The evidential problem of evil argues that the amount, distribution, and apparent gratuitousness of suffering in the world constitutes strong probabilistic evidence against the existence of an omnipotent, omniscient, and perfectly good God. Unlike its logical counterpart, this formulation does not claim that God and evil are strictly incompatible, but rather that the specific patterns of evil we observe—particularly instances of seemingly pointless suffering—make God's existence highly improbable. The argument typically proceeds by identifying cases of intense suffering that appear to serve no greater good, then inferring that an all-powerful and loving God would likely prevent such evils, thus their existence provides evidence against theism.
The modern evidential formulation emerged prominently with William Rowe's 1979 article "The Problem of Evil and Some Varieties of Atheism," which introduced the famous example of a fawn dying slowly in a forest fire. Rowe argued that such apparently gratuitous evils lower the probability of God's existence. Paul Draper advanced the argument in "Pain and Pleasure: An Evidential Problem for Theists" (1989), employing Bayesian probability to argue that naturalism better explains the distribution of pain and pleasure than theism. Michael Tooley's "The Problem of Evil" (2019) provides comprehensive statistical analyses of suffering patterns. Earlier precursors include David Hume's Dialogues Concerning Natural Religion (1779), which questioned whether the world's evils cohere with divine benevolence.
Theistic responses have developed along several lines. Skeptical theists like William Alston ("The Inductive Argument from Evil," 1991) and Michael Bergmann ("Skeptical Theism and Rowe's New Evidential Argument," 2001) argue that human cognitive limitations prevent us from assessing whether apparently gratuitous evils truly lack justifying reasons. Peter van Inwagen's "The Problem of Evil" (2006) contends that we cannot make reliable probability judgments about what God would permit. Richard Swinburne's Providence and the Problem of Evil (1998) offers theodicies explaining how various evils enable greater goods like free will, soul-making, and moral responsibility. Marilyn McCord Adams' Horrendous Evils and the Goodness of God (1999) argues that God can defeat evils through each person's overall life narrative.
The evidential problem differs from the logical problem of evil, which claims strict incompatibility between God and any evil whatsoever. While the free will defense effectively addresses the logical version, it less clearly handles evidential concerns about excessive or seemingly pointless suffering. Natural evil arguments focus specifically on suffering from non-human causes, while the evidential problem encompasses all evil types. Skeptical theism responds specifically to evidential arguments by challenging our ability to assess divine reasons, rather than offering positive theodicies like soul-making approaches that explain evil's purpose.