A 2011 confernce held at Cornell University presented a series of academic papers that focused on viewing biological systems in terms of informational questions. The authors drew parallels between numerical searches and biological systems. By applying conclusions about searches to biological information, conclusions can be drawn about how the capabilities of biological systems to produce practical results. The proceedings were published in Biological Information: New Perspectives, Marks, et. al., eds., (London: World Scientific, 2013).
One paper in this volume is Pragmatic Information by John W. Oller, Jr. In this paper, Oller attempts to show the difficulty of filtering a useful result from a quantity of random noise. Oller uses information found in the English language to illustrate the task needed to draw useful results by searching through non-useful noise.
Such a search is relevant to darwinian evolution, which purports to work by random mutations being filtered by natural selection to provide useful results. The many non-useful mutations are eleminated and the few useful mutations are kept.
Oller shows the difficulty of a search method finding a useful result in a large amount of random data. The 26 letters of the English alphabet give us about 35 available sounds. The word “strengths” gives eight phonetic segments. A random change in these eight segments by the 35 available sounds gives a possible combination of 35[8th]*, or about 2.25 trillion combinations (Biological Information, p.76). Only a miniscule fraction of such a number would result in any meaningful sylables allowed in English. By moving our attention to sentences, the numbers get much worse:
The number of meaningful 12 word sentences would be about on the order of 10[12th] enabling us to estimate that the ratio of meaningful 12 word sentences in English to all the strings that could be formed from all the words in the Oxford English Dictionary: It comes out to be about 4.59×10[-58th] Finding the few meaningful strings by chance in a heap of such nonsense would be a little like trying to find some very tiny needles in a really huge haystack . . . Consider next that if we move the combinatorial explosions up several notches to the length of a short novel, say 30,000 words . . ., the number of possible strings explodes to 600,000[30,000th]. . . At the level of a short novel, the ratio of meaningful strings to possible ones has diminished to a complete vanishing point for all practical purposes.(ibid.)
Practical research by biochemist Michael Behe supports Oller’s conclusion. Behe lays out his research in his book The Edge of Evolution: The Search for the Limits of Darwinism. In studying malaria and the drugs used to combat it, Behe has a large population in the millions of people infected with a disease over many, many years, and a large number of the population that has been injected with drugs to fight the disease. In the case of malaria, darwinian evolution seems to be limited to filtering through a very few mutations, and those only in terms of degrading existing information, such as sickle cell disease, not creating new biological information that was not there in the first place.
If Oller’s and Behe’s work is valid, we have no good evidence to show that natural selection working on random mutations has the power to filter through the vast quantity of biological data required for darwinian evolution to work on a scale as large as entire species. The level of information in a short novel is quite small compared to that required to change from one specie to another. It does not seem reasonable for the biologist to continue to tell us that since such huge improbabilities are not impossible, they must be true.
*wordpress has difficulty presenting superscript numbers clearly, so I have used brackets to indicate scientific numbers.