Chapter 4

Eugene Lashchyk, Scientific Revolutions, 1969

Chapter IV
Conflicting Views on the Nature of Science

For centuries thinkers have speculated on the nature of science. They marveled at the developments in medicine which rescued man from disease and premature death. They marveled at the science of mechanics, developments which brought man machines to transport him by land, sea, and air with ease and comfort. Above all, no area of human endeavor has proved so successful in satisfying the natural curiosity of man, or in finding so many universally accepted answers to questions about nature. The obvious question arises, why has science been so successful in arriving at universally recognizable answers to some of the most perplexing questions about nature? It seems that the majority of scientists are too busy doing science to worry about such a question. But among the great scientists, those giants that have been responsible for making innovations in theories as well as in experimental techniques, there are always a few, like Pierre Duhem, Sir James Jeans and Albert Einstein, who feel it their duty to explain to the non-scientists the method employed by the scientists.{l}
Besides such scientists there were and still are many philosophers who speculate on what science really is, and in particular they discuss what method the scientists are or ought to be following.

(A) Inductivist View

Many have felt that the enormous success of science must be due to some mechanical, routine method analogous to a sausage-making machine. Perhaps the oldest "sausage-making machine" view of science can be traced to Francis Bacon who is the father of the famous "inductive method" in science. One article describes the so-called "Baconian method", which scientists ought to follow in order to arrive at truth in matters concerning nature, as follows:

First Rule: Compile the facts.

Second Rule: Classify these facts into three classes and then list:

instances of the presence of the characteristic in question.
instances of its absence.
instances of its presence in various degrees.

Third Rule: Determine what is and what is not connected with the phenomena under investigation.{2}

Carl Hempel who also criticizes the narrow inductivist method quotes the following modern description of this method:

If we try to imagine a mind of superhuman power and reach, but normal so far as the logical processes of its thought are concerned. . . . would use the scientific method, the process would be as follows: First, all facts would be observed and recorded, without selection or a priori guess as to their relative importance. Secondly, the observed and recorded facts would be analyzed, compared, and classified, without hypothesis or postulates other than those necessarily involved in the logic of thought. Third, from this analysis of the facts generalizations would be inductively drawn as to the relations, classificatory or causal, between them. Fourth, further research would be deductive as well as inductive, employing inferences from previously established generalizations.{3}

This emphasis on the collection of facts, while an improvement on the medieval method of citing authorities, can be overemphasized to the point of obliterating the contributions of the creative imagination of man. Furthermore, the motto "observe and compile all the facts" is demanding the unattainable. No person or persons can ever satisfy such a demand, for there is an infinite number of "facts" (i.e., phenomena) to be observed. If the first rule is modified to read "make some observation of facts," then the obvious question creeps up "how much is some?" and furthermore how can we know that a reliable sample has been obtained. If, however, the demand is that a "representative sample of facts is to be compiled," what method or theory will guarantee that a representative sample has been extracted?{4}
To put the finger on the difficulty with the above proposals, it can be said that all of them lack criteria of relevance. Such criteria of relevance are usually supplied by the problems under investigation. Rational inquiry begins with problems, puzzles, questions and not with a collection of so-called facts. Once the problem is clear, hypotheses are proposed for the solution of the problem or puzzle. Compilation of data in support of a proposed hypothesis usually comes at this point. Now the criterion of relevance is clear; instances of the generalized hypothesis are relevant factual material for the confirmation or falsification of a generalized hypothesis. Take for example, the hypothesis that "all crows are black" which can be symbolized as (x)(Cx --> Bx). Such a hypothesis will be falsified by an instance of a crow that is not black -- in symbolic logic, the sentence S1 (Ca & ~Ba). Thus, S1 can be said to be a negative instance of hypothesis h or a falsifying instance of this hypothesis. A scientific report S2 of a black crow (Cb & Bb) can be said to be a positive instance of hypothesis h or confirming instance.
The method attributed to Bacon, therefore, overstresses observation or "fact" gathering without giving proper role to man's judgment in the formulation of hypotheses by his creative imagination.
Throughout the history of science there were other versions of the Inductivist Method. For example, Sir John Herschel stated the inductive method as follows: the "whole of natural philosophy consists entirely of a series of inductive generalizations, commencing with the most circumstantially stated particulars and carried to universal laws or axioms."{5}
Many inductivist philosophers and scientists before and after Herschel's time refer to inductive generalizations, but I have yet to see an adequate exposition of this type of generalization. It seems that scientists have made discoveries under varying circumstances using no universally accepted method. Herschel's proposal is thus unilluminating unless he specifies how these inductive generalizations are to be found.
Bertrand Russell also in similarly vague terms describes the scientific method as follows:
The scientific method, although in its more refined forms it may seem complicated, is in essence remarkably simple. It consists in observing such facts as will enable the observer to discover general laws governing facts of the kind in question. The two stages, first of observation and second of inference to a law, are both essential, and each is susceptible of almost indefinite refinements.{6}

