Knowledge and Bias: A First Response to Tom Clark

Several weeks ago Tom Clark commented here on a blog entry about dualism. Clark is the director of the Center for Naturalism and is (I believe) also responsible for a related website, He speaks nationally on naturalism and has authored many articles on the topic. I’ve read several of these articles and exchanged a couple of emails with him, and I’ve found him to be both gracious and thoughtful. If I were a naturalist or atheist, I would rather have Tom Clark for a spokesperson than some of the more prominent writers like Richard Dawkins or Sam Harris.

Here at Thinking Christian he referred us specifically to his article Reality and Its Rivals: Putting Epistemology First. In it he argues (as stated in the abstract),

Being epistemically responsible—not taking appearances at face value and seeking external confirmation for belief—inevitably pushes us toward intersubjectivity and science. This in turn increases the plausibility of the claim that there’s nothing over and above the natural world, what science shows to exist.

As reflected in its title, the article’s primary focus is epistemology: how we know what we know, how we know that we know it, and what practices one ought to follow in order to gain knowledge most reliably. He touches also on matters of ethics and meaning. I plan to respond to the latter two topics in future posts here. For this one and another soon to come I will concentrate on what he has to say about knowledge.

He opens with this:

About the most crucial distinction we can make as cognitive creatures is between appearance and reality, between how things seem and how they really are, between subjectivity and objectivity.

He proceeds to describe how science has succeeded in developing human knowledge over the past several centuries, and contrasts science’s reliability and success with what comes of “disdaining empiricism,” especially “religiously inspired anti-empiricism,” which he equates with “contempt for intellectual prowess.” Anti-empiricism, he says, is “driven by ideology and profit as well as by its role as a cultural identifier.”

Christian evangelicals, who believe the planet and its life forms are God’s creations, and that Darwin’ [sic] theory of natural selection inevitably leads to moral decay, have a religious stake in science being wrong about evolution and cosmology. They work tirelessly to inject young earth creationism and its offspring, intelligent design, into public school curricula.

There are several errors of fact in these short bits quoted so far, which I trust others will be able and ready to comment on. I am mostly interested in what Clark calls “the central argument of this paper:”

that when it comes to representing reality, there is no coherent ethically responsible substitute for science and other empirical disciplines. The alternatives—faith-based religions, empirically unfounded secular ideologies, and commercial agendas hostile to evidence—often claim to be objective representations of how the world is in various respects, but have no entitlement to such claims. The only reliable basis for knowledge, the only route from subjectivity, is to relentlessly subject a belief, then to allay the doubt (or confirm it) by gathering evidence that’s independent of one’s commitment to the belief…. We must put epistemology first and get it right, and make no bones about it.

So how does he propose to get it right? He places primary emphasis on two cornerstone principles. First is what he calls the insulation requirement:

To back up our claim that experience captures reality we must rule out such influences [intensity of experience, strong expectations, lifelong immersion in a religious culture, etc.], insulating our beliefs as best we can from subjective bias and possibly mistaken conventional wisdom…. We must do our best to insulate beliefs in [G]od, the soul, and the supernatural from sources of potential bias.

I note in passing that Clark generally knows how to capitalize proper nouns. Other than the occasional (and understandable) typo, he does it for every proper noun in every article I’ve read of his; except there is one such word he never capitalizes, for some strange reason. I don’t quite understand the rule of grammar that calls for “God,” when used as a proper noun, to be written all in the lower case. Suppose I were to capitalize every proper noun in this blog post except for “Clark.” I think he might take it as a very intentional yet rather puerile personal swipe against him. If he has some reason other than disrespectful dismissiveness for not capitalizing “God,” I would be interested to know what it is. For my part, when I quote him in sentences that include the word I will correct his grammar, as I have just done.

Back to the argument. His second cornerstone principle is what he calls the public object requirement:

Unless there’s intersubjective data, a public object of some sort we can all in principle see or sense in some fashion and thus agree exists, it doesn’t matter how many millions of individuals report subjective experiences of [G]od and the soul: they could all be mistaken, just as all those reporting experiences of alien abduction could be (and likely are) mistaken.

By these two criteria, he is quite sure that religious claims are “perilously unsupported,” extremely likely to be false, and irresponsible to hold as knowledge claims. Science, on the other hand, can be insulated from bias and rests entirely on intersubjective, public-object methodologies, and is therefore reliable as a means of gaining knowledge.

What can be said about this? Is there value in his two central principles? Quite obviously there is, especially for dealing with the natural world; there’s no disputing that science has built its enormous success on these and other methodological principles. But should we take them as normative for all knowledge? We run into serious problems if we try. I’ll start with the insulation requirement. Clark supposes that (a) all religious knowledge claims are tainted by bias, and (b) science can be (and often enough is) free of bias. He insists that

science as it’s commonly practiced manifestly does not make any commitment to naturalism…. science can’t be accused of dogmatism…. Science isn’t in the business of defending or rejecting a worldview, whether naturalism or supernaturalism, scientology or Briantology….

There’s nothing of worldview naturalism in any of this, only a quintessentially natural desire for trustworthy grounds for belief. Non-empirical ways of knowing fail to meet worldview neutral standards of epistemic accuracy….. Were they to champion empiricism as the most reliable route to objectivity, science-friendly organizations wouldn’t thereby be promoting naturalism.

Now, I certainly think that science can operate as nondogmatically, with reference to naturalism, as he describes here. Naturalism is hardly essential as a foundational belief either for scientists or for the practice of science. Science has to hold that there is a natural world and that it operates generally in ways that can be studied, predicted, and understood, but it does not have to hold that the natural world is the only reality. That science has this worldview freedom is not to say that it always exercises it, or that Clark himself is consistent on the point. His own naturalistic bias is quite evident.

