Scott Alexander (host, Astral Codex Ten)

I'm not saying there aren't long, scholarly-sounding papers with 27 authors from the psychiatry departments of top medical schools called things like a rubric for the emotional support animal evaluation that makes sense.

I'm saying that when you take out all the legalese, the executive summary is think really hard about whether this animal really helps this person, then think really hard about whether it will cause trouble, and if it helps the person and won't cause trouble, sign the letter.

Here's a typical case.

you've been seeing a patient with depression for three years.

You prescribe them medication, maybe they get a little better, maybe they go up and down randomly in the way of all depression patients.

Then they say, my roommate is leaving so I need to move to a new apartment, but almost nowhere allows dogs, and the only place that does allow them charges more than I can afford.

Please write me an emotional support animal letter, or else I'll lose my beloved Fido, the light of my life.

So you say, okay, I've got to do an evaluation to see if you're really depressed.

They say, you've been treating me for depression for three years.

You've prescribed me six different antidepressants.

Come on.

You say, okay, fine, I'll skip that part.

But I've got to do an evaluation to see if your animal really helps you.

They say, I feel so much better whenever I'm with Fido, he really brightens up my day.

You ask the same question several times, in the manner of all psychiatrists, and your patient always gives the same answer.

Then you say, I've got to evaluate whether your animal is safe, and he says, oh yeah, Fido is such a good boy, he would never hurt a fly.

Now what?

could keep evaluating harder.

You could make them bring Fido into your office, good luck, and observe him.

The observation would look like your patient petting a dog for a half-hour appointment, for which you charge them $200.