Science in Christian Perspective
From: JASA 17
(March 1965): 8-15.
The history of science reveals at least three major at titudes as to the nature of scientific theories. Popper
calls these essentialism, instrumentalism, and genuine conjecture. This paper begins by outlining the disad vantage of the first two and sides with the third. Illustrations of each attitude in the history of astron omy are also sketched. The latter portion of the paper explores contemporary cosmological theorizing, in the light of the above survey, so as both to show the presence of these attitudes today and to point up why cosmologists give their theories the bent which they do. Certain lessons of an apologetic sort are men tioned in conclusion.
It is not the purpose of this paper either to explore the competing methods of the philosophy of science in
a systematic manner or to examine in detail their theoretical meta-philosophical orientation. Rather, we
shall set our sights toward several more readily achiev able goals within the broad landscape above. Precisely,
an attempt will be made to delineate certain positions widely held as to the nature of scientific theories, to
illustrate these from past and present speculations in physics and cosmology, and to search out some of the
controls operating in such choices together with a few consequences arising from these influences.
In his Logik der Forschung published in Vienna in 1934 (with the imprint '1935') and which has, since
1959, been available in English as The Logic of Scien tific Discovery, Karl Popper first enunciated several
useful distinctions among the attitudes of scientists as to the character of their theoretical work.1 The divis ions which he suggests center upon the ideas that in contrast to the highly critical attitude requisite in the
pure scientist, the attitude of instrumentalism (like that of theories aim at ultimate explanation by essences, that applied science) is one of complacency at the success of theories serve merely as instruments, and that theories applications. Thus it may well be responsible for the recent are conjectures aiming at severe testing. Since this stagnation in theoretical physics schema, is to my way of thinking, valuable we shall utilize it in our exploration after pointing out what Popper says about each of the divisions he names.
Essentialism he finds to consist of three doctrines. It claims that the scientist's goal is to find a true descrip tion of nature and of its laws which also explains the facts of observation, It proclaims that scientists can succeed in establishing the truth of these theories be yond reasonable doubt, and it states that the best theories describe the reality lying behind appearances. Popper accepts the first belief but concludes that the second fails to see that tests (or even a new theoretical discussion) can lead to discarding a theory. Theories all remain conjectural. The third idea, that science aims at ultimate explanation, he also rejects, saying "(Whether] essences exist or not, the belief in them does not help us in any way, and indeed is likely to hamper us; so that there is no reason why the scien tist should assume their existence."2 The belief in essences, whether true or false, may create obstacles
to our posing new and fruitful questions; it isn't part of science since we could never be certain of even a
theory which might fortuitously succeed in describing the essence of something, and it isn't "one of those
extra-scientific beliefs (such as faith in the power of critical discussion) which a scientist need accept."3
Instrumentalism holds that theories are merely instru ments helping us to explain why one symbolic repre sentation of reality leads to another symbolic representation.+ Theories are sets of instruction for deriv ing one logically singular statement from another.5 Popper claims, however, that theories are not compu tation rules. In summary, his view is that
logical relations which may hold between theories and com
putation rules are not symmetrical; and they are different
from those which may hold between various theories, and
also from those which may hold between various computation
rules. The way In which computation rules are tried out
is different from the way In which theories are tested; and
the skill which the application of computation rules demands
is quite different from that needed for their (theoretical)
discussion, and for the (theoretical) determination of the
limits of their applicability.6
Thus instrumentalism leaves us unable to account for the actual progress of science which involves attempt
ed refutation and not just the establishing of theories as having differing ranges of applicability.
[By] neglecting falsification, and stressing application, in strumentalism proves to be as obscurantist a philosophy as essentialism .... It is only in considering how [in science] various theories stand up to tests that it can distinguish be tween better and worse theories and so find a criterion of progress. 7
If theories are merely predictive instruments they can not be falsified; indeed they may, as ad hoe hypotheses,
rescue a scientific theory threatened by falsification. Popper, as a consequence, rejects it as an alternative
Finally, Popper describes his own attitude regarding scientific theories.
*Thomas H. Leith is Associate Professor of Natural Science, York University, Toronto, Ontario.
They are genuine conjectures-highly informative guesses about the world which, although not verifiable ... can be submitted to severe critical tests. They are serious attempts to discover the truth . . . even though we do not know, and -ay, perhaps, never know, whether [they are] true or not. 9
As a result, this view is unlike essentialism in that it does not claim to describe, even partially, some real world behind appearances but rather postulates a world for each theory, described in their turn by further theories more universal and testable. "The doctrine of an essential or ultimate reality collapses together with that of ultimate explanation."10 New theories, like older ones, "are genuine attempts to describe these further worlds ... as equally real aspects, or sides, or layers, of the real world."11 A piano is no more real than its molecules or atoms. The theories delineating all three are equal in "their claims to describe reality, although some of them are more conjectural than others."12
The view is also unlike instrumentalism in that while Popper calls a state of affairs "real" if, and only if, the statements describing it are true, this doesn't diminish the claim of such a conjectured state of affairs to describe something real. Conjectures may be true and thus describe a real state of affairs or they may be falsified and thus contradict what is really the case.13 Testable theories assert that something cannot happen and thus they assert something about the real world. Hence the more conjectural a theory is, the higher should be its degree of testability. There is undoubtedly much which we don't know at all that is real! Therefore, "[with the belief] that science can make real discoveries I take my stand . . . against instrumentalism."14 In sum, true science can predict events of a type already experienced (e.g. eclipses) and also new kinds of events (e.g. the birth of a galaxy). Instrumentalism cannot handle the second of these. Discoveries are guided by theories which do not offer only predictions from the known but which aim at creating novel situations for new tests.
