inferential habits is
more likely to survive than is someone who has bad inferential habits.
Although it may seem natural to suppose that evolution can explain this,
I’m not wholly convinced of this. We will return to consider this issue
further in the next section, on the general relationship between
evolutionary theory and the argument from reason.
There is a popular illustration associated with the last
objection from Anscombe and Flew. The illustration is that of the computer.
The operations of computers, it is suggested, are fully explicable in
naturalistic terms, and yet a computer is more than capable of performing
calculations and inferences according to the rules of mathematics and
logic. This, it is claimed, shows that the two systems of relation can both
apply to the same series of events … they are not incompatible. And if the
two systems are not incompatible, then Lewis’ argument fails.
This, however, moves far too quickly. If Anscombe and Flew
have interpreted Lewis correctly, and if any argument from reason must
proceed in the same fashion, then computer illustration may be sufficient
to undermine those arguments. It seems to me, however, that the Lewisian
argument has a little more to it than we have yet seen. One natural way of
reading Anscombe and Flew is as claiming that if a person can, when asked, adduce
good reasons in support of a belief that he holds, then we can pronounce
that belief is rational no matter what its causal history. If this is how
Anscombe and Flew were arguing, then I think they were mistaken.47
This is because such an account of what it means to hold a
belief rationally allows for no distinction between reasons for holding a
belief and rationalisations of that belief. Suppose that chancy Charlie
decides what to believe on a certain subject through a game of chance (by
associating the various positions that might be held on the subject with
the different possible outcomes of the game). Even though Charlie, being an
intelligent and creative fellow, can produce formidable arguments for the
position he adopts he surely does not hold that belief rationally. The
problem here is that the reasons that Charlie offers in support of his
belief are not really his reasons. To count as his reasons
those reasons must at least partially explain why Charlie believes
as he does. That is to say, the reasons are to justify Charlie’s belief
those reasons must be part of what brings it about that Charlie
believes as he does.48 It may have been
something like this concern that Lewis was voicing in the central
passage, quoted above. What does it take for a person’s reasons to be a
part of what brings it about that they believe as they do? It seems to me
to take, at least, the truth of both P4 and P5, which
to remind the reader, ran as follows.
P4) The apprehension of logical laws
plays an explanatory role in the acceptance of the conclusion of the
argument as true.
P5) The state of accepting the truth of a proposition
plays a crucial explanatory role in the production of other beliefs, and
propositional content is relevant to the playing of this role.
To help us in our thinking about the relationship between naturalism
and these two maxims, it will be worth doing a little more thinking about
naturalism itself. Firstly, we will distinguish between two different forms
of naturalism, and secondly we shall consider the naturalistic
understanding of the laws of nature.
Two Forms of Naturalism in the Philosophy of Mind
The naturalistic positions we want to distinguish
will be referred to as Reductive and Supervenient Naturalism.
Reductive naturalism holds that mental states (events), such as beliefs,
are identical to certain physical states (events) – usually to states
(events) that obtain in the brain of the person who enjoys that mental
state. Unlike Reductive naturalism, Supervenient naturalism holds that
mental and physical realities exist, as it were, on two levels. On the
ground floor, we have the physical realities, on the first floor the
mental. This counts as a kind of naturalism because according to
supervenient naturalism there is an asymmetric dependence relation between
the ground floor and the first floor. The psychological situation depends
upon the physical one, and there can be no psychological differences
between situations that are physically identical.49
For our purposes, this distinction between two varieties of naturalism will
simply mean that we will occasionally have to reword our arguments so as to
apply to both positions.50 While my
discussion could proceed perfectly well without reference to these two
positions, keeping them in view will help us to relate Lewis’ argument to
current philosophical concerns.
Naturalism and the Laws of Nature
According to naturalism of all stripes, the physical realm governed
by the laws of nature. Such laws govern all causal interactions. Each of these
laws may be expressed in something like the following form.
Under conditions C, entities (or systems) of kind K, exhibit
behaviour B.51
These laws fall into two kinds: (i) basic laws and (ii) derived laws.
