24 November 2002
Testing Darwinism versus Cosmic Ancestry What'sNEW
Most of the past theories of science are already suspected of being false: there is presumably every reason to anticipate that current theories of science will suffer a similar fate — Larry Laudan (1)
Where We Agree With Darwinism
We do not doubt that natural selection, operating on the phenotypic results of mutation, recombination, and gene acquisition, plays a very important role in evolution. We accept the evidence that natural selection can use genetic changes to produce evolutionary results over time as follows —
- Small mutations can allow proteins to become optimised for different environments and different species.
- Point mutations and other simple mechanisms can switch existing programs off and on.
- Simple mutations could cause incompatabilities that lead to reproductive isolation, or speciation.
- Symbiogenesis has been important in evolution, as Lynn Margulis says.
- Horizontal gene transfer can deliver from various sources the components of large genetic programs such as the ones for photosynthesis.
Where We Disagree With Darwinism
Darwinists extrapolate from evidence for the above-listed processes to the claim that these processes can invent wholly new genetic programs and can produce sustained macroevolutionary progress over time. This leap is unsupported by evidence or logic.
Recently, Darwinists have begun to accept the evidence that horizontal gene transfer can deliver the components of large genetic programs. However, this makes the original source for them even more difficult to explain. How would independently evolving genes, exons, or other lengthy DNA strands gradually acquire the correct sequences needed to serve as components of a genetic program whose function will only be performed when all the needed strands are installed — in a species different from the one(s) in which the strands were written? The correct mutations would not be neutral, so if the strands were active in the donor species, natural selection would surely hinder the process. If the strands were silent while gradually acquiring the correct sequences, the monkeys that can write Shakespeare must have done it.
Cosmic ancestry holds that wholly new genetic programs cannot be invented by the above-listed processes over any period of time. Rather, the strands of new genetic programs must always be imported into a given species by gene transfer mechanisms. (That they were ever invented is an untested assumption.) After transfer, the strands may become properly grouped or assembled by DNA repair and recombination mechanisms using logic analogous to grammar- and syntax-checking in word processing programs. If the imported programs provide an advantage, natural selection will promote the acquiring species. But without a source for new programs, natural selection will soon reach the maximum evolutionary potential of the available programs, and macroevolutionary progress will halt. The full implications of cosmic ancestry are discussed elsewhere on this website. Here we want to consider scientific tests that could distinguish between the two theories.
Test One: Biology
If new programs can be invented by strictly darwinian processes, it should be possible to demonstrate sustained macroevolutionary progress in a genetically closed system. Such a demonstration would confirm darwinism's strongest claims and remove the need for cosmic ancestry's open system.
Unfortunately for darwinism, no such demonstration has yet succeeded. No wholly new genetic programs have been seen in even the best series of biological experiments in genetically closed systems. These have now run for more than 25,000 generations of E. coli, in experiments managed by Richard Lenski of Michigan State and others (2).
This failure is usually ignored, or else explained away — there wasn't enough time, the system wasn't large enough, etc. But the simple fact is that the surest demonstration for vindicating darwinism over cosmic ancestry has not succeeded. This failure supports our contention that sustainable macroevolutionary progress requires a genetically open system.
Test Two: Computer Models
If time were the problem, then a computer model of evolution should be the solution to vindicate darwinism. Computer models can run a generation in seconds or minutes, instead of the hours or weeks needed for the fastest biology experiments. Yet computer models to date have also failed to produce unmistakable macroevolutionary progress, after sufficiently many trials (3). So the second-best test for vindicating darwism over cosmic ancestry also fails. Most biologists are unaware of this failure.
Test Three: Reconstructing the Past Genomically
Third-best are historical reconstructions of actual biological evolution based on genomic studies of existing species. To get any result one must assume that today's version and the ancient version of a given species are genetically very similar, and that gene transfer and gene loss will not obscure the remote past. Then, if new programs are produced by the darwinian method, their precursors should be present in the genomes of today's species that are ancestral to the one posessing the full program under study. These precursors should exhibit a pattern of gradual construction, one or a few nucleotides at a time, as one ascends through the phylogeny of the program. However, if new programs are imported by cosmic ancestry, the gradual intermediate steps will not be found. Instead, one would observe nearly identical programs, or nothing similar, in the ancestral species (4).