It might be helpful to look at a historical example of each of the above three attitudes. Anyone familiar with the monographs in the history of science will be able to add many others for himself. Since Popper discusses Galileo, whose view of scientific theory is largely essentialistic, we may take Kepler as our example of that position for the sake of variety.
Some of Kepler's early thought, in the Mysterium Cosmographicum, which he published when only twentyfive years of age, is familiar to most scientists. It will
be recalled that in that work Kepler provided his first
suggestion as to the mathematical reason for the number and distances from the sun of the planets; a solution he found in solid geometry. The shell of Mercury's orbit fits inside an octahedron whose points
touch the shell of Venus' orbit. This in turn fits an
icosahedron touching the shell of the orbit of the earth.
Then between our shell and that of Mars is a dodecahedron; between Mars and Jupiter he fits a tetrahedron;
and between Jupiter and Saturn a cube.
In the subsequent examination of Tycho Brahe's records of planetary motion, Kepler both revised the above schema, on the assumption that God might use other principles in laying out the paths than the above geometry suggests, and developed his three famous laws. Behind this work, as in his early ideas, lies the belief that the preferred theory is that one which best ties together previously distinct facts, and experience has shown this to be of some mathematical nature. He concluded from this that the world of the planets must therefore itself be a mathematical order; he argues that we do not accept a theory of planetary motion because of its utility or even its mathematical e egance but rather because the planetary world is a beautiful order in its essence.15 He is quite explicit in arguing that theories are not merely instruments of calculation, and hence mere formal aids in science, but that to be true, they must fit the facts of experience which mirror the structure beneath them. Thus, while we know a priori that the universe is mathematical and that true knowledge of it must therefore be mathematical too, Kepler argues that only experience shows us the exact nature of this harmony.
The major problem here is that Kepler did not prove his case as to the mathematical essence of reality. To be sure we do find him offering mystical analogies for the purpose in his Harmonice Mundi of 1619, such as the attribution of different vocal ranges to the various planets16 or the treatment of the sun as God the Father, the fixed stars as God the Son, and the intervening ethereal medium, through which the power of the sun is communicated to impel the planets around their orbits, as the Holy Spirit.17 But the fact that a mystical metaphysic agrees with an elegant mathematical arrangement empirically found for the planets, is not sufficient to provide credibility to what is only an analogy. Also, because a religious view predicts a rational and mathematical planetary system, even if such a prediction were exact (which it was not, for Kepler), this is not sufficient to obviate other theoretical foundations for the same data. Kepler commits the fallacy of arguing that, by affirming certain empirical consequences, one can thereby affirm a precise mystical or religious antecedent. His essentialism fails, as it must always fail on such argument, to prove its case.
Instrumentalism is well illustrated in Bishop Berkeley. A most severe and intense critic of the fundamental concepts of Newtonian mechanics, Berkeley argued that Newton's absolute space, time, and motion were without physical meaning. Rather, such concepts as space and time are neither perceived by the senses nor proven by reason, and motion as people know it is always relative. Similarly, to talk of "force" taken as the cause of motion is to introduce into physics an occult quality. While a physics based upon forces does have a utilitarian value, since it leads to correct results, we must never confuse it with metaphysical explanation. As a consequence, Berkeley firmly espouses a formalistic or instrumentalistic attitude toward science. Its theories are instruments for calculation and prediction of the regularities of nature.
Popper calls Berkeley's principle of analysis "Berkeley's razor" since it allows us a priori to eliminate from physical science all essentialist explanation. If they have a mathematical and predictive content, they may be admitted qua mathematical hypotheses (while their essentialist Interpretation Is eliminated). If not, they may be ruled out altogether. This razor Is sharper than Ockhamls: all entities are ruled out except those which are perceived.19
Jammer summarizes this in one sentence. "All that natural science can supply is an account of the relations among symbols and signs; but the sign should not be confused with the vera causa, the real cause of the phenomenon."20
This sort of thinking shows up repeatedly in later phflosophy of science. In Ernst Mach, Berkeley's rejection of absolute space, time, and motion and of hypothetical forces is carried over to include the idea of cause and effect. Science was to discard such concepts entirely. He also rejects all metaphysical and theological tradition where Berkeley, ever the good churchman, used these ultimately to explain phenomena. "While Berkeley says that there can be nothing physical behind the physical phenomena, Mach suggests that there is nothing at all behind them."21 Or, as Einstein in his obituary to Mach put it, "Concepts have meaning only if we can point to objects to which they refer and to the rules by which they are assigned to these objects."22
Instrumentalism re-appears clearly in William Clifford, and in his editor Karl Pearson.23 "The terms matter and force, together with the ideas associated with them, [should be] entirely removed from scientific terminology- [reducing], in fact, all dynamic to kinematic."24 Finally, we may note its appearance on Henri Poincar6 for whom the laws of physics became arbitrary stipulations, expedient conventions, about how words like "force" and "straight lines" are to be used.25 This view is really an extreme logical wing of positivism as opposed to the extreme empirical wing represented by Macifs view that laws are summaries of experimental facts.
