“The usual analogy is a room full of monkeys with typewriters eventually writing Shakespeare. The problem with that is that the text would not be coherent, you would have to know Shakespeare to “find” it amongst the millions of other incoherent letters and punctuation marks.”
The analogy is popular, but as a comparison to evolution there are at least two types of problems. First, to the question of “how do you know Shakespeare to recognize which is the desired text?”, the answer is that the environment is what determines if a particular genome is a good enough one. Arguments that particular complexity is a problem for evolution like to point out that a computer program that is looking for a specific character sequence has to include the target sequence within the program if the program is self-checking. But the functionality of the genome is tested by the existing laws of nature; it does not have to include the test within it. And all the other genomes are simultaneously tested against the environment; it’s not necessary to read one at a time. Successful genomes go on to the next round; unsuccessful ones don’t. Secondly, for English, the number of combinations of symbols that actually produce words is much less than the number of possible combinations, even if we’re lenient about spelling. But any sequence of DNA bases has a level of biological activity. Any amino acid sequence has some biological activity. The sequence only needs to be good enough to function. Even a tiny amount of a new function is more ability than none at all; the beginning does not have to be at a high level of function. Evolution takes what’s good enough to survive, copies it with varying levels of errors, and sees if the descendant variants survive. If you imagine that the chimps are trying to produce any work in any language, anyone able to read can take a look and inform the chimps what part is on target, and the chimps can copy that part, trying again, then you get closer to the situation for evolution.
The diversity of life shows that there is a very wide range of DNA sequences that are functional as directions for living things. One way that DNA code is like English words is that there are often quite different ways to give the same result, and a small change can give a very different result but often doesn’t make much difference. For example, Dawkins’ quote from Hamlet “Methinks it is like a weasel”, if treated as an attempt at mammal identification, is quite similar to the meaning from Homer and Jethro of “There was a smart guy, from the city, and he picked up a striped kitty…We held our nose as we buried his clothes”. But “A weasel thinks it is like me” is quite different in meaning.
Given that the steps needed between making basic biochemical building blocks like amino acids, sugars, fats, and nucleotides and making organisms are highly uncertain, it’s not very practical to assess probabilities. If we start with a simple bacterium (no nucleus), it probably doesn’t have the greatest error checking, and there are plenty of mutations in the following generations; some survive and some don’t. Particularly given that prokaryotes tend to readily pick up bits of DNA, and may join up to form symbiotic relationships, there is quite complex mixing and matching of DNA sequences to potentially produce new useful options. Indeed, it is not clear that evolving from bacteria to us is actually “uphill”, as the range of DNA sequences is becoming more random with each new kind of organism.
“There is an old adage: “If it ain’t broke, don’t fix it””
That aptly explains why “how come there are still ____?” in evolution. If an organism is already successful, changes are likely to be detrimental, and so evolution favors not changing in that situation. But environments change and the set of associated organisms changes, so eventually a change becomes advantageous – exactly what produces the pattern of punctuated equilibrium. A more gradual pattern of change happens if there is less severe competition or less environmental pressure, so that there’s a high proportion of survivors. Contrary to Lamarck, organisms are not inherently trying to become improved. But if organisms are becoming more variable, some of the variants will likely be more complex and others more simple. For example, the first muricid snails (the source of Tyrian purple) were small and had no spines. Today, many species are large and many have lots of spines. But if a snail has no spines, it can’t go any lower in spine number. Random variation in spine number would lead to an increase in spine number in this case. Likewise, if the first ones were near the small end of the possible size range, random variation will lead to an increase in average size. Biologically, different organisms have different strategies that work for them – there’s no biological definition of overall “improvement”, merely “being better at doing ___”.
“All these things rely on two other things
- diagnosis
- reinvention (or at least "tweaking)”
Which, as noted above, does not require intelligence. Diagnosis requires some sort of criterion. In the case of natural selection, both the standards and the testing against the standards are done by the laws of nature. Gravity is not intelligent; it merely provides certain constraints on what works and negative consequences for trying to disregard those constraints. Reinvention requires moderately reliable but not absolutely perfect copying of the original. DNA copying fits that requirement.
