go back to Zen Teachings: Koan studies


The Koan as Ritual Performance
Hakuin's Daruma

Barry Stephenson
Journal of the American Academy of Religion June 2005, Vol. 73, No. 2, pp. 475–496
© The Author 2005

Study of the Zen kōan tradition has been dominated by psychological and textual methods and theories. But kōans not only constitute a literary genre, they are part of the liturgical order of Zen Buddhism. In this article I argue that psychological and textual approaches offer a limited understanding of the ritual and performative contexts of kōan practice and, drawing on ritual and performance theory, develop an alternative approach to the kōan tradition.

As Buddhism has made the trek from Asia to Europe and North America, understandings of it have been shaped in part by the tools and lenses prevalent in the academy. The reception and interpretation of the Zen kōan tradition has been dominated by certain theoretical and methodological trends. These trends may be conveniently divided into two camps, one populated by experientialists, the other by textualists. Experientialists approach kōans from the perspectives of psychology, mysticism, and religious experience. Textualists employ philological, historical-critical, and hermeneutical methods and theories in dealing with kōan collections. Between the two groups there are methodological and interpretive tensions, often centered on the question of language, and put in a nutshell by Steven Heine, a supporter of the textualist camp. “What,” Heine asks, “is the nature and function of a kōan? Is it a psychological device that defeats language or a literary tool that fosters textuality?” (1994: 4). Similarly, D. T. Suzuki, representative of the experientialist camp, states that “there are two approaches to satori: the one may be termed metaphysical, philosophical or intellectual, and the other psychological or conative” (1972: 89). Heine’s questions and Suzuki’s statement indicate methodological self-consciousness within the field of kōans sscholarship; each approach to [p476] the kōan directs the gaze in a certain way, unearthing features and insights the other is incapable of seeing and providing distance from which to generate critique.

My aim in this article is to begin fashioning a third standpoint from which kōans study may proceed. Kōans may be understood as psycholinguistic riddles capable of generating a certain type of experience or as a literary genre with a complex history embodying centuries of religious and philosophic discourse. But kōans, since the development of ritualized [1] forms of kōan practice in the early Sung period, have held a prominent place in Zen’s liturgical order. Kōans are ritual performances — they are staged, worked on, enacted, watched, and judged. Psychological and hermeneutical theories and methods offer limited tools for dealing with the kōan conceived as a rite. Ritual and performance theory, an interdisciplinary approach exploring connections between everyday behavior, dramatic performance, and ritual action, and associated with such scholars as Erving Goffman (sociology), Victor Turner (anthropology), Richard Schechner (theatre studies), Ronald Grimes (religious studies), and Roy Rappaport (anthropology) offers another framework, another set of ideas, and terms from which to study the Zen kōan and kōans scholarship. Ritual and performance theory is not a unified school of thought; of the scholars mentioned, only Schechner explicitly refers to his work as “performance theory.” But scholars interested in the study of ritual have come to realize that ritual action and performative action share features that make speaking of them in the same breath both reasonable and methodologically fecund.

Although some scholars have spoken to the liturgical context of kōans , there has been no attempt to frame the discussion of kōan practice in terms of ritual and performance theory. My direction in this article is primarily theoretical, bringing the perspectives of another camp to the study of the Zen kōan tradition. To parachute in questions and viewpoints from another camp to the study of the kōan is to follow Jonathan Z. Smith’s method of juxtaposition, in which various interpretations, ideas, and approaches are placed side by side and compared in the spirit of play, fostering interplay and difference rather than resolution and sameness; [p477] treating kōans as ritual performance emphasizes the embodied and enacted dimensions of the Zen kōan . Following a brief summary of the approach to kōans taken by the experientialist and textualist camps, I move on to show how these models offer a limited understanding of ritual. Lastly I develop a performance approach to the study of kōans .

Mind And Text

A psychological understanding of kōans has long dominated both scholarship and popular accounts of Zen. The Harper Collins Dictionary of Religion , for example, describes kōans as follows: “Devised as pedagogical tools, kōans pose paradoxical problems, the non-intellectual, non-conceptual resolution of which represents a spiritual breakthrough.” The absurd behaviors and illogical, paradoxical statements typical of many kōans are said to pose an intractable problem that frustrates discursive thought. The application of reason to the solution of a kōan brings the mind to the boiling point. In this state the monk or student is open to the method of direct pointing, and a certain phrase or action from the master can jar loose one’s intuitive powers of perception. Zen, chant the experientialists, is a special transmission outside of scripture, with no dependence on words and letters. According to D. T. Suzuki, through kōans study “the searching mind is vexed to the extreme as its fruitless strivings go on, but when it is brought to an apex it breaks or explodes and the whole structure of consciousness assumes an entirely different aspect” (1974: 61). Experientialist C. G. Jung, in his “Foreword” to Suzuki’s Introduction to Zen Buddhism , speaks of the “opacity” of the “strange conception called satori.” “Kōans,” claims Jung, “aim for the complete destruction of the rational intellect . . .” (1969: 539). Alan Watts, another important figure in the experientialist camp, claims “the basic position of Zen is that it has nothing to say, nothing to teach” (163).

The psychological and mystical model of Zen and the Zen kōan offered in the studies of experientialists such as Suzuki, Jung, and Watts [2] have been so influential that the word “kōan ” has now become colloquial English. The Longman Dictionary of the English Language (1991) defines “kōan ” as a “paradox to be meditated on that is used to train Zen Buddhist monks to abandon reason and develop intuition in order to gain enlightenment.” As a last example of the experientialist model of the kōan , consider the description offered by another prominent figure in the transmission [p478] of Zen to Europe and North America, Ruth Fuller Sasaki: “Kōans study is a unique method of religious practice which has as its aim the bringing of the student to direct, intuitive realization of reality without recourse to the mediation of words or concepts” (1965: x).

