Testing in historical science

Keith B Miller (kbmill@ksu.edu)
Wed, 12 Nov 1997 21:32:26 -0600 (CST)

There have been several recent posts that directly or indirectly argue that
the historical sciences (called "origin science" by some) deal with
questions that are beyond test or verification. By contrast, theories in
the "hard" sciences (eg. physics) are assumed to be subject to experimental
test and are in a different (more sure) category of scientific study. This
dichotomy is a false one, as George Murphy has already indicated. Below is
the abstract of an ASA presentation of mine from two years ago that
addresses this issue.

Frequent claims appear in the Christian science/faith literature that the
historical sciences are fundamentally different than the "hard" sciences,
and that their scientific conclusions are less rigorous and less testable.
Scientific claims about Earth and biological history are then dismissed as
untestable speculation. This represents a serious misunderstanding of both
the nature of experiment and theory testing, and the character of
scientific "proof."
Experiments may be constructed to test predictions of proposed models or
theories, or to gather information on a system that is not well understood.
In sciences such as chemistry or physics the events or processes being
studied are rarely directly observable. The investigator examines selected
records of experimental events to infer unobservable processes. In the
geological sciences, the records of past events are also examined to infer
causal processes. But in this case the preserved record is controlled not
by the investigator but by nature. And rather than repeating an
experiment, the geologist can return to the same record and look at it in
new ways - utilizing new technologies or simply focussing on previously
ignored or unrecognized aspects of the record. Historical sciences are
also just as predictive as the "hard" sciences. Each new observation is a
test of the predictions of existing models and theories.
All theories are accepted based on their predictive and explanatory power.
The validation of a scientific theory is not like a legal proof in which
truth must be established beyond a reasonable doubt. No scientific theory
will be without unresolved problems, inconsistent evidence, or unexplained
phenomena. Both the "hard" and historical sciences are on equal ground
here. Science pursues truth, but never claims to have it fully within its


Keith B. Miller
Department of Geology
Kansas State University
Manhattan, KS 66506