On the manned space program

Matthew Yglesias writes:

My only beef with [the civilian space program] is that the program has been disproportionately focused on the idea of manned space exploration. Human beings, being fragile creates who evolved on the planet earth, turn out to be hard to send into space. They also, being humans, tend not to be interested in taking extremely long trips even though many interesting things in space are very far away. Under the circumstances, it’s just not very practical to send human beings into space unless there’s something important that only human beings can do. And in recent decades, there just [haven't] been the sort of compelling projects that justify the difficulties of manned space flight. Instead, we’ve been making up missions — most recently the preposterous idea of a manned mission to Mars — in order to justify the human-oriented space program.

All true. To which I would add, not only are manned space missions expensive and impractical, they are also incredibly risky. The shuttle has something like a 1% failure rate. That is, of the two hundred or so shuttle missions we've launched, two have resulted in the loss of the spacecraft and all its occupants. That's two orbiters worth $1.7 billion each and 14 astronauts with an enormous amount of specialized training. The shuttle's predecessor, Apollo, had two catastrophic failures, as well, one of which took the lives of three astronauts and one of which nearly did the same. In what other fields, short of the military and maybe police and firefighting, would we find such mortality rates remotely acceptable?

Send robots. They like the work, and they're cheap.