Though there is historical precedent for a psychological understanding of kōans [3] textualists are crying foul and have in recent years have offered important critiques of a strictly psychological approach to kōans. Kōan texts and commentaries, argue textualists, are filled with Buddhist ideas such as impermanence, non-duality, emptiness, and no-self. The textualists certainly have a point. Jung, for example, cites the following kōan as an example of the “opacity” of the “strange conception called satori”:

A monk once went to Gensha, and wanted to learn where the entrance to the path of truth was. Gensha asked him, “Do you hear the murmuring of the brook?” “Yes, I hear it,” answered the monk. “There is the entrance,” the master instructed him. (1969: 539)

But this kōan is less opaque in light of tropological analysis of water imagery in Buddhist texts, water being a common motif for the flow and impermanence of life. “Discourse analysis,” writes Heine, “reorients the issue of the effectiveness of kōans as a religious symbol in terms of textuality and rhetorical strategies” (1994: xiv). For Heine kōans are “rhetorical devices or literary symbols that utilize fully the resources of language in highly creative and original ways indicating that verbal expression supports rather than obstructs the attainment of Zen enlightenment” (1994: 4). Kōans scholarship, partly in reaction to several decades of psychological and mystical interpretations of kōans, is increasingly drawing on hermeneutical methods and theories. A key shortcoming in the first wave of contemporary kōans scholarship, argues Heine, is that these studies failed to take into account “the literary traditions of Chinese and early Japanese religion . . . which tend to stress the efficacy of poetic metaphor or scriptural recitation in disclosing spiritual attainment” (1990: 358–359). A recent collection of essays published under the title The Kōan : Texts and Contexts in Zen Buddhism , edited by Steven Heine and Dale Wright, “employ a variety of methodological perspectives, such as textual analysis and literary criticism, philosophical hermeneutics and phenomenology, and social historical and history of religions approaches” (6) and is representative of the new trend in kōans scholarship.

[p479] But battles between experientialists and textualists are not really what I am interested in here. I want to open up a different front altogether and put the experientialists and the textualists in the dock on the count of ritual abuse. Then, drawing on ritual and performance theory, I will suggest another way to reorient the issue of the effectiveness of kōans .

Ritual Abuse

The body is conspicuously absent in D. T. Suzuki’s writings on kōans : “intellectual experience,” “intellectual quest,” “metaphysical quest,” “vexed mind,” “mental impasse”— these are the watchwords of Suzuki’s heady Zen (1974: 22–58). With satori, Suzuki writes, “the whole structure of consciousness assumes an entirely different aspect” (1974: 61) and “one’s mental construction goes through a complete change” (1974: 24). Suzuki plays this consciousness altering experience off against ritual on the one hand and the textualist’s concern for learning on the other: “[Zen] has faithfully transmitted and upheld its tradition by upholding satori against ritualism and erudition” (1974: 23). In Suzuki’s hands Zen is iconoclastic and antiritualistic.

Suzuki’s elevation of mystical experience at the expense of “ritualism” is well suited to Jung’s psychological approach to Zen, and Jung clearly did not appreciate formal ritual practice. “To the extent that Zen is a movement,” he writes, “collective forms have arisen in the course of centuries. . . . But these concern externals only” (1969: 548). And, commenting on Eugen Herrigel’s Zen and the Art of Archery , Jung writes,

An even greater distance, seems to separate the satori of meaningful kōans from Zen archery . . . It is just pathetic to see a man like Herrigel acquiring the art of Zen archery, a non-essential if there ever was one, with the utmost devotion — but, thank God, it has obviously nothing to do with the inner life of man! (1973: 601–602)

What is important for the experientialist is the inner experience obscured by surface forms. Experientialists typically set up a dualism between form and emptiness (or ritual and mysticism) that is foreign to Zen. The central theme of the Heart Sutra, chanted daily in Zen practice, is that emptiness and form entail one another; satori presupposes ritual, and we do violence to the form-emptiness identity if we view Zen ritual as a means or device to attain satori. Textualists often make the point that there is no pure experience, all experience being situated and mediated in and through language. The same may be said of the body: all experience is bodily situated and mediated.

[ p480 ] If experientialists are prone to reject ritual in favor of pure, formless experience, textualists offer a limited understanding of ritual. Heine: “[Zen has] created a cluster of linguistic and visual symbols as well as symbolic actions—or discourse—to communicate its vision of spiritual fulfillment” (1994: 15). Here Heine suggests ritual action is a symbolic discourse, a kind of language that communicates information. The fundamental metaphor employed by textualists in dealing with performative action, is, not surprisingly, action is like a text ; the body has a language, its grammar and meanings can be read. The discipline of religious studies is dominated by texts and textualism, and hence the study of religious phenomena tends to be dominated by linguistic metaphors: all things become “texts” to be “read.” When applied to ritual, textual metaphors often turn rites into symbolic “structures,” the architectural metaphor of structure connoting such qualities as solidity, stability, and immovability. The metaphor of structure also leads to representational thinking, the structure of a particular rite pointing outside of itself to social relations, metaphysical principles, or ethical systems.

Ronald Grimes suggests that when applied to the study of Zen the notion of ritual as symbolic action may miss the mark because many Zen gestures and forms “do not mean, refer to, or point to anything” (1995: 106). In the writings of those scholars whom I am considering under the rubric of “performance theory,” meaning is not located in a structure or a pattern that ritual action both derives from and points to, but is rather acted out or performed; the notion of performance emphasizes the processual, dynamic qualities of ritual, often focusing attention on the play of overt action rather than what that action symbolizes or represents. Fritz Staal goes so far as to define ritual as “pure activity, without meaning or goal” (1979: 14). “To performing ritualists, rituals are like dance, of which Isadora Duncan said: ‘If I could tell you what it meant there would be no point in dancing it’ ” (quoted in Staal 1989: 115–116). Staal tells us that participants are not really interested in the deep, symbolic meanings of ritual but with surface properties, with matters of staging, performance, and doing. Perhaps this observation is especially important to the study of Zen ritual where sitting, as we are told over and over again, is just sitting. Certainly ritual symbols may refer, but semantic notions of ritual set us looking outside of ritual for meaning (see Grimes 1993: 19–22). If the ruling metaphor is “body as text” we will be enticed into finding out what a gesture or act points to or means. This is perfectly valid, but ritual as performance generates other questions—questions about style, fittingness, and efficacy. Ritual need not be treated as a symbolic form.

[p481] Ritual Performance

The Zen kōan tradition constitutes a unique literary genre, but kōans are as much part of Zen liturgy and practice as Zen literature. As John Daido Loori, abbot of Zen Mountain Monastery, writes, “Zen teachers don’t routinely analyze kōans in terms of their psychology, philosophy, structure, or dynamics. We do kōans ” (xvii). It is this doing of a kōan that is of interest here — the enacted, embodied dimension of kōan practice. [4] The origins of kōan training are in the Sung era of Chinese Ch’an.

