Citation metadata

Editor: Donald M. Borchert
Date: 2006
Encyclopedia of Philosophy
From: Encyclopedia of Philosophy(Vol. 1. 2nd ed.)
Publisher: Gale
Document Type: Topic overview
Pages: 4
Content Level: (Level 4)

Document controls

Main content

Full Text: 
Page 92


In the most general use of the term, "agnosticism" is the view that we do not know whether there is a God or not. Although the history of agnosticism, in this general sense, is continuous with that of skepticism (thus reaching back to the ancients), the term itself was coined by T. H. Huxley and its distinctive philosophical bearings emerged in the course of the nineteenth-century debate on religious belief. Participants in that debate often used the word in a strong and specific sense: To be an agnostic was to hold that knowledge of God is impossible because of the inherent, insuperable limitations of the human mind. To assert confidently either the existence or the nonexistence of a deity with definite and intelligible attributes was to transgress these limits.

This consciousness of limitation is classically expressed in the "Transcendental Dialectic" of Immanuel Kant's Critique of Pure Reason (1781). There is a continual temptation, Kant stated, to raise questions about the totality of things; but these questions, he argued, are demonstrably unanswerable. Contradictions are encountered, for instance, whether it is assumed that the world is finite in space and time or infinite in space and time. Or, in another instance, one event may properly be called the cause of another event, but such a concept cannot be used to assert that something (a First Cause) is the cause of the universe as a whole. Of this "whole" one has, and can have, no experience. The main line of agnostic argument in the nineteenth century followed Kant closely in his criticism of cosmological reasoning, although many agnostic writers were not thoroughgoing Kantians. Nor did they have to be Humeans to have their metaphysical assurance called in question by David Hume's famous (or notorious) criticism of speculation in An Enquiry concerning Human Understanding (1748): "If we take in our hand any volume; of divinity or school metaphysics, for instance; let us ask, Does it contain any abstract reasoning concerning quantity or number? No. Does it contain any experimental reasoning concerning matter of fact and existence? No. Commit it then to the flames: for it can contain nothing but sophistry and illusion."

A person who calls himself agnostic commonly judges that he cannot have both agnosticism and, say, Christian belief. Yet the main positions of nineteenth-century agnosticism were in fact worked out and held by "religious agnostics," writers who argued that a very high degree of ignorance concerning the deity was nonetheless compatible with a religious commitment of some kind. In fact, if not in name, this view was also found in the twentieth century; it is essentially the view of those who disclaim metaphysical knowledge of God, but yet stake all upon "faith," "authority," or Christianity as a practical way of life. Kant may again provide the archetypal model: Having denied that theoretical reasoning could furnish arguments for the existence of God, he nevertheless claimed that God had to be "postulated" in order to make sense of moral experience.

In his most influential article, "Philosophy of the Unconditioned" (Edinburgh Review, 1829), Sir William Hamilton tersely introduced themes that were to be developed, refined, and repudiated by writer after writer to the end of the century and well beyond. "The mind," he wrote, "can … know only the limited, and the conditionally limited." To attempt to think the unconditioned or absolute is to think away "those very conditions under which thought itself is realized." "Loath to admit that our science is at best the reflection of a reality we cannot know, we strive to penetrate to existence in itself; … But, like Ixion, we embrace a cloud for a divinity."

H. L. Mansel, in his Bampton Lectures, The Limits of Religious Thought (1858), tried to show in detail that alleged knowledge of the Absolute is self-contradictory at many points. One attributes personal qualities to God, for instance, and yet one cannot think through the notion of personality without the idea of limitation; thought must be distinguished from thinker, and so on. But limitationPage 93  |  Top of Article is incompatible with infinite and absolute deity. The conclusion, however, is not a total religious skepticism. For although speculation about the divine nature is a vain attempt to escape the inescapable conditions of human thought, yet through the "feeling of dependence" and in moral conviction faith may still operate where speculative reason cannot.

Herbert Spencer in his First Principles (1862) accepted this picture of a limited human reason, aware of its limits and yet (in his view) aware also that those limits are decidedly not the limits of the real. Science and religion could, in fact, be reconciled by realizing that each of them testifies to a mystery, to an inscrutable Absolute, quite beyond the frontiers of knowledge or conception but yet not mere negation or nothingness.

