An empirical test of the existence of sensus divinitatis in atheists

Paul Manata, in a comment on the post about Dan the agnostic who claims there aren’t really atheists, linked to a paper he wrote arguing for a similar conclusion. The position he takes in that paper, following Greg Bahnsen, is that atheists really do believe in God–everyone does–but that they have a second-order belief about themselves that they do not believe in God. That is, they are self-deceived about not believing in God, and their professions of atheism are based on the erroneous second-order belief.

Although Paul has a section in his paper labeled “Empirical support for NA” (NA meaning the No Atheists claim), the evidence he appeals to is weak (that most people alive today and who have lived in the past have been religious) or non-empirical (that there exist arguments which–he claims–prove that the Christian God is a necessary presupposition of logic, science, and ethics).

But if there is really an innate universal belief in God–or a natural direct perception of God, a sensus divinitatis, a mental faculty that allows direct basic knowledge of God’s existence–it seems to me that we should be able to find much stronger empirical evidence of it.

Evan Fales (in “Critical Discussion of Alvin Plantinga’s Warranted Christian Belief“, NOUS 37 (2003):353-370) raises the question of whether independent empirical evidence regarding such a sensus divinitatis (SD) and guidance of the Holy Spirit (HS) can be obtained (even granting the presumption that Christian theism is true):

However, maybe we can acquire independent evidence of the existence of the SD/HS mechanisms. After all, we have abundant evidence for the existence of sensory faculties that put us in touch with the world in reliable ways; also memory and reason. We are beginning to decipher the neurological substrata of these cognitive processes. Of course, our evidence here is evidence produced by these very faculties, so its evidentiary status begs the question against skepticism; but at least these faculties are not epistemically isolated from one another, or from
utilization in investigations that yield an increasingly ramified picture of the very processes thus employed, and of the grounds for trusting them.

Can anything comparable be done on behalf of the SD and HS? So far as I know, nothing of this sort has been done; and, short of some much more detailed specification of the alleged structures and processes, it is hard to see how it could be done. Perhaps there are no such structures; perhaps these “faculties” consist in ways God directly stimulates our thoughts. Absent a physical substratum, there may be no independently detectable evidence to be had. But if so, then so much the worse for the epistemic status of the claim that there are such faculties.

It’s this suggestion that I’d like to pursue, but first let me note that Fales presents three arguments based on empirical evidence which cast doubt on the existence of a sensus divinitatis–first, the divergence of claims and beliefs by those who claim to have one (lack of reliability, even within Christian sects). Second, the lack of a demonstrated superior moral life by Christians versus non-Christians (similar comparisons can be drawn between Christians and nontheists). Third, the presence of Bible verses–accepted by most varieties of Christian as authored by men inspired by the Holy Spirit, presumably with their sensus divinitatis functioning properly–in which “God performs, commands, accepts or countenances rape, genocide, human sacrifice, pestilence to punish David for taking a census, killing David’s infant to punish him, hatred of family, capital punishment for breaking a monetary promise, and so on”?

I agree with Fales that all three of these are defeaters for a sensus divinitatis (with the lack of agreement being the one I find most persuasive–surely a divinely created faculty would work reliably across the set of believers), but I would like to pursue his suggestion of examining evidence of empirical science. I think this would qualify as the sort of “theistic science” that Plantinga has elsewhere advocated. The example I will give is meant to be illustrative rather than a well-defined experiment–I’ll mention some problems with it at the end.

While flying on a plane today, I was reading V.S. Ramachandran’s book, Phantoms in the Brain. Chapter 7 of this book is titled “The Sound of One Hand Clapping,” and is largely about a syndrome that occasionally occurs in individuals who have suffered damage from a stroke to the right parietal lobe, causing paralysis of (parts of) the left side of the body. The specific phenomenon he discusses is called anosognosia, or lack of awareness of a defect, and is also often accompanied by neglect, the failure to attend to the left side of the body. (He discusses neglect in chapter 6, “Through the Looking Glass.”)