During the Sung period kōan collections were distilled from the encounter dialogues, parables, anecdotes, biographies, and sermons recorded during the preceding T’ang dynasty. The word kōan originally meant something like “public document” or “public case.” Each kōan presents the monk or student with a case to which a past master had given a precedent-setting solution or answer or, perhaps more accurately, response. During the Sung period kōans were heavily ritualized and formalized, laying the ground for the rise of kōan rites. In Japan the origins of kōan practice are found in the medieval period, roughly from 1300 to 1600 C.E. The medieval kirigami documents (literally “paper strips”) of Japanese Zen include funeral and other ceremonial procedures, lineage charts, and prayers. These documents also include “secret curriculum manuals of acceptable responses to kōans ” and mark the rise of “kōan Zen” in Japan, “a kind of Zen training emphasizing the study of traditional kōan cases” (Rikizan: 238). The medieval period in Japan was, like the Sung era in China, a time of intense ritualizing.

If Heine would have us attend to the complex literary traditions of China and Japan, so too must attention be given to the performative and aesthetic traditions in understanding and interpreting kōans. The rise of kōan Zen in Japan coincides, for example, with the development of N o theatre through the fourteenth and fifteenth century under the leadership of the distinguished performer-playwrights Kannami and his son Zeami. Kirigami documents also include “forms of communication in other medieval aesthetic traditions such as the tea ceremony,” which were “based on an intense apprenticing, master-disciple relationship” (Heine and Wright: 10). Any full account of the Zen kōan tradition must attend to its origins in the aesthetic and performative ethos of the Sung era and medieval Japan. “In a Japanese Rinzai Zen monastery not only are ritual and ceremony indeed important parts of Zen training . . . all [p482] aspects of monastic life—sleeping, eating, running, chewing one’s food, reaching for tea cup—are ritualized and choreographed” (Hori: 250). The meaning and significance of the kōan tradition are located as much in performativity as in textuality or experience.

Heine suggests a progression in the development of the kōan tradition: spontaneous oral exchanges were collected in texts, which were then ritualized and performed as part of Zen training (1990: 360–361).

Originally, in the early T’ang, the transmission dialogues were apparently truly spontaneous occasions of profound existential encounter [but] . . . during the late T’ang and early Sung . . . the formalized, catechistic kōan began to represent a shift from originality and creativity to the performance and ritualistic quality in terms of how the disciple delivered to the master in ceremonial fashion an already known response to a paradigmatic case. (Heine 1994: 47–48)

John McRae challenges this developmental view of kōan as devolution from word to text to rite. “Ch’an encounter dialogues derived not (or perhaps not solely) out of spontaneous oral exchanges but rather (perhaps only in part) out of ritualized exchanges.” The idea of spontaneity, argues McRae, is “inscribed within the heavily ritualized context of Sung dynasty Ch’an,” drawing into question the “distinction between the ‘classical’ age of T’ang dynasty Ch’an, when encounter dialogue was spontaneous, and the subsequent ritualization of dialogue with Sung dynasty Ch’an” (65). Whatever the complex historical relations between kōan as textual record of a decisive, spontaneous encounter and kōan as ritualized exchange, kōan practice during the Sung era did emerge as a kind of catechism: known responses to paradigmatic cases came to be performed as part of Zen training. The catechistic qualities of kōan training were further developed in medieval Japan. Later, Hakuin (1685–1768), as part of his efforts to revitalize monastic practice in Japan, established a graded kōan curriculum of some 200 kōans ; this curriculum forms the basis of contemporary kōan practice in Japanese Rinzai Zen.[5]

But Heine, in contrasting profundity, originality, spontaneity, and creativity with ritualistic performance in a ceremonial fashion, expresses a common conception of ritual. As Catherine Bell writes ritual is often understood “as particularly thoughtless action — routinized, habitual, [p483] obsessive, or mimetic — and therefore purely formal, secondary, and mere physical expression of logically prior ideas” (19). According to Heine for the kōan to be an effective “means of spiritual training . . . it must not be turned into a formula, conceptual crutch, or object of dependence—i.e., ritualized so that the mere repetition diminishes spontaneity” (1990: 360). The tacit assumptions here are that ritualists are pre-critical and uncreative; this fails to recognize the critical, creative, and even cognitive potential of the ritualized body. What if, contra Heine, the effectiveness of kōan practice lies precisely in its ritualized, repetitive, catechistic qualities?

The first kōan encountered in contemporary kōan practice is typically Hakuin’s One Hand: “In clapping both hands a sound is heard; what is the sound of the one hand?” (Hoffmann: 47). Hakuin’s One Hand has a formally accepted answer, as do all catechisms. [6] The term “catechism,” however, suggests rote responses to questions of doctrine or dogma; kōans are perhaps better understood as scripts or scenarios rather than as purely formulaic responses. McRae uses the image of the kōan text as a “liturgical skeleton.” The kōan literature of Ch’an, suggests McRae, “was prepared as skeletal notations upon which teachers and students could improvise” (71). McRae’s reference to improvisation is, I believe, key insight, a notion to which we shall return. A kōan text provides teacher and student with a scenario or script as the basis for ritual performance. The script for Hakuin’s One Hand takes a few minutes to read. In response to the original question the student “faces his master, takes a correct posture, and without a word, thrusts one hand forward” (Hoffman: 47). As the script proceeds the student responds to questions with quotations from poetry and other Zen texts, engaging in verbal and bodily sparring, slipping through traps set by the master. The performance is embodied throughout, the student thrusting out his hand, slapping the master’s face, shading his eyes, and looking down on the world as if standing on Mt. Fuji, imitating sounds he hears, bowing, and so on. Although this kōans script is relatively short, it may take months to develop an adequate performance.

