Must I Split the Difference with an Equally Reliable Disagreer?

Two friends independently calculate how to split a restaurant bill. They are normally equally reliable at mental arithmetic, have seen the same bill, and reach different answers. Once each learns of the disagreement, simply insisting “I did the calculation, so I must be right” looks dogmatic. Yet philosophers disagree about whether both must reduce their confidence to exactly the halfway point.

Peer disagreement is not just any opposing opinion. It concerns people who are roughly equal in relevant ability, access to evidence and care. The Equal Weight View says the other person’s judgement should count as much as one’s own, and that disagreement itself cannot be used to dismiss the peer as unreliable. Steadfast views allow the original, first-order evidence sometimes to sustain one’s position. Between them, total-evidence approaches say we must consider both the evidence for the answer and the new higher-order evidence that another reliable judge reached the opposite conclusion.

My view is that peer disagreement should at least trigger rechecking, but not mechanical averaging. If the calculations were independent and equally transparent, and no further check is available, suspending judgement may be sensible. If one person copied the other’s intermediate figures, the two answers are not independent evidence. If an itemised till record can be checked, that new evidence may break the symmetry. Intellectual humility does not count every opinion as one vote; it acknowledges another person’s reliability while asking how the evidence is related.

https://plato.stanford.edu/entries/disagreement/
https://plato.stanford.edu/entries/higher-order-evidence/