It seems apparent that in reacting to essentialism this positivistic instrumentalism, still very common today, has gone too far. Its results are debilitating to scientific advance. If we consider hypotheses in science to be nothing but instruments for predicting phenomena we must ignore the fact that there are no phenomena which are not themselves hypothetical. All "observed facts" are actually theories about reality-statements about instruments and clock readings, for example, are convenient bases for testing theories but they are really only corroborated theories themselves and hence ever open to further testing. Are we never to be permitted to revise the meaning of our prior interpretations of phenomena in the light of new theories? Are we not to be permitted to devise competing theories, each explaining all current data, but differing on their predictions as to future data so we may test among them? Are we to assume that the idea of a hierarchy of explanatory theories isn't useful in science and that the historical development of these hierarchies has not arisen because people sought a deeper and deeper insight into the character of the real world? Are we not to be permitted to suggest explanations for the phenomena we observe in science, however much the mistakes of essentialism have shown us we should be cautious in belieting what we speculate? Surely not.
Finally, as an example of the conjectural method let us look at Newton. It must be admitted that Newton was in many ways an instrumentalist in his conception of scientific theory.26 A mathematical framework which adequately describes the data of contemporary physics usually appears to satisfy him, without feigning additional hypotheses to explain such things as the observed fact of gravity. But inconsistencies in applying this reveal that there were also essentialistic: elements in his thought. The inability to leave apparent action at a distance unexplained, for example, led him to demand an ultimate explanation of gravity in terms of the action of a Divine Spirit.27 While empiricism could not accurately determine such action, Newton believed this to be an ultimate explanation, that is, an explanatory theory not needing further explanation; not indeed capable of being further explained.
But Newton was quite aware that the mathematical and mechanical elements of his model would be subject to the test of future applicability in his new situation and in the light of -new information. In this sense Newton operated within what Popper would call the conjectural framework-the method of tentative hypothesis (in the current, not Newton's, use of this term) and test. Any hypothesis must be scientifically acceptable, capable of experimental verification, and be subject to sacrifice if it conflicts with the data of physical experience.28
Indeed, taking some liberties with Strong's analysis, we might outline Newton's method fairly readily as follows:
1) A query asks if some deduction from a proposed theory is or is not a fact. It seeks to settle some doubt about the theory.
2) The proposed theory arises originally by abstraction rooted in the analysis of past data of observation and experiment.
3) A query, when no contrary evidence is submitted and to which no exceptions are taken, corroborates the theory (or part of it) subject to further queries suggested by new empirical data or by the logic of the theory.
It seems that a strong case can be made for such an attitude, refined of course to handle the minutiae of logic and method necessary to satisfy the sophisticated contemporary philosopher of science and suitable to everyday theoretical science. Of course it is impossible to develop the theme here, but we might summarize a few of the major attitudes of this type of approach.
1) Scientific theories (which are not just laws summarizing observation) are justified, not by their in duction from experience, but by their ability to withstand severe testing when consequences deduced from them are tried out.
2) If a theory is not potentially refutable by some conceivable event it is not scientific at all. To the degree that it is testable in new and risky situations it is to be preferred over a less testable alternative. If a theory is in error we should want to discover that it is, regardless of how well it has survived past testing.
3) Our prior convictions tend to influence our attitudes regarding the worth of specific tests and even to prejudice what theories we bother thinking about. This biasing can best be reduced by requiring that tests be reproducible so others may check, by not attempting ad hoc escapes from a test which has seemed to falsify our pet theory, and by consciously trying endlessly to refute rather than verify known theories and to think up more falsiflable alternatives.
4) This sort of methodology will prevent the usual dogmatism of the essentialist and the artificialities of the instrumentalist. To the degree that any theory is falsified it will tell us some character which the world does not have. To the degree a theory is corroborated by severe tests it tells us what the world might be like; hence, as we move to more testable theories for future work we must try to retain whatever we can of this prior insight.29
We may now turn to some interesting aspects of the above three views on the character of scientific theories as revealed by recent cosmological theorizing. Our findings will be utilized, in the final section of the paper, to define the basis for the conclusions noted there.
Looking first at essentialism, it appears that in current cosmological constructs this position appears in a weakened form from that of Kepler, who we chose as our early representative. At present essentialism is usually held only with regard to certain fundamental factors in the cosmological theory and not for the theory as a whole. For example, in the steady state model of Bondi and Gold,30 the perfect cosmological principle is conceived essentialistically although the detail of the model is open to a sizeable variation and change. It will be recalled that this principle, a modern combination of the traditional uniformity and economy principles, assumes that any permissible theory of the nature of the universe must be controlled by the realization that the large-scale description of the universe remains constant indefinitely through space and time.
