God commands the Israelites to commit genocide against the Canaanites, including killing women and children, which is straightforwardly monstrous and incompatible with a good God.

Why people find it persuasive

Deuteronomy 20:16-18 and Joshua's conquest narratives describe herem, total destruction of Canaanite cities, explicitly including noncombatants. No amount of theological sophistication changes the fact that a text presents God ordering the killing of infants. If this is what divine command looks like, it is hard to see how the God of the Bible is meaningfully different from any other tribal deity used to justify conquest.

Important distinctions

• • Ancient Near Eastern hyperbolic conquest rhetoric versus literal, complete extermination
• • Herem as a unique, non-repeatable command tied to a specific historical judgment versus a general policy
• • Archaeological evidence about the actual conquest versus the text's rhetorical genre
• • Divine command theory's scope (can God command anything?) versus this specific historical claim
• • The text's own internal tension, since Canaanites are shown surviving throughout Judges

Christian response

• • Old Testament scholars including Kenneth Kitchen and, from a more critical angle, William Dever, note that conquest narratives use standard ancient Near Eastern hyperbolic military rhetoric. Phrases like 'utterly destroyed' and 'left nothing that breathed' appear in other ancient inscriptions (such as the Moabite Mesha Stele) describing battles where the enemy demonstrably continued to exist afterward. Judges 1-3 shows Canaanites still living throughout the land immediately after the conquest, which is textually incoherent if Joshua is read as literal total extermination rather than the genre's standard exaggeration for a decisive military victory.
• • Scholars like Paul Copan, in Is God a Moral Monster? (2011), argue the primary function of herem language was to drive Canaanite political and religious structures out of the land rather than to mandate house-by-house killing of every individual, pointing to the fact that the same texts allow for Canaanites like Rahab and the Gibeonites to be spared and incorporated. This is a contested reading, and not all scholars, including many evangelical ones, accept it fully.
• • The text situates the conquest as a specific, bounded historical judgment on entrenched Canaanite religious practices, including child sacrifice to Molech (Leviticus 18:21, 2 Kings 3:27), rather than establishing a general principle that any nation may be exterminated for any reason; Genesis 15:16 frames the timing explicitly around the 'iniquity of the Amorites' not yet being complete, suggesting judgment rather than arbitrary conquest, and the same standard is later applied to Israel itself when it is expelled from the land for comparable practices.
• • Even granting all this, many contemporary Christian philosophers, including Randal Rauser and William Lane Craig himself in different ways, acknowledge this remains among the most morally difficult passages in scripture, and legitimate Christian scholarship disagrees sharply on whether hyperbole-based readings fully resolve the difficulty or merely soften it; some, like Old Testament scholar Eric Seibert (Disturbing Divine Behavior, 2009), argue the text should be read as ancient Israel's flawed theological interpretation of history rather than an accurate record of divine command.
• • What can be said with more confidence is that this is a bounded, historically situated narrative, not a template scripture presents for ongoing conduct; nothing in Christian ethics from the prophets onward, through Jesus's teaching to love enemies (Matthew 5:44), extends herem as an abiding moral instruction, unlike the way it functioned in later conquest ideologies wrongly borrowing biblical language.

Christian counterresponse

Reading ancient military texts through their known genre conventions is standard historical method, not selective special pleading; the same interpretive move is used by secular historians on Egyptian, Assyrian, and Moabite inscriptions with comparable hyperbolic destruction claims, and the internal contradiction between 'total destruction' language and immediately following narratives of ongoing Canaanite presence is the text's own evidence that hyperbole is the intended genre.