“The best [counter-argument] is probably the issue of deep mining,” Schmidt said. “We are mining very old deposits for metals and ore, and if these had been depleted by a prior civilization, it might have been noticed. These place limits on the scope of a prior civilization, but don’t eliminate the possibility.”
This has interested me before, and I think it's good evidence to the contrary.
I am not a geologist, but from what I know, "Very old" means the precambrian shield in Canada, for example, which is among, if not the, oldest rock on earth (~3.6Gy). Oldest, because before that time, the planet was molten on its surface. We're mining that today, and the geology can be fit to geological models: afaik, the geology being encountered in the very old rock is what people expect to see from naturally-occurring processes.
I do still wonder, though, what rock would look like after ~billions of years, if parts of it had been mined out, and whether that mining would be detectable. The rock is hard rock; would you see a corridor of rubble in solid rock, from a collapsed tunnel? Or are there still enough geological forces acting on the rock to turn the rubble from a collapsed tunnel into something resembling metamorphic placer (river-borne) rock, for example?
Interesting question, either way.