What is the difference between Swinburne's method (Bayesian cumulative argument) and Plantinga's method (Reformed epistemology)?

This question places us before the most important methodological division in contemporary analytic philosophy of religion. Richard Swinburne and Alvin Plantinga, both prominent Christian philosophers, represent two radically opposing approaches to the question of religious knowledge. The first builds a Bayesian evidential structure, the second rejects the need for argumentation altogether. Understanding this disagreement is necessary for comprehending the contemporary philosophical landscape.

Inadequate responses to avoid

From some defenders of faith:

"Both prove God's existence, so the disagreement is merely technical." A misleading oversimplification. The disagreement is not in the conclusion but in the very nature of religious knowledge. Swinburne sees belief in God as needing evidential justification; Plantinga sees it as basic knowledge requiring no justification.

"Plantinga's method is stronger because it doesn't depend on arguments that could be refuted." An error in understanding epistemic strength. Not depending on arguments doesn't necessarily mean greater strength. It might also mean inability to convince those who don't share basic intuitions.

From some critics:

"Swinburne uses probabilities in religious matters, which is illogical." A superficial criticism. Bayesian probability theory is a valid mathematical tool for any domain of knowledge, including religion, as long as hypotheses are formulated precisely.

"Plantinga makes faith a purely subjective matter." A misunderstanding of his theory. Reformed epistemology doesn't mean subjectivity, but rather that some beliefs can be justified without argumentation, like belief in the external world or the past.

Swinburne's Method: Bayesian Cumulative Construction

Swinburne in his book "The Existence of God" (1979, second edition 2004) presents a method that can be summarized in five steps:

Step One: Formulating the hypothesis. The theistic hypothesis (h): There exists one God, eternal, omnipotent, omniscient, perfectly good, creator and sustainer of the universe.

Step Two: Identifying the evidence. The existing universe, its mathematical laws, fine-tuning, existence of consciousness, religious experience, miracles, etc.

Step Three: Applying Bayes' theorem. P(h|e&k) = P(e|h&k) × P(h|k) / P(e|k)

Where:
- P(h|e&k): probability of hypothesis given evidence
- P(e|h&k): probability of evidence given hypothesis is true
- P(h|k): prior probability of hypothesis
- P(e|k): total probability of evidence

Step Four: Individual arguments. Each piece of evidence is analyzed Bayesianly:
- Cosmological argument: universe's existence more likely under theistic hypothesis
- Fine-tuning argument: physical constants calibrated for life
- Consciousness argument: emergence of consciousness in material universe
- Religious experience argument: widespread religious experiences across cultures

Step Five: Accumulation. Evidence accumulates. Each piece may be weak alone, but together they raise the total probability of the theistic hypothesis above 0.5, making it "more likely than not."

Swinburne is explicit: this doesn't provide certainty, but rational probability (rajḥān ʿaqlī).

Plantinga's Method: Reformed Epistemology

Plantinga in his epistemic trilogy (Warrant: The Current Debate, 1993; Warrant and Proper Function, 1993; Warranted Christian Belief, 2000) presents a radically different method:

Basic principle: Not all beliefs need evidence. Some beliefs are "properly basic." For example: "I see a tree," "I remember eating breakfast," "Others have minds." These are beliefs we don't argue for, but form directly within us.

Belief in God can be basic. Through what he calls "sensus divinitatis" (divine sense), an innate cognitive capacity that produces belief in God under certain circumstances (seeing nature, feeling guilt, religious experience).

Extended Aquinas/Calvin model. The Holy Spirit works with the divine sense to produce genuine Christian knowledge. This knowledge is "warranted" if it is:
- Produced by cognitive faculties functioning properly
- In appropriate environment
- According to design plan aimed at truth

No need for arguments. The believer doesn't need arguments to be rationally justified in faith, just as one doesn't need arguments to believe one sees a tree.

Arguments are defensively useful. Despite not being necessary for justification, arguments (like the ontological argument Plantinga developed) are useful for "defeating the defeaters" and showing that atheism has no rational superiority.

Radical Differences

Nature of religious knowledge. Swinburne: evidential, needs argumentative construction. Plantinga: basic, can be immediate.

Role of evidence. Swinburne: necessary for rational justification. Plantinga: unnecessary (though potentially useful).

Success criterion. Swinburne: raising probability above 0.5. Plantinga: showing that faith can be warranted.

Target audience. Swinburne: any rational person, believer or non-believer. Plantinga: primarily believers (for reassurance) and philosophers (to defeat their objections).

Criticizability. Swinburne's method is open to detailed criticism (are the probabilities correct?). Plantinga's method is harder to criticize (how do you prove the divine sense doesn't exist?).

Strengths and Weaknesses

Swinburne:
- Strength: provides general argument open to rational debate
- Strength: deals with counter-evidence (like evil) within Bayesian framework
- Weakness: depends on probability estimates that may be subjective
- Weakness: may not reach sufficient certainty for living religious faith

Plantinga:
- Strength: protects faith from impossible evidential demands
- Strength: aligns with believers' actual experience
- Weakness: provides no way to convince non-believers
- Weakness: method could be applied to any religious belief (Great Pumpkin Objection)

Contemporary Debate

The division between these methods reflects a deeper split in epistemology: between classical foundationalism and reformed foundationalism. Swinburne is closer to the former, Plantinga pioneers the latter.

Contemporary philosophers attempt synthesis or mediation. Michael Sudduth integrates natural arguments with religious experience. Paul Moser proposes "transformative knowledge" combining reason and heart.

Connection to god-database Method

The cumulative rational probability (rajḥān ʿaqlī) method is closer to Swinburne in relying on evidence, but it:
- Doesn't claim to reach precise numerical probabilities
- Accepts religious experience as epistemic source (closer to Plantinga here)
- Sees the six masālik as accumulative without need for explicit Bayesian calculation

Philosophical Conclusion

The disagreement between Swinburne and Plantinga isn't merely technical, but reflects two different visions of religious rationality. Swinburne sees religion as hypothesis needing support; Plantinga sees it as basic experience. Both offer valuable contributions: Swinburne in building rational bridges with non-believers, Plantinga in protecting believers from unrealistic evidential demands.

For Advanced Reading

─ Advanced level: Plantinga's critique of natural theology, and Swinburne's response
─ Swinburne, The Existence of God (2nd ed., 2004)
─ Plantinga, Warranted Christian Belief (2000)
─ The Analytic Theist (collection of essays on Plantinga, 1998)
─ Page "Taxonomy: Natural Theology vs Reformed Epistemology"