Historical reconstructions like that by Raymond et al. (5) look hard for evidence of gradual construction. Claims of such finds are very few. Instead, geneticists are often surprised to find evidence that new programs were delivered to a species by gene transfer (6). Evidence for transfer, with program reassembly, favors cosmic ancestry over darwinism for the reasons stated above.
No Good: Reconstructing the Past Morphologically
In the fossil record, some evolutionary traits, like increasing size in horses, appear to arise gradually. Such evidence is used to support the darwinian claim that random genetic processes can invent wholly new genetic programs. But this case is highly inferential. Some morphological changes, like larger size, may not require new genetic programs, but only slight changes in existing ones. And changes that do require new genetic programs might still appear fairly gradual in the fossil record, even if the programs have been acquired by transfer. Conversely, when the morphological record refutes darwinism, as examples of punctuated equilibrium do, darwinists discount the evidence in various ways.
What'sNEW
 3 Apr 2008: What is the origin of eukaryotic RNA polymerases?
 Elliott H. Margulies and Ewan Birney, "Approaches to comparative sequence analysis: towards a functional view of vertebrate genomes" [open access login], doi:10.1038/nrg2185, p 303-313 v 9, Nature Reviews Genetics, Apr 2008. "...It is inherently impossible to know the correct alignment for any region...."
 Dmitri Novikov's reply prompts a brief restatement of the need for closed-system tests, 24 Dec 2007.
...Fresh fossil evidence is the subject of a Reply from Stan Franklin, 13 Dec 2007
Keeping an eye on evolution is the subject of a Reply from Stan Franklin, 4 Dec 2007
26 Sep 2007: The genomes of 17 species of fungi have been analysed to reconstruct gene duplication....
18 Jun 2007: In The Edge of Evolution, Michael Behe makes even stronger claims....
1 Dec 2006: Evolutionary Dynamics, by Martin Nowak, Belknap Press, 2006.
12 Nov 2006: The Making of the Fittest, by geneticist Sean B. Carroll, W. W. Norton, 2006.
20 June 2006: Bats and horses are closely related, according to a genomic study using retroposon (L1) analysis.
15 Apr 2006: The Evolution Prize will be launched hopefully at ALife X.
31 Aug 2005: Americans think public schools should teach creationism alongside evolution.
3 Feb 2005: Complex early genes.
Guoping Zhu et al., "The Selective Cause of an Ancient Adaptation" [abstract], doi:10.1126/science.1106974, p 1279-1282 v 307, Science, 25 Feb (online 13 Jan) 2005. "Rigorous tests of hypotheses regarding adaptation are needed to identify the mutations responsible for changes in function and to demonstrate that they are selectively advantageous under the conditions specified. This has yet to be achieved for any reconstructed gene."
9 Sep 2004: The ring of life!
2003, 27 October: Stan Franklin wonders if the morphological steps in the evolution of eyes can be reconstructed well enough to count as convincing evidence that new programs can be invented by strictly darwinian processes. Subsequently we have added the paragraph, "No Good: Reconstructing the Past, Morphologically," to this page.
2003, February 18: Humanoid gene arose abruptly?
2003, February 4: The latest results from a closed-system biological experiment.
2002, December 31: No evolutionary progress in a closed system!
References
1. Larry Laudan, Progress and Its Problems: Towards a Theory of Scientific Growth, University of California Press, 1977. p 126.
2. Lenski's experiments are described with references in WhatsNEW, 15 July 1999.
3. Four CA webpages, beginning with Computer Models of Evolution, discuss the failure of computer models to demonstrate sustainable macroevolutionary progress.
4. An experiment specifically designed to seek evidence of gradual construction versus importation is described on the CA webpage, New genetic programs in Darwinism and strong panspermia, 7 Apr 2002.
5. A recent analysis by Jason Raymond et al. is noted in WhatsNEW, with links to original sources, 24 Nov 2002.
6. Additional instances of horizontal gene transfer are noted on the CA webpage Viruses....
|