Michael Mohr offers the image of the kōan as a “screen, on which students can focus their mind, [serving] as a surface onto which to project their understanding” (246). But this image is very cerebral; given the embodied exchange that takes place during dokusan , the kōan may be better considered as a stage where a play or performance is conducted. The image of projecting understanding does not match the physicality of the actual encounter between roshi and student. Victor Turner, with his [p484] penchant for etymology, shows that the word “play” carries the meaning of a “danced-out or ritualized fight” (33). A kōan may be a psychological device or a literary tool, but it is also a play in precisely this sense. In a roundtable on Zen published in The Eastern Buddhist , Kyoto School scholar Nishitani Keiji commented that sanzen encounters are “different from the dialogues that take place in schools or elsewhere . . . [they are] direct body attacks,” whereas Shibayama Zenkei compared dokusan to “two swordsmen fighting with real swords” (cited in Faure: 213). Robert Aitken, one of the deans of American Zen Buddhism views the kōan in terms of dance. The mondo writes Aitken is “[best] described as a drama or a dance in which the players or dancers bring forth the dynamic being of the Dharma” (106–107). Victor Hori states that “there is something correct about the performative account, for clearly in many kōan the proponents engage in shouting, bowing, slapping, putting sandals on one’s head, and so on—all performances” (286). And these performances are not simply found in kōan texts but in kōan practice.

Roy Rappaport defines ritual as “the performance of more or less invariant sequences of formal acts and utterances not entirely encoded by the performers” (24). Rappaport stresses the importance of the performative dimension of ritual.

Ritual is not simply one of the number of more or less equivalent ways in which the material embodied in liturgical orders may be expressed, presented, maintained or established. . . . The act of performance is itself a part of the order performed, or, to put it a little differently, the manner of “saying” and “doing” is intrinsic to what is being said and done. (37–38)

To consider ritual as an alternative, secondary medium for expressing what could otherwise be (perhaps more easily) expressed is to miss what is distinctive about ritual: a rite requires performance; if it is not performed, there is no rite. And the form of the performance adds something to the substance. In the case before us here, the kōan text is one thing, its performance another. The “manner” of the performance is all important, and the difficulty of capturing or presenting a kōan in the right manner is the hard nut to crack in kōan practice, as in any disciplined practice. Hakuin’s One Hand may be a formalized catechism (or liturgical script), but that need not mean its performance lacks originality, spontaneity, or creativity—the manner, the style of its doing is central, and certain meanings and effects are entailed within the performance. In other words performance itself has noetic value, and this knowledge is corporeal (in contrast to cerebral), active (not simply contemplative), and potentially transformative (not merely speculative) (see Jennings). [p485]

Zeami, in a treatise titled “Teaching on Style and the Flower,” raises the question of the relationship between text and performance, and in doing so points to the value and function of performative action.

What is the relation between movement and text in a N o performance? Answer: This matter can only be grasped through intricate rehearsal . . . [O]ne must project feelings that are in accord with the words being spoken. . . . When the idea of observing some object is suggested in the text, the actor performs a gesture of looking . . . when a sound is to be heard, the actor assumes an attitude of listening. (Zeami: 157)

A similar dynamic can be detected in the script of Hakuin’s One Hand. For example, at one point the master asks,

The Mt.-Fuji-summit-one-hand, what is it like? Answer: The pupil, shading his eyes with one hand, takes the pose of looking down from the summit of Mt. Fuji and says, “What a splendid view!” naming several places to be seen from Mt. Fuji—or others would name places visible from where they happen to be. (Hoffmann: 48–49)

For theatre director and performance theorist Richard Schechner, in good acting “the doing of the action of a feeling is enough to arouse the feeling both in the doer and in the receiver” (1990: 41), an observation similar to that of Zeami above. We may extend Schechner’s observation and add that the doing of the action of an idea can arouse the idea in the doer and receiver. Through the enactment of a kōan, feelings and ideas central to Zen are embodied, driven, as it were, into the body. Ideas are not simply absorbed through consciousness but formed, given body through enactment. Satori entails not simply D. T. Suzuki’s transformation of consciousness but bodily knowing—the body is the vehicle of the dharma, [7] and the liturgical order of Zen (which by definition is necessarily performed) coalesces with the establishment of Zen. A liturgical order (or liturgical system) “establishes the actuality of the order represented and, possessing the truth of things, that order becomes sovereign for those accepting it. . . . Cosmic orders are made in correspondence to the complex representations of liturgical orders in their entireties” (Rappaport: 345–346).

[p486]Robert Sharf has developed an important critique of the rhetoric of experience in Buddhist studies and reception. Kōan collections, Sharf argues, are not simply, or even primarily, to do with records of satori experiences to be replicated in students, but liturgical texts. [8] Sharf criticizes the

Tendency to psychologize ritual acts—to assume that the cultural significance and religious intent of a particular practice is best located in the subjective experience it is intended to induce, rather than in a mode of behavior perfected through rehearsal and repetition. (235)

“Traditional Ch’an and Zen practice,” concludes Sharf, “was oriented not towards engendering ‘enlightenment’ experiences, but rather to perfecting the ritual performance of Buddhahood” (243). This perfecting is not to be achieved through mere textual study, but through, to use Zeami’s words, “intricate rehearsal.” We are pursuing here in part a genre question; the kōan is neither simply a psycholinguistic riddle nor meditative and philosophical text but liturgy. Just as the performance of a Catholic mass is not to be reduced to the mass text, so too is kōan as liturgy different from kōan as text.

Heine speaks of the kōan text as a “vehicle for overcoming the very obstacles that verbal expressions create.” He continues: “the barriers to enlightenment that arise from language can only be broken through by means of language”—language is “at once the lock and key to authentic understanding” (1994: 200–201). Rappaport is also concerned with the limitations of language. “Humanity is a species that lives and can only live in terms of meanings it itself must invent” (8). The symbolic power of language to construct meanings is both virtue and vice, because “the linguistic capacity that is central to human adaptation makes it possible to give birth to concepts that come to possess those who have conceived them, concepts like god, heaven and hell” (9). Attention to the possessive and delusional power of words and concepts has been a central theme of the Zen tradition, though perhaps at times overstated by interpreters, particularly those in the experientialist camp. But whereas Heine points to overcoming the limits of language in language, through the kōan text , carrier of true words and spontaneous occasions of profound existential encounter, Rappaport points to the centrality of ritual as a forge in fabricating true [p487] words, in distinguishing truth from falsehood, and in releasing us from the tendency to reify language.