Should science find public, reproducible evidence for intelligent design, including a specification of the designer and a clear account of its mode of operation, all this would perforce be incorporated into our best intersubjective picture of the world. The more reliable and convincing this account, and the more integrated with the rest of what we reliably know, the less tempted we would be to call such design supernatural. By illuminating the connections between phenomena of vastly different scales and types, science is inherently monistic in showing the unity of reality. So it’s hard, perhaps impossible for purportedly supernatural phenomena to survive clear explanation and empirically-based understanding; instead, they get naturalized.

A passage like that is hardly free of worldview bias. It predicts that if any truth about God or spirituality is ever reliably discovered, it must push us toward naturalistic interpretations of God and the spiritual world. Elsewhere, similarly and with no less bias, Clark argues for “causal closure, the idea that a scientific examination of bodily action leaves no explanatory room for anything non-physical.'”

But there is a deeper problem with the insulation requirement, if I have understood correctly. Its purpose is to insulate knowledge from bias. Its effect, however, is to rule out non-natural knowledge by definition. It cannot permit a person to come to a non-natural conclusion. It is thoroughly biased toward naturalism. The bias of which Clark accuses religious knowledge is peanuts by comparison. He says that it has great difficulty overcoming preconceptions, desires, and prior expectations. His epistemology, on the other hand, does not just have great difficulty overcoming its expectation of non-supernaturalism, it cannot ever, by definition, overcome its prior commitment to non-supernaturalism.

More specifically, it is biased against any knowledge toward which the knower holds a personal relationship of concern or commitment. Clark says that those who claim to know God ought to be able to do so dispassionately, unconcernedly, if they are to trust their own claim. He supposes, I guess, that God would want us to relate to him as we do to a laboratory experiment; that our doctrine of God might be of no more personal concern than our doctrine of quarks. If God is personal, a being of love and holiness who relates to us as such, then Clark says such a God cannot be known, because inevitably we would care about that which we think we know, and as soon as we care about it we must conclude we don’t know it after all. The insulation requirement does not insulate us from bias, for it is itself inherently biased. It rules out even the bare possibility that there is a knowable personal God.

I have two principles of my own to suggest to Tom Clark. (I’m sure he already knows them, but if my analysis is correct he has not applied them here.) One: any rule of knowledge that by definition rules out even the possibility of knowing something in matters as important as God, is biased and cannot be used to judge whether there is a God. Two: if you propose a universal rule of knowledge that violates its own strictures, it would be wise to discard it.

Now to Clark’s “public object requirement.”

Not only must we do our best to insulate beliefs in [G]od, the soul and the supernatural from sources of potential bias, we must find evidence for them outside private subjective experience, evidence that’s publicly observable by those who haven’t experienced [G]od’s embrace.

His discussion on this is pointed primarily toward disagreements he has with a theologian, John Haught, whose religious epistemology is (according to Clark) based on internal experience alone. I have not read Haught, but I have seen enough quoted from him in secondary sources to doubt that he represents historic Biblical Christianity—he seems to hold to something like process theology. Clark also speaks of similar discussions he has had with Stuart Goetz and Charles Taliaferro, who I believe are more orthodox in their Christianity. Nevertheless I share the view that one contributing factor to Christian knowledge is one’s personal encounter with the personal God. I don’t see anything the least bit incoherent about that. My experience with God is not evidence for you to believe in, but it certainly can be evidence for me, supporting my belief. Not the only evidence, but part of it.

Clark makes his own case absurdly simple, though, by pointing to Haught in the way that he does. He represents Christianity as if it had nothing more to offer on its behalf than Christians’ internal experience (or Intelligent Design, mentioned elsewhere in his paper). This is hardly the case! There is the historical record of Israel, Christ, and the church; there are strong philosophical arguments in favor of God, and also in favor of the soul and free will (see below); and there are strong existential arguments in favor of Christianity. There is evidence for Christian belief outside private subjective experience. If Clark addresses this anywhere, I have yet to find it. I haven’t read all of his papers, so maybe he can point us toward one that’s relevant. In this article he certainly treated it as if there were only the two ways of knowing about God.

Clark also says,

Just as belief in[G]od needs support by public observation to be warranted, so too does belief in the soul and free will.

What could be more publicly observed than people (one’s self and others) leading lives by our own decisions, exercising free will? And what basis could there be then for denying free will, other than a strong naturalistic bias?

Finally, the public object requirement seems rather too stringent. In a paper devoted entirely to Haught’s religious epistemology, he puts the requirement this way:

Given the fallibility of human experience, its potential to misrepresent the world, it seems reasonable to ask for further justification for the claim that religious experience reveals the truth about things. This normally involves producing evidence for the claim that’s independent of the experience itself, something other than the mental state of the experiencer, such as a publicly observable object or measurement.

It seems to me that a consistent application of this requirement would lead to a rejection of any knowledge that was not, in principle, available for public observation and measurement. Thus I cannot know what I am thinking at this moment, for my thoughts are quite inaccessible to any other person’s verification, or for observation by any instrument. Now, perhaps there is some valid principle whereby the public object requirement is set aside for knowledge of one’s own thoughts and internal state. If so it ought to be articulated; for Clark certainly seems to make it a universal test for reliable knowledge:

I’ve called this the public object requirement, and along with the insulation requirement it constitutes basic epistemic good practice, without which no factual claim about the world has credibility (emphasis added).

So although Tom Clark’s paper is well written, and his tests of knowledge are interesting, the standard they set is one they fail to meet, if my analysis is correct. They are fatally self-contradictory. And there is yet one additional major sense in which I think they fall short, which I will save for my next post on this topic.