John Loftus and The Outsider Test for Faith

I am reviewing Why I Became an Atheist, by John W. Loftus. Other posts are here.

In my previous post I offered some criticisms of chapter 4 of John Loftus’s book, Why I Became an Atheist. Loftus presents what some take to be a relatively unique piece of atheist apologetic, “The Outsider Test for Faith.” In my review of chapter 4 I said the following:

Whether Loftus thinks skepticism is a presumption or a “control belief”, something we start with or something we conclude with, is mysterious.

In the comments, Loftus writes:

If you want further thoughts on the OTF then see the links here. My more developed argument dealing with all of these concerns including yours is found in my forthcoming “The Christian Delusion.”

He not only links me to revisions, but refers to future books. But I’m reviewing this book. Also, why should I want to read “further thoughts” if the initial thoughts were so disappointing? My critiques offered in the previous review apply to the “Outsider Test” as a general project.

I will go ahead and embark upon the Sisyphean task of  hitting the ever-moving target of Loftus’s philosophy.

Outsider Test #1: 2/1/2006

Long ago, Loftus asked, “How is it possible [for a Christian] to rationally evaluate the Christian faith when the Christian can only do so from within the presuppositions of that faith in the first place–presuppositions which he or she basically accepted by the “accidents of history.”

Part of Loftus’s problem is vague language. No one does anything “within” presuppositions. Think of presuppositions like hidden assumptions. If we assume, say, the truth of the Bible (as we interpret it), then it will be very hard to come to the conclusion that, say, God doesn’t exist. Why? Because the existence of God is an immediate inference from much of the content of the Bible. Whether or not any Christian “evaluating” her faith would actually reason in such a silly fashion is unlikely. Usually people want to know if things in the Bible are actually true, or if their experiences of God are hallucinations, etc. But let’s say someone does the rationally unthinkable: assumes the truth of the Bible, and then happily discovers that the Bible verifies the existence of God. John Loftus has taken the courageous stance that we shouldn’t commit logical fallacies. Loftus proposes what he takes to be a better method than begging the question, “the outsider test”:

So let me propose something I call The Outsider Test: If you were born in Saudi Arabia, you would be a Muslim right now, say it isn’t so? That is a cold hard fact. Dare you deny it? Since this is so, or at least 99% so, then the proper method to evaluate your religious beliefs is with a healthy measure of skepticism. Test your beliefs as if you were an outsider to the faith you are evaluating. If your faith stands up under muster, then you can have your faith. If not, abandon it, for any God who requires you to believe correctly when we have this extremely strong tendency to believe what we were born into, surely should make the correct faith  pass the outsider test. If your faith cannot do this, then the God of your faith is not worthy of being worshiped. [bolding in original]

You might ask, why should I take this “test”? The motivation for doing this is supposed to be that if you (a Christian) were born in Saudi Arabia, then it is “at least” 99% likely that you would be a Muslim instead. So Loftus says, evaluate your faith as if you were in an “outside” situation. This early presentation of the Outsider Test raises the following issue. Loftus starts out by saying that if you were outside our faith, you wouldn’t have your faith (tautology alert, set condition one throughout the ship!). His next step is to say: imagine that you are in such an outside situation. His final step is to say: if your faith “fails” this test (wait, what test?!), then your God is not “worthy” of worship. We have to translate that into “then you don’t have fully rational belief,” since Loftus prefers puffy rhetoric to clarity and rigor.

So it’s not at all clear what 2006 Loftus has in mind. He doesn’t even say what the Outsider Test is, he just tells us to take it – a very mystical approach. It has something to do with a “healthy” amount of “skepticism.” And this skepticism has something to do with imagining we are in situations where we would not hold our current religious beliefs, and then doing the hard work of figuring out if what we just postulated remains hypothetically true over hypothetical time t, after we’ve postulated, in the imagined world.