A key feature of ritual practice allowing for testing and revealing the limitations of language is that of reflexivity. Schechner refers to ritual as a process consisting of “restored behavior” or “twice behaved behavior, behavior that can be repeated, rehearsed” (1990: 41). The repetitiveness and formality of a rite supply ongoing occasions for reflexivity. Kōans are enacted over and over again until done right, and then one moves onto the next kōan in the curriculum. The “liturgical skeleton” of the kōan provides the basis for repeated exchanges during which the monk or student presents their understanding to the roshi. Schechner argues that ritual efficacy depends not just on ritual as an action or a doing, but on ritual as “a showing of a doing” (1977: 65). Ritual performance involves display, it is meant to be observed, the ritual act is shown to someone, even if that someone is an internalized self; and a rite is not necessarily a one time, static event but, as is the case with kōan practice, an ongoing process.

What transpires through the performance process, according to Turner, is that “what is normally sealed up, inaccessible to everyday observation and reasoning, in the depth of sociocultural life, is drawn forth—Dilthey uses the expression Ausdruck, ‘an expression,’ from ausdruck , literally, ‘to press or squeeze out’ ” (13). For Turner ritual is potentially a crucible for reflexivity. Rappaport also highlights the reflexive capacities of ritual performance. In ritual action two broad classes of messages are transmitted: canonical and self-referential (52–54). Ritual actors express and reveal their own physical, psychological, spiritual, or social states, but they also transmit messages not entirely encoded by themselves, the invariant features of the accepted liturgical order.[9] The student in kōan practice is constantly displaying subjective states—to the roshi, other students, and oneself—while at the same time displaying, transmitting, and establishing the canonical features of Zen Buddhism. The structural features of Zen practice—zazen, kōan work, dokusan , communal life, textual study—bring self-referential and canonical messaging into a tightly interwoven relationship; [10] both types of messages are repeatedly “squeezed out” and thereby made accessible to everyday, ongoing observation, critique, assimilation, integration.

[p488] Rappaport’s view of ritual as a forge fits McRae’s suggestion that kōan texts emerged through embodied, ritualized exchanges; it is these exchanges that generate and objectify certain ideas, beliefs, and practices and in doing so establish the paradigmatic cases (the true words) that become the basis for kōan collections. Rappaport discusses the relationship between performance and what he terms Ultimate Sacred Postulates. These “crowning bodies of religious discourse” (281) . . . are not merely claimed in ritual but “constituted by the performativeness intrinsic to liturgical orders themselves” (278). If we take, for example, the non-duality of reality as one of Zen Buddhism’s foundational Ultimate Sacred Postulates, it is not so much that the performance of a kōan reflects or mirrors this already established idea but generates and demonstrates repeatedly and effectively the idea, the experience, the substance of non-duality. It is one thing intellectually to comprehend or be able to explain non-duality, another to embody and establish the idea. Once again Zeami offers a relevant observation on the relationship between the objective and subjective dimensions, between canonical and self-referential messaging.

In regard to the style of dance in the No, Zeami first defines two opposing basic forms of Self-Conscious Movement (shuchi) and Movement Beyond Consciousness (buchi), and places above them Mutuality in Balance (sokyokuchi), which in turn comes into being as the unification of the two. (xli)

Zeami refers to this unification as internalization , a “state in which the flow of movement has become so well assimilated that the actor loses even the consciousness of controlling it” (xli). The kōan, as a kind of dance or play, may have as its end something of this sort. The rigor of kōan training is potentially transformative; it is meant to do something, or rather two things: establish Zen but also transform the individual.[11] At some point, we imagine these two are not really different, as when Sharf speaks of the “ritual perfection of Buddhahood,” or Rinzai of the “True Man without any rank” (Izutsu: 152). This ritual perfection entails neither self-conscious movement—putting on, feigning, trying too hard—nor [p489] pure subjectivity—movement beyond consciousness—but the seamless integration of the public and private, object and subject.[12] Hori notes that “kōan after kōan explores the theme of non-duality” (289).[13] From a performative perspective, what is sought in kōan practice is something akin to Zeami’s notion of internalization which, I suggest, is similar to the fusion of Rappaport’s notions of canonical and self-referential messaging.

If this description of kōan practice sounds idealistic if not naïve, it is perhaps because ritual is so often associated—particularly in pragmatic, positivist, Protestant culture—with pretence and formality, in which case a notion such as “the ritual perfection of Buddhahood” is tantamount to fakery at best or indoctrination at worst. Fakery—or, to put the matter bluntly, lying—is a very real problem in religious discourse and life. But this need not mean that liturgical orders are in the end reducible to other orders—whether these be economic, social, political, or gender—functioning as the hidden means by which these other orders are established and covertly maintained. Nor should we ignore the possibility that a religious system has built-in means of self-criticism and reflexivity that attempt to overcome the problem of lying. Such is precisely Rappaport’s claim. He argues that religious concepts and practices play an important role in “the adaptive processes of humanity” (408).[14] Heine, as we have seen, attributes this reflexive, truth-making function to the kōan text and its study; I am suggesting, following Rappaport’s emphasis on ritual action as the grounds from which “the sacred, the numinous, the divine and the holy . . . are fabricated” (406), that this function may be found in kōan practice, one aspect of which is reading.

If fakery, cheating, and lying are potentially inherent in performance, and there seems little doubt this is the case, so too is the overcoming of the inauthentic. Schechner notes that the “devices of performer training go beyond the physical into realms of simulation, feigning, pretending, play around with—all kinds of ‘as-ifing.’ ” Moreover “[e]very performer knows that this kind of playing around is a dangerous game verging on self-deception-accepted-as-truth” (1990: 41). The kōan , as a performative genre, takes part in this dangerous game. As Hori writes, because “the [p490] responses for kōan have now become standardized, it is possible for a monk who learns the standard answer to play-act his way through a session with the roshi without having any real insight” (294). But the context and structure of kōan practice work to ameliorate feigning. A prolonged and, one may assume, intense master–disciple relationship coupled with zazen, textual study, and the intricacies of communal life work to obviate fakery and inauthenticity.