The characteristic behavior of this problem, which is usually temporary, is that the individual is in complete denial that his or her left side is at all paralyzed. The chapter begins with the example of Mrs. Dodds, who had suffered a stroke two weeks previously and been confined to her bed or a wheelchair since that time:

“Mrs. Dodds, how are you feeling today?”
“Well, doctor, I have a headache. You know they brought me to the hospital.”
“Why did you come to the hospital, Mrs. Dodds?”
“Oh, well,” she said, “I had a stroke?”
“How do you know?”
“I fell down in the bathroom two weeks ago and my daughter brought me here. They did some brain scans and took X rays and told me I had a stroke.” Obviously Mrs. Dodds knew what had occurred and was aware of her surroundings.
“Okay,” I said. “And how are you feeling now?”
“Can you walk?”
“Sure I can walk.” […]
“What about your hands? Hold out your hands. Can you move them?”
Mrs. Dodds seemed mildly annoyed by my questions. “Of course I can use my hands,” she said.
“Can you use your right hand?”
“Can you use your left hand?”
“Yes, I can use my left hand.”
“Are both hands equally strong?”
“Yes, they are both equally strong.” (p. 128)

Ramachandran describes multiple patients with this disorder, and different forms of denial–some would make excuses for failure to move paralyzed limbs, while others would insist that they were, in fact, moving them.

He hypothesizes that the left hemisphere, where the functions of speech, syntax, and semantics reside, is also where beliefs get integrated into a coherent worldview and anomalous data is rejected via semi-Freudian mechanisms such as denial, reaction formation, repression, and humor, while the right hemisphere acts as “Devil’s Advocate,” presenting anomalous information and acting as a check on the reliability of the overall worldview. The damage caused by the stroke results in a failure on the part of the right hemisphere to get its point across that something is wrong.

Now here’s where things get interesting–there’s another phenomenon, called nystagmus, where the eyes make rapid involuntary movements, scanning rapidly in one direction (saccade) and then slowly back in the opposite direction (vestibulo-ocular reflex). This phenomenon occurs naturally in circumstances where you are observing motion, such as looking out a car window at telephone poles you pass, but also can occur in other circumstances. It can also be artificially induced by pouring cold or warm water into one ear canal, to create a

temperature gradient, which stimulates the vestibulocochlear nerve. (This is called calorically induced nystagmus.)

A side-effect of inducing nystagmus (with cold water in the right ear or warm water in the left) is that anosognosia temporarily disappears. The individual no longer denies being paralyzed, but recognizes the paralysis. Ramachandran attributes this effect partially to a stimulation of the right hemisphere, and also speculates that the phenomenon may be related to REM sleep (and predicts that “patients with denial should dream that they are paralyzed” and “if they are awakened during the REM episode, they may continue to admit their paralysis for several minutes before reverting to denial again” (p. 148)).

But about 30-40 minutes later, the denial returns, and when questioned about the interactions that occurred while the denial was gone, the sufferer will engage in further denial, confabulation, etc. to describe what happened.

Ramachandran speculates that the distorted body image of a sufferer of anorexia nervosa might similarly be changeable with such a test.

Now, one problem common in science is overgeneralization from a minimum of data, and this is almost certainly what is going on here. This very issue was discussed in 2001 in an email discussion archived here, which I think makes it unlikely that the details of Ramachandran’s speculation about the mechanism are correct. But if it were correct, my proposed test would be to use the calorically induced nystagmus test to see if atheists temporarily stopped professing nonbelief or disbelief in God–if they did, that would be very interesting empirical evidence in support of the “atheists actually believe in God but are engaged in self-deception” thesis. And that is, I think, the sort of thing that Manata should believe is possible to do if his thesis is correct.

This “no atheists” thesis, even if true, would not prove that atheism is false–we already know in other contexts that the heuristics used in normal everyday reasoning often lead to falsehoods. One of the conclusions of Pascal Boyer’s Religion Explained is that religiosity is an expected outcome of the templates of representing and reasoning that human beings use. But if the “no atheists” thesis were proven true, I think it would add considerable weight on the ledger for theism and against atheism, and place the burden on the atheist side to explain why such a belief would universally occur in the absence of a God.

One further note–if there were empirical evidence in support of a sensus divinitatis through such a test, the next step would be to determine where it is localized and find what specific beliefs correlate with it, in order to address the reliability/incompatible belief issue raised by Evan Fales and determine whether the universal underlying belief is pantheism, deism, Judaism, some flavor of Christianity, or something else entirely.