Science in Christian Perspective
Science and Finite Imagination
Norman E. Shank
Department of Chemistry
Grantham, Pennsylvania 17027
From: JASA 34 December 1982): 98-99.
Although the physical sciences are principally concerned with a study of the physical universe, their impact goes far beyond the fields of their direct investigation. This is because the physical sciences, especially astronomy and geology, have given us information about the extent of the universe in both space and time. This in turn has given man a perspective of his place in the universe, and has been important in shaping his world-view. Thus modern man's self-perception is influenced by the findings of the physical sciences.
However these findings of the physical sciences often come from very complicated and highly technical experiments. Few people are able to read and evaluate all investigations of the physical sciences in the original research literature. Most of us are forced to follow the progress of science by reading condensed articles in periodicals and textbooks. Such condensed accounts are not always careful to distinguish clearly between experimental facts on the one hand, and theories and speculations on the other hand. Furthermore, such accounts are often written from a rather noncritical point of view. One is faced with the task of evaluating such secondary accounts for oneself, and of judging their impact on one's own world-view. This communication seeks to give a perspective to this task.
We begin by reviewing the main points of procedure and analysis in a scientific investigation.' For this purpose it is helpful to consider an example which is as nontechnical as possible. Yoder, Suydam, and Snavely' give a useful example that is used here with modification. Suppose we were to find a heavy, sealed steel box with two small holes. There is a rope coming out of each hole. Assume we have no means of breaking or opening the box. We wish to do a scientific investigation of the box and ropes. We label the two ropes A and B respectively. We begin by pulling on rope A. We observe that it can be pulled out of the box. We also observe that as rope A is pulled out, rope B comes out as well. We begin to measure the lengths of rope involved and find that the length of rope B that comes out is always three times the length of rope A. Conversely, we find that if we pull rope B, the length of A that comes out is always one-third the length of rope B. We conclude that no matter which rope is pulled, three lengths of B come out for every length of A. This conclusion is based entirely on observable, measurable facts.
By this time we are curious about the interior of the box. Since
the box is sealed, we begin to imagine what might be inside. We
make a list of possible models for the interior of the box:
1. Two spools of rope connected by gears with a three-to-one ratio.
2. Two spools on the same axle, but with one spool having three times the diameter of the other.3. One spool wound with two ropes.
4. Two spools on two battery-powered motors. The motors are connected to a small computer. The computer senses motion of one rope and then runs the other motor at the appropriate speed to maintain a three-to-one ration.
The list of models could go on until our imagination runs dry. Clearly our imagination, not logic, generates the initial list of models.
We next use analysis, together with our experimental data, to try to eliminate all models that are inconsistent with the facts. In this case we can eliminate Model 3 because it would feed out the ropes at the same rate. On the other hand, Models 1, 2, and 4 are all consistent with all known facts. We may prefer Model I or Model 2 because of its simplicity, but Model 4 is possible and has not been disqualified. Thus we have shown that Model 3 is inconsistent, but we have not established any model as the correct one.
In scientific investigations of more complicated systems, one may be able to do more experiments and to eliminate all but one of the models. In such situations there is a strong temptation to regard the one remaining model as having been demonstrated to be correct. However, this is a fallacy. The original list of models is constructed from the investigator's imagination. Since his imagination is finite, there is no guarantee that he has thought of all possible models. Therefore the elimination of all but one model does not establish the correctness of the remaining one.
It should be noted here the "proof by elimination" is often used in mathematics. For example, if a real number can be shown to be not negative and not zero, then it must be positive. In this case one knows what all of the possibilities are, namely negative, zero, and positive. Therefore a "proof by elimination" is valid. By contrast, when investigating the physical universe, there is no way of knowing all of the possible models. The range of models we work with is limited by our finite imagination. Thus there can be no "proof by elimination."
