Participants in the “ontological commitment” debate would benefit from distinguishing two different ways of understanding the concept. If the question at hand is “what a theory says it is” or “what a theory claims exists,” we are discussing an “explicit” commitment, whereas if we are asking about the ontological costs or truth preconditions of a theory , we investigate “implicit” commitment. I defend a conception of ontological commitment as implicit commitment; I also develop and defend an account of existentially quantified idioms in natural language that sees them as implicitly, but not explicitly, demanding. Finally, I use the distinction between two types of ontological commitment to diagnose a flaw in a widely used argument that existential quantification is not ontologically committed. The question of ontological commitment is the question of "what does a theory say that it exists". All this is familiar to any student of Quine. See Quine, "On What There Is," repr. in his From a Logical Point of View (Cambridge: Harvard University Press, 1980), pp.1-19, especially p.15ff. However, the theory of ontological commitment remains in poor shape, and consensus is lacking even regarding the most basic questions: how should we give a precise formulation to the notion of ontological commitment, and should we treat existentially quantified idioms as ontologically committing? Without agreement on fundamental questions like these, ontology is an impossible discipline, because unless we understand which phrases in the language of our theory can be used in ontologically demanding ways, we cannot know whether the theories advanced by aspiring ontologists have had success. the ontological meaning they want them to have. In fact...... half the paper ...... in any sentence that makes ontological claims about s but says about some named objects that they are s. On some not unreasonable views of situational semantics, in particular, that sentences containing names are not existential quantifications in disguise, and that there can be no truth expressed by a sentence of the form "a is a" where the name a fails to refer to. The theory cannot be true unless there are s, yet the theory contains no sentence like "There are s" that would count as an ontological claim about s. Although s are not among the theory's explicit commitments, it is certainly true that the theory is committed to s in some sense, because if the theory is true then s must exist, and anyone who holds the theory still doesn't believe in it. Its existence deserves to be be censured for failing to recognize the ontological cost of his theory.