The kōan curriculum, however, is long. There is always the next kōan , and the next kōan , and one cannot fake one’s way through the entire curriculum. . . . [A] roshi will often demand that a monk repeat and repeat his response to a kōan until he is able to perform it with genuineness, real conviction and personal flair. A roshi can usually spot play-acting, but if play-acting is fakery, then let it be genuine fakery. (Hori: 294)

The verbs “to perform” or “to act” mean “to do,” but also “to pretend,” and so performing and acting are often thought of as being filled with pretense, particularly when associated with ritual. Ritual is a serious business, it is sacred; it is not a mere performance, not mere play. But “to pretend” means also to intend, to design, to plot, to attempt, to hold before one, to extend. We need not divorce the serious, or the sacred, from play and performance (see Turner). Jonathan Z. Smith defines ritual as “a means of performing the way things ought to be in such a way that this ritualized perfection is recollected in the ordinary, uncontrolled, course of things” (63). This is a less cynical way of looking at ritual; ritual as an imaginative space, a perfected performative zone, where things can be imagined, staged, watched, and done right. If the crux of the Zen tradition is overcoming subject–object duality, the means by which this is achieved is “the very activity of play-acting” (Hori: 294).

The notion of ritualized perfection should not, however, be taken too literally, for it undervalues or obscures the messiness involved in playing around—the presence of the accidental, the spontaneous, and the disjunctive inherent to play; indeed, these qualities may be essential in overcoming subject–object duality. Rappaport’s emphasis on ritual as the means by which liturgical orders are established privileges the social, formal, and conservative features of performance. In the end the notion of self-referential messaging is swallowed under the canonical impulse of the ritual act, these messages referring to one’s acceptance (or rejection) of a liturgical order and the conventions of obligation and tradition. This formalist conception of ritual and its relation to liturgical order—implicit, for example, in Sharf’s notion of the ritual perfection of Buddhahood— devalue or ignore the improvisational, ludic, playful qualities inherent to [p491] kōans. The sense of the term “performance” in Rappaport’s work parallels that of Austin’s speech act theory; performance is a doing, the establishing of social or liturgical conventions or postulates. But the notion of performance can be taken more literally and related more closely to its theatrical sense. Kōan practice is not merely functional work, it is play, aesthetic, dramatic activity.

Grimes (1995: 40–57) delineates various modes of ritual sensibility: ritualization, decorum, ceremony, magic, liturgy, and celebration. These modes are not necessarily types but sensibilities, and various modes may be present in a single rite. The liturgical mode involves “re-presenting events” and “event-ualizing structures.” Liturgy “aspires to answer every question, to declare, ‘This is the way things are’. . . . [H]uman liturgies . . . are the means by which spirit is made present” (1995: 52). Rappaport’s notion of liturgical orders echoes Grimes’s description of the liturgical mode. Celebration, in contrast, “is distinguished by its roots in play and its seeming to be unmotivated and spontaneous. . . . In our robes and drapes of holiness we are always something of a clown” (Grimes 1995: 53). One of the enduring and endearing features of kōan texts is the presence of playfulness, and this is undoubtedly a central feature of kōan rites. Kōan practice, as I have tried to stress, is playful and dance-like, improvisational, as McRae suggests. Here, I would extend Rappaport’s notion of self-referential messaging to include the unique, nuanced, individualized performance of a rite. Rappaport, defining ritual as “the performance of more or less invariant sequences,” emphasizes the formalist, conservative qualities of ritual enactment. Grimes’s notion of celebration tempers this overly formalist conception of ritual and allies performance more with the arts than with the doing of liturgical work. “Celebration rites,” Grimes notes, “arise from expressive culture; hence, their link to the arts. They are subjunctive, and their ‘as-if’ quality, like that of good fiction, must be at once convincing and specially framed” (1995: 54). Celebration is highly reflexive, feelings and ideas are framed, staged, cultivated; choreography and spontaneity are wedded. Kōan practice, to recall Hori’s suggestive phrase, entails “genuine fakery,” which I take to mean authentic play and to include both creative and critical capacities.

Schechner plots the difference between ritual and theatre in terms of an efficacy–entertainment dyad. Both ritual and theatre entertain and have an effect, but ritual emphasizes efficacy, theatre emphasizes entertainment. This binary pair is performance: “performance,” writes Schechner, “includes the impulse to be serious and the impulse to entertain” (1977: 87). When a performance is efficacious, Schechner speaks of transformance— performance as the “means of transformation from one status, identity or situation to another” (1977: 71). One kōan where this dynamic can be [p492] seen is Bodhidharma’s “Skin, Flesh, Bones, Marrow,” one of Dogen’s favorite kōans and a principal case for Heine’s discourse analysis. This case deals with the first Chinese patriarch’s process of selecting his successor from among his top four disciples.

Bodhidharma, sensing that his time has come, asks his disciples to succinctly state their understanding of the Dharma. The first three disciples use some kind of metaphorical or philosophical expression, and the master’s response is that they express, in succession, his “skin,” “flesh,” and his “bones,” thereby indicating a progression of understanding from superficiality to depth . . . that still falls somewhat short. . . . The final disciple, Hui-k’o . . . bows three times and remains silent. (Heine 1994: 4–5)

As Heine points out, such a kōan plays easily into psychological interpretations in which silence and intuitive insight take precedence over analysis and speech (4). Heine, following Dogen, who “maintains . . . that language is a necessary and effective means of conveying the Dharma,” teases out the various philosophical implications of this kōan for religious praxis and discourse. But keeping Schechner’s linking of performance, transformance, and efficacy in mind, Hakuin’s bows can be viewed as an expression of a developed ritual sensibility.

Bowing, whatever symbolic referents it may carry, is first and foremost an act. Ronald Grimes has observed that in Zen practice bows often “function as commas and periods to mark transitions between ritual phases” (110). Using this simple though subtle observation of surface properties as an interpretive tool, Hui-k’o’s three prostrations to Bodhidharma are not an arbitrary sign that Hui-k’o understood the ineffability of the dharma, nor is bowing a literary trope filled with meaning pointing beyond itself; rather, the three prostrations exemplify Hui-k’o’s ritual sensibility. Although the other three responses may indeed have valid didactic significance within Zen discourse, the verbal answer to Bodhidharma’s question was not as important as recognizing the significance of the question—it was time for Bodhidharma to step aside, the old giving way to the new, a rite of passage. Hui-k’o’s prostrations were the means by which this particular transition was, within the liturgical system of Zen, appropriately marked and effected: after all, Hui-k’o did get the job. “Performance,” Turner writes, “does not necessarily have the structuralist implication of manifesting form, but rather the processual sense of ‘bringing to completion’ or ‘accomplishing’ ” (91). It is as if Hui-k’o asked himself not What shall I answer?, but How shall I act? Hui-k’o recognized the play he was in, and he gave a very efficacious performance. Kōan performance involves accomplishment, and the kōan curriculum can be viewed as a protracted series of rites of passage.