This sounds ridiculous, even confusing. Surely part of the reason theistic religious believers hold their religious beliefs is because of compelling experiences that seem to be with God, collections of evidence involving books, peers, etc., developed thought over a lifetime, and so on. If we consider ourselves “outside” our faith, do we eliminate these reasons for our beliefs as well? Who, after all, is the “you” being transferred to Saudi Arabia? Say person X has some experience E with God, which yields religious belief. It’s obvious that removing E would also remove the religious belief (I am excepting plausible scenarios where, if God exists, he would reveal himself by other routes). But this is not any sort of “test.” It’s just a way of isolating the reasons why someone believes something. Growing up in Saudi Arabia, for example, removes a person from the entire set of circumstances in which their religious belief arose, thereby taking away, very possibly, all of the reasons why a person might hold their religious beliefs. Loftus tries to be tricky by attributing this solely to demographic concerns, but it doesn’t take even a B.A. in philosophy to realize that demographics only change religious belief in light of other contributing features. Loftus likes to talk about “parents.” He might also talk about, for example, available religious institutions, social groups, political setting, available reading material, education system, and on and on.

Brief note: There is of course one thing that remains across demographic boundaries, and that is the ability of God to reveal himself; correspondingly, we find belief in gods of some sort to be surprising pervasive across cultures in time and space. Unsurprisingly, these rudimentary religious experiences are filtered and expressed through available cultural and institutional forms.

Outsider Test #2: 2008

This is the Outsider Test presented in Loftus’s book, Why I Became an Atheist. I reviewed it in the previous post.

Outsider Test #3: 3/20/3009

Here Loftus makes clear that his Outsider Test for Faith is, instead of an argument against the rationality of religion, “one of several arguments I use to demonstrate that when examining the evidence for a religious set of beliefs the predisposition of skepticism is warranted.” So, according to Loftus, before examining the evidence for religious claims, we should employ the Outsider Test. This should lead us to  “presuppose” skepticism, one of Loftus’s many underdefined terms. But skepticism seems to be a negative term, meaning something akin to “Don’t assume the Bible is true,” or something very much like that. It’s unclear whether Loftus has anything positive in mind for skepticism as an actual investigative methodology for the epistemology of religion. But he must think it is involved in “examining” religion, since the Outsider Test is supposed to lead us to skepticism in exactly that context.

Loftus repeats his book’s claim that sociological data is a “background” fact to be used in assessing religious “truth claims.” It is important for the non-specialist reader to realize that, usually, sociological or cultural arguments are relevant only to the rationality of religious believers themselves. Say Joshua’s only reason for believing in God is because his trusty uncle told him so. This wouldn’t cast doubt on the proposition “God exists,” but it would cast doubt on the proposition, “Joshua believes in God for good reasons.”

Loftus lists a bunch of shady beliefs we would have if we were in different cultural contexts. To pick one of his ridiculous examples that isn’t even true, “If in today’s world we were born in the Palestinian Gaza strip, we would hate the Jews and probably want to kill them all.” Let’s say this racist stereotype is true. According to the Outsider Test, we should question – with the handy tool “skepticism” – our belief in not killing all the Jews. After all, had we been born elsewhere, we’d want to kill them. But it seems to me that we are perfectly within our rights in not spending any time whatsoever evaluating whether or not we should kill all the Jews. So Loftus must convince the religious believer that she in particular should evaluate whether she is hallucinating every day. Loftus is unaware that a belief could be culturally likely and based on evidence and good reasons. In fact, to the chagrin of the Outsider Test, sometimes cultural and sociological facts can help as much as hinder someone’s ability to adequately consider the evidence. Human beings can prevent one another from all sorts of actions, and these actions include intellectual pursuits. But they can also create societies where intellectual endeavors flourish.

Loftus assures readers of his intellectual integrity. “I have investigated my faith from the inside as an insider with the presumption that it was true.” But presuming the truth of something precludes the possibility of investigating it to find out if it is true. Loftus is misusing terms like “presumption” again, and must mean something entirely different. It almost seems like Loftus is saying he had a “true until proven false” mentality, akin to the presumption of innocence.  But he then says that “from the inside” Christianity actually does seem true, right after saying that from the inside he couldn’t continue to believe it.

