= 27d567bc2d48d9f40e9ccc20ea370b15)
Torello comments on Rationality Quotes Thread February 2015 - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (127)
Maybe I should have included the whole paragraph:
"And even when "truth" can be clearly defined, it is a concept to which natural selection is indifferent. To be sure, if an accurate portrayal of reality, to oneself or to others can help spread one's genes, then accuracy of perception or communication may evolve. And often this will be the case (when, say, you remember where food is stored, and share the data with offspring or siblings). But when accurate reporting and genetic interest do thus intersect, that's just a happy coincidence. Truth and honesty are never favored by natural selection in and of themselves. Natural selection neither "prefers" honesty nor "prefers" dishonesty. It just doesn't care."
He's talking about the "maps" that humans/animals may carry in their brains. These maps don't need to match the territory to be adaptive (I think your criticism of the quote hinges on how you would define "significantly"). But there's quite a bit of space where a "bad map" does not prevent adaptive behavior.
For example, some non-venomous snakes "copied" the color patterns of venomous snakes. It's still adaptive for animals to avoid all snakes with this coloring (just to be safe) without needing to know the truth about which snake is dangerous and which isn't. And natural selection is "rewarding" the non-venomous snake for lying about how dangerous it is.
Partially that, but also partially about the direction of the gradient. First, maps never match the territory perfectly precisely, they are always simplified models. In that sense, of course, a map not "matching" the territory is not a obstacle to surviving and prospering.
However I would claim that the greater the mismatch between the map and the territory, the greater disadvantage in the natural selection game does the creature accrue. If, magically, you get a choice between getting a more accurate map or a less accurate map, you should always choose the more accurate map.
That is not true -- you set up the question wrong. There are three maps involved: map 1 does not recognize venomous snakes at all; map 2 confuses venomous and mimicry-using snakes; and map 3 successfully distinguishes between venomous snakes and mimicry-using ones.
Map 3 matches the territory better than map 2 which matches the territory better than map 1. The natural selection would give advantage to an animal with map 3 over the one with map 2, and the one with map 2 over the one with map 1.
I think that point he's trying to make is that natural selection doesn't magically get a choice between maps. In general, a more accurate map will only become available to the mind of some creature if it happens to be adaptive for genes in a particular population in a particular environment.
Think of all the creatures with really bad maps. In terms of reproduction, they are doing just fine. For some species, their relative reproductive success can be improved with more accurate maps, but that's a means to the end of reproductive success,
There are many ways to "make a living" in evolutionary terms, and having a mind with accurate maps is only one of them.
Think of all the creatures who don't have "maps" as humans do. They are still being acted upon by natural selection.
This seems to be conflating possessing truth and sharing truth. The former is almost always valuable. The latter is an interesting bit of game theory, that can go either way.
As it has been said, truth may be spoken as events dictate, but should be heard on every occasion.
The idea that it is just "a happy coincidence" makes me think Lumifer's criticism still applies.
This additional context does help; thanks.
Yes, this could be adaptive, but not costless. An animal that avoids all snakes that look venomous misses out on some opportunities (e.g., foraging for food in a tree occupied by a harmless but dangerous-seeming snake). The opportunity cost, in reproductive terms, might be negligible, or it might matter, depending on the specifics. (Here I'm agreeing with you when you point to the importance of the term "significantly.")
Because the truth, even in small matters like snake coloration, can make a difference, the original quotation is an overstatement.
All natural selection "cares about" is genes copied. Claws, peacock tail feathers, and "maps" can all "make a difference," but natural selection only acts on the results (genes copied); natural selection itself doesn't favor any particular kind of adaptation, that's why I think the original quote is not an overstatement.