Be careful about your assumptions. Most individual animals have six legs plus wings, but no learning capacity. Most animal species do as well.
Here is something similar to what you were describing, although probably not the specific work you had in mind. This seems like it was a relatively simple effort. Nevertheless, after a bunch of random instruction sequences were assessed for how ordered they made the test lists, and “children” were preferentially spawned based on sorting success along with random mutations in the code, they say they achieved perfect sorting after about ten generations.
In this case, however, the resulting algorithm wasn’t “incomprehensible”: if you look at the decision tree, it describes the classic bubble sort. Which itself I think may have some interesting implications. The bubble sort works and achieves the objective simply and reliably, but it’s horrendously inefficient on very large lists. But if the lists involved were all short, as indeed they were, then who cares? I wonder if one can use this as a sort of allegory for the view that evolution isn’t geared to producing optimum results, but rather to achieving “good enough” as simply and with as few iterations as possible. Once there’s no survival advantage to further optimization, there’s no evolutionary driver for it, and further changes to those traits become largely random.
Well, yes, but the organism has to recognize that it has to be reassembled.
I am an IT guy. The last biology class I had was in ninth grade. I have some inkling about this stuff but don’t really understand how most biological processes work. I have this simplistic idea that DNA is an instruction set for how to build an organism, but I don’t know how the as-built organism knows, biologically, that its tail was bitten off or that its arm is scratched or that its bone is broken, and needs to have the instructions reapplied.
I expect this is rather complicated and I have no aspirations to make up for lost time in the bio lab. Note title of this subforum. However, I am open to dumbed-down explanations if anyone wants to bother.
Some of them are lying intentionally, and some may be lying unintentionally, like saying they were atheists because they hated God.
But any big name apologist who says he was an atheist is almost certainly lying intentionally.
I don’t know if this is still a hot topic, but 30 years ago there were tons of papers on various search space heuristic. One was genetic algorithms, one was simulated annealing, and I recently saw a paper which simulated how bees find flowers. They all terminate when they find a good enough answer, and they all find local optima by hill climbing, and most have some way of doing radical changes to try to get to another hill that is higher.
Natural evolution certainly finds local optima - look at the peacock, for example.
The genome, encoded in DNA and contained within the chromosomes, is actually a set of instructions (genes) for building proteins and regulating their functions. When you say that you don’t really understand how these biological processes work, you are standing in good company, because while a lot is known about the details of genes and the proteins they code for, the understanding of how gene expression is controlled “just so” to produce a functional organism rather than a collection of randomly differentiated cells is very limited. Those same processes of development are involved in repair of trauma or other damage although they are less effective in a fully developed organism (e.g. producing scar tissue instead of complex structure epithelium) and degrade with age, which is why young children can quickly heal broken bones and will often heal cuts with minimal residual scarring but it takes adults much longer to heal and often not as completely.
In fact, most adaptations aren’t even “local optima” in any specific sense. A lot of adaptations are just adequate in the mean, and often reflect both the specific pathway to get them there and a balance of competing influences. The recurrent laryngeal nerves are often cited as evidence for evolution because their convoluted path is not a choice any ‘intelligent’ designer would make; they are, of course, a result from adaptation from the earliest tetrapods, and function just fine even in long-necked quadrupeds like the giraffe, so having a phenotype that is optimized in one particular way isn’t necessarily beneficial for general fitness.
Stranger
Yes, this is a real difference from natural evolution. But in a sense, evolution always has a goal–better reproductive success in a given environment.
This is absolutely correct. Evolution solves the problems it faces and not any future problem.
I think I understand what you’re saying, and I think you are speaking metaphorically. But I think many people have the mistaken idea that evolution is literally a goal-oriented process moving towards some target state. Evolution is not a process, it’s a consequence of DNA mutation.
I’ve heard this viewed as a tautology - that which reproduces more has the best reproductive success. Which is as far away from a goal as you can get.
And too often ‘I was an atheist when I was insert age between 12 and 19 here but then I found Christ.’
IANAB, but as I see it, reproductive success is just a subset of the larger set of traits that contribute to success in the survival of a species. Successful reproduction, avoiding predators, disease, and other causes of fatalities, and the ability to source sufficient food, are all survival traits. I don’t really see a tautology here. Survival itself is very much a goal.
I think that when people say “goal”, they mean that intent is somehow involved. When you drop a ball, is its goal to hit the ground? Like gravity, evolution is just something that happens.
