In this post, I want to revisit an argument from silence used by Victor Stenger against the existence of God based on the absence of scientific evidence for God.

In his 2010 debate with William Lane Craig, Stenger argued that “the absence of evidence for God is evidence of absence” of God. In his words, “If God plays such an active role in the universe, then his actions should surely have been observed by now.” As I understood him, he offered four examples of scientific evidence which could have been found but is lacking. I summarize those four examples below in the Prediction Table below.

Topic	Theistic Prediction	Naturalistic Prediction	Observation(s)
Revelation	God talks to people and occasionally provides information which they could not have known any other way.	Not one verified revelation from God.	Not one verified revelation from God.
Prayer	God occasionally answers prayers.	No scientific confirmation of answered prayers.	No scientific confirmation of answered prayers. No evidence that prayer has any benefit whatsoever on health.
Design	Life shows evidence of design.	Life does not show evidence of design.	Science has shown how order can arise from complexity — Darwin’s theory of evolution by natural selection. Biological life is imperfect and contains useless DNA. Hence, it does not show evidence of design; instead, it looks just like what we would expect from an unguided evolution.
Cosmology	The universe should have possessed some degree of order at the moment of creation.	If the universe has a beginning, it should have begun with no structure or organization.	The universe “began with complete chaos.”

Stenger’s Formulation of His Argument

Following Ted Drange‘s “Lack of Evidence Argument” (LEA), Stenger explicitly formulated his argument in his book, God: The Failed Hypothesis (p. 22), as follows.

(1) Probably, if God were to exist, then there would be good objective evidence for his existence.

(2) But there is no good objective evidence for His existence.

(3) Therefore, probably, God does not exist.

In his 2010 debate with Craig, Stenger states that this argument shows “beyond a reasonable doubt” (!) that God does not exist.

I don’t think Stenger manages to substantiate either (1) or (2), much less establish them “beyond a reasonable doubt.” Consider (1). What reason is there to believe (1) is true? One reason is that “God wants us to believe in Him and so would provide good objective evidence of His existence.” But it’s far from obvious that reason is correct. Why couldn’t God use other methods, such as religious experience? According to Stenger, a deity who did so “would not be perfectly loving” (p. 22). But this answer just misses the distinction between, on the one hand, God’s goal or end (belief in Him), and, on the other hand, God’s method or means for bringing about that goal (such as objective evidence, subjective evidence, and so forth). If God makes Himself known to humans through religious experiences, why would that fact call into question whether God is “perfectly loving”? Stenger never says.

As for (2), I have two comments. First, this premise is probably false given a common definition of “evidence” known as the positive relevance view of evidence.  To say that evidence E favors hypothesis H1 over H2 is just to say that we have more reason to expect E on H1 than we do on H2. Notice that this definition of evidence does not require that E greatly favor H1 over H2; all that is required is that E is at least very slightly more probable on H1 than on H2. With such a weak threshold for evidence, however, it should not be surprising to find evidence both for and against God’s existence.

Second, not only can there be evidence both for and against God’s existence, but there can be evidence both for and against God’s existence within the same domain or discipline. Thus, broad, sweeping statements (such as “There is no evidence whatsoever that prayer has any benefit on health”) are riskier than narrower, more precise statements (such as “There is no evidence whatsoever that prayer has ever healed an amputee”). Why make the riskier claim if it’s unnecessary? (As an aside, I should mention that I find Stenger’s sweeping claim about prayer to be counter-intuitive: even on the assumption that metaphysical naturalism is true, we would expect prayer to have at least some benefit on health, however small, not because prayers are answered by God but because prayers can induce a placebo effect.)

I conclude, therefore, that the LEA is unsound. This does not mean, however, that Stenger’s points have no evidential value whatsoever. Stenger could greatly strengthen his case by focusing his evidence statements on what we do know, rather than what we don’t know, as summarized below in Table 2.

Topic	Original Evidence Statement	Revised Evidence Statement
Revelation	Not one verified revelation from God.	1. Many people never receive a divine revelation. Those who do almost always have a prior belief in God or extensive exposure to a theistic religion. 2. The content of alleged divine revelations are notoriously contradictory. 3. The accuracy and informativeness of alleged revelations is just what we would expect on the null hypothesis and no such revelations are actual revelations from God.
Prayer	No scientific confirmation of answered prayers.	1. So much in medical science is intelligible without any appeal to answered prayers or other forms of supernatural agency. 2. The history of medicine contains no examples of answered prayers replacing a naturalistic explanation for healing.
Biology	Life does not show evidence of design.	1. All known life forms are the gradually modified descendants of earlier organisms. (Descent with modification). 2. Pain and pleasure are fundamentally biological rather than moral phenomena. 3. Facts about the flourishing and languishing of sentient beings: 3.1. Only a fraction of living things, including the majority of sentient beings, thrive. In other words, very few living things have an adequate supply of food and water, are able to reproduce, avoid predators, and remain healthy. 3.2: An even smaller fraction of organisms thrive for most of their lives. 3.3: Almost no organisms thrive for all of their lives.
Cosmology	The universe “began with complete chaos.”	1. Our best scientific evidence indicates that the universe began with time, not in time. 2. The universe began with complete chaos.