Loftus shifts again:

“The outsider test is simply a challenge to test one’s own religious faith with the presumption of skepticism, as an outsider.”

Recall the remark Loftus opens with, that the Outsider Test is

“one of several arguments I use to demonstrate that when examining the evidence for a religious set of beliefs the predisposition of skepticism is warranted.”

Loftus can’t get clear about what is arguably his only original argument. Of course many people don’t know what they are talking about, e.g. Kent Hovind. But it is especially tedious when someone doesn’t even know what they are trying to talk about. Is Loftus trying to directly give us a way of evaluating our own religious beliefs – where we use the Outsider Test to evaluate them? If so, exactly how do we take the Outsider Test? Do we just do a thought experiment where we eliminate our reasons for believing and then say to ourselves, “Now I have no reason to believe”? Or is he trying to convince us that “skepticism” (a method? an attitude?) is what we should use to evaluate our religious beliefs? Either option is clear, even if the first is trivial. In the next passage, Loftus defines the presumption of skepticism to be that we should presume skepticism because one’s religious beliefs are “probably” wrong. But this is the conclusion, not presumption, of atheism, plausibly characterized as the belief that “there (probably) is no God.” So there’s an assertion of atheism within the justification for the Outsider Test, which both contains and is within arguments for skepticism, using skepticism as its method all along, without defining it.

Loftus says that the “amount” of skepticism required (when doing what? taking the outsider test? before taking the outsider test? after? I thought the outsider test itself argued for skepticism anyway?) is, via some hidden epistemological mathematics, proportional to various facts, such as how many “rational” people disagree, how the beliefs in question were adopted (but I thought we already knew this, and that’s why we were going outside in the first place), what kinds of evidence are available for investigating the beliefs, and so on. I agree with the bulk of this. Although the point about beliefs being “probably false” assume the probably false notion that beliefs are determined by chance, one good reason we might have for questioning out own beliefs, is when people we take to be epistemic peers or better don’t agree. Another good reason to challenge certain beliefs is if we suspect that we believe for bad reasons (e.g. my trusty uncle told me there were angels, so I believe there are angels).

Loftus has a paragraph where he lists things the Christian aren’t allowed to presume while taking this elusive outsider test. He picks weird examples, e.g. “no more quoting the Bible” to show how Jesus can be 100% man and 100% God. Instead, the Christian theist must “try to make sense of this claim.” Of course if there was a good explanation of this alleged doctrine within the Bible, then quoting it would be just like quoting any other book (say, one of the many books that tries to do what Loftus is asking). The only way to investigate beliefs is to try and find out if they are true without assuming them. So investigating a Biblical claim cannot consist in reading the claim over and over again in the Bible. Who does this anyway? It seems to me that all tests of any beliefs are “outsider” tests, in the intuitive sense where “outsider test” is just trying to find out if a belief is true (since the very endeavor precludes assuming the belief to be true).

Loftus says certain rather conspicuous things given his own projects, such as “Christian, just ask yourself whether the initial reasons you had for adopting your faith were strong ones.” Yet Loftus doesn’t want us examining his initial reasons for adopting his atheism, and to instead examine the new arguments in his book. Remember, not only did Loftus have dubious reasons for adopting his atheism, but had he been born in Saudi Arabia, he likely would have never become an atheist at all. So we have to decide which Loftus to go along with. The Loftus that at the beginning of his book doesn’t think initial reason are relevant for analysis of present beliefs, or the mid-book Loftus who thinks they are very important for evaluating the Christian’s present beliefs.

Loftus invents six objections to his Outsider Test.

Objection 1

Some Christians will say that sociological data just helps us see that not all religious claims can be true, not that some particular claim isn’t true. Loftus responds by saying that the chances are low that any particular claim is the lucky one. Thus he misses the force of the objection he raises. The point is that religious believers – Mormons, Muslims, Hindus, and everyone else – have various reasons for holding their religious beliefs. To successfully counter their beliefs one must at least counter their reasons. You can’t just show them what they already know – that other people disagree with them. Loftus thinks the Christian can’t “rationally justify such luck.” But the Christian (or Muslim, or…) isn’t justifying luck, they are trying to justify their beliefs.

