Science in Christian Perspective



Charles S. Peirce, Scientific Method, and God

Terry G. Pence*

Northern Kentucky University
Highland Heights, KY 41099-2200

From: PSCF 49 (September 1997): 156-161.

Charles S. Peirce, the founder of American pragmatism, wrote extensively about a form of inference called abduction, or more familiarly, reasoning to the best explanation. He claimed that it was essential to the growth of science. In this article, I examine Peirce's theory of abduction to see if he thinks that this essential form of reasoning precludes appeals to a supernatural agency. I argue that it does not and that Peirce himself defends just such an abductive inference in his "A Neglected Argument for the Reality of God."

Charles S. Peirce (1839˝1914), founder of American pragmatism, claimed that there were three kinds of reasoning: deductive, inductive, and abductive. The last of these is more familiarly called reasoning to the best explanation. Two important claims which he made about abduction were that it was the only ampliative kind of reasoning and that it was essential to science. In calling it the only form of ampliative inference, Peirce meant reasoning which can give us more information than is contained in the premises. It is essential to the growth of science because "every plank of its advance is first laid by retroduction (i.e., abduction) alone."1

The question I wish to address in this paper is whether abduction or reasoning to the best explanation, as understood by Peirce, must preclude the supernatural from the explanans. Does Peirce, for example, believe that we should rule out appeals to supernatural agency on grounds of simplicity or requirements of empirical consequence from the explanation? The thesis of this paper is that he does not. I will try to explain Peirce's theory of abduction and show that it has no restrictions which would a priori eliminate appeals to the supernatural. I cite Peirce's own piece of abductive reasoning in "A Neglected Argument for the Reality of God" as primary evidence that, for him, there is no essential hostility between the most essential aspect of scientific reasoning and broadly theistic conclusions.

Peirce's Theory of Abduction2

Hypothesis Generation

Induction and deduction are commonly acknowledged kinds of reasoning, but Peirce claimed that, besides these, there is a third kind that he variously called abduction, hypothesis, hypothetic inference, retroduction, and presumption. In his mind, this was not so much a discovery as a recovery of a type of reasoning that Aristotle had mentioned.3 Peirce argues that in classifying types of reasoning, one should aim at bringing out the amount and kind of certainty that each form of reasoning affords and "bring out the possible and esperable uberty [fruitfulness], or value in productiveness of each kind."4 Regarding abduction and its relationship to deduction, he says:

From the first type (deduction) to the third (abduction) the security decreases greatly, while the uberty as greatly increases...I don't think the adoption of a hypothesis on probation can properly be called induction; and yet it is reasoning and though its security is low, its uberty is high.5

Abduction is an ampliative kind of reasoning. It is the only one which can introduce novel ideas differing in kind from those found in the premises or explanandum. This sort of reasoning takes place at the very beginning of scientific inquiry. In fact, "all the ideas of science come to it by way of abduction."6 Abduction takes place when:

Upon finding himself confronted with a phenomenon unlike what he would have expected under the circumstances, he looks over its features and notices some remarkable character or relation among them, which he at once recognizes as being characteristic of some conception with which his mind is already stored, so that a theory is suggested which would explain (that is render necessary) that which is surprising in the phenomena.7

Peirce also notes an interesting and controversial aspect of abduction when he says:

Abduction, although it is very little hampered by logical rules, nevertheless is logical inference, asserting its conclusion only programmatically or conjecturally, it is true, but nevertheless having a definite logical form.8

This logical form is the following: "The surprising fact C, is observed. But if A were true, C would be a matter of course; Hence, there is reason to suspect that A is true."9

Abduction then is an explanatory inference from certain data. Peirce claims that it is an inferential process because reasons can be adduced for the hypothesis.10 Although at one point he says: "Reasoning, properly speaking, cannot be unconsciously performed...For reasoning is deliberate, voluntary, critical, controlled, all of which it can be if it is done consciously,11 Peirce also describes abduction as a kind of guessing instinct. Abduction, he says: "...tries what el lume naturale [the light of nature]...can do. It is really an appeal to instinct.12 "The abductive suggestion comes to us like a flash. It is an act of insight, although of extremely fallible insight."13