The sources of nineteenth-century agnosticism—particularly the agnosticism of those who abandoned organized religion—were, however, more numerous and complex than has been indicated so far. It is rare indeed that a single line of philosophical argument produces by itself either religious conviction or disillusionment. At least three additional sources should be mentioned.

First, a growing mass of data and theory supplied by the physical sciences was prima facie at variance with biblical history and cosmology. There was the new time scale of geology, the impersonal and amoral Darwinian evolutionary theory, and the radical textual, historical criticism of the Bible itself.

Second, once the strong initial resistance to systematic and searching criticism of Christian teaching had been overcome, it was possible to express openly a good many moral misgivings about the Christian conception of God and his governance of the world. J. S. Mill declared it was impossible for a thoughtful person to ascribe "absolute perfection to the author and ruler of so clumsily made and capriciously governed a creation as this planet" (Three Essays on Religion, 1874). He found "moral difficulties" also in "the recognition … of the object of highest worship, in a being who could make a Hell" and create creatures whom he foreknew to be destined to suffer in it eternally. No less morally repugnant to many writers was the insistence of the orthodox that their dogmas required sheer unswerving acceptance, and that breakdowns in argument or intelligibility were simply occasions for the exercise of an intensified faith. T. H. Huxley was forthright. In "Agnosticism and Christianity" (1889) he wrote, "I, and many other Agnostics, believe that faith, in this sense, is an abomination." In "Agnosticism" (1889) he said, "I verily believe that the great good which has been effected … by Christianity has been largely counteracted by the pestilent doctrine … that honest disbelief in their more or less astonishing creeds is a moral offence, indeed a sin of the deepest dye."

Third, the same authors were vehemently critical of the standards of evidence and reasoning normal in theology, and contrasted them with the severe, rigorous, and dispassionate criteria of the sciences. To Mill, "The whole of the prevalent metaphysics of the present century is one tissue of suborned evidence in favour of religion." If one considers the nature of the world as one actually observes it, the very most one could dare to hazard is the existence of a good but finite deity; and Mill put forward even this possibility with a characteristically agnostic tentativeness. For Huxley agnosticism was "not a creed but a method, the essence of which lies in the rigorous application of a single principle": Reason should be followed "as far as it can take you," but undemonstrable conclusions should not be treated as if they were certain. "One may suspect," he said, "that a little more critical discrimination would have enlarged the Apocrypha not inconsiderably." In a similar vein, Leslie Stephen protested against theologians who ventured to define "the nature of God Almighty with an accuracy from which modest naturalists would shrink in describing the genesis of a black beetle" (An Agnostic's Apology, 1893).

It is not the purpose here to estimate how far theologians remedied, or could ever remedy, the deficiencies in their arguments that offended their agnostic critics. Some permanently valuable lessons can be learned, however, from the course of the controversies. An obvious one is the odd instability or ambiguity of certain agnostic positions. Let us suppose—as did many of the writers just quoted—that one ceases to find convincing the arguments for the existence of a deity. Experience, one now judges, is limited to the observable world; and reason, although it may lay bare the conditions and presuppositions of that experience, cannot extend our experience of what is. A religiously minded person, in this situation, is tempted to divide reality into the knowable and the unknowable and to attribute to the latter many of the lineaments of deity. Thus, "negative theology" and a religiously toned agnosticism can be the closest of relatives. No sweeping philosophical criticism can demonstrate that all such positions are untenable or involve a cryptotheism; each case must be scrutinized individually. Certain religious attitudes toward the unknown or unknowable—attitudes, for example, of wonderment and awe—can be perfectly appropriate and invulnerable to criticism, whereas others—such as the expectation of personal encounter with the unknown—are obviouslyPage 94  |  Top of Article most vulnerable. One can turn to history for some examples.

In 1896 James Ward delivered his Gifford Lectures, Naturalism and Agnosticism (published in 1899), at Aberdeen University. These contained a vigorous attack on the basic presuppositions of the Hamilton-Mansel-Spencer approach. The sciences, Ward said, do not form a whole that floats in a surrounding "nescience." The world we know does not consist of "appearance" concealing an "ultimate reality" that lies behind or beyond it. In any case, nescience is nescience. "Where nescience is absolute, nothing can be said; neither that there is more to know nor that there is not." Spencer and like-minded writers had, however, said a good many mysterious things about their Absolute, things that, by their own account, were strictly unsayable.

