Allchin's Shoehorn, or Why Science is Hypothetico-Deductive