I read something similar about connecting together real physical integrated circuits. Evolution eventually led to a variety of connections that made no sense, at least at first inspection – things like unused parts of the circuit winding up connected to other unused parts. After some investigation they realized some of the imperfections in the integrated circuits, such as the assumed but not quite perfectly realized isolation between sub circuits, wound up contributing to the final design.
Evolution is an intriguing, if slow and often inadvertent, way of getting someplace.
So, there’s a number it’d be interesting to comprehend. Supposing that the random genetic changes that feed evolution can only happen at some moment that an organism reproduces an imperfectly copied genetic offspring, how many times has that moment occurred? How many bacteria, in total, have reproduced? We’d have to include that great majority of mutations that lead to an unsuccessful organism, all those “reproductions” that failed to survive. How many bag shakes have there been, since the very beginning???
Except that, unlike a ball falling to the ground, evolutionary processes are complex systems in the specific sense that they are, among other things, adaptive and emergent. The evolutionary “goal” of survival is inherent in the fact that organisms that successfully procreate, multiply, and survive are going to be dominant over those that don’t, whether or not there is any explicit agent of intentionality.
Exactly, and there were billions of years when all there were were bacteria. Unimaginable numbers of bacteria over unimaginable numbers of years in countless environmental conditions splitting, combining, reproducing and mutating all the time. Not really so surprising that something more “useful” arose. And it only had to happen once.
OK, why not try to calculate the lower boundary of that number?
The oceans cover 361 million square kilometer. Assume this has been the average through the ages.
Life started 3.5 billion years ago.
There are about one million bacteria (and over ten times more viral particles, let’s leave them aside) per mililiter seawater in the upper 30 m (cite, page 3, conservative estimate). This number is an equilibrium, I postulate that it has been like that since it was first reached, and it was reached fast. That is the vital capacity of planet Earth (no cite, just a guess).
Bacteria divide every 20 minutes.
Disregard life outside the oceans (that is conservative).
Assume that there are as many bacteria below 30 m as there are above (that is really conservative, oceans are 4 km deep on average).
Multiply: 361 x 106 x 106 (to make it sqm) x 30 (the depth arbitrarily chosen) x 109 (to make ml out of cubic m) x 106 (bacteria per ml) x 2 (to add the bacteria below 30 m) x 3.5 x 109 (years) x 8766 (hours in a year) x 3 (rate of division of bacteria).
Bacteria have divided at least 2 x 1045 times so far.
I guess I am missing some zeroes.
A self-organising electronic circuit has stunned engineers by turning itself into a radio receiver.
Organisms are continually reassembling themselves. Each cell is maintaining homeostasis–constantly regenerating proteins and pushing molecules into and out from itself. Cells are organized into tissues, and tissues into organs, and organs into organisms. At each level, there is maintenance–deviation is counteracted against, but it all comes down to cellular homeostasis. Cells recognize something is off within themselves and the chemical reactions adjust to restore things.
But consider the species where the male dies immediately after impregnating the female. Survival, except insofar as it aids reproductive success, clearly isn’t very important.
Evolutionary success is measured by the biomass of the species, which is determined by reproductive success, and that is determined by features which are selected for which increases reproductive success.
Other features may indirectly support this, and get selected for, some are selected for by accident, temporarily at least, but it all boils down to reproductive success.
It’s a very interesting point but ultimately I don’t think it refutes in any way my assertion that in the general case, selecting for individual survival is a very important evolutionary driver, particularly in higher-level species. The list of traits that contribute to human survival is very long indeed – things like fight-or-flight reactions, our extreme sensitivity to motion in the peripheral vision, etc. etc.
I think the basic answer to your point is that to the extent that we understand the phenomenon of either or both sexes dying after reproduction – a phenomenon called semelparity – it tends to occur in special cases having unique tradeoffs between reproduction and survival. Specifically, you tend to find it in plants and low-level invertebrates where adult survival rates are inherently poor, so evolution favours mobilizing all available resources to maximize reproductive output. Thus your comment about “evolutionary success is measured by the biomass of the species” is quite astute but doesn’t contradict my point in the general case. This can also happen when the resource cost and effort of reproduction is extremely high.
But for the vast majority of the higher animals, as I said, individual survival is a vital evolutionary driver. And it seems that the higher level the creature – ultimately culminating in humans – the more obvious these survival traits are.
I found this to be quite a good article on semelparity:
https://www.nature.com/scitable/knowledge/library/semelparity-and-iteroparity-13260334/