And now a confession.

Like a million years ago, when This Week With George Stephanopolopili was still This Week With David Brinkley, George F. Will produced a critique of somebody’s tortured logic. The fellow’s arguments were contradictory and inconsistent, Will said, and in my memory — recall that this must have been more than 20 years ago, probably while I was still an undergraduate and thus, Sunday falling when it does in the week, perhaps hung over — he called the argument a “bucket defence.” I have remembered the argument fondly, albeit vaguely, ever since, so I used the term yesterday in response to the Conservatives’ shameful forest of stupid rebuttals to Richard Colvin’s deeply disturbing testimony about treatment of prisoners in Afghanistan.

But the thing is, I have googled “bucket defence” many times and got nowhere. Maybe I dreamed the whole thing.

Today a reader of Parker Donham’s blog solves the riddle. ‘[A] “bucket defence” might be more familiar to some of your readers as Freud’s “kettle logic.” However, for Freud, the defence is used when the denier has difficulty accepting the truth, not simply difficulty admitting it.’

Here’s the best account I can find of Freud’s “kettle logic.”

“In the first place, he had returned the kettle undamaged; in the second place it already had holes in it when he borrowed it; and in the third place, he had never borrowed it at all. A complicated defence, but so much the better; if only one of these three lines of defence is recognized as valid, the man must be acquitted.”

The memory of George Will on that long-ago Sunday is thus redeemed. In future I will keep using the term “bucket defence,” but you’ll all know the term is, in part, a reference to the haziness of punditological memory.