At first glance, this would seem to disqualify abduction as a form of reasoning on Peirce's own account. But while Peirce believes that reasoning is a conscious process, he nevertheless holds that:

All that is necessary is that we should, in each case, compare premises and conclusion, and observe that the relation between facts, expressed in the premises involves the relation between facts implied in our confidence in the conclusion.14

In other words, the fact that Peirce defines reasoning as conscious, voluntary, controlled, and capable of being criticized at every point does not, as yet, rule out abduction as a form of reasoning, because one isn't required to be conscious of the whole process. Thus, although a certain hypothesis may occur as a flash of insight, the flash of insight is not justification. It is something which can be supported by reasonsˇsometimes good, sometimes bad.

A historical example will illustrate this point. In 1879 Louis Pasteur noticed a "surprising fact." He injected some chickens with bacillus that had been around for several months. Instead of dying as expected, the chickens became only slightly ill and then recovered. Pasteur concluded that the old cultures had spoiled. So he obtained a new culture of virulent bacilli from chickens afflicted with a current outbreak of cholera. Then he again injected the chickens along with some new ones. In due time, all the first-time injected chickens died. Those previously injected with the old "spoiled" stuff lived. When Pasteur was told what had happened, he, according to an eyewitness, "remained silent for a minute, then exclaimed as if he had seen a vision: `Don't you see that these animals have been vaccinated!'"15 Nearly a hundred years earlier, Edward Jenner had seen the connection between cowpox and smallpox. Even the term "vaccination" derives from the Latin word for cow, "vacca." The germ theory of disease was more recent. Pasteur was the first to connect the two and give birth to modern immunology. The point, however, is that Pasteur's theory was an abductive inferenceˇwhether it occurred to him in a minute, ten seconds, or a flash of insight. The surprising fact was that the previously inoculated chickens did not die. If his hypothesis, "These animals have been vaccinated," was true then, of course, they did not die. But Pasteur could equally have supposed any number of hypotheses that could have accounted for the fact that these chickens did not die. For instance, it may have been that the assistant forgot to inoculate them again; or, these chickens were just heartier than the new batch; or, it could have been something in their diet; or, they were charmed, etc. All these are abductions. Whether they occur as flashes of insight, they are supportable by reasonsˇsome more plausible than others.

Hypothesis Selection

Another aspect of Peirce's theory of abduction concerns whether it meant to include not only hypothesis generation but also a theory of hypothesis selection.16 In hypothesis generation, abduction can be a kind of insight or intuition that is "little hampered by rules." In hypothesis selection, abduction seems to be a logical inference that involves some explicit considerations that govern the process of selecting a hypothesis.17 Although these are perhaps separate issues, they are not separable processes in practice, nor are they kept apart in Peirce's thought.18 There are passages in which Peirce notes the considerations which go into hypothesis selection. These may be summarized as follows:

1. It must explain the facts.

2. It must be experimentally verifiable.

3.It must be economical.

The first consideration is the essential feature of abduction. Peirce writes in many places that, "Abduction ...amounts observing a fact and then professing to say what it was that gave rise to the fact..."19 and "abduction consists in studying facts and devising a theory to explain them."20 

In hypothesis generation, abduction can be a kind of insight or intuition that is "little hampered by rules." In hypothesis selection, abduction [is] a logical inference that involves some explicit considerations that govern the process of selecting a hypothesis.