Returning to the example of the box, we may summarize as follows. The facts are those things which are observable outside the box. If the box is not opened, we can never establish what is inside the box. The best we can hope to do is imagine one or more models that are consistent with the facts. We now apply these principle to several areas of physical science.
1. Atoms. The observable facts about atoms include such things as ionization potentials, emission and absorption spectra, chemical reactions, etc. One particularly important experimental observation was that of Rutherford' when he bombarded gold atoms with alpha particles and measured the angle of scattering. However, no one has made any observations of the inside of an atom. Thus the inside of the atom corresponds to the "inside of the box". All that we can "know" about the inside of the atom is to imagine possible models for what might be in there. For a model to be acceptable it must be consistent with the scattering angles measured by Rutherford, the emission spectrum of the atom, etc. One acceptable model is that the atom has a heavy, positive nucleus which is surrounded by electrons in orbitals. However we cannot be sure that this is the only acceptable model, or that this is the correct model. (It is true that this model is a useful idea in that it helps us "make sense" out of a large amount of experimental data.)
As a specific example, consider the reaction of a supercritical mass of uranium-235 in an atomic bomb. This is a so-called "nuclear" reaction. The observed facts are that a large amount of energy is released, the uranium disappears, and other elements (e.g. barium and krypton) are produced. The model is that uranium atoms have nuclei which are split. The experimental data demonstrates that uranium-235 has a large internal energy and that it can be transformed into other elements. The data do not establish that uranium atoms have nuclei.
II Earth. Observations of the earth include such things as seismographic data, the location of mountains, volcanoes, and earthquakes, etc. No one has made observations of the earth's interior below the bottom of deep mines and wells. Thus the interior of the earth corresponds to the "inside of the box". We seek a model for the earth's interior that is consistent with our observations. One model is that the earth has a crust, mantle, and core; and that the crust is made up of slowly moving plates. However, because of our finite imaginations, we cannot be sure that this is the only model consistent with our observations.
III. Geology. Geologists observe and measure such things as rock strata, argon-potassium ratios, fossils, etc. No one has made any observations of earth's prehistoric past. Thus a chronological account of earth's past corresponds to the "inside of the box". Imagining an acceptable model (historical account) for earth's past has not been easy, and controversy on this continues. Such a model may include statements about lengths of time, large-scale ecological changes, etc. However, regardless of what model is proposed, it is still subject to the same principles as a modelsnamely that it can never be unambiguously shown to be the correct model. There is no way of knowing that we have thought of, and eliminated, all other possibilities.
IV. The Universe. Astronomers observe and measure the angular position, brightness, color, red shift, etc. of starlight as it strikes the earth. No one has observed or measured actual distances between stars or between galaxies. Thus the size of the observable universe corresponds to the "inside of the box". One model for the size of the observable universe suggests it is at least many millions of light-years across (and probably much bigger). Furthermore, if the speed of light is assumed constant, this immediately implies that the observable universe is millions of years old. This model is certainly useful in organizing a vast amount of astronomical data. The correctness of the model, however, cannot be determined.
The above four examples are sufficient to illustrate the distinction between observable facts on the one hand, and the models we construct on the other hand. Keeping this distinction clearly in mind can be useful in reading and evaluating the secondary scientific literature. In particular, it is the models, not the observed facts, which tell the extent of the universe in space and time. Thus it is the models that influence man's self-perception of his place in the universe.
Such models may be highly useful in organizing data, planning
experiments, etc. They may also be useful, at times, in developing
one's world-view. However, as we have seen, the correctness of a
scientific model can never be confirmed beyond doubt. Therefore
it can never be used to establish the truth of any particular worldview.
1Karl Popper, The Logic of Scientific Discovery, New York: Harper & Row, 1968, pp. 27-48.
2Claude H. Yoder, Fred H. Suydam, and Fred A. Snavely, Chemistry (2nd Ed.), New York, Harcourt Brace Jovanovich, 1980, pp. 3-6.3Lord Rutherford, Phil. Mag., 21 (1911), p. 669.