In the spirit of dissolving questions (like Yvain did very well for disease), I wanted to give an off-fthe-cuff breakdown of a similar contentions issue: use of the term “design”, as in “cats are designed to be good hunters [of small animals]” or “knives are designed to cut”.
Generally, people find both of those intuitive, which leads to a lot of unnecessary dispute between reductionists and anti-reductionists, with the latter claiming that the former implicitly bring teleology into biology.
So, here’s what I think is going on when people make a “design” evaluation, or otherwise find the term intuitively applicable: there are a few criteria that make something seem “designed”, and if enough of them are satisfied, it “feels” designed. Further, I think there are three criteria, and people call something designed if it meets two of them. Here’s how it works.
“X is designed to Y” if at least two of these are met:
1) Goodness: X is good at Y. 2) Narrowness: The things that make X better at Y make it worse at other things normally similar to Y. 3) Human intent: A human crafted X with the intent that it be used for Y. (Alternately: replace human with “an intelligent being”.)
(You can think of it as a “hub-and-spoke” neural net model where each of the criteria’s being activated make the “design” judgment stronger.)
“A knife is designed to cut” meets all three, and we have no problem calling it designed for that. Likewise for “A computer is designed to do computations quickly”.
Now for some harder cases that create fake disagreements:
“A cat is designed to hunt small animals.” Cats are good at hunting mice, etc, so it meets 1. They weren’t human-designed to catch small animals so they fail 3. Finally, a lot of the things that make them good at catching mice make them unsuited for other purposes. For example, their (less-)social mentality and desire to keep themselves clean makes them harder for mice to detect when hunting, but it prevents them from using hunting tactics that dogs use on larger animals (e.g. “split the pack up and have one of them chase the prey downwind to the rest of the pack”).
Thus, the cat example meets 2, and this “narrow optimality” gives us the “feel” of it being designed, and has historically led humans to equate this with 3.
How about this one: “a stone is designed to hurt people”. Stones are good at hurting people, so they pass 1. However, they are not narrowly good, nor were they specifically crafted with the intent of hurting people. Thus, we generally don’t get the feeling of stones being designed to hurt people.
Try this test out for yourself on things that “feel” designed or non designed.
Sorry, fixed. I was trying to remove the google crap from the URL when I got the search results but overdid it to the point that even the anchor text disappeared.
In the spirit of dissolving questions (like Yvain did very well for disease), I wanted to give an off-fthe-cuff breakdown of a similar contentions issue: use of the term “design”, as in “cats are designed to be good hunters [of small animals]” or “knives are designed to cut”.
Generally, people find both of those intuitive, which leads to a lot of unnecessary dispute between reductionists and anti-reductionists, with the latter claiming that the former implicitly bring teleology into biology.
So, here’s what I think is going on when people make a “design” evaluation, or otherwise find the term intuitively applicable: there are a few criteria that make something seem “designed”, and if enough of them are satisfied, it “feels” designed. Further, I think there are three criteria, and people call something designed if it meets two of them. Here’s how it works.
“X is designed to Y” if at least two of these are met:
1) Goodness: X is good at Y.
2) Narrowness: The things that make X better at Y make it worse at other things normally similar to Y.
3) Human intent: A human crafted X with the intent that it be used for Y. (Alternately: replace human with “an intelligent being”.)
(You can think of it as a “hub-and-spoke” neural net model where each of the criteria’s being activated make the “design” judgment stronger.)
“A knife is designed to cut” meets all three, and we have no problem calling it designed for that. Likewise for “A computer is designed to do computations quickly”.
Now for some harder cases that create fake disagreements:
“A cat is designed to hunt small animals.” Cats are good at hunting mice, etc, so it meets 1. They weren’t human-designed to catch small animals so they fail 3. Finally, a lot of the things that make them good at catching mice make them unsuited for other purposes. For example, their (less-)social mentality and desire to keep themselves clean makes them harder for mice to detect when hunting, but it prevents them from using hunting tactics that dogs use on larger animals (e.g. “split the pack up and have one of them chase the prey downwind to the rest of the pack”).
Thus, the cat example meets 2, and this “narrow optimality” gives us the “feel” of it being designed, and has historically led humans to equate this with 3.
How about this one: “a stone is designed to hurt people”. Stones are good at hurting people, so they pass 1. However, they are not narrowly good, nor were they specifically crafted with the intent of hurting people. Thus, we generally don’t get the feeling of stones being designed to hurt people.
Try this test out for yourself on things that “feel” designed or non designed.
That’s the most literally dissolved question I’ve ever seen.
Sorry, fixed. I was trying to remove the google crap from the URL when I got the search results but overdid it to the point that even the anchor text disappeared.