In this paper I wish to address two questions about reference that are among the most fundamental issues in the theory of meaning and intentionality. They are 1) what makes different external (e.g. causal) relations count as semantic relations, count as reference; and 2) whether reference is objectively indeterminate or inscrutable and whether a positive answer subverts our commonsense conceptions of semantic facts. It appears to me that the two matters are deeply connected, and I will propose a more or less simple idea that permits a unifying answer.