The notion of the sacred or the numinous as a category for understanding religion was substantially launched by Rudolf Otto (1869-1937) in his classic Idea of the Holy (Das Heilige) in 1917. Otto, indeed, coined the term 'numinous,' which has now become part of common usage. Otto's influence on thought about religion extends from C.G. Jung (1875-1861) to the 'Chicago School' of history of religion founded by Mircea Eliade (1907-1986). On the other hand, Otto's influence on the philosophy of religion has been less strong, perhaps because he was professionally more of a theologian (not rigorous enough for philosophers) and is too easily misunderstood and dismissed as describing some kind of mysticism. Even in the history of religion, Otto's own analysis often does not persuade because of his clear preferences for Christianity and his devaluation of religions that do not measure up to Christian paradigms.

The history of Rudolf Otto's theory of the sacred begins, however, more than a century earlier, with the great philosopher Immanuel Kant (1724-1804). Otto's own confidence in his ability to talk about God has its origin in Kant's own theory about the basis of the concept of God in human reason. In the Critique of Pure Reason (1781) Kant had reworked the traditional distinction between the immanent (within the world) and the transcendent (outside the world) by distinguishing between phenomena and things-in-themselves. 'Phenomena' are how it is that objects appear in our own conscious minds. We do not have access to the world outside of the experiences we enjoy through our own consciousness, and Kant believed that consciousness itself, or the possibility of conscious experience, imposes certain conditions on the manner in which phenomenal objects appear to us. Among those conditions are the forms of space and time and the abstract forms of connections between events and objects such as the concept of substance and the relation between cause and effect. David Hume (1711-1776) had challenged philosophers to show why it is that we believe in principles such as the one that every event must have a cause. Kant's answer, then, was that the mind itself constructs a phenomenal reality according to just such a rule.


Eliade, of course, is coming out of the kind of theory found in Otto, which is only minimally different from the philosophy of religion in Nelson and Fries; and Nelson and Fries are in many ways still four-square Kantians. Otto's Das Heilige is famous for dealing with the 'irrational' in religion, but he himself cautions that this is only a supplement to the rational treatment of religion that he had provided in Die Kant-Friesische Religions-Philosophie. Consideration of the 'Numen ineffabile' only follows upon the 'devoted assiduous and serious study' of 'the 'Ratio aeterna'.' What the rational side of religion largely contributes is morality, and there is little attenuation of the power of the Moral Law in Fries or Nelson or Otto from Kant. Otto himself establishes a hierarchy of religions according to the degree to which they have become morally 'schematized': the basis of his clear preference for Christianity as the supreme religion. It is possible for us to argue which religion has the better moral conceptions and to disagree with Otto about that, but we may not disagree with him in principle to suppose that some religions do have superior moral conceptions to others.

On the other hand, supposing that some religions are morally superior to others implies that others are inferior. Given some religion that is not 'morally schematized' very well, we can then ask what makes it a religion in the first place and what kind of meaning it had for its adherents, even if to us it is morally intolerable. That is the question that particularly interested Otto (and Eliade). This does not imply, as accusations of Eliade would imply, that the morally un-schematized religions are superior to the morally sophisticated ones. That would be standing the whole theory on its head.


The purpose here is not a general review of Rem B. Edwards' Reason and Religion, An Introduction to the Philosophy of Religion. Instead, this book is merely taken as a representative example of how Rudolf Otto and his theory of numinosity are treated in a standard philosophy of religion book, a book that Edwards himself says was developed over eight years of teaching undergraduate courses in philosophy of religion [p. v].

Edwards' principal discussion of Otto occurs in his chapter, 'Religious Language and Experience: MYSTICISM' [p. 289]. Otto himself is introduced in a subsection, 'The Object and Vocabulary of Mysticism Experience' [p. 311], under the heading, 'Rudolf Otto's Idea of the Holy' [p. 312]. Edwards says that the 'holy' is 'a thing of fascinating mystery and terrifying magnitude,' and that 'when we speak of fascination and terror, we are speaking about the psychological response of the self that has the experience, but mystics insist that it is the object of experience that is holy...' [p. 312].