The second consideration is equally insisted upon, as for example, in this passage:

The principle rule of presumption (abduction) is that its conclusion should be such that definite consequences can be plentifully deduced from it of a kind which can be checked by observation.21

It is consideration three, however, which brings in processes that are "little hampered" by logical rules. Peirce believes that "before you try a complicated hypothesis you should make quite sure that no simplification of it will explain the facts equally well"22 or "try the theory of fewest elements first; and only complicate it as such complication proves indispensable to the truth."23

Abduction and God

Given this sketch of the theory of abduction, it would be pertinent to ask whether explanations referring to God as part of the explanans would be prohibited. As a theory of hypothesis generation, the theory of abduction seems to be no bar to the God hypothesis. It is an ampliative form of inference which can bring forth any kind of explanan. Abduction as a theory of hypothesis selection, however, does seem to present some problems. Would a God hypothesis meet the criteria of simplicity and empirical verifiability? I believe that Peirce would answer "Yes." The best evidence for this claim is his discussion of just these points in "A Neglected Argument for the Reality of God."24

In this article, Peirce presents an abductive argument for the reality of God. He defends the appropriateness of making this sort of inference from playful musing speculation on such surprising facts as the variety, homogeneity, interconnectedness, and beauty in the cosmos; our ideas; and our active powers to connect our ideas to facts and things. What is surprising is that he goes out of his way to suggest that this inference meets the canons of scientific inference, when it was surely open to him to say that the inference is justified, but not science.

Consider the ways in which he defends the process and the project. It seems to me that at every turn, he could have accepted the objection and sidestepped it by saying that theology has its own canons of evidence and domain so it need not meet those of science. He could have said this, but he does not.

A priori Barriers to Inquiry

Peirce dismisses those who would suggest there is no point or profit in speculating along certain lines of inquiry. He colorfully calls them "tribes of Sir Oracles, colporting brocards to bar off one or another roadway of inquiry."25 Auguste Comte is named as a tribal leader. The examples he cites of short-lived maxims were: "No science must borrow the methods of another" (an a priori methodological restriction) and "It is not the business of science to search for origins" (an a priori restriction on the proper domain of science). Peirce thinks that history has shown that these types of a priori roadblocks to inquiry are laughably obsolete.

Peirce could have said that such restrictions are all fine and dandy for science, but we need not heed them because we are doing something different. It is a different game; it has different rules. He could have said this, but he does not.

The Problem of Simplicity

Whatever the God hypothesis explains can be explained or explained away on some other hypothesis. Since naturalistic explanations are going to be used anyway, doesn't any appeal to God as an explanatory hypothesis needlessly proliferate a theoretical entity and violate one of Peirce's own rules of economy of research, namely, of two hypotheses, the simpler is to be preferred?

[Pierce] argues that the God hypothesis does meet the simplicity criterion.

Peirce could dodge this issue by arguing for the inapplicability of the simplicity criterion. Instead he argues that the God hypothesis does meet the simplicity criterion. Now if you just count explanatory assumptions, the God hypothesis is more complicated, but is this how simplicity should be understood? Peirce does not believe that fewest is always the truest. He only insists that the uncomplicated hypothesis be tried first, since it may be the easiest to refute. But which hypothesis is the simplest? Is there some objective criteria to determine it? Consider this warning by a contemporary philosopher of science, Carl Hempel:

Any criteria of simplicity would have to be objective, of course; they could not just refer to intuitive appeal or to the ease with which a hypothesis or theory can be understood or remembered, etc., for these vary from person to person.26

and then consider Peirce's answer:

Modern science has been built after the model of Galileo who founded it on il lume naturale. That truly inspired prophet had said that of two hypotheses, the simpler is to be preferred, but I was formerly one of those who, in our dull self-conceit fancying ourselves more sly then he, twisted the maxim to mean the logically simpler, the one that adds the least to what has been observed....It was not until long experience forced me to realize that subsequent discoveries were every time showing I had been wrong, while those who understood the maxim as Galileo had done, early unlocked the secret, that the scales fell from my eyes and my mind awoke to the broad and flaming daylight that it is the simpler Hypothesis in the sense of the more facile and natural, the one that instinct suggests, that must be preferred; for the reason that, unless man may have a natural bent in accordance with nature's, he has no chance of understanding nature at all.27

Thus, in elaborating what is meant by economy, Peirce describes two sorts of simplicity. The first he calls "logical simplicity." This is the hypothesis that "has the fewest elements" is least complicated, or is easiest to refute. Another type of simplicity goes unnamed but might be called "natural simplicity." This is the type of simplicity just described in the quotation above. The relative importance of these two senses of simplicity is captured in this remark: "I do not mean that logical simplicity is a consideration of no value at all, but only that its value is badly secondary to that of simplicity in the other sense."28 What is more important, the God hypothesis has this type of natural simplicity to the highest degree.

