Oh, I found a problem in your clean solution: nobody knows how to detect, in non-humans, the presence or absence of a mind sufficiently capable to pass your test. Or have you been interviewing all the animals you eat?

It's pretty easy to do some basic tests for intelligence in social animals. The test concept is quite sophisticated, so you can definitely rule out species. Humans raise a lot of dumb animals. Just don't ask me to eat dolphin.

The question you propose to ask is "can this species understand its own fairness to other species?"

That is absolutely not testable by our current means, not for any rigorous definition of knowledge. There are numerous instances, for example, of mothers of one species nursing the young of another species. Sometimes even of predator mothers nursing prey young. So, prima facie, assuming a reasonable definition of "fairness" it would seem that a vast number of species are capable of exhibiting this trait, including many of the ones you eat.

Second, it's also completely arbitrary. Why test for fairness? Why not ability to generate a profit on the stock market, or to do square roots, or tie shoes? It's a stupidly chosen metric that, at its heart, embodies circular logic -- choose a human trait, and then apply a human-centric test (which we don't even do by the way) to see if the species can be eaten. What do you expect the result would be? Even dolphins can fail this exam.

The bottom line is, your criteria is arbitrary and untestable. And it cannot even be applied consistently, as you would have to admit the consumption of severely encephalitic fetuses and the like -- which is something only a person arguing an abstraction could endorse. No socially well-adjusted person can seriously admit that they want such a thing to be permissible in their civilization.

So it's ok to eat sufficiently mentally-retarded humans?

I already answered that one way http://news.ycombinator.com/item?id=5329829

Another way of going about it would be to allow/disallow in aggregate, not in specific cases. AKA all humans or no humans. In which case it wouldn't be allowed even with a braindead person.

There exist people who aren't braindead, but who don't have sufficient intelligence to have a concept of "being fair". I don't know where you got the idea that they don't exist. (You may counter that even the least intelligent person will get angry if you attack them, and that's largely true; but the same is true of any animal, so we must be talking about a more complex understanding of fairness than that.)

And you could say that being part of a species gives you special moral standing if other members of your species have the attributes required for it, even if you don't personally have them, but it would be a completely arbitrary thing to say.

They exist but they are far rarer than mere retardation. Rare enough that I'm just going to punt. The answer for them doesn't matter in the big picture.

Maybe not, but whether it's ok to eat animals does, and so does consistency. If the criteria you previously gave is really the criteria you believe makes it ok to kill/eat something, why aren't you willing to say it's ok to eat humans who meet the criteria?

I'm resistant because it's so hard to do a prudent level of testing on a single individual. If you have some method of determining that there is no shred of full-functioning in their brain, that they truly have the mind of a lower animal, that you can show this to others, then okay it's viable to eat them.

But it still gives a misleading impression when someone hears you ate "a human".