Bayesian Apologetics
The application of Bayes’ Theorem to argue for the probability of theological or supernatural claims — most famously the resurrection of Jesus. This approach attempts to quantify faith claims using formal probability calculus.
Key Proponents
- richard-swinburne — the most prominent modern advocate. In The Resurrection of God Incarnate, he calculated a 97% probability that Jesus was God incarnate and rose from the dead. 1
- William Lane Craig — uses probabilistic and inductive frameworks (less strictly mathematical than Swinburne). 1
The Contradiction: Opposing Conclusions
The same Bayesian framework has been used to argue diametrically opposed positions:
- Pro-resurrection: Swinburne’s 97% probability
- Anti-historical Jesus: Richard Carrier used Bayes’ Theorem to argue Jesus likely never existed
This divergence is itself cited as evidence of the method’s fundamental problems — the outputs depend entirely on subjective prior assignments. 1
Core Critiques
See also: subjectivity-of-priors, dwindling-probabilities, humes-argument-against-miracles
-
subjectivity-of-priors — The priors P(H) and P(E|H) are entirely speculative when H involves a supernatural being. The psychology and intentions of a divine entity cannot be modeled statistically.
-
dwindling-probabilities — alvin-plantinga’s critique: complex theological claims require chains of hypotheses, and multiplying probabilities across the chain makes a high posterior nearly impossible.
-
humes-argument-against-miracles — david-hume’s foundational argument that the evidence against miracles (uniform experience of natural law) always outweighs the evidence for them (fallible human testimony).
-
Historical Methodology Clash — Historians like Bart Ehrman argue that miracles, as violations of natural law, are inherently the least probable events. Historical methods cannot validate what defies probability itself. 1
-
post-hoc-probability-fallacy — Probability tests specified after the fact are a form of confirmation-bias reasoning: some outcome had to occur, and “the probability of X occurring, given that X has been observed to occur” is unity. Unlike the critiques above, which reduce evidential weight, this one nullifies the argument — design-style probability calculations built on already-known outcomes are its home territory. 2
-
material-theory-of-induction — Norton’s material theory of induction denies that any formal calculus supplies universally applicable inductive rules; in particular, his “completely neutral support” result (Ch. 10) shows that objectively “neutral” background support cannot be defined — the indifference route to objective priors provably fails. For apologetic arguments this undercuts the standing hope that quantified priors for supernatural hypotheses can be read off from ignorance alone. 3
Academic Consensus
There is no academic consensus supporting Bayesian apologetics. The approach is dismissed across disciplines:
- Historians: Viewed as a category error — math cannot rescue historically uncorroborated claims
- Secular philosophers: Mathematical theater dressing subjective theology in statistical language
- Theistic philosophers: Many prefer reformed-epistemology, viewing Bayesian quantification of faith as misguided and reductive
See Also
- reformed-epistemology
- richard-swinburne
- david-hume
- alvin-plantinga
- reasoning-methods — Where Bayesian confirmation theory fits among the modes of inference
- fine-tuning-argument — The most prominent Bayesian-style design argument; Carroll’s Within Reason critique and the measure problem
- post-hoc-probability-fallacy — Probability tests specified after the fact (critique 5)
- kolmogorov-probability-axioms — the axiomatic foundation of the calculus these arguments apply: it fixes the mathematics of probability, not its interpretation.
- material-theory-of-induction — the “no universal rules” critique, including why neutral priors cannot be defined