Pascal’s Wager
Posted: Wed, Oct 1, 2025
Epistemic vs. pragmatic reasons for belief
- Epistemic reasons for belief have the right kind of connection to truth or knowledge.
- Pragmatic reasons for belief are merely practical.
Example: I Saw the TV Glow is a brilliant movie.
- An epistemic reason to believe this: A lot of my friends say so, and they know what they are talking about.
- A pragmatic reason to believe this: It will make my girlfriend happy.
Pascal’s move: We have a compelling pragmatic reason to believe in God.
Decision under uncertainty
| Friend @ Prospect Park (25%) | Friend @ MoMA (75%) | |
|---|---|---|
| Go to Prospect Park | 50 | 25 |
| Go to MoMA | 25 | 100 |
Expected utility: How good/bad the outcome is expected to be.
- EUPP = 25% × 50 + 75% × 25 = 31.25
- EUMoMA = 25% × 25 + 75% × 100 = 81.25
Applied to belief in God
| God (x%) | No God (100% − x%) | |
|---|---|---|
| Believe in God | ∞ | 0 |
| Don’t believe in God | −∞ | 0 |
The value of x does not matter:
- EUBelieve = x% × ∞ + (100% − x%) × 0 = ∞
- EUDon’t believe = x% × −∞ + (100% − x%) × 0 = −∞
You may think there is a cost to believing in God: The argument goes through so long as the cost is not infinite.
A third row?
| God (x%) | No God (100% − x%) | |
|---|---|---|
| Believe in God | ∞ | 0 |
| Wager for God | ? | 0 |
| Don’t believe in God | −∞ | 0 |
Too many gods?
You can run the argument over and over again for many, many Gods—this can even go beyond actual world religions.
Omnibenevolence?
| God (x%) | No God (100% − x%) | |
|---|---|---|
| Believe in God | ∞ | 0 |
| Don’t believe in God | ∞ | 0 |
Inverted God?
What if God is a philosopher who wants you to believe only what you have good epistemic reasons to believe?
| God (x%) | No God (100% − x%) | |
|---|---|---|
| Believe in God | −∞ | 0 |
| Don’t believe in God | ∞ | 0 |