Home » Library » Modern Library » Michael Martin Schlesinger

Michael Martin Schlesinger

[Editor’s Note: This article was originally published in International Journal for Philosophy of Religion 16 (1984): 257-262. The page numbers below show the position of the text within that pagination scheme.]



Does the Evidence Confirm Theism More Than Naturalism?

Michael Martin

Boston University


A positive answer is given by George Schlesinger in Religion and Scientific Method[1] to the question posed in the title of this paper. I will first outline Schlesinger’s argument for this answer and then show that his argument fails.

Schlesinger’s Argument

Schlesinger’s argument is based on Principle E, a principle he claims is an elementary principle underlying scientific method. This principle can be stated as follows:

Principle E: When a given piece of evidence E is more probable on H than on H’ then E confirms H more than H’.

According to Schlesinger, Principle E "should appear very reasonable to everyone."[2] It can be illustrated as follows: Let

hl = All ravens are black.

h2 = 99% of all ravens are black.

e = Of the five ravens hitherto observed all were black.

Since, according to Schlesinger:

p(e/h1) > p(e/h2)

then by Principle E

e confirms h1, more than h2.

Schlesinger is quick to point out that just because evidence E confirms H more than H’ it does not mean that we should subscribe to H rather than H’ since the initial credibility of H’ may be higher than H.[3]



The application of this principle to the confirmation of theism is as follows. According to Schlesinger, it is a widely accepted fact that

R: The laws of nature governing the Universe and the initial conditions in it are such that creatures capable of responding to the Divine are permitted to exist.

Let ‘T’ be the theistic hypothesis and ‘N’ be the naturalistic hypothesis. Now according to Schlesinger :

p(R,T) > p(R,N).

This, according to Schlesinger, is because if T is true then R must be true. However, if N is true, then R may or may not be true. N is compatible with a Universe in which no creatures could exist or if they exist, they would not be capable of responding to the Divine. Consequently by Principle E, R confirms T more than N.

This is basically Schlesinger’s argument why T is more confirmed than N. To be sure, Schlesinger goes on to try to argue that T is better confirmed than other alternatives to N using another elementary principle of scientific method, a principle he calls Principle A. But he seems to believe that it is "not very crucial"[5] that he succeed with these additional arguments. This is because, on his view, for the last few hundred years N has been taken to be the real threat to T and not these other alternative hypotheses.

Critique of the Argument

The first thing to notice about Schlesinger’s argument is that even if T is confirmed more than N by R, this does not mean that one should subscribe to T rather than to N. As we have noted, Schlesinger admits that the initial credibility of the hypothesis affects whether the hypothesis should be subscribed to. T may have less initial credibility than N. Indeed, some atheists have attempted to show that the concept of God is inconsistent or incoherent. If any of these arguments are successful, T may be initially less credible than N. Significantly, Schlesinger does not attempt to refute these arguments. Furthermore, even assuming the equal credibility of T and N, the question of whether T is confirmed more than N by the available evidence is not settled by showing that T is confirmed more than N by R.

What needs to be assessed is whether T is more confirmed than N by the total available evidence. Schlesinger makes no attempt to do this.

To be sure, he attempts to answer the plausible objection that the existence of evil counts against T.[7] But even if he is successful in this attempt,[8] other evidence not considered by Schlesinger may confirm N more than T, and this may cancel out any confirmation of T over N by R and any other evidence brought up by Schlesinger.



In the remainder of this paper I will bring up some evidence that seems to confirm N more than T if R confirms T more than N. I will also question whether R does confirm T more than N. I will show that Schlesinger’s argument does not establish that p(R,T) > p(R,N) since T per se does not entail R. I will also raise questions about Principle E, the principle on which Schlesinger’s entire argument is based.

Theism and Its Various Versions

According to Schlesinger, if T is true, then R must be true. No doubt one could understand T in such a way that makes this claim true. However, as theism is usually understood, it is simply the belief that a personal God exists. So understood, theism is contrasted with deism and polytheism. The mere existence of a personal God does not necessitate R. In order to derive R from the existence of a personal God, one must assume that this personal God has particular desires and powers. For example, one must at least assume all of the following:

(1) God wants some of His creatures to respond to Him.

(2) God has the power to create laws that make it possible for some creatures to respond to Him.

(3) God does not have other wants that conflict with His desire to have some creatures respond to Him.

If one builds (1), (2), and (3) into the concept of a personal God, it may be possible to argue that T necessitates R. Put in a slightly different way, T per se does not necessitate R. What is true is that a particular version of T necessitates R. Let us call this version T*. No doubt Schlesinger assumes T*.