R. Flint (Agnosticism, Croall lectures, 1887–1888, published in 1903) also denounced the equivocations (as he saw them) of a religious agnosticism. "All that the mind can do on the side of the Unknowable is to play at make-believe, to feign faith, to worship nothingness." "Call your doubts mysteries," said Stephen, satirizing the complacent, "and they won't disturb you any longer."

Is it possible for a reflective person to be an agnostic in the present time? Logical positivists have answered "No." In Language, Truth and Logic (1936), A. J. Ayer claimed that since "all utterances about the nature of God are nonsensical," the agnostic's statements about God are no less nonsensical than the theist's. Both assume, wrongly, that "the question whether a transcendent God exists is a genuine question." According to positivism and postpositivist logical analysis, the theological problem is not a problem of evidence and argument, but a problem of meaningfulness. If "God" is a meaningless word, the sentence "Perhaps God does not exist" is also meaningless.

In stating the situation thus, positivism was dramatically drawing attention to what it believed to be distinctive in its approach, but it simultaneously obscured some important lines of continuity with the earlier debate on agnosticism. Before the nineteenth century had ended, Flint had written, in criticism of Hamilton, "Credo quia absurdum can be the only appropriate motto of a philosopher who holds that we may believe in a God the very idea of whom we can perceive to be self-contradictory." The possibility of internal illogicality in the very notion of deity, the risk of the absurd and nonsensical, were well enough recognized. Spencer, wrestling with the problems of the world's origin and beginning, said that the questions here are not questions of credibility but of conceivability. Notions such as self-existence and creation by an external agency "involve symbolic conceptions of the illegitimate and illusive kind." The logical positivist tethered his theory of meaning to the demands for observational verification and falsification of our claims about existents. Compare Spencer once more, writing in 1899: "Intellect being framed simply by and for converse with phenomena, involves us in nonsense when we try to use it for anything beyond phenomena." It must, of course, be added that the positivists and later analysts carried out their austere program with far greater thoroughness and consistency than did their predecessors. But the lines of continuity are there; and they are—once more—those same lines that reach back to Kant's "Transcendental Dialectic" and to David Hume. They justify the use of "atheist" to describe one who rejects the performances and attitudes of religion on the grounds that talk about God is unverifiable talk, or that the concept God contains inner illogicalities.

But is there still room for agnosticism as undogmatic dubiety or ignorance about the existence of God? A case for saying that there is still room can be made on the following lines. Where one gives an account of an expression in our language, and where that expression is one that refers to an existent of some kind, one needs to provide not only a set of rules for the use of the expression, but also an indication of how the referring is to be done—through direct pointing, perhaps, or through giving instructions for an indirect method of identifying the entity. Can this be done in the case of God? Pointing, clearly, is inappropriate, God being no finite object in the world. The theologian may suggest a number of options at this point. He may say: God can be identified as that being upon whom the world can be felt as utterly dependent, who is the completion of its incompletenesses, whose presence is faintly adumbrated in experience of the awesome and the numinous. Clear direction-giving has here broken down; the theologian may well admit that his language is less descriptive or argumentative than obliquely evocative. Does this language succeed in establishing that statements about God have a reference? To persons susceptible to religious experience but at the same time logically and critically alert, it may seem just barely to succeed, or it may seem just barely to fail. Some may even oscillate uneasily between these alternatives without finding a definite procedure of decision to help them discriminate once for all. A person in this last category is surely an agnostic. His agnosticism takes full account of current linguistic criticisms of religion; it is in the course of his reflections upon meaning that he sees the necessity of relating the linguistic to thePage 95  |  Top of Article extralinguistic, and his answers to this problem, the problem of reference, plunge him into the deepest uncertainty.