To show why we interpret Bondi's and Gold's position here to be essentialism, it is necessary to show that they conceive of the perfect cosmological principle as a presently achieved true description of nature. Of course there is a variety of ways a believer in the principle might defend his thesis. He might have taken the principle to be self-evident, but this can hardly be the case logically, since its contradictory is not intern ally inconsistent, or in fact, since it is argued that the empirical consequences of applying it are what weigh so heavily in its favor. Or, it might be claimed that the principle is simply an extrapolation of what has been so far observed to be the case in large-scale astronomy into regions of future observation. However, changes that have been necessitated by researches since 1948 (when the principle was first enunciated) into just what the observations in fact are that we must extrapolate in future, show that the content of the principle may require yet further revision and that what we have presently is therefore hardly the last word.31
Another tack might be to treat the principle as an axiom in some theory, but if this is done it must have deducible consequences. Now it is obvious that the models which the steady-state school present do have testable, or at least potentially testable, consequences. But the perfect cosmological principle is not one of the statements involved in the complex making up these models, for they can be constructed without it. It serves rather as a regulative principle in deciding what physical laws and theories the steady-state people will permit in their models, either from the start (as in Bondi and Gold) or after some later point (as in Hoyle).32 Indeed its advocates use it to define what laws and theories may be given the appellation "scientific". These choices are logically arbitrary (however preferable they may seem, a priori, from past results or in comparison to present alternatives) and their value will have to be proven out against competing ideas, equally arbitrary on logical ground. Thus we may test their steady-state models, but some other construct, "unscientific" by their definition, may prove to be preferable. At that point we may look back and decide which regulative principles were not justified in suggesting the restriction of science to certain spec ified classes of descriptive relation (laws and theories).
Hence, we see that the perfect cosmological principle is really a regulative meta-theoretic concept defining the scope of what is science and what cosmological models are permissible. But the steady-state people illicitly tend to treat what should be merely a proposal as a true and necessary prescription, hence as essentialistic. While dogmatism here is surely undesirable, fairness requires that we point out that treating the perfect cosmological principle as if it had to be true has served as an incentive to exploring models based on it and to developing some very suggestive astrophysical ideas. But we must also point out that this essentialistic attitude has had several deleterious consequences. One is that, in arguing that on any other ground than the perfect cosmological principle, astrophysicists must remain agnostic about the universe in its distant reaches and its far past, it has ignored possibilities that avoid such a distasteful consequence offered by competing cosmological principles.33 Another has been that apologetes for particular antireligious arguments and synthesizers who saw in the steady-state ideas a correspondence with their metaphysical predilections have suffered a sort of mental petrifaction. as a result of tying their theses in part or entirely to such a particular cosmological assumption as if it were final.
Instrumentalism, while very common in many areas of contemporary science and particularly physics, is not too common in cosmological theory. But it does appear in the early Einstein, in later discussion on the famous cosmological constant in the field equations, in kinematic relativity (particularly its artificial time scale, in the odd relativities of Eddington and Jordan, in Hoyle's revised field equations, and above all in the purely descriptive side of Dirac and Jordan's work and the untestable aspects of Godel's, Omer's, and Heckmann's models.
For the sake of those unfamiliar with the G8del-Heckmann-Omer models in cosmology let us quickly describe them. 34 Godel's models involve the possibility of an absolute rotation of matter, a model with a general rotation of all galaxies relative to what he calls the "compass of inertia". This model is non-isotropic as a consequence and thus both deviates from the usual field equations of general relativity, with their concommitant inertial ideas, and from the usual isotropic models of cosmic theorizing. The Gadel idea involves only local times defined by the motion of matter in various regions of the model which cannot fit together to give a single time-history for the entire universe. It thus avoids a singularity (a beginning) for the present universe at some past, moment.
Omer and Heckmann discuss, on the other hand universes which are more complex than the homogeneous "fluid" models of radiation and matter commonly found in cosmological theory. Omer's model involves a general expansion, but with a velocity differing, at a given time, from place to place. Hence each region has its own beginning in a dense environment expanding toward what is known as a de Sitter state in infinite time. Heckmann postulates the presence of some shear and rotation everywhere in the motion of matter. His models seem, under certain conditions, to permit the evidence of a past time singularity.
There is no doubt that the above ideas suggest important future work in theorizing from novel directions. But we are forced to call them instrumentalistic because at present they are mere mathematical exercises and show little possibility of being tested in the foreseeable future. For they do not attempt to explain the origin of their unusual conditions nor do they give us any idea how to detect the inhomogeneities and anisotropies of the magnitude suggested. Those who look on them with favor, therefore, do so because they provide a means of avoiding a beginning for the entire present universe and in spite of their lack of likely testability. Philosophical or methodological predilections seem to be determinative in giving them serious consideration.