However, it is also possible to interpret N in such a way that N necessitates R. Let us call this version of N N*. So construed, N* is no less confirmed by R than T* is confirmed by R. Thus, if Schlesinger can interpret T as T*, so can a naturalist interpret N as N* so that R is necessitated. To be sure, a naturalist who would interpret N as N* would be going beyond N as it is usually understood. But, a theist who interprets T as T* would be going beyond T as it is usually understood.

T* and Protective laws

However, let us grant that Schlesinger’s understanding of T as T* is not idiosyncratic, and let us also allow that N* is indeed an idiosyncratic understanding of N. Surely there may be other implications that can be drawn from T* that are not supported by the available evidence.

Consider: If God has the desire to create creatures that can respond to the Divine, and God is all powerful, and God does not have conflicting desires, and



so on, then God would create laws that would protect human beings from any natural disasters that would prevent human beings from responding to Him. For example, super novas, colliding comets and like events, events that are completely out of the control of God’s creatures, may well prevent human beings from responding to God since the human race may become extinct because of these events. Further, if God wanted to create creatures that could respond to Him, via the process of evolution, and God was all powerful, and so on, then He would create laws that would stop certain (perhaps fortuitous) events from preventing the evolutionary development of human beings. If not prevented, these events, completely outside the powers of God’s creatures, could disrupt the evolutionary process and prevent the development of any creatures capable of responding to the Divine.

Let us call such laws "protective laws" and let us call the hypothesis that such protective laws do not exist in the Universe R’. The evidence for R’ is strong. There is every reason to think that there are no laws protecting the extinction of the human race from natural disaster and no laws protecting the evolutionary process that led to the human species from being nipped in the bud. Indeed, Schlesinger himself refers to human beings as "precarious system."[9]

However, since it is not unreasonable to suppose that R’ is incompatible with T* and that R’ is compatible with N, then

p(R’,N) > p(R’,T*).

Consequently, by Principle E

R’ confirms N more than T*.

Since R and R’ pull in different directions in terms of their confirmation of T* over N, it is not implausible to suppose that the confirmatory powers of R and R’ cancel each other out in terms of T*. So even if R confirms T* more than N when R’ is taken into account, T* is not more confirmed than N.

The reasonableness of Principle E

So far we have uncritically accepted Principle E which is the foundation of Schlesinger’s argument. Without this principle his argument that T is confirmed more than N does not succeed. We are told that this principle should appear to be very reasonable. But no argument is given for this claim and it seems to be a dubious one in the light of the following counter examples. Let

h1 = All ravens are black because the Devil made them black.

h2 = 99% of all ravens are black.

e1 = Of the five ravens hitherto observed all were black.



Now according to Principle E, since

p(e1,h3) > p(e1,h2)

then el confirms h3 more than h2. But this seems absurd.

Schlesinger might attempt to get around this problem by suggesting that strictly speaking el does confirm h3 more than h2, but that h3 should not be subscribed to since it has a lower initial credibility than h2. But other counter examples to Principle E can be generated where this reply will not work. Let

h4 = All ravens are black and some roses are red.

Surely there is nothing initially less credible about h4 than h2. Nevertheless, since

p(e1,h4 > p(e1,h2)


el confirms h4 more than h2.

But this seems absurd.

I conclude that, despite Schlesinger’s assurances that Principle E should appear very reasonable, it is not a reasonable principle and should be rejected.


Schlesinger’s argument for the confirmation of Theism over Naturalism fails. Principle E on which it is based is unacceptable. But even if Principle E were acceptable the argument would not succeed, for other evidence, not considered by Schlesinger, pulls in the other direction cancelling out the confirmatory effect of the evidence he does cite. Further, granted Principle E, it is not clear that theism as it is usually understood makes the evidence he cites more probable than naturalism; consequently he is not justified in concluding that the evidence he cites confirms theism more than naturalism. Finally, Schlesinger fails to take into account the possibility that the initial credibility of theism may be less than naturalism since theism may be incoherent and naturalism may not be; he also fails to take into account the total available evidence relevant to the confirmation of theism and naturalism.




[1] George Schlesinger, Religion and Scientific Method (Dordrecht-Holland: D. Reidel Publishing Co., 1982).

[2] Ibid., p. 157.

[3] Ibid., pp. 157-160.

[4] See ibid., Chapter 23.

[5] Ibid., p. 185.

[6] See for example George H. Smith, Atheism: The Case Against God (Buffalo, New York: Prometheus Books, 1979).

[7] Schlesinger, Op.cit., Chapters 9 and 10.

[8] It is dubious that he is successful. See for example Peter H. Hare’s review of Religion and Scientific Method by George Schlesinger, Metaphilosophy 11, 1980, pp. 292-295.

[9] Schlesinger, Op.cit., p. 182.