Objection 2

Some Christians will respond to Loftus by pointing out how minorities of people in Muslim countries choose to be Christian. Loftus doesn’t develop exactly why this is an objection to him. But he doesn’t hesitate to respond that in his opinion such converts are being irrational. His innuendo against all missionaries and converts is presented without evidence, so I’ll comment on something else. This objection shouldn’t be about minorities in Muslim countries. It should be about huge majorities across Europe and elsewhere who accept the teachings of a 1st century Jewish sect. The cross-cultural appeal of Christianity (and Islam, I would say) goes against suggestions that they are demographically limited or determined.

Objection 3

This objection of hypothetical Christians is similar to the first. Some Christians will want to point out that disagreement among rational people does not mean we should doubt our beliefs. Loftus says that our doubt of beliefs should be in some way proportional to the amount of disagreement. It is important to realize that unless beliefs are randomly selected in a lottery, there isn’t a direct relationship between quantity of disagreeing views and which one is true. But if we add the concept of epistemic peers, I don’t think the claim is controversial. If some people you intellectually respect disagree with your views on something, that provides at least some reason to question your views. However, this reason for doubt doesn’t seem to go as far as Loftus wants it to. All of us have all kinds of disagreements with untold numbers of equally rational people. That’s not a reason to abandon our beliefs, or even be suspicious of them. It’s just a good reason to consider them more carefully than we would have otherwise. This is especially tricky in cases of religious belief, since one of the reasons why people have religious beliefs is palpable experiences that seem to involve God. How many non-religious people also claim to have palpable experiences that strongly seem to involve God, a supernatural agent outside of themselves? We can safely say: less than the total number of non-religious people. Thus the total number of “rational” objectors to religious belief will not be as compelling to religious believers as Loftus wants it to be, since they usually don’t claim access to the same set of evidence. In any case, the number of non-supernaturalists in the world is incredibly small, and the number of rational supernaturalists sufficiently high, so that it is reasonable for believers to not have too much doubt on these grounds.

Objection 4

Some Christians will say Loftus’s argument is self-defeating. Loftus responds by saying it is not self-defeating. Loftus provides no argumentation for this counter-assertion. He just points out that he uses anthropological data to show that we should doubt our beliefs. I can’t detect any real argument here. My own view is also that his argument is not self-defeating, since for an argument to be self-defeating, its application to itself would have to show itself to be in some way deficient. But the most Loftus’s argument can do to itself is show us that we should have skepticism about the argument, and not beg the question in favor of the argument. Since Loftus has defined virtually none of the terms he is using, it could be that self-defeat is lurking, but there’s no reason to think so yet. The worst case scenario is that Loftus’s argument yields an infinite regress of Outsider Testing. Shouldn’t we do Outsider Tests with respect to the beliefs that led us to take the Outsider Test? And so on.

Objection 5

Here Loftus addresses the objection that perhaps he is committing the “genetic fallacy,” doubting a proposition on the basis of how it was originally adopted by a believer. He says first that he doesn’t think the genetic fallacy is a big deal. This is interesting, since Loftus really doesn’t want us to attack his original reasons for deconverting – he wants us to address his new sophisticated arguments that take place in his books. But Loftus says that even though the genetic fallacy is irrelevant, he hasn’t committed it anyway, since he isn’t questioning whether or not Christian belief is true. Rather, he’s trying to argue that we should have skepticism about beliefs arrived at through cultural means. As Alvin Plantinga has shown, this sort of thing reduces to attacking the truth of belief, in those cases where the belief system includes beliefs or models about how the system came about. For example, Christians will often hold that the spirit of God gets much of the credit for their belief. So if Loftus is correct about cultural determinism, then he actually is objecting to the truth of Christian belief.