Turning, finally, to Popper's view that theories are to be seen as conjectures used for the purpose of severely testing their assumptions we shall not point out the very numerous applications of this principle by others in cosmology. Let me instead report rapidly some conclusions arrived at by the writer who used Popper's attitude in assessing some cosmological problems. First, and perhaps of most -general interest, is the fact that there are indications that the steady-state models may be in trouble as may the traditional Einstein-de Sitter model and the kinematic relativity model of Milne. However, the whole question of the value permitted for the cosmological constant, indeed even whether it exists at all, and of the curvature permitted for various models is still wide open. Secondly, and of more methodological import is the need for denying the well-known pulsating models any serious scientific consideration and the same applies to a lesser degree to the venerable Lemaitre exploding atom thesis. These are simply untestable or involve too many ad hoe theses to be credible. Thirdly, the question of continuous and/or initial creation turns out to be both untestable, and thus outside scientific discourse, and also an unsuitable criterion for even philosophical choice among models. Finally, the entire question of the age permitted the universe by most models is quite uncritical for choosing among these.35
In the closing section it will be our purpose to point up the kind of controls which seem to operate, though they are of an unscientific sort, on the problem of choice among cosmological theories. It is these metascientific criteria which are influential both in choosing among the three views of theories discussed earlier and upon the detailed application of even Popper's laudable scheme. Of course we can only examine a few of the many influences of this type, but we shall attempt to center our remarks around a unifying thread which one might call religio-metaphysical.
Cosmology is perhaps one of the most interesting theoretical and empirical fields for the purposes of a paper such as this because it reveals the more abstract type of scientific thinker at work and because it is also closely related to matters of philosophical and religious concern. Its only rivals in this regard seem to be certain biological theories and the indeterminancy problems of microphysics. Consequently, such a science is most affected by ideological concerns which can operate upon the scientist and by the predispositions of the scientist himself.
The classic example of the former is the control upon ' cosmological speculation within Soviet Communism since here the ideology, unlike Naziism with its dominant sociological dogma, is restricted by an imposed metaphysic. Mikulak and others36 have studied this problem most carefully. Mikulak notes that Engels set the tone of later dogma by espousing an uncreated universe consisting of innumerable worlds in infinite space, existing eternally and continually renewed by extinct stars falling together, heating up, and producing new stars. Codification of the basic aspects of this model was rapid so that theoreticians began to demand that science under dialectical materialism accept the following theses: (1) that there is an infinite space and an eternal universe; (2) that matter and motion are without beginning and are inexhaustible and indestructible; (3) that the universe cannot run down to a heat death; and (4) that one cannot extrapolate physical laws discovered locally or in closed system to the universe as a whole.
Imposition of these axioms was also rapid. By 1937-1941 Soviet philosophers were voicing intense criticism of those who utilized relativistic approaches in cosmology. They rejected what they felt was an implied creationism in these, the fictitious nature of their "curved" space,37 their finitude, and their assumed "heat death". A report in 1938 went so far as to claim that fascist agents were at the root of attempts to infiltrate Soviet astronomy with such ideas and that astronomers could not remain indifferent to the ideological battle.
Again In 1948 a major conference was held to examine postwar cosmology in the light of the reigning ideology. Milne, Lemaitre, Eddington, and Jeans were criticized for extrapolating special relativity into the entire universe, and an expanding universe was held to be anthropomorphic. In fact a closed, expanding universe was taken to be the major ideological enemy of materialistic science! Mikulak concludes that such ideas have apparently set the tone for Soviet theory up to the date of his writing, since there was an almost complete absence of positive contributions in Russian literature to the concept of an expanding universe. This is still the case. Instead, Soviet cosmology prefers to stress ideologically "safe" problems and research so that, as a result, the literature is largely observational.
As examples of the biases offered by personal predilections in cosmology, we will note only certain dominant religious and metaphysical influences. Religious attitudes first become active in choosing among models already available for the choosing. Milne's choice was predicated on the thesis that God could not create a universe wtih finite and therefore arbitrary mass or with arbitrary laws.38 Lyttleton and Hoyle both find the possibility of a creator in expanding models to be distasteful.39 Similar religious criteria are apparent in Lemaitre and in several other Christian writers.40 Secondly, religious attitudes are a factor in developing one's own model. Kepler exhibits such an influence, as do Leibniz and Newton in earlier astronomy, but Einstein or Milne or Hoyle are fine contemporary examples.41
A third role shows up in the varied interpretations of deductions from cosmic constructs. This very common, and probably inescapable, relating of science to one's philosophy has a vast literature.42 It ranges from attempts to prove the existence of God from cosmological ideas to just as serious attempts to disprove the same idea and from attempts to argue to creation or purpose to opposition to these theses. Finally, religious pre-disposition affects the evaluation of test data itself. One problem here is the introduction, often surreptitiously of teleological terminology, in describing empirical data. Terms such as "the plan of the universe", "the world is so constructed", and "the universe apparently purposes to" are rather easy to use but they tend to personify nature or beg the question they seek to answer.43 The language in which we scientists formulate our conclusions and its images and analogies lead to exponents of varied philosophies assuming that their view is demonstrated or denied.