The derived laws are as they are because the basic laws are as they are.
The basic laws simply are as they are, and their being that way cannot be
explained. To illustrate, there are laws that govern the refraction of
light as it passes from one medium to another. From these laws (and perhaps
various others besides) we could produce a law about the conditions
necessary for the production of rainbows. The laws about the refraction of
light are “more basic” than the laws about rainbows.
In keeping with our earlier definition of naturalism, it
seems to me that if naturalism is true the most basic laws will only refer
to physical entities, systems and properties. That is to say, even if some
physical states turn out to be identical with certain mental states, in so
far as the basic laws refer to such states they will refer to them under
their physical (and not their mental) descriptions. If there are any laws
governing mental states as mental states, these must be derived
laws. Returning to the schema above, in the most basic laws the
placeholders C, K and B must be filled out by terms that refer to only
physical realities (and refer to them under physical descriptions).
With this understanding in place, we may offer the following
argument against reductive naturalism. Reductive naturalism clearly allows
that one mental state can cause another. This is because physical states
can cause one another, and according to reductive naturalism mental states
just are physical states. No mystery there. But suppose that state s1
causes state s2, and in fact both of these states are mental
events, beliefs say. If our animadversions on the laws of nature were
accurate, then the reason that s1 causes s2 has
nothing to do with the fact that these states are beliefs with particular
content. That s1 causes s2 is fixed by the purely
physical properties of those two states. Consider the following from
eminent philosopher of mind, Jaegwon Kim
[Reductive naturalism] fails
to do full justice to psychophysical causation in which the mental qua
mental has any real causal role to play. … [W]hether or not a given event
has a mental description (optional reading: whether it has any mental
characteristic) seems entirely irrelevant to what causal relations it
enters into. Its causal powers are wholly determined by the physical
description or characterstic that holds for it; for it is [first and
foremost] under its physical description that it may be subsumed under a
causal law.52
The same argument can be deployed, mutatis mutandis, against supervenient
naturalism. If laws that are blind to those supervening states govern the
physical states on which the mental supervenes, then the existence of the
mental is irrelevant to what happens on the physical level. Furthermore, it
is hard to see how the mental states themselves can play any causal role at
all. If one mental state, m1, supervenes on the physical state s1
and a second, m2, upon on s2, the fact that s1
causes s2 does nothing to produce any further causal relation
between m1 and m2. On supervenient naturalism, then,
everything that exists on the “first floor” seems to be absent any power to
influence the actual course of events. If this is right, then neither form
of naturalism can accommodate P4 or P5.53
Evolution
and the Argument from Reason
If the argument of the last section is correct, the ramifications
for Flew’s response to Gellner are enormous. There we concluded that on the
assumption of naturalism the existence of mental states makes no difference
to the course of events. But if this is so, then evolution will be
completely blind to their existence. The result will be that evolution
cannot “select for” organisms because they have mental states, and more
importantly, cannot select for them because their mental states accurately
represent the world. In short, if the argument of the previous section is
sound, evolutionary theory cannot be used to combat the argument from
reason, and if evolution cannot be used in this way then the “coincidence”
version of the argument from reason remains undefeated.54
While I think this defence of the argument from reason has
much to be said for it, it will be interesting to see how far we can get
while allowing the naturalist to assume that mental states do effect the
course of events. To this end, we
will investigate the relevance of evolution a little further. By way of
recap, then, the “coincidence” version of the argument from reason, could
be expressed like this …
When logic says a thing must
be so, Nature always agrees. No one can suppose that this can be due to a
happy coincidence. A great many people think that it is due to the fact
that Nature produced the mind. But on the assumption that Nature is herself
mindless this provides no explanation. To be the result of a series of
mindless events is one thing: to be a kind of plan or true account of the
laws according to which those mindless events arose is quite another. Thus
the Gulf Stream produces all sorts of results: for instance, the
temperature of the Irish Sea. What it does not produce is maps of the Gulf
Stream. But if logic, as we find it operative in our own minds, is really a
result of mindless nature, then it is a result as improbable as that. The
laws whereby logic obliges us to think turn out to be the laws according to
which every event in space and time must happen. The man who thinks this an
ordinary or probable result does not really understand. It is as if
cabbages, in addition to resulting from the laws of botany also gave
lectures in that subject: or as if, when I knocked out my pipe, the ashes
arranged themselves into letters which read: ‘We are the ashes of a
knocked-out pipe.’ But if the validity of knowledge cannot be explained
that way, and if perpetual happy coincidence throughout the whole of
recorded time is out of the question, then surely we must seek the real
explanation elsewhere.55
The evolutionary response has it that creatures inveterately wrong
in their inferences “have a pathetic but praiseworthy tendency to die
before reproducing their kind.”56
Cognitive faculties that enable us to reliably make (deductively and
inductively) valid inferences, are more conducive to survival than are
faculties that encourage faulty reasoning.