The temper of mind just outlined, with all its inner turbulence and anxiety, is probably the most creatively fruitful of the many varieties of agnosticism. Where there is no temptation to believe, there can be little philosophical interest in not believing. Where there has been little or no religious experience, no sense of the haunting strangeness that makes the believer wittingly violate language and logic to express it, there can be little incentive to explore minutely the possible interpretations—theistic, pantheistic, naturalistic—of that experience. As a matter of history, agnostics of this temper are to be found far more rarely today than at the height of the agnosticism controversy a century ago. For the great writers of that controversy were in most cases brought up within the Christian faith, had identified themselves with it, and subsequently suffered a bewildering disorientation. Yet, if one is to take seriously today the problems of philosophical theology, there must be some suspension of disbelief, at least an imaginative venture, in order to see why the believer feels compelled to use the extraordinary language he does use. He knows well enough that it is extraordinary; but he deems that it is ordinary language that is found wanting, and not his experiences and the interpretations he puts upon them. The agnostic knows that sometimes ordinary language needs to be violated, as a poet often violates it. He knows also that to disturb our linguistic apparatus in so radical a way can obscure some movements of thought of a very questionable (or downright invalid) logic. Has this happened in the particular case of theism? Searching in this obscurity, the agnostic reports that he cannot tell. For the health of philosophy and theology, it is well that he should continue to search.



Hume, David. Dialogues concerning Natural Religion. London: Robinson, 1779.

Hume, David. Enquiry concerning Human Understanding. London: A. Millar, 1748. Especially sections X and XI.

Kant, Immanuel. Kritik der reinen Vernunft (1781). In Berlin Academy, Complete Works. Berlin, 1902–1955.


Flint, R. Agnosticism. Edinburgh and London, 1903.

Hamilton, Sir William. "Philosophy of the Unconditioned." Edinburgh Review (October 1829).

Huxley, T. H. "Agnosticism and Christianity" (1889). In his Collected Essays, Vol. V. London: Macmillan, 1894. Entire volume is relevant.

Mansel, H. L. The Limits of Religious Thought. London, 1858.

Mill, J. S. Three Essays on Religion. London: Longmans, Green, Reader, and Dyer, 1874.

Spencer, Herbert. First Principles. London: Williams and Norgate, 1862.

Stephen, Leslie. An Agnostic's Apology. London: Smith Elder, 1893. First published as an essay in 1876.

Ward, James. Naturalism and Agnosticism. London: Macmillan, 1899; 2nd ed., with alterations, 1903.


Armstrong, R. A. Agnosticism and Theism in the Nineteenth Century. London: Green, 1905.

Britton, K. John Stuart Mill. London: Penguin, 1953.

Burtt, E. A. Types of Religious Philosophy. New York: Harper, 1951.

Collins, J. God in Modern Philosophy. Chicago: Regnery, 1959.

Garrigou-Lagrange, R. Dieu, son existence et sa nature; solution thomiste des antinomies agnostiques. Paris, 1915.

Hájek, Alan. "Agnosticism Meets Bayesianism." Analysis 58 (3) (1998): 199–206.

Kenny, Anthony. Faith and Reason. New York: Columbia University Press, 1983.

Lightman, Bernard. The Origins of Agnosticism: Victorian Unbelief and the Limits of Knowledge. Baltimore: Johns Hopkins University Press, 1987.

Marty, Martin E. Varieties of Unbelief. New York: Holt Rinehart Winston, 1964.

McGrath, P. J. "Atheism or Agnosticism." Analysis 47 (1987): 54–57.

Morris, Thomas. "Agnosticism." Analysis 45 (1985): 219–224.

Oppy, Graham. "Weak Agnosticism Defended." International Journal for Philosophy of Religion 36 (1994): 147–167.

Packe, M. St. John. John Stuart Mill. London: Secker & Warburg, 1954. Biography.

Passmore, J. A Hundred Years of Philosophy. London: Duckworth, 1957. Especially Ch. 2.

Pyle, Andrew, ed. Agnosticism: Contemporary Responses to Spencer and Huxley. Bristol, U.K.: Thoemmes, 1995.

Stephen, Leslie. English Thought in the Eighteenth Century. London: Smith, Elder & Company, 1876.

van Fraassen, Bas C. "The Agnostic Subtly Probabilified." Analysis 58 (3) (1998): 212–220.

Willey, Basil. Nineteenth Century Studies. New York: Columbia University Press, 1949.

Ronald W. Hepburn (1967)

Bibliography updated by Christian B. Miller (2005)

Source Citation

Source Citation   

Gale Document Number: GALE|CX3446800032