Sociobiologist and journalist Robert Write has a terrific essay that explores how a "higher purpose" could be compatible with strict, Darwinian evolution.
What I enjoyed most was his grappling with the question of how consciousness could have evolved at all:
How exactly does a consciousness that was previously epiphenomenal, incapable of exerting influence on the organism harboring it, suddenly acquire causal force, provoking discussions about itself?To which he offered this answer by analogy:
Suppose there was an organism that had no eyesight, no perceptual sensitivity to light whatsoever. Then suppose it developed eyesight and every once in a while looked at its shadow and reacted to its shadow. Well, at this point its shadow would have gone from being epiphenomenal—having no effect on the organism’s behavior—to having an effect on the organism’s behavior. Strictly speaking, nothing about the nature of the shadow would have changed—the changes all came in the physical constitution of the organism—and yet the shadow would now have a new capacity for interaction with the organism.
This analogy is a brilliant insight on consciousness and very much as persuasive as earlier arguments about altruism. The old argument used to be that "Evolution is red in tooth and claw and so altruism serves no purpose." If survival is all that matters, screw everyone else. Yet, examples of altruism kept being observed in nature, from army ants to chimpanzees. So, the revised consensus became "Well, individual survival may be important but altruism can also confer advantage to the gene pool." That helps explain the widespread occurrence of homosexual behavior in many species even though strictly non-reproducing members won't pass their genes directly to future generations. Your gay uncle or aunt shares genes with you and by contributing to your survival increases the chance that a version of their DNA is carried on, nonetheless.
Given the fall of the anti-altruism theory, "consciousness serves no purpose" starts to sound like an equally short-sighted argument. So, how could consciousness, a useless "shadow" on the wall of our mind possible effect evolutionary fitness? Or, an even harder puzzle, how could the seed of partial, incomplete consciousness take root at all and be worth passing down to your progeny?
Well, take one feature of consciousness, self-awareness. At a base level, primitive self-awareness kicks in to prevent a lizard from eating its own tail, for example. At a higher level, if a mammal's mental model includes itself, it might better be able to able to recognize its place in the pack status hierarchy and allocate its energy less to sparring with the pack leader and more on hunting and mating. Also, a short hop from self-modeling is wariness and empathy toward other members of your species. Predicting who your allies and rivals are could be a powerful secret weapon against those who lack the ability. Which of these traits, modeling of others or modeling of self, came first might not matter since either would yield great dividends.
So, just as Dawkins step-by-step disproved the "What good is a half-evolved eye?" argument, it should be possible to show that half-evolved consciousness (say, self-awareness alone) could confer significant evolutionary advantages.
As for finding higher purpose, it seems like a pretty direct line from "modeling self" to "modeling others" to "modeling the environment" to "modeling the universe."
As Carl Sagan said so profoundly, "We are a way for the Cosmos to know itself."
Given the fall of the anti-altruism theory, "consciousness serves no purpose" starts to sound like an equally short-sighted argument. So, how could consciousness, a useless "shadow" on the wall of our mind possible effect evolutionary fitness? Or, an even harder puzzle, how could the seed of partial, incomplete consciousness take root at all and be worth passing down to your progeny?
Well, take one feature of consciousness, self-awareness. At a base level, primitive self-awareness kicks in to prevent a lizard from eating its own tail, for example. At a higher level, if a mammal's mental model includes itself, it might better be able to able to recognize its place in the pack status hierarchy and allocate its energy less to sparring with the pack leader and more on hunting and mating. Also, a short hop from self-modeling is wariness and empathy toward other members of your species. Predicting who your allies and rivals are could be a powerful secret weapon against those who lack the ability. Which of these traits, modeling of others or modeling of self, came first might not matter since either would yield great dividends.
So, just as Dawkins step-by-step disproved the "What good is a half-evolved eye?" argument, it should be possible to show that half-evolved consciousness (say, self-awareness alone) could confer significant evolutionary advantages.
As for finding higher purpose, it seems like a pretty direct line from "modeling self" to "modeling others" to "modeling the environment" to "modeling the universe."
As Carl Sagan said so profoundly, "We are a way for the Cosmos to know itself."