Evidence and a cumulative case for God › The hypothesis being assessed

Suppose several observations each seem to favor an explanation. One concerns why anything exists, another why nature follows intelligible laws, and another why conscious beings exist. Does putting them together strengthen a case, or merely repeat an assumption in different forms? Richard Swinburne's The Existence of God asks how considerations of this kind might combine in an argument for God.

His project is a cumulative argument. He examines whether theism provides a probable explanation of broad features of experience. The conclusion is intended to remain open to uncertainty. Following the argument requires separating a consideration that supports a hypothesis from evidence strong enough to make that hypothesis more probable than not.

Swinburne treats the existence of God as a claim about reality that can be assessed through explanatory reasoning. The proposed God is a personal agent with unlimited power and knowledge, subject to logical constraints, who creates and sustains the universe. Divine goodness is part of the religious conception he develops. The proposal is more specific than saying that the universe has an unidentified cause.

The personal element matters. An explanation in terms of a person appeals to capacities, beliefs, intentions, and reasons for acting. It differs from explaining an event through preceding physical conditions and laws. Swinburne asks whether a personal explanation can account for features that a scientific explanation presupposes.

Consider a hypothetical garden. One explanation of its arrangement concerns soil conditions and the growth of plants. Another concerns a gardener's intention to arrange paths and flowers. The explanations answer different questions and need not compete. This example illustrates personal explanation; it does not establish that the universe resembles a garden in the ways required to support a divine gardener.

The argument must do that additional work. We cannot infer a divine person merely from noticing that some ordinary events have personal explanations. Swinburne proposes criteria for assessing explanatory hypotheses and argues that theism performs well under them. The credibility of those criteria and their application form part of the case.