[p493] Performance theory also speaks to the question of experience, a category that has been central in kōan interpretation. Typically discussion of kōan practice by experientialists has located experience in the subjectivity of the practitioner. Certainly kōan practice entails various kinds of experiences, including Heine’s notion of the kōan as a spontaneous occasion of profound existential encounter, but we need not locate this experience strictly in subjective cognition. Turner’s use of Wilhelm Dilthey’s notion of Erlebnis is helpful here. Literally the term means “what has been lived through,” and Dilthey uses the notion to counter Kant’s suggestion that experience is formless. The key point for Turner is that lived experience entails presentation, communication: “an experience is never truly completed until it is ‘expressed,’ that is, until it is communicated to someone.” An implication of Turner’s fusing an anthropology of performance with an anthropology of experience is that we know ourselves not through subjective introspection but through the products of material culture; “we can know our own subjective depths as much by scrutinizing the meaningful objectifications ‘expressed’ by other minds, as by introspection” (14). Such observations seem especially relevant to an understanding of kōan practice. The kōan , as a “public record” or “case” manifests within its performance the Ultimate Sacred Postulates of Zen Buddhism. Students scrutinize through performance a set of cases (or liturgical scripts)—the kōan curriculum—through which they come to establish and participate in the “lived experience” of Zen. Viewed this way the kōan does not serve as a launching pad into the depths of subjective experience but is a product of performative culture, and hence necessarily public.


If understanding of the Zen kōan tradition is to be deepened, greater attention will need to be given its performative, ritualized dimensions. The two dominant concerns of the discipline of religious studies since its inception— the category of experience and the hermeneutics of texts—have dominated the scholarly study of the kōan tradition. A performative approach can knock these two meta-texts around, loosening their somewhat tyrannical hold on the reception, conception, and study of the kōan. Interest in performative culture has generated a wealth of theory, across disciplines, on performance, and ritual action; this work could be fruitfully brought to bear on study of the performative basis of the Zen kōan . I have suggested how experiential and textual perspectives import limited conceptions of ritual to their respective understandings of the ritual contexts of the kōan text, and that the application of ritual and performance theory to the study of the kōan makes good sense given the ritualized ethos of kōan practice.

[p494] Although a performative approach can offer some insight into questions of the historical relations between text and rite and even the hermeneutics of the kōan texts, a fuller appreciation of the kōan tradition calls for fieldwork. Given the privacy of the dokusan encounter, it is unlikely scholars will ever have direct access to dokusan , but ethnographic reports and descriptions of kōan work are a real possibility, and these would prove invaluable in the study of the kōans . We would then have another “text” to place alongside kōan collections, and this would surely complicate and alter conceptions of the kōan . There are public forms of dharma combat in the Zen tradition and a study of these rites may prove valuable in better understanding kōan practice. I have also suggested connections between aesthetic traditions such as N o and kōan practice, and a more detailed comparative analysis of the two may also prove insightful.[15]


1 . In my usage of terms I follow Grimes. A “ritualized form” is the result of “ritualizing,” the act of deliberately cultivating rites (see Grimes 1995: 659–662). The term “rite” “denotes specific enactments located in concrete times and places.” “Ritual . . . refers . . . to the general idea of which a rite is a specific instance” (1990: 9–10). A rite then entails a sequence or sequences of actions rendered special within a community or tradition by virtue of their elevation and stylization, generally named and set off from “ordinary” behavior by virtue of their being localized in special times and spaces. The notion of ritual is derived from actions characterized by a family of qualities: performed, condensed, patterned, stylized, and so on.

2 Psychological interpretations of the kōans tradition have moved far beyond these early figures. See Heine 1994: 87–93 for a brief summary of psychological studies. The kōans as ritual practice, however, remains of little interest to the experientialist camp.

3 See, for example, Hakuin’s notion of great doubt (discussed by Dumolin: 258–260). See also Yuan Wu’s commentary from the Blue Cliff Record on the kōans Three Pounds Flax (case 12) for an even earlier example of a psychological understanding of kōans.

4 Performative approaches to kōans to date have been limited to applying John Austin’s speech act theory. Victor Hori, for example, drawing on Austin, distinguishes between descriptive and performative language in arguing that kōans embrace both of these dimensions (305). The analysis, however, remains more with the kōans text than with kōans as ritual performance.

5 According to T. Griffith Foulk, kōans practice in “Japanese Rinzai Zen monasteries today [is] . . . best interpreted as ritual reenactments of certain formal relationships that are established in the kōans literature” (16). True as this may be, it leaves open the question how these formal relationships were originally established. McRae, as noted, suggests ritualized exchanges may have played an important role in this regard.

6 See Hori: 290–291 for a description of the “script” for the well-known kōan “Mu.”

7 Grimes offers a brief and valuable discussion of bodily knowing, arguing that the body is not “merely a tool for obtaining or communicating knowledge” but is itself “knowledge and communication”; “the centrality of the human body implies that meaning is embodied in overt action, posture, and gesture and that both culture and psyche can be ‘somatisized’ . . .” (1990: 148).

8 Zen training involves “perfecting the elaborate ceremonial repertoire incumbent upon a Zen Roshi. In the case of Rinzai Zen this repertoire includes mastering vast selections from Zen kōans collections and commentaries so as to be able to guide students through ritualized kōans exchanges. Here too prescriptive religious texts are treated not so much as practical guides for meditation, but as liturgies to be memorized for ritual performance (234).

9 Sharf, in his efforts to counter the rhetoric of experience in Buddhist studies, emphasizes the canonical features of Zen practice: “the injunction to practice Zen—to embody or instantiate the Buddha-dharma by participation in monastic ceremony and ritual—is not equivalent to the injunction to attain some sort of enlightenment experience” (266).

10 A very tight braid—at least, potentially so—between canonical and self-referential messaging is perhaps one of the unique features of Zen practice when compared with other ritual systems.