Another difficulty in the same area arises from arguing too much from one's data. Lecomte du Nouy's Human Destiny and Cressy Morrison's Man Does Not Stand Alone are classic examples, but cosmology provides many others. The danger lies, not in claiming to prove a theological thesis, for most arguments of this sort claim only a high probability for their conclusions, but in too quickly assuming that one knows all the principles in the area "modelled" and in not yet having sufficient data to ascertain if the model is significant. But there is equally the danger of arguing too little.44 Because of a strong reaction to the restrictions enforced by historical religions, many scientists consciously or (worse) unconsciously claim an objectivity in their analyses which places value judgments, not out clearly where they can be analyzed, but into the realm of the intuitive, unrecognized, or haphazard. The extensive use of what are really ad hoe arguments to preserve cosmological theories in the face of contrary observational data is often based on an unspoken desire to protect a model in the face of the alternative of choosing other models really considered, rightly or wrongly, to have religious overtones of what the theoretician considers an objectionable nature.
We have stressed the role of religion, but similar comments might be made on more general metaphysical attitudes.45 Questions of whether scientific language is adequate to express ideas on all fields, the demarcation of ontological concepts from hypothetical entities, the differing criteria of judgment in philosophy and science, the precise theoretical framework necessary for discussing scientific concepts, and of how one discovers when experiment will induce a change in an accepted idea so that what was inconceivable becomes "thinkable" all are pertinent here. So are current debates on metaphysical directives on science and the relation of theory construction to prior value judgments.
To this point we have shown both the various pre-theoretic.views as to the character of theories together with certain consequences and also certain controls of other types which affect one's interpretation even within these perspectives. It seems apparent that cosmology finds these attitudes toward theory and the conditioning influences of religion and metaphysics operative to a high degree because of its close affinity to matters of rather deep human concern. But, in concluding, we might do well to suggest a few lessons to be learned from all of this.
The first of these is that stultifying biases to science, historically arising from religion and metaphysics, are least serious when operating in a Popperian view of the task of scientific theories. But, while it is less serious its presence even there must be recognized. If instrumentalism cannot avoid a judgment of what we may loosely call a metaphysical sort even when it claims to be ontologically agnostic or even anti-metaphysical, because these claims are really themselves ideological and because the predictable character of an instrumentalist view of theories has implicit in it the implication of a predictable world of interrelated phenomen a, the same kind of thing must affect a view which claims to seek the nature of reality. Can one ever claim only to attempt a description of nature -without at the same time really assaying to understand or explain it in some more fundamental way?
This brings up a second point. For those retaining the more common historical attitude toward the Scriptures, as seen in the Judaeo-Christian tradition, the control of scientific theorizing by interpretations placed upon certain passages has long been notorious. It is, for scientists in this tradition, no escape from the tendency to repeat this to move either toward instrumentalism or essentialism. The former dichotomizes one's faith and scientific practice and the latter cannot help but leave one dogmatically affirming some theory apparently agreeing with a specific exegesis of Scripture while remaining stubbornly obscurantistic to alternative theories and their argument. Surely it is far better to treat one's science as conjectural and to bend every effort in the direction of its critical examination, all the time aiming at a knowledge of God's world which one's faith otherwise affirms.
But this necessitates an interplay of hermeneutics and scientific theorizing. Interpretations of Scriptural passages are often frozen into the pattern of past and frequently unsatisfactory, metaphysical theories or scientific ideas. One's theories as to what a passage means must remain sensitive to scientific advance. However, were we to say no more, this would leave us with no other criterion of correct exegesis in some parts of Scripture than that of current scientific theorizing and leave us endlessly subject to the charge of arbitrary ad hoc adjustment to preserve the writings we revere in the face of new concepts and hypotheses. Thus, while it would seem that Biblical scholars would do believing scientists, and indeed their own religion, a great service if they didn't usually appear to lag science by a generation or more but rather honestly stated just how broad the possibilities for scientific interpretation usually are in Scripture, this breadth does necessitate placing some exegetical boundaries. For if there are no boundaries, one's view of Scripture as authoritative is left appearing to be purely the product of faith, since no portions of it can be potentially falsifiable in the light of experience. To attract the critical modern mind, the Judaeo-Christian teachings must be open to possible disconfirmation by whatever is suitable evidence.
At present, as we remarked earlier, cosmology cannot do more than say that the ideas of creation and of an age for the universe are meaningful in different ways in different models. But they will never be directly testable as a logical analysis will show. Only as we falsify the models or otherwise reject them as poor scientifically, can we falsify what they entail, and this might include their ideas of origin and age. As yet this is frequently an open question awaiting future work. Also, it seems that hermeneutical principles permit a broad spectrum of cosmological possibilities. Consequently, the falsification or corroboration of Scripture by cosmology appears quite dubious in the foreseeable future. Other sorts of scientific studies are far more pertinent to that task.
Thirdly, and lastly, the relationship of theism itself to the Judaeo-Christian faith must be mentioned. The type of God taught in the Bible is one who need not have created any particular type of universe, and whose laws in any universe He did create would be solely the product of His nature and not otherwise determined. Consequently, whatever occurs in the universe, or even the non-existence of any universe, would of necessity bear witness to such a God. Of course, were there no world in which God might reveal Himself, there would be no knowledge of this fact by human beings; but whatever kind of universe seems best corroborated in science can never be used to falsify the Judaeo-Christian God.46 Hence, while it is possible that we might falsify the Judaeo-Christian faith, and the Scriptures in which it is rooted, we cannot, oddly enough, falsify the existence of the God which that faith and those Scriptures logically presuppose.
