Full Idea
If we go for the necessity-of-origins view, A and B are different if the origin of A is different from the origin of B. But one is left with the further question 'When is the origin of A distinct from the origin of B?'
Gist of Idea
If you individuate things by their origin, you still have to individuate the origins themselves
Source
Cover,J/O'Leary-Hawthorne,J (Substance and Individuation in Leibniz [1999], 7.4.1)
Book Reference
Cover,J/O'Leary-Hawthorne,J: 'Substance and Individuation in Leibniz' [CUP 1999], p.273
A Reaction
There may be an answer to this, in a regress of origins that support one another, but in the end the objection is obviously good. You can't begin to refer to an 'origin' if you can't identify anything in the first place.