The Problem of Direct Verification and Empirical Consequence

It could be argued that the God hypothesis posits a theoretical entity which is not directly observable and has few empirical consequences. Peirce rejects the former criterion and tries to finesse the latter.

One rule for hypothesis selection noted above was that a hypothesis should have empirical consequences. That remark should not, however, be interpreted as support for positivism or the claim that only what is empirically verifiable can be included in a hypothesis. Peirce actually repudiates the descriptive positivism found in Comte and others.29 In a review of James' Principles of Psychology, Peirce chides him for rejecting theoretical entities. He claims that this is a matter of James' personal taste:

 Nor is it in the least true that physicists confine themselves to such a "strictly positivistic point of view." Students of heat are not deterred by the impossibility of directly observing molecules from considering and accepting the kinetical theory; students of light do not brand speculation on the luminiferous ether as metaphysical; and the substantiality of matter itself is called in question in the vortex theory, which is nevertheless considered as perfectly germane to physics. All these are "attempts to explain phenomenally given elements as products of deeper-lying entities." In fact this phrase describes, as well as loose language can, the general character of scientific hypotheses.30

Does the God hypothesis have empirical consequences? Here I think that Peirce has three responses. The first is to say that we can't expect too much in the way of empirical consequences from the God hypothesis because "the hypothesis can be apprehended so very obscurely that in exceptional cases alone can any definite and direct deduction from its ordinary abstract interpretation be made."31 Second, the hypothesis has pragmatic consequences of a sort in the way it regulates and commands influence over the life of the believer.32 Lastly, the God hypothesis does have empirical consequences after all. At the end of the article, Peirce says that the God hypothesis "is connected so with a theory of thinking that if this be proved so is that."33 Peirce thinks that since the theory of the nature of thinking has empirical consequences, the God hypothesis can claim some of its empirical credit.


In this paper, I have tried to show that the most important kind of reasoning which takes place in science is abductive reasoning or reasoning to the best explanation. Peirce viewed abductive inferences which appeal to God as an explanation as legitimate as any scientific inference. I have construed his "A Neglected Argument for the Reality of God" as arguing for this point. Now I would like to offer some ideas as to why Peirce thinks this.

Peirce viewed abductive inferences which appeal to God as an explanation as legitimate as any scientific inference.

Two reasons occur to me. The first is that Peirce is a critical realist. He believes that scientists have made correct guesses about the nature of reality and that God is real as well. Unless you are prepared to argue that parts of reality can only be discovered by incommensurable means or that there are incommensurable domains of knowledge, one seems driven to the conclusion that claims about reality should meet some common, or at least similar, standards. It is not just that science and its methods can epistemologically dictate the grounds of rational acceptability, it also goes the other way around. That is, where it appears that the canons of scientific evidence or method have apparently precluded what we know to be true, it is time to reform the science. The adequacy of proposed methods is to be measured against what we know to be true.

My second speculative thought about why Peirce wishes to conform the argument for God into a scientific inference is this: the God hypothesis and materialistic explanations are, at times, in competition for explaining the same facts. If this is the case, then it would make sense that the God hypothesis should try to best the materialist hypothesis on the criteria it thinks it meets.



1Charles Sanders Peirce, Collected Papers of Charles Sanders Peirce, 8 vols., Charles Hartshorne and Paul Weiss, eds. (Cambridge, MA: Harvard University Press, 1958). Citations in the notes will cite only the title, the volume number, the date where there is one, and then the paragraph. Subsequent references will cite title and paragraph. This reference is from "A Neglected Argument for the Reality of God," Vol. 6 (1908), 477.