Just as in previous formulations, so in Russell's one wonders what form of "inference to a law" he has in mind. To my knowledge there is no method of inference which would lead a scientist to arrive at the laws of nature.
It is annoying how often such a formula is repeated by scientists and philosophers without their ever explicitly identifying the magical form of inference which is to lead the scientists to laws of nature.
There is an inductive argument or form of inference which some inductivist philosophers and scientists might have in mind as a method of discovering laws of nature; it proceeds in the following way. Compile instances of C which have the property E: Thus:

C1 is B
C2 is B
C3 is B
C4 is B
.
.
.
Cn is B
Therefore, all C's are B.

This type of inductivist argument has been called induction by enumeration of its instances. Inductivism as a philosophy of science, therefore, could be described as the view which holds that all scientific hypotheses are discovered by inductive enumeration and it is some times added that scientific hypotheses can also be justified by inductive enumeration.{7}
Also if Reichenbach{8} is right that all inductive inferences are reducible to inferences of induction by enumeration, then all the above references by Bertrand Russell, Herschel and others who refer to such inductive inference are subject to the following criticism.
Karl Popper has rightly criticized the claim that inductive enumeration is always used by scientists as a method of discovery. The history of science does not bear out such a claim.
Scientists arrive at hypothesis by many means. Only a relatively small number of hypotheses are arrived at by inductive enumeration. These are usually those hypotheses that generalize experimental results. For many hypotheses this method of induction couldn't possibly have been used for the simple reason that these hypotheses contain theoretical terms whose instances either have not as yet been observed or even more importantly for some cases, observation is in principle impossible.
Table of Contents -- Next

Notes
{1} See in particular the following works: Pierre Duhem, The Aim and Structure of Physical Theory, translated by P. P. Wiener (Atheneum, New York, 1962); Sir James Jeans, The Mysterious Universe (New York: 14acMillan Co., 1930); Albert Einstein and Leopold Infeld, The Evolution of Physics (New York: Simon and Shuster, 1938). [Back]
{2} See for example the article on Francis Bacon by Charles Singer in the Encyclopedia Britannica (1944), II, p. 886. [Back]
{3} Carl G. Hempel, Philosophy of Natural Science (Englewood Cliffs, N.J.: Prentice Hall, Inc., 1966), p. 11. Above quotation is from A. B. Wolfe, "Functional Economics" in The Trend of Economics, edited by R. G. Tugwell (New York: A. Knopf, Inc., 1924), p. 450. [Back]
{4} For further development of this argument, see Karl Popper's, Conjectures and Refutations (New York: Basic Books, 1963), p. 46. [Back]
{5} Quoted by J. Agassi in his monograph Towards A Historiography of Science, note 49 from Herschel's A Preliminary Discourse on the Study of Natural Philosophy (London) 1831, p. 104. [Back]
{6} Bertrand Russell, The Scientific Outlook (New York: W. W. Norton and Co., Inc., 1962), p. 13; originally published in 193l. [Back]
{7} See for example James D. Carney and Richard K. Scheer, Fundamentals of Logic (New York: The MacMillan Co., 1964), p. 352. [Back]
{8} Reichenbach claims:
"At this place it may suffice to remark that modern analysis has shown all forms of inductive inference to be reducible to induction by enumeration, a result which makes it permissible to restrict the discussion of inductive method to this simplest form as Hume did." (Hans Reichenbach, The Rise of Scientific Philosophy (Berkeley and l.os Angeles: University of California Press, 1954), p. 86.) [Back]
Table of Contents -- Next