Christianity Without Certainty is OK

I think it’s possible for a person to be certain in their own Christian faith, but I also think it’s reasonable for a person to identify as Christian without needing to demonstrate that Christianity is true with certainty.

More generally, I think it is valid to claim that a statement is true without needing to show that the statement is certainly true.

Christians are often asked by non-Christians, and also by each other, about how one might be certain that Christianity is true. I think probably the best answer is found within presuppositional apologetics, but this school of thought is not accepted by many people. A second approach, which I prefer because it follows a commonly accepted pattern of argumentation, is to argue that Christianity is rationally preferred even without certainty.

Here's one of many articles where I make an argument for Christianity without requiring certainty. Most arguments I make in favor of Christianity with optional certainty fall into one of two types of argument. First, if it is more likely that Christianity is true than it is likely that Christianity is not true, then it's rationally preferred to believe that Christianity is true, even without certainty. Second, even with a very small chance that Christianity is true, identifying Christianity as true is rationally preferred when the expected value of doing so is high, and that is always the case under salvation by faith alone, because of the payoff from obtaining salvation is high.

Suppose that we ignore axiomatic, necessity-grounded, presuppositionalist, and all other forms of argument for Christian belief with certainty. Let’s take seriously the calculation of the likelihood that Christianity is true while assuming uncertainty. I think there are two basic ways to get at this. Both involve what I’ve previously called ordinal logic, and the discussion which follows is also under the influence of bayesian logic. There is perhaps a sort of veil of ignorance approach in some of the points made.

Start by defining the possibility set and stop before weighting any particular possibility. Assuming equal likelihood of any outcome, a first-glance probability is obtained. At first glance, Christianity is either true or untrue. Without weighting either option, these two options would be equally likely, and Christianity would have a 50% chance of being true.

From the baseline at 50%, two different moves serve to increase or decrease the final calculation of the likelihood that Christianity is expected to be true. Either directional move constitutes one of the two ways to get at the question to which I earlier referred. One move is to count the number of successful arguments for and against Christianity. This move seems to strongly favor high likelihood of the truth of Christianity. A second move is to recognizing several omitted intermediate factors between a generalized truth of Christianity and a personalized, potentially unique, individual cost-benefit calculation. This move results in the observation of a significantly reduced likelihood that any particular Christian holds a true belief set.

There are many good arguments for Christianity, but let's see the mathematical impact of acknowledging even just one. If there is  one good argument for Christianity, for now ignoring the number of good arguments against Christianity, there is a 75% chance that Christianity is true. A statement is expected to be more likely to be true when it is supported by a good argument than when it is not supported by a good argument. The likelihood of the supported claim will now be at least what is was without the supporting argument, and at best it is now proven with certainty. By way of expected value, then, (.5+1)/2 = .75.

In general, when we ignore the number of good arguments against Christianity and we include the number of good arguments supporting the notion that Christianity is true, the first-glance likelihood of Christianity is the chance that Christianity is true is 1 - (.5^(n+1)), where n is the number of arguments in favor. Now notice that every good argument for has an equal power to every good argument against, so that we can use the same formula above and redefine n as the net number of arguments in favor. With n of 5, the chance Christianity is false is about 1.6%. In the real world, I don’t know of any good arguments against Christianity, and I know of many more than 10 good arguments for it, so I think a 5 is very conservative value for n.

Now let's calculate the result of the move in the opposite direction, generating a worst-case, but still plausible, calculation of the likelihood of the truth of Christianity. The approach here is to recognize that Christianity being true is insufficient to yield a complete cost-benefit calculation for a particular Christian. Christianity writ large includes a diversity of beliefs. Many of these differences are unimportant for the purposes of cost-benefit calculation, but in a few cases the differences are critical.

Interestingly, that dogma which is often considered essential or critical from a theological perspective is often different from that dogma which is critical from a benefit-cost perspective. For example, the death and resurrection of Christ is essential under most approaches to systematic Christian theology, but a cost-benefit analysis obtains without regard to whether Christ died in a literal way, a metaphorical way, a physical way, a spiritual way, in accordance with modern medical definitions of death, according to ancient Jewish legal definitions of death, or at all.

Salvation theology, on the other hand, is critical to cost-benefit calculation. If an individual is not identified as eligible for the payoff from salvation, that individual faces a drastically reduced incentive to prefer Christianity over other worldviews. Likewise, even if an individual is considered eligible for the payoff from salvation, if the payoff itself can be identified as finite or small relative to some alternative payoff from a mutually exclusive worldview, the individual under consideration faces an incentive to rationally prefer the non-Christian worldview.

To perform a concrete calculation, the truth of the doctrine of salvation by faith alone is considered critical. It doesn't need to be the case that identification of an individual as eligible for payoff is considered a distinct factor from the factor representing the chance of the truth of salvation by faith alone, but we will grant it as a second distinct factor in order to ironman the anti-Christian view. A third factor is the chance that the payoff is infinitely large.

Given that each of these factors have an independent chance of being true at 50%, multiplied by the chance that Christianity itself is true, the result is obtained. The result is that there is about a chance of 6% that not only is Christianity true, but the specific theological approach outlined previously is true, yielding an end-to-end cost-benefit calculation.

Notice that the specific theological approach outlined previously is actually a preferred theological set in real-world Christianity. One would expect that the preferred theologies among Christians and expert Christian theologians would be more likely then there are alternatives, not co-likely. This means that a reasonable estimate of the likelihood that Christianity is true would be greater than the 6% noted previously. This reasonable estimate becomes even more favorable toward Christianity if the number of factors is reduced from 4 to 3 by consolidating individual identification as eligible for the salvation payoff into the salvation theology factor itself. Under this more favorable calculation the likelihood that Christianity is true comes out to .5(.75)(.75) = .28.

In summary, when opposition to Christianity is ironmanned, it remains difficult to argue that the probability that Christianity is true is less than 6%. The expected value of preferring a Christian worldview remains comparably preferred to other options even with this 6% result. Alternative calculations identify a 98.4% chance of Christianity being true as a conservative number. The cross-approach average, then, is in excess of 50%. This indicates that it is more reasonable to expect that Christianity is true than it is reasonable to expect that Christianity is false, even before multiplying by expected payoff, and even when arbitrarily ruling out the many good arguments for Christianity with certainty.