2The following is an account of the more salient features of Peirce's theory of abduction. Most of Peirce's doctrines show signs of development and this theory is no exception. For my purposes, however, I will ignore the subtleties of this development and refer mainly to the mature theory. For an account of this development one may consult Arthur Burks, "Peirce's Theory of Abduction," Philosophy of Science (October 1946): 301˝306; Francis E. Reilly, Charles Peirce's Theory of Scientific Method (New York: Fordham University Press, 1970), 31˝35; and especially K. T. Fann, Peirce's Theory of Abduction (The Hague: Martinus Nijhoff, 1970).

3"Lessons from the History of Science" Vol. 1 (c. 1896), 65.

4"A letter to F.A. Woods," Vol. 8 (1913), 384.

5Ibid., 387˝88.

6Harvard Lecture V, "On Three Kinds of Goodness," Vol. 5 (1903), 145.

7"Syllabus," Vol. 2 (1903), 776. Cf. "A Letter to Paul Carus," Vol. 8 (1910), 229.

8Harvard Lecture VII, "On Pragmatism and Abduction," Vol. 5 (1903), 188.

9Ibid., 189.

10See "On the Natural Classification of Arguments," Vol. 2 (1867), 511n.

11Minute Logic, Chapter 2, Section 2, "Why Study Logic?" Vol. 2 (1902), 182. For a valuable discussion of these four marks of reasoning (deliberate, voluntary, critical, controlled), see Maryann Ayim, "Retroduction: The Rational Instinct," Transactions of the Charles S. Peirce Society 10 (Winter 1974): 34˝43. See especially pp. 37˝38.

12Cambridge Lectures, Lecture 1, "Philosophy and the Conduct of Life," Vol. 1 (1898), 630.

13Harvard Lecture VII, "On Pragmatism and Abduction," Vol. 5 (1903), 181.

14Minute Logic, 183.

15Cf. Arthur Koestler, The Act of Creation: A Study of the Conscious and Unconscious in Science and Art (New York: Dell, 1964), 112˝114.

16The question is explicitly addressed by Harry G. Frankfurt in his "Peirce's Theory of Abduction," The Journal of Philosophy 55 (July 3, 1958): 593˝597.

17Cf. "On the Logic of Drawing History from Ancient Documents Especially from Testimonies," Vol. 7 (1901), 220.

18This is the solution to the problem that K T. Fann suggests in Peirce's Theory, p. 41.

19Cambridge Lectures, Lecture III, "The First Rule of Logic," Vol. 5 (1898), 581.

20Harvard Lecture V, "On Three Kinds of Goodness," Vol. 5 (1903), 145. Cf. "A Letter to Paul Carus," Vol. 8 (1910), 229.

21From Contributions to Baldwin's Dictionary of Philosophy and Psychology, Vol. 2 (1901), 786. Cf. also Harvard Lecture VII, "On Pragmatism and Abduction," Vol. 5 (1903), 196.

22Harvard Lecture III, "On Phenomenology," Vol. 5 (1903), 60.

23Grand Logic, "The Essence of Reasoning," Vol. 4 (1893), 35.

24"A Neglected Argument for the Reality of God" Vol. 6 (1908), 452˝491. This article was originally published in Hibbert Journal 7 (1908): 90˝112.

25Ibid., 460.

26Carl G. Hempel, Philosophy of Natural Science. Foundations of Philosophy Series (Englewood Cliffs, NJ: Prentice-Hall, 1966), 41.

27"A Neglected Argument," 477. It is interesting that although Hempel insists upon an "Objective Criteria" for simplicity and Peirce denies there is one in Hempel's sense, Hempel ends on this note: "Thus, while all the different ideas here briefly surveyed shed some light on the rationale of the principle of simplicity, the problems of finding a precise formulation and a unified justification for it are not as yet satisfactorily solved" (Natural Science, 45). A more extensive treatment of the simplicity criterion can be found in Fann, Peirce's Theory, 47˝51; and Reilly, Scientific Method, 38˝41.

28"A Neglected Argument," 477.

29Lowell Lectures, Lecture VIII, "How to Theorize," Vol. 5 (1903), 597.

30Review of William James' The Principles of Psychology , Vol. 7 (1891), 60.

31"A Neglected Argument," 489.

32Ibid., 490.

33Ibid., 491.