We now return to the question of self-reference and self-identification, and examine issues which required a full delineation of IEM: the relation between IEM and guaranteed reference of "I"; the opposition between direct and indirect reference theories of "I"; and whether there is genuine self-identification in IEM cases. Proponents of direct reference, who hold that the self-reference rule fully captures the meaning of "I", hold that the apparent redundancy of identifying knowledge shows that the reference of "I" is unmediated by a sense. Fregean proponents of indirect reference, in contrast, deny that grasp of the self-reference rule is sufficient to characterise self-consciousness, and require in addition apprehension of a sense for "I", which they must reconcile with apparent absence of identifying knowledge. This chapter attempts to undermine the debate between direct and indirect reference by drawing on Wittgenstein's "no-reference" view of "I", and arguing that IEM-judgements do not involve genuine self-identification.