Identity is not like love. One can be in love with a person at one moment, not in love with that person at another moment, and perhaps again in love with that person at another, even later moment. Not so with identity: one thing (e.g., you reading this) cannot be identical with a thing (e.g., a certain person) at one moment and not identical with that thing at a later moment.Footnote ¹ If you are not identical with a certain person at this moment, you never were and you never will be. And if you are identical with a certain person, you are stuck with yourself.

With this is in mind, consider the following, which I've adapted from Allan Gibbard:Footnote ² I have before me a piece of clay, which I decide to name ‘Lumpy’. Out of Lumpy I fashion (at time t) a statue bearing a great likeness to Margaret Thatcher, and I call the statue ‘Maggie’. It may seem that Lumpy = Maggie, but by Leibniz's Law, that is, the Law of the Identity of Indiscernables, Lumpy and Maggie do not share all properties. While they physically occupy the same spatial region and are constituted out of the same stuff, they do not share the property of coming into existence at time t – only Maggie has that property; Lumpy came into existence as Lumpy (that is, as that particular, individual piece of clay) at sometime before t.

So I want to alter things just a bit. I have before me two pieces of clay. I fashion one into a statue of the lower half of Margaret Thatcher, the other into a statue of the top half of Margaret Thatcher. Putting the two pieces together at time t, I at once create a lump of clay (which I name ‘Lumpy’) and a statue (which I name ‘Maggie’). Unlike our previous example, Lumpy and Maggie share all their properties in common. Not only are they physically occupying the same spatial region and constituted out of the same stuff, they now share the property of ‘coming into existence at time t’.

Now let's say that this particular brand of clay is a quick-drying clay with a precise drying time – indeed, it hardens in an instant – such that I know that at exactly 5:00 this afternoon, the clay out of which Maggie is made will harden, and to such a degree that to destroy Maggie will be to destroy Lumpy. In other words, after 5:00, in shattering Maggie I will break Lumpy into shards that cannot again be combined to form a lump of clay (as they are too hardened). And let's say that I let the 5:00 hour pass, and it is only at 5:15 that I destroy Maggie and Lumpy with a blow of a hammer. It would seem that it is correct to say that Lumpy = Maggie: they came into existence at the same moment, occupied the same space throughout their existence, were made of the same stuff, and went out of existence at the same moment. In other words, all their properties were held in common in a way that satisfies Leibniz's Law.Footnote ³

But let's alter things once more. Imagine now that at 4:55 I destroyed Maggie, in this case by smushing (as opposed to smashing, for the clay has not hardened and is still soft). Lumpy, though, survives just fine. In doing so, I have made it such that Lumpy = Maggie is false, for now they do not share the property of ‘going out of existence at 4:55 on such-and-such a day’.

So where are we? Had I done nothing until after 5:00, Lumpy = Maggie is true and was true during its span of existence. But in smushing Maggie, I rendered it such that Lumpy = Maggie is false. Yet if it is false at 4:55, it was always false, going back to time t – as said in the opening, something cannot gain or lose identity with something over time. Something that is identical to something was identical to that something since the time it became that thing and will continue to be identical with that thing until that thing perishes. Conversely, if two things are distinct, they cannot have been identical in the past, nor can they possibly become identical in the future. In this case, however, at 4:55, the question of whether Lumpy = Maggie does not have an answer. Should I give into my urge to smush Maggie, then Maggie and Lumpy are not identical, and thus never were identical from t to 4:54:59; if I temporarily resist the urge and let the 5:00 hour pass, only to give into a similar urge at 5:15, Lumpy = Maggie is true, and thus always was true, including the period from t to 4:54:59. So the question is, how can something which will occur after 4:54:59 (either a smushing at 4:55 or a smashing at 5:15) determine, retroactively, whether from time t to 4:54:59 Lumpy = Maggie is true or not?

Two further observations: there is something odd about the fact that from time t until 4:59:59, the question of whether Lumpy = Maggie is unanswerable.Footnote ⁴ Further, though identity or non-identity between Lumpy and Maggie should depend on the intrinsic relation between them (viz., whether they share all properties in common) and on no one else, in this case the question of whether they are identical depends on something extrinsic to the Lumpy-Maggie relation, namely me and my ability to control my destructive urges.

Of course, these concerns go away once we abandon Gibbard's view (it was David Lewis' view as wellFootnote ⁵) that when the statue and a lump are coincidental throughout ‘their’ histories, we have only one thing before us. And that these concerns go way when we abandon the view goes, I think, quite a way toward thinking we should so abandon it.