Yet, in remarking on controlling principles in science, such as the cosmological principles or even Popper's own suggestion for how to look at theories, we have noted that, under certain conditions, they can be falsified by failure to aid in the derivation of usable scientific consequences. What then, - is there about the nature of the Judaeo-Christian God that makes this different?
May we suggest that the difference lies in the fact that the faith in or denial of such a God is more primary (indeed to the point of being an ultimate commitment) than any Judaeo-Christian belief? For you see if I argue only the theory that "If God exists, then Judaeo-Christianity will be the case" and I find the consequent false, I would falsify my theory only if the antecedent were true. And if I argue only the theory that "If Judaeo-Christianity is the case then God exists" and I find the former false my theory will be true regardless of the truth or falsity of the latter. Hence, only if I claim on faith the truth of the biconditional thesis "that if God exists, then Judaeo-Christianity is the case and if Judaeo-Christianity is the case, then God exists" would the falsification of the religion I hold involve the falsity of the existence of the God I claim. I may always avoid this by arguing that this strong faith was erroneous, in which case the biconditional would still leave the truth of the existence of such a God as I claim possible. It is the kind of faith we have which is the determining factor here.REFERENCES
1. Since then, the division he makes has been elaborated in
his Open Society and Its -Enemies (London: Routledge & Kegan
Paul, 1957) and in various articles now compiled in his Conjectures and Refutations
London: Routledge & Kegan Paul,
1963). A recent discussion of Popper's views may be found
In Mario Bunge (ed.), The Critical Approach to Science and
Philosophy, Free Press, Glencoe, 1964.
2. K. R. Popper, CAR, p. 105. These Ideas reveal two facets of Popper's use of the term "essence". The second argues that we can never be sure that whiteness is always characteristic of being a swan while the third implies that an a priori belief in anything as being necessary uniquely to swan-ness entails the belief in ultimate explanation and the premature cessation of questioning. Compare his LSD, pp. 430-341.
3. K. R. Popper, CAR, p. 107. See also his LSD, p. 252.
4. K. R. Popper, LSD, pp. 144-145.
5. Ibid., p. 373. See also his "A Note on Berkeley as a Precursor to Mach," CAR, pp. 166-174 for a discussion of this In practice.
6. K. R. Popper, CAR, p. ill.
7. Ibid., p. 113.
8. K. R. Popper, CAR, p, 114.
9. Ibid., p. 115.
10. Loc. cit.
11. Loc. cit.
12. Loc. cit.
13. Popper thus believes In a world independent of theories, but claims that we can be certain only about characteristics it does not have and only reasonably confident in the correctness of descriptions arising from theories that have withstood severe testing. Where such theories overlap in their "realm of discourse" we presume that we are likely describing different facets or layers of complexity in reality.
14. K. R. Popper, CAR, p. 117.
15. See Derek Price, "Contra-Copernicus" in M. Clagett (ed.). Critical Problems in the History of Science (Madison: Univ. of Wisconsin Press, 1959) and Ralph Blake et. al. (eds.) Theories of Scientific Method (Seattle: Univ. of Washington Press, 1960), pp. 38-43.
16. See Kepler's Epitome Astronomiea Copernicanae.
17. Compare Edwin A. Burtt, The Metaphysical Foundations of Modern Science (New York: Humanities Press, 1951), p. 48.
18. Compare Max Jammer, Concepts of Force (Cambridge: Harvard University Press, 1957), pp. 203-208 and K. R. Popper, CAR, pp. 107-109,166-174.
19. K. R. Popper, CAR, p. 171.
20. M. Jammer, CF, p. 205.
21. K. R. Popper, CAR, p. 173. Sources in Berkeley's writings are conveniently collated in this chapter.
22. Quoted In Philipp Frank, 1%instein, Mach, and Logical Positivism" in Edward Madden (ed.), The Structure of Scientific Thought (London: Routledge & Kegan Paul, 1960), p. 84.
23. Wm. K. Clifford, The Common Sense of the Exact Sciences (New York: Dover Publications, 1955) and K. Pearson, The Grammar of Science (London: Everyman's Library, 1937).
24. W. Clifford, CSES, P. LXV in Pearson's preface.
25. See his Science and Hypothesis, trans. W.J.G. (New York: Dover Publications, 1952), his "Principles of Mathematical Physics" reprinted in the Scientific Monthly, 82 (1956), 165175, and Ernst Nagel, The Structure of Science (New York: Harcourt, Brace & World, 1961), pp. 260-267.
26. Compare Mary Hesse, Forces and Fields (London: Nelson, 1961), pp. 147-148.
27. Compare Alexandre Koyr46, From the Closed World to the Infinite Universe (Baltimore: Johns Hopkins University, 1957), p. 234.
28. See E. W. Strong, "Hypotheses Non Fingo" In H. M. Evans (ed.), Men and Moments in the History of Science (Seattle: Univ. of Washington, 1959), p. 163. Compare Ralph Blake, "Isaac Newton and the Hypothetico-Deductive Method," In Ralph M. Blake et al (eds.), TSM, pp. 119-143.
