The year was 1994. I was in my final semester at Talbot School of Theology, and wrestling with the question of predestination or free will. I was stuck. 

That semester I was taking a class on the metaphysics of freedom. We spent much time debating the pros and cons of the Libertarian and Soft Determinist views of freedom (discussed here). 

Part of the discussion focused on whether our choice to follow Christ was free, determined, or somehow both (discussed here). I became convinced that none of the four answers I discussed last week are adequate. Then I discovered a fifth way forward which proved very helpful.

Learning to Ask Additional Questions

I’ve discovered when stuck without a clear way forward in solving a puzzle, the solution is often to step back and ask if there may be other questions which should be asked. This can help reframe the issue in ways that illuminate the path toward a solution.

I found this was the case in solving the “predestination or free will” dilemma. The standard questions asked result in us only seeing the four positions outlined last week.

But that semester, as I wrestled with this question, I also had a class in the philosophy of science. In scientific study we also confront dilemmas when data seems to partially confirm two contradictory hypotheses. A fruitful way forward in solving such dilemmas in science is equally fruitful in solving dilemmas in theology such as the “predestination or free will?” conundrum.

Solving Dilemmas in Science

In the study of science, we collect data. We then ask questions about which hypothesis the data confirms. These are empirical questions, focusing on the specific results of our experiments. This data results in empirical problems for one hypothesis, more so than the other.  

Sometimes this is all that is needed to answer a scientific question. The empirical data counts strongly in favor of one theory and strongly against its rival. 

However, sometimes this is not the case. The empirical data is not conclusive. Perhaps roughly half the data supports Hypothesis A, and the other half supports Hypothesis B. Or proponents of both hypotheses look at the data and interpret it to support their differing views.

At this point, the scientist may be tempted to take one of four paths similar to those outlined last week:

• She may decide she is going to be a supporter of Hypothesis A, dig her heels in and work to explain away data supporting Hypothesis B.

• She may decide she is going to be a supporter of Hypothesis B, dig her heels in and work to explain away data supporting Hypothesis A.

• She may just decide for pragmatic or sociological reasons to favor Hypothesis A or B. Perhaps she can get more funding for further research if she adopts Hypothesis A. Or maybe her colleagues will think better of her if she supports Hypothesis B. So, based on this she chooses Hypothesis A or B.

• She may decide both Hypothesis A and B are correct, but it is simply a mystery how this can be so.

Scientists do sometimes take one of these four pathways when faced with contradictory data. The reason is that these appear to be the only options if the empirical data are all that one considers.

However, philosophers of science have developed two additional questions to raise at this point. These questions have proven to be very effective in adjudicating between inconclusive and competing hypotheses which are both supported by the data. 

Applying This To Theology

We can learn much about how to find a way forward in the “predestination or free will?” debate (as well as other theological debates), by asking these same questions. This is because, as in the study of science, in the study of theology we also begin by collecting data (biblical passages) which help confirm one theological hypothesis or the other (Calvinism or Arminianism, for instance).

And, like in scientific investigation, in some cases this is all we need. For instance, the biblical data is conclusive that Jesus is divine. But in other cases, such as whether the Calvinist or Arminian hypothesis is correct, the data is less conclusive. As we have seen, some passages (data points) seem to support the Calvinist hypothesis, while other passages (data points) seem to support the Arminian hypothesis.

I suggest the path forward is the same as when scientific data supports two rival hypotheses. 

 The Second Question to Ask

At this point in scientific investigation a second question is asked: 

Are there any internal conceptual problems that count against one hypothesis or the other?

If so, this helps “tip the scales” in favor of the position with fewer internal conceptual problems.

Each of these terms is important to understand. The term “internal” refers to the field of study to which the question pertains. For instance, if you are wrestling with a question of chemistry, and the data from your experiments is inconclusive, you may consult other knowledge in the field of chemistry to help you adjudicate between the rival hypotheses. In this way this second question is internal—it is still within the field of chemistry.

“Conceptual” refers to the fact that these questions are not about the data directly (the data from the experiments done to answer the question). Rather, these are other truths known in the same field of study which may have a bearing on how to interpret the data obtained through the experiments.

For instance, as Warren Heisenberg discovered, one cannot determine the location and the momentum of a quantum particle.* If the location is determined, the momentum cannot be determined, and vice versa. From this one may develop the hypothesis that, at the quanta level, reality itself is indeterminate (the location and velocity of a particle are not fixed, but determined by the observer). 

However, other data from physics at the macro level indicates that reality is determinate, and not dependent upon the observer. This fact is then conceptual data, still within the field of physics, which is a problem for the view that quanta reality is observer-dependent. 

Finally, “problem” refers to the fact that the internal conceptual data counts against, or is a problem for, one of the two hypotheses supported by the experimental data, as in the illustrations above.

In summary, internal conceptual problems arise when something else in the same field of study is known to be the case, or most probably the case, which contradicts one of the hypotheses in question. This then counts against one of the hypotheses (and, therefore, adds support to the other). 

Conclusion

In the following weeks I will apply this question to the discussion at hand. Are there internal conceptual problems that help tip the scales in favor of Arminianism or Calvinism? I believe there are. 

Until then, grace and peace.

*At least not with our current instrumentation.