How does Robin Collins formulate the fine-tuning argument in Bayesian terms, and what degree of support does it provide for theism?

Robin Collins's Bayesian formulation of the fine-tuning argument represents one of the most rigorous attempts to transform intuitions about "cosmic wonder" into a disciplined probabilistic argument. Collins, a philosopher of science specialized in theoretical physics, has presented since 1999 a formulation that moves beyond general impressions to precise mathematical analysis.

Inadequate Responses to Avoid

From some believers:

"Fine-tuning proves God's existence decisively." This is a methodological error. Even Collins himself emphasizes that his argument is probabilistic, not demonstrative. The argument increases the probability of theism, but it doesn't provide mathematical certainty.

"Every physicist who acknowledges fine-tuning believes in God." This is incorrect. Many physicists acknowledge fine-tuning as a phenomenon, but they explain it in different ways (multiverse, physical necessity, anthropic principle). Acknowledging the phenomenon doesn't mean accepting the theistic explanation.

From some naturalists:

"Fine-tuning is an illusion resulting from our ignorance of fundamental physics." This is a strong claim without sufficient evidence. Most specialized physicists (including atheists like Leonard Susskind) acknowledge that fine-tuning is a real phenomenon requiring explanation, even if they differ on the explanation.

"String theory will solve the fine-tuning problem." This is currently unjustified hope. String theory at best transfers the problem from one level to another (why this solution among 10^500 possible solutions?). Moreover, the theory itself remains experimentally unproven.

Why These Responses Are Inadequate

They share a failure to engage with the probabilistic structure of the argument. Collins doesn't claim decisive proof, but presents a Bayesian analysis that compares relative probabilities. A serious response requires understanding the Bayesian framework first.

Basic Bayesian Framework

Bayes' theorem in probability theory allows updating the degree of belief in a hypothesis based on new evidence. The basic formula:

P(H|E) = P(E|H) × P(H) / P(E)

Where:
- P(H|E) = probability of hypothesis after evidence (posterior)
- P(E|H) = probability of evidence if hypothesis is true (likelihood)
- P(H) = probability of hypothesis before evidence (prior)
- P(E) = total probability of evidence

In the fine-tuning context:
- H₁ = theistic hypothesis
- H₂ = atheistic naturalism hypothesis
- E = evidence of fine-tuning of constants

Collins's Formulation of Fine-Tuning

Collins focuses on six main examples of fine-tuning:

1. Cosmological constant (Λ): Fine-tuned to one part in 10^120. If slightly larger, the universe would expand too rapidly to allow galaxy formation. If smaller (negative), the universe would collapse on itself.

2. Strong nuclear force: If 50% weaker, no atoms heavier than hydrogen would form. If 2% stronger, all hydrogen would burn to helium in the first minutes.

3. Mass ratio (proton/electron): Fine-tuned to 0.02%. A slight change prevents complex chemistry.

4. Electromagnetic constant (α): A 4% change prevents star formation.

5. Initial conditions of the universe: Astoundingly low entropy (probability 1 in 10^10^123 according to Penrose).

6. Dimensions of space: Only three spatial dimensions allow stable orbits and proper wave propagation.

Detailed Bayesian Analysis

Collins analyzes the probabilities as follows:

P(E|H₁) - probability of fine-tuning under theism:
Collins argues this probability is "not excessively low." A God wanting to create conscious life has reasons to create a fine-tuned universe. An exact number cannot be determined, but Collins estimates it as "reasonable" (perhaps 10^-2 to 10^-1).

P(E|H₂) - probability of fine-tuning under naturalism:
Here lies the core of the argument. In a single random universe, the probability of obtaining all fine-tuned constants together is astronomically low. Collins estimates this at less than 10^-100 (and possibly much lower).

Likelihood Ratio:
P(E|H₁) / P(E|H₂) > 10^50

This means fine-tuning is at least 10^50 times more probable under theism compared to naturalism.

Addressing Main Objections

Multiverse objection:
Collins acknowledges that the multiverse could explain fine-tuning, but he raises the "second-level fine-tuning problem": the universe-generating mechanism itself requires fine-tuning. For example, eternal inflation requires a fine-tuned inflation field.

Anthropic principle objection:
"We observe a fine-tuned universe because we wouldn't be here if it weren't fine-tuned." Collins responds with the "firing squad" analogy: if 50 rifles were fired at you and you survived, it's true you'd only observe survival, but this doesn't eliminate the need to explain why all shots missed.

Physical necessity objection:
Perhaps the constants can only be as they are. Collins responds that this:
1) Is mere speculation without evidence
2) Contradicts the independence of constants in current theories
3) Even if true, raises a deeper question: why are necessary laws fine-tuned for life?

Degree of Support for Theism

Collins is cautious in his conclusions:

1. The argument supports "design" not classical theism directly. The transition from designer to the God of Abrahamic religions requires additional arguments.

2. The support is probabilistic, not decisive. Even with a likelihood ratio of 10^50, this doesn't mean certainty. If your prior probability for atheism is very high, you might remain atheist despite the evidence.

3. The argument is part of a cumulative case. Collins sees fine-tuning as part of a set of evidence (cosmological, moral, religious) that cumulatively supports theism.

Post-Collins Developments

Luke Barnes (2012) developed a more rigorous analysis of fine-tuning in physics.
Geraint Ellis and Joe Silk (2014) acknowledged the strength of the fine-tuning problem from a physics perspective.
Atheist philosopher Thomas Nagel (2012) acknowledged that fine-tuning poses a genuine challenge to naturalism.

Where We Stand Today

Collins's Bayesian formulation remains among the strongest contemporary formulations of the fine-tuning argument. Even critics acknowledge its technical rigor. Current debate revolves around:

1. Accuracy of probability estimates
2. Reasonableness of the multiverse hypothesis as an alternative
3. Relationship between fine-tuning and other design arguments

The balanced position — according to the "rational preferability" (rajḥān ʿaqlī) method — is that Collins's argument provides considerable probabilistic support for design, without reaching decisive proof. Its strength lies in combining with other evidence in a cumulative framework.

For Advanced Reading

- Robin Collins, "The Teleological Argument" in The Blackwell Companion to Natural Theology (2009)
- Luke Barnes, "The Fine-Tuning of the Universe for Intelligent Life" (2012)
- Advanced level: Sean Carroll's Bayesian critique and Collins's response
- "Fine-Tuning Argument: Bayesian Formulation" page on the website