My neighbor has been stealing my newspaper. And when I confront him, he apologizes with a sarcastic, condescending air, as if to say that he’s surprised I can read at all. I find it impossible to forgive him, but then I learn that he’s about to lose his job. He’s an aging executive with a large family to support, and I’m sure that this misfortune will soften his scorn and make him more sincerely apologetic. I decide to forgive him when all this happens.

This seems odd — if I’m sure that he’ll lose his job and express real contrition for stealing the paper, why do I have wait for this to happen? Why can’t I forgive him now?

Another twist: I learn that I (and only I) can save his job. This would amount to doing him a large favor, so I feel justified in withholding my help until I’ve forgiven him. But is this fair? Can I refuse to help him until I get a sincere apology, knowing that this will happen only after he loses his job?

Xanthippe is angry that Socrates is late, but she knows that he’ll apologize when she starts making dinner. Knowing this, can’t she skip the dinner and just forgive him? “In other words,” asks Tennessee State University philosopher James Montmarquet, “knowing that he would apologize, may she not still forgive him — having elected, for quite good reasons, not to allow conditions apt for his apology even to take place?”

(James Montmarquet, “Planned Forgiveness,” American Philosophical Quarterly 44:3 [July 2007], 285-296.)

In January 1942, Soviet Air Force lieutenant Ivan Mikhailovich Chisov was serving as navigator on an Ilyushin Il-4 bomber when an attack by Messerschmitt fighters forced him to bail out.

He left the plane at 6,700 meters and decided to forgo opening his parachute until he’d dropped below the level of the battle. But due to the thin atmosphere he passed out before he could pull the ripcord.

At an estimated 200 kph he struck the edge of a ravine whose steep sides were covered in deep snow. He tumbled to the bottom, where cavalrymen found him alive and still wearing his unopened parachute. He spent a month in critical condition with a broken pelvis but was flying again three months later.

This map shows mountain symbols above a river symbol. Suppose that, in the part of the world that the map represents, there really are mountains in the location that the map indicates. But suppose that there are also mountains on the other side of the river — where no mountains are indicated on the map. Would we say that the map is inaccurate?

People tend to say yes — in general, if a marker appears on a map, then we tend to think that the absence of the marker reflects an absence of that feature from the corresponding location.

But this is very different from linguistic representation. “After all,” writes Rutgers philosopher Ben Bronner, “if I say that there are mountains north of the river, the accuracy of my assertion doesn’t depend on whether there are mountains south of the river.”

And suppose I drew a map on which some national capitals were indicated, but not all. We wouldn’t take this to mean that the absent capitals don’t exist, merely that the map is incomplete. So how can we make sense of the intuition?

(Ben Bronner, “Maps and Absent Symbols,” Australasian Journal of Philosophy 93:1 [2015], pp. 43-59.)

The town of Bozouls in the south of France sits at the edge of a horseshoe-shaped canyon 300 feet deep, the product of 2 million years of erosion of the region’s limestone plateau by rivers and glaciers.

Because the outcrop at the center of the horseshoe is accessible only from the south, it makes an ideally defensible position, and a castle was built there in the 9th century, of which only ruins remain. In medieval times guards in towers monitored the approach 24 hours a day.

One historic building still survives: The 12th-century St. Faustus church sits right on the edge of the cliff, looking over the river.