29. This listing roughly generalizes Popper's Ideas and those of the writer in his "Popper's Views of Theory Formation Compared With the Development of Post-Relativistic Cosmological Models", Boston University Graduate School Ph.D. dissertation. Much of the rest of this paper is distilled from small portions of the latter.
30. H. Bondi, "Review of Cosmology", Mon. Not. Roy. Astron. Soc., 108 (1948), 104-120; H. Bondi & T. Gold, "The SteadyState Theory of the Expanding Universe," Mon. Not. Roy. Astron. Soc., 108 (1948), 252-270; H. Bondi, Cosmology (Cambridge: Cambridge Univ. Press, 1952); and H. Bondi et al, Rival Theories of Cosmology (London: Oxford University Press, 1960).
31. See editorial, London Times Science Review, (Slimmer, 1961), 1-2; the report in Discovery, (March, 1961). 91-92; G. C. McVittie, Fact and Theory in Cosmology (London: Eyre & Spottiswoode, 1961). pp. 155-166; and the extensive analysis in the author's PVTF.
32. Fred Hoyle, "A New Model for the Expanding Universe," Mon. Not. Roy. Astron. Soc., 108 (1948)3 372-382; "On the Cosmological Problem," Mon. Not. Roy. Astron. Soc., 109 (1949), 365-371; "Stellar Evolution and the Expanding Universe," Nature, 163 (1949), 196-198; "A Covarlant Formulation of the Law of the Creation of Matter," Mon. Not. Roy. Astron. Soc., 120, (1960), 256-262; and his Nature of the Universe (Oxford: Blackwell, 1960).
33. See T. H. Leith, PVTF, pp. 297-302, 313-339, 536-M.
34. Kurt Mdel "An Example of a New Type of Cosmological Solution of Einstein3s Field Equations of Gravitation," Rev. Mod. Physics 21 (1949), 477 ff. . See also his paper In P. A. Schllpp (ed.), Albert Einstein, (New York: Harper & Bros., 1959), 11, pp. 557-562. Also see discussion in J. L. Synge, Retativity: The General Theory (Amsterdam: North Holland, 1960), pp. 331-8 and G. J. WhItrow, The Natural Philosophy of Time, (London: Nelson, 1961), pp. 257-261. 0. Heckmann's lectures in London are reported in The Observatory, 79 (1959), 130-131.
35. See my PVTF, pp. 313-339, 346-351, 545-5W for a detailed discussion on these points.
36. Maxim Mikulak, "Soviet Cosmology and Communist Ideology," Sci. Mon., 81 (1955), 167-172; his "Soviet PhilosophicCosmological Thought," Phil. of Sci., 25 (1958), 35-50; David Joravsky, Soviet Marxism and Natural Science, 1917-1932 (New York: Columbia University Press, 1961); and a continuing definitive source of analysis, Studies in Soviet Thought, a quarterly journal published by D. Reidel Pub. Co., Dordrecht, Holland.
37. However, this could be reinterpreted, for in 1957 Kalman argued that Lobachevsky's geometry was based upon a materialist view of the world whereby spatial forms and laws were established from the movement of material bodies existing independent of consciousness. See report in Nature, 179 (1957), 1176.
38. E. A. Milne, Modern Cosmology and the Christian Idea of God (Oxford: Clarendon Press, 1952).
39. R. Lyttleton, The Modern Universe (London: Hodder & Stoughton, 1956), pp. 205-206; and F. Hoyle In R. Thruelsen et al, Adventures of the Mind (New York: Vantage Books, 1960), pp. 180-181 and his NU, p. 100.
40. G. Lemaitre in R. Stoops (ed.), La structure et Vevolution de 11univers (Solvay Congress in Bruxells, 1958), pp. 5,7; R. E. D. Clark, The Universe: Plan or Accident? (London: Paternoster, 1961); and J. C. Monsma (ed.), The Evidence of God in an Expanding Universe (New York: G. P. Putnam's Sons, 1958).
41. Gerald Holton, "Johannes Kepler: A Case Study on the Interaction of Science, Metaphysics and Theology," The Philosophical Forum, 14 (1956), 21-33; H. G. Alexander (ed.), The Leibniz-Clarke Correspondence (New York: Philosophical Library, 1956); F. Hoyle in M. Stockwood (ed.) Religion and the Scientists (London: SCM Press, 1959), esp. p. 60; and Hoyle's Einsteinian Spinozism in Man and Materialism (London: George Allen & Unwin, 1957), pp. 72, 157.
42. See bibliography in my PVTF, pp. 501-502.
43. Compare A. J. Bernatowicz, "Teleology In Science Teaching," Science, 128 (1958), 1402-1405.
44. See Richard Rudner In P. Frank, VST, pp. 24-28.
45. See my PVTF, pp. 505-513.
46. Of course we are assuming the most consistent and best corroborated thesis as to what the Bible teaches regarding God. Alternatives seem to either argue to the character of God on other standards which must, by the nature of the case, eliminate this view of God to start with or to a thesis regarding revelation which flies In the face of much available evidence.