However, in so far as truth in beliefs and validity in
reasoning are things that selection pressures will encourage, this is only
because these things are instrumentally connected with other more obviously
relevant features of the organisms in question. The evolutionary story is
only directly concerned with features of organisms that in some way
contribute to the “fitness” of that organism. To be fit in this sense, an
organism must be well adapted to survive and reproduce in its environment.
For evolution, then, the value of truth in belief and of validity in
reasoning is entirely instrumental. This creates problems for the
evolutionary response in a variety of ways. The most important of these is
best approached somewhat obliquely.
Philosophers, and especially philosophers of science, have
long asserted that theory is underdetermined by data. What they mean by
this assertion is that given any amount of information about the way the
world is, there will always be more than one way to account for that
information. That is, there will always be more that one theory that will
accommodate any given set of data. In fact, given that there will always be
more than one such theory, it follows that there will always be an infinite
number of theories that will accommodate our data.
For instance, the theory that the Earth is flat is not conclusively
refuted by observations from space which seem to indicate otherwise. The
“flat-earther” can always put this appearance down to, say, distortions
caused by the upper atmosphere. Indeed, the “flat-earther” could find ways
to accommodate any conceivable piece of evidence against his theory. (This
is not to say that his theory is rationally acceptable, but only that its
rejection is based on something more than “the data”.)57
The under-determination of theory by data has a pleasant illustration
within mathematics, in which data is symbolised by points plotted on a
graph and theories by equations whose respective lines run through those
points. The theorem runs thus: for any finite number of points plotted on a
graph, there will always be an infinite number of equations whose
corresponding lines, when drawn on the graph, would run through those
points.58
From the under-determination of theory by data, several important
thinkers have concluded that any preference for one theory over another
that accommodates all the same data must be either irrational or purely
pragmatic. This, it need hardly be said, cannot be the position of the
naturalist. If naturalism is to be defended by rational argument and not
endorsed merely as a useful hypothesis but as true then there must be some
reliable method of choosing between two competing theories each of which
accommodates the data we possess. Not only this, but we must be able to
account for the reliability of these methods within the naturalistic scheme.
Christopher Hookway puts the issue well.
[S]ince there is no limit to
the number of hypotheses that can fit a given body of data, what reason
have we to suppose that we are capable of producing, and finding plausible,
an hypothesis that is on the right lines?59
Hookway goes on to point out that if the success
of our theorising is not to be attributed to mere luck, then we must
suppose that “there is, in any particular case, an affinity between our
sense of plausibility and the nature of reality.”60
In chapter 13 of Miracles, “On Probability,” Lewis makes a similar
assertion, stating that our judgement of the probability of a claim often
hinges on “some innate sense of the fitness of things.”61 To what extent will evolutionary theory
reassure us that the “sense of fitness” is a reliable guide? Could
evolution underwrite a real “affinity between our sense of plausibility and
the nature of reality”? Hookway continues
that a faculty was necessary
for the commonsense inquiries which facilitate survival and reproduction is
no guarantee that it will help us to describe reality. Science has no
survival value, and we have to rely upon our sense of plausibility in areas
remote from the vital concerns of everyday practice.62
Supposing, contrary to our earlier argument, that
evolutionary theory can assume the causal relevance of humans having
certain beliefs, it would appear that at best that theory can only explain
why our cognitive faculties are apt to yield reliable conclusions on
subjects closely connected with our everyday concerns. It seems hard to see
how accepting a true scientific or metaphysical theory could significantly
affect one’s chances of survival, for the acceptance of such a theory has
only a minimal impact on our behaviour. But if the acceptance of true
scientific or metaphysical theories does us little evolutionary good,
neither do faculties tuned to enable us to reach such conclusions.
