Many scientists who study origin of life and evolution bemoan public skepticism about their work. They say, “Look, we are making progress, we will get there eventually.”
The problem, as it currently stands, is that nobody knows how to make a living cell. We actually don’t know all the requirements. We tend to think that we know, but the progress of science in this field has largely been to show us all the new requirements we hadn’t been thinking about before.
But, even assuming that a living thing could be built from scratch, there is a bigger problem. Life reproduces. Life produces life. We often view this as a mundane fact about organisms but it is actually quite profound. A lot goes into the ability to replicate. A living organism must distinguish self from non-self. Otherwise, how would it know what to replicate? It must know what parts to replicate. A part that is replicated needs sufficient means to reproduce everything else, including the reproduction mechanism. That is, if we get our living thing to reproduce something but it isn’t enough to make an entire new living thing, then we haven’t gotten anywhere.
In other words, the whole system must be in place for anything to work at all. It can’t “evolve” to this point, because, before this point, there is nothing that is capable of evolution. This is the reason for much of the public skepticism of origin of life research.
The extreme difficulty of scaling production of new technology is not well understood. It’s 1000% to 10,000% harder than making a few prototypes. The machine that makes the machine is vastly harder than the machine itself.Elon Musk (@elonmusk) September 22, 2020
Indeed, whatever the difficulty of creating life in the lab, making individual prototypes is not nearly as problematic as making “the machine that makes the machine,” which all reproducing living cells can do. That is, the ability of an organism to reproduce is at least an order of magnitude harder that the ability of an organism to just live.
But, the fact is, organisms require both.
Without the ability to reproduce, any metabolism that develops will be ephemeral, as the original cannot persist in the environment indefinitely. Interestingly, evolution faces a similar problem, which we learn from computer science rather than manufacturing.
In the Journal of Advanced Computational Intelligence and Intelligent Informatics, a paper, “The Search for a Search: Measuring the Cost of Higher Level Search,” points out (with rigorously documented proof) that successful searches for searches are exponentially less likely to be productive than a search itself. Similarly, we know from biology that many of the beneficial mutations that we see are predisposed by the genome to occur. That predisposition is sometimes called the implicit genome. In other words, certain types of mutations are implicit in a genome configuration that is predisposed to them. But what causes that predisposition?
The “Search for a Search” paper shows that biologists are mistaken if they assume that the predisposition that makes beneficial mutations more probable does not require any explanation. Just as Musk pointed out that building a machine to make the machine is at least an order of magnitude more difficult than building the machine in the first place, “Search for a Search” shows that finding a process which finds the right predispositions for future changes is always going to be significantly harder than finding the right change directly. Thus, the same type of bootstrapping problems that exist in origin of life research rear their heads in evolutionary research as well.
It may turn out that the public skepticism is wrong about either one or both of origin of life and evolution. However, the skepticism shouldn’t be viewed as a mere naive reactionary stance, but rather the product of a deep intuition that matches much that we know about complex systems and how they behave.
Note: “The Search for a Search: Measuring the Information Cost of Higher Level Search”, William A. Dembski and Robert J. Marks II, Journal of Advanced Computational Intelligence and Intelligent Informatics, Vol.14 No.5, 2010 is open access.
You may also enjoy:
Can computers evolve to program themselves without programmers? How much computing power would we need to evolve the programmer’s intelligence via Darwinian evolution? Computer scientist Roman Yampolskiy explains why Darwinian evolution in computers wouldn’t likely produce an AI superintelligence.