It is not clear from this alone why Otto is even included in the section. Is the quoted description Otto's definition or characterization of mystical experience? Well, no. It is Otto's description of the experience of anything that is numinous, which can mean God in a direct, mystical experience of God, but more generally means anything that is holy or uncanny. Edwards trips up on his unfamiliarity with Otto's Friesian epistemology, and exhibits the confusion that necessarily occurs when Otto is assumed only to be talking about God (and about mystical experience) and not about ordinary religious objects and practices. The 'psychological response' of a person to the numinous or the holy is necessarily the object of Friesian 'Critique,' but the objects, whether of ordinary experience or of religious, even mystical, experience, can only be described, in Otto's Friesian terms, by the rational theory of objects, whether immanent (Wissen, 'knowledge,' Friesian epistemology of experience and science) or transcendent (Glaube, 'belief,' Friesian metaphysics, a form of Kantian 'transcendental idealism'). Later, Edwards says:

The mystical experience is more than feeling: It is a kind of immediate knowledge by acquaintance with a supreme, self-sufficient religious object. To be sure, the feeling is also there, but it is not the only element... Otto's main emphasis, however, was on the holy object and not on the feelings of the mystic subject. [p. 313]

Unfortunately, the use of the term 'immediate knowledge' must imply much more for any reader of this text than the idea of Ahndung, 'intimation,' possibly could for Otto himself. Ahndung is more than feeling (Kent Richter translated it 'aesthetic sense'), but it is not 'immediate knowledge' of transcendent objects the way we have immediate knowledge of the ordinary objects of experience; nor is it the reason why we might say that a mystical experience is an experience of God, since the concept of God comes from rational Glaube, in four-square Kantian terms as an Idea, not from the experience of Ahndung. Otto's emphasis, therefore, is not on the 'holy object,' but on the special qualia that religious sense or experience attributes to its objects, whether mundane and superlunary.

Edwards is unaware of the Kant-Friesian background of virtual rationalistic natural theology in Otto, even though Otto warned at the very beginning of the Idea of the Holy: 'And I feel that no one ought to concern himself with the 'Numen ineffabile' who has not already devoted assiduous and serious study to the 'Ratio aeterna'.' In the rest of Edwards' treatment, there can be no sense, much less a critical sense, of how any religion, or religious experience, let alone a mystical experience, will be a tangle of rational and non-rational elements.

Since Otto is treated strictly in a mystical context, the rest of Edwards' book contains no hint that Otto might actually have a general philosophy of religion or that 'numinosity' is about something in ordinary religion and religious experience. Thus, Edwards considers the character of God as holy [p. 185] and considers 'sacred objects and places' [p. 33] as one of the defining traits for religion itself but ignores Otto in relation to these questions, even though Otto may be said to have contributed an all but unique treatment of holiness as such, supplemented by later examinations like Eliade's The Sacred and the Profane. Eliade, in that book, with its theory of sacred space and sacred time, is clearly not talking about mysticism.

Indeed, the brevity of Edwards' treatment of 'sacred objects and places,' and his sense of the limitations of the 'sacred' as defining for religion itself, both betray the incompleteness of his conception [p. 33]. If the 'sacred' necessarily contains a non-rational element and a transcendent (in the Kantian sense) reference, then the sacred objects of Leninism or Maoism, the Gods of rationalists like Aristotle or Spinoza, the religion substitutes of rationalistic 'humanists,' and the completely metaphorical application of the term to 'success, wealth, golf, and fishing' are all easily distinguishable from holy things in genuine religions. In confusing all these applications of 'sacred,' or in confusing religion with rationalistic or political substitutes, Edwards also must overlook the whole issue of ritual pollution, desecration, purification, and so the ritual purity required on the proper occasions by all ancient, and most modern, religions. Since mystics rarely have mystical experiences of pollution (unless, I suppose, they see demons or Satan), this indicates how unnaturally restricted the 'sacred' is in a mystical context. [Cf. 'The New Friesian Theory of Religious Value'.]

A philosopher of religion like Edwards will be unlikely to simply accept the metaphysics and epistemology behind Rudolf Otto's 'idea of the holy'; but what we really find is that he appears unaware of what they are and misconstrues the kind of theory that Otto offers. This loss and misunderstanding is unfortunate for the whole philosophical treatment of religion, since the Kant-Friesian tradition is one of the few ways available in modern philosophy for a non-reductionistic evaluation of religion which does not at the same time simply accept the authority of one particular religious practice or revelation -- Otto's own theism can be easily defused if the Kantian approach to the concept of God as a 'postulate of practical reason' (Friesian Glaube) is seen to fail and we are left with the Antinomies, which sound like nothing so much as the Buddhist fourfold negation (where we might say, 'there is neither a God, nor not a God, nor both, nor neither'). To both accept the plurality of religions without trivializing them, and without reducing them to popularized presentations of rational, moral, political, or aesthetic truths, is a challenge that is really only met with the conceptual tools provided by Kant, Fries, and Otto.