To make the point slightly more concrete, consider our preference
for simple theories over complex ones. A neat example of this is that given
points plotted at (x,y) = (1,1), (2,2), (3,3) and (4,4), the plot of either
of the following formulas will run through these points.
y = x
y = (x-1)(x-2)(x-3)(x-4) + x
If we allow the plotted points to represent our “data” and the
formulas our “theories”, then it is clear that our choice of theory is
underdetermined the data. Nevertheless, if this data is all we have to go
on, it is obvious that we ought to prefer the first of these theories. Our
preference for that theory is due to its simplicity. It would seem then,
that in deciding between theories we take it that, other things being
equal, a simple theory is more likely to be true than a complicated one. It
is, however, very difficult to see why simplicity should be an indicator of
truth. On the other hand, arising from the fact that they are easier to
work with there are obvious, and evolutionarily relevant, pragmatic
benefits to preferring simple theories over complex ones.63
These considerations, then, may be used to sustain a
best-explanation argument from reason based upon P6. According
to Alvin Plantinga, this kind of thinking can also be used to bolster a
self-defeat argument. His argument is that if a hypothesis about the
origins or provenance of our cognitive faculties confers a low probability
on the proposition that those faculties are reliable, then adherence to
that hypothesis renders belief in the reliability of our faculties
irrational (and visa versa). Using arguments not unlike those above,
Plantinga contends that evolutionary naturalism is just such a hypothesis.64 Many objections to this argument have been
voiced, but it is far from clear that any of these objections are
successful. Whatever we think about Plantinga’s argument, and I shall not
be evaluating it here, we should surely be sceptical of the evolutionary
response to the argument from reason.
Conclusion
While the argument is ostensibly an argument against naturalism, if
naturalism is considered the most plausible variety of atheism, the
argument will – if successful – also offer support for theism. While unsure
about just how to evaluate the arguments presented in this chapter, the
argument from reason is not easily dismissed. Thomas Nagel neatly
summarises the worry for naturalism:
[T]he idea of a natural
sympathy between the deepest truths of nature and the deepest truths of the
human mind, which can be exploited to allow gradual development of a truer
and truer conception of reality, makes us more at home in the
universe than is secularly comfortable.65
As a final offering, I ask you, the reader, to consider the fact
that if naturalism is true, then not only are all our thoughts fully
explicable by the operation of non-rational causes … all your thoughts
are explicable in that manner. Indeed, if naturalism is true, the very
thoughts you’ve had while reading this chapter can be so explained. No doubt,
the shape of the marks on these pages plays an important role in this
causal story, but so too does the physical constitution of your cognitive
faculties and various other things besides. While in a moment of
abstraction I can nearly bring myself to think the naturalists causal
explanation of a person’s thinking consistent with a reasons based explanation,
I cannot but agree with C.S. Lewis, that “it is, to me at any rate,
impossible” to regard my own thinking that way and at the same time,
to regard it “as a real insight into external reality.”66
Notes
7. Christopher W. Mitchell, “University Battles: C.S.
Lewis and the Oxford University Socratic Club” p. 330, in Angus J.L. Menuge
(ed.), C.S. Lewis: Lightbearer in the Shadowlands
(Wheaton, Illinois: Crossway Books, 1997).
8. The famous philosophers who spoke at the Socratic
while Lewis was president included the following: G.E.M. Anscombe, J.L.
Austin, A.J. Ayer, F.C. Copleston, Michael Dummett, A | 