11 One need not understand this transformation in a mystical sense. Schechner observes that in theatrical training “what happens is that a person enters training or workshop as a ‘fixed’ or ‘finished’ or ‘already made’ being. “The training consists of specific methods of “breaking down the neophyte, of rendering him/her psychophysically malleable . . . quite literally, the performer-intraining (or workshop) is taken apart, deconstructed into bits” (1990: 41). Zen training can perhaps be imagined along similar lines. The performativity of Zen practice involves more than ritual formality; it is a practical method for “breaking down the neophyte.”

12 “Zen students early catch on to the fact that they must perform in front of the roshi, and at first they assume that any kind of physical movement will do. After a few rounds with the roshi, they learn that nonsense action is just as wide of the mark as purely intellectual explanation. Their response must be a performance but one that is appropriate to the context of the kōans” (Hori: 306).

13 “[T]he very activity of play-acting is a training in overcoming subject and object duality” (Hori: 294). Grimes arrives at a similar conclusion; the “whole point of Zen practice is to eliminate the split between practice and play, preparation and execution, symbol and referent” (1995: 107).

14 Turner makes such a claim for ritual action in his theory of social drama. Rappaport does not ignore the potentially pathological or maladaptive features of ritual and religion.

15 Grimes 2003 has provided a potential model for such inquiry in his comparison of western proscenium theatre and the Roman Catholic mass.


Aitken, Robert 1996 Original Dwelling Place . Washington, DC: Counterpoint.

Bell , Catherine 1992 Ritual Theory, Ritual Practice . New York: Oxford University Press.

Dumolin, H. 1979 History of Zen Buddhism . Tokyo: Weatherhill.

Faure, Bernard 1993 Chan Insights and Oversights: An Epistemological Critique of the Ch’an Tradition . Princeton, NJ: Princeton University Press.

Foulk, T. Griffith 2000 “The Form and Function of k ō an Literature: A Historical Overview.” In The k ō an : Texts and Contexts in Zen Buddhism . Ed. by Steven Heine and Dale S. Wright. New York: Oxford University Press.

Grimes, Ronald L. 1990 Ritual Criticism . Columbia: University of South Carolina Press.
__________ 1993 Reading , Writing and Ritualizing: Ritual in Fictive, Liturgical and Public Places . Washington, DC: Pastoral Press.
__________ 1995 Beginnings in Ritual Studies . Columbia: University of South Carolina Press.
__________ 2003 “Ritual and Performance.” In Encyclopaedia of Religion and American Cultures . Ed. by Gary Laderman and Luis Léon. Santa Barbara, CA: ABC Clio.

Heine, Steven 1990 “Does the k ō an have Buddha-Nature?” Journal of the American Academy of Religion 58: 357–387
__________ 1994 Dogen and the k ō an Tradition: A Tale of Two Shobogenzo Texts . Albany: State University of New York Press.

Heine, Steven, and Dale Wright, eds. 2000 The k ō an : Texts and Contexts in Zen Buddhism. New York : Oxford University Press.

Hoffmann, Yoel, ed. 1975 The Sound of the One Hand: 281 Zen k ō ans with Answers. New York : Basic Books.

Hori, Victor 2000 “kōan and Kensho in the Rinzai Zen Curriculum.” In The k ō an : Texts and Contexts in Zen Buddhism, 280–316. Ed. by Steven Heine and Dale S. Wright. New York: Oxford University Press.

Izutsu, Toshihiko 1977 Toward a Philosophy of Zen Buddhism . Boulder, CO: Prajna Press.

Jennings , Theodore W. Jr. 1982 “On Ritual Knowledge.” History of Religions 62/2:111–127.

Jung, C. G. [1939] 1969 “Forward to D. T. Suzuki’s An Introduction to Zen Buddhism .” In The Collected Works of C. G. Jung , vol. 11. 2d ed. Princeton, NJ: Princeton University Press.
__________ 1973 Letters . Ed. by Gerhard Adler and Aniella Jaffe. Trans. by R. C. F. Hull. Princeton, NJ: Princeton University Press.

Loori, John Daido 1994 Two Arrows Meeting in Mid Air: The Zen k ō an . Boston: Charles E. Tuttle Co.

McRae, John R. 2000 “The Antecedents of Encounter Dialogue in Chinese Ch’an Buddhism.” In The k ō an : Texts and Contexts in Zen Buddhism, 46–74. Ed. By Steven Heine and Dale S. Wright. New York: Oxford University Press.

Mohr, Michael 2000 “Emerging from Non-Duality:kōan Practice in the Rinzai Tradition since Hakuin.” In The k ō an : Texts and Contexts in Zen Buddhism, 244–279. Ed. by Steven Heine and Dale S. Wright. New York: Oxford University Press.

Rappaport, Roy 1999 Ritual and Religion in the Making of Humanity . Cambridge: Cambridge University Press.

Rikizan, Ishikawa 2000 “Transmission of Kirigami (Secret Initiation Documents): A Soto Practice in Medieval Japan.” In The k ō an : Texts and Contexts in Zen Buddhism, 233–243. Ed. by Steven Heine and Dale S. Wright. New York: Oxford University Press.

Sasaki, Ruth Fuller, and Isshu Miura 1965 The Zen k ō an . New York: Harcourt Brace.

Sharf, Robert 1995 “Buddhist Modernism and the Rhetoric of Meditative Experience.” Numen 42: 228–283.

Schechner, Richard 1977 Essays on Performance Theory, 1970–1976 . New York: Drama Book Specialists.
__________ 1990 “Magnitudes of Performance.” In By Means of Performance:Intercultural Studies of Theatre and Ritual , 19–49. Ed. by Richard Schechner and Willa Appel. Cambridge: Cambridge University Press.

Smith, Jonathan Z. 1982 Imagining Religion: From Babylon to Jonestown . Chicago: University of Chicago Press.

Staal, Fritz 1979 “The Meaninglessness of Ritual.” Numen 26: 2–22.
__________ 1989 Rules without Meaning: Ritual, Mantras, and the Human Sciences . New York: Peter Lang.

Suzuki, D. T. 1972 Living by Zen . London: Rider.
__________ 1974 Essays in Zen Buddhism: Second Series . London: Rider.
__________ 1989 The Way of Zen . New York: Vintage.

Zeami 1984 On the Art of the No Drama: The Major Treatises of Zeami. Trans. by J. Thomas Rimer and Masakazu Yamazaki. Princeton, NJ: Princeton University Press.