Objection 6

Objection 6 is oddly similar to objection 4, with vague edges. This is the accusation that Loftus’s argument is “circular.” But it seems like here he is really talking about self-defeat, not begging the question. Anyway Loftus just repeats that he has given arguments for his Outsider Test – therefore he doesn’t need to apply the Outsider Test to it. But the Christian can just say, “Hi John Loftus. I’ve given arguments for my Christian belief. Therefore I don’t need to apply the Outsider Test to it.” Loftus claims he doesn’t have to test his test, because his test has good grounds for it, despite the indisputable cultural data he keeps talking about, that shows Loftus would not have developed the Outsider Test if he was born on Mars. But the Christian also thinks she has good ground for her beliefs, despite the indisputable cultural data.

Outsider Test #4: 3/25/2009

Victor Reppert, author of C.S. Lewis’s Dangerous Idea, objected to Loftus’s Outsider Test. Loftus responded here.

In this development of the Outsider Test Loftus becomes clearer that the Outsider Test is especially applicable to religion, and not to, say, science. There is a somewhat ad hoc reason fro this: we can, for example, test scientific claims in a specific fashion. It seems to me that we can do something like test, play around with, etc., religious claims. I wonder where Loftus draws the line. Surely historical beliefs are less testable, in the scientific sense, than religious beliefs. We should keep in mind the issue of “epistemic fit” when talking about these issues. Methods such as (certain technical forms of) testing, as well as consensus-building, have good epistemic fit with the natural sciences, because these methods are well-suited to the pursuit of truth in the natural sciences – they are the most likely methods to yield accurate results. Technical forms of testing in the context of natural sciences, however, do not apply well to, say, the field of ancient history. We can find some vague resemblances in historiography (e.g., we might have some softly generalizable historical trends), but there’s not much. Testimony, for example, is quite useful in history, and not in natural science.

What Loftus really needs to ask is what sorts of investigations are useful in religious affairs. It seems like he has not asked this question. Instead, he keeps repeating that because religious beliefs are determined by culture, then we should have lots of skepticism about them. Then he employs several ad hoc principles to avoid applying this principle to other areas of human concern. Unsurprisingly he is most resistant in applying the Outsider Test to his own beliefs. Loftus should first ask why culture might be correlated with religious beliefs. In virtue of what facts does culture influence religious belief? Then he should notice that a couple of religions actually do exist cross-culturally, and are spreading. Specific propositions don’t pop up randomly, they spread through the usual means – associations, teachers, palpable experiences, etc. Lastly Loftus should realize that we need to figure out the best way to investigate religious claims. To me it seems to be that proper methodology will be to put ourselves in situations where we would be likely to perceive religious objects – say, God. For example, it would be unreasonable to write off Islamic belief on the basis of reading a book about it, when Muslims claim to have palpable experiences of God. If we really want to investigate Islam so that we can justifiably say it is groundless, we at least need to include some experiential investigation. That’s just a quick idea, but Loftus doesn’t seem even remotely curious about this extremely important issue. Instead, he just wants us to read books and be “skeptical.”

Outsider Test #5: 4/01/2009

Here is a weird post that appears to be another statement of the Outsider Test. It’s hard to perceive anything new or interesting beneath the annoying format and refrain of that post, “Deny this if you can.”

Loftus says:

Let me put it to you this way, if you read everything that I have read and experienced everything that I have experienced, then you would think on these issues exactly the same way I do. Deny this if you can.

I have an even better suggestion. If I became John Loftus, my beliefs would be those of John Loftus. Deny this if you can!

Unless Loftus or someone else can point out what Outsider Test #5 adds to discussion, I won’t discuss it further.

Conclusion

I sincerely doubt Loftus’s discussion in the upcoming book, or new editions of the current one, will be much better than his discussions thus far. For one thing, he keeps repeating himself. Second, the whole project of the Outsider Test is either trivial (don’t assume what you’re trying to prove) or incoherent (are we using skepticism? proving that we should use it? what is skepticism in the first place? etc.) or grossly inapplicable to investigating religious beliefs (the sociological data don’t tell us about truth-propositions, we have non-sociological reasons for beliefs anyway, and investigating religious systems involves holistic engagement, not just vague skeptical attitudes).