Evolution makes lots of testable predictions.

  The question of whether or not evolutionary theory makes testable
predictions has been raised a number of times -- most recently on this
list in the last week or two.

  It seems to me that "historical science" theories (Big Bang cosmology,
geology, biological evolution) can and do make lots of testable
predictions of this general sort:

  Historical theory AAAAA says that system BBBBB was produced by a set of
natural processes CCCCC. So if you perform experiment DDDDD on object
EEEEE (a portion of system BBBBB), then based on inferences FFFFF, the
results must fall in range GGGGG.
  If the results don't fall in range GGGGG, then -- assuming experiment
DDDDD was performed properly and assuming inferences FFFFF are valid --
then it's safe to say that object EEEEE was not formed exclusively by
natural processes CCCCC. Other processes must have been involved.
  Often, this sort of testing can be used to distinguish between competing
historical theories AAAAA1, AAAAA2, and AAAAA3.

  Big Bang cosmology and Inflation makes predictions about features of the
cosmic microwave background and early galaxies, to name a few.
  Geology makes predictions about the structure and composition of tiny
rocks and large geographical features.
  Evolutionary biology, in particular, makes predictions about genomes of
species.

  Evolutionary theory -- that modern species were produced by natural
processes of mutation and natural selection -- was developed before
anything was known about DNA. It was developed based upon fossils records
and geographical distributions of species and observation of mutation and
natural selection at work today.
  Once an understanding of DNA was developed several decades later, a new
and independent way of testing evolutionary theory became possible.
  Consider species BBBBB. Something of its evolutionary history can be
hypothesized from its form, its geographical location, and fossils.
  Once something is known about the genomes of species related to BBBBB,
evolutionary theory makes a prediction about the genome of species BBBBB.
If evolutionary theory is true, BBBBB's number of chromosomes, the
arrangement of genes on the chromosome, the sequences of those genes, and
the sequences of non-coding introns and pseudogenes on the chromosomes,
etc., all ought to fall within a certain range GGGGG. If you sequence its
genome and find some results which fall well outside that range, then you
should be able to conclude that some other processes (supernaturally
miraculous, alien genetic manipulation, unknown natural mechanisms?) were
partially responsible for species BBBBB. (As always, these conclusions
depend upon certain inferences FFFFF. However, the more that is
understood about how DNA works and how mutations can happen, the more
tightly constrained and the more certain those inferences can be.)

  The number of species whose genomes are being sequenced is steadily
growing, and will probably continue to grow for some decades. It seems to
me that one of the things this will do is provide massive confirming (or
disconfirming) evidence for evolution - since evolutionary theory really
does make predictions about the possible ranges of those results.

Loren Haarsma
Received on Fri Dec 17 17:22:06 2004