Does the Sacred Make a Difference? Category Formation in Comparative Religion

entities, but are inseparably connected to the corporeality and territoriality of human beings. For example George Lakoff and Mark Johnson argue that the structures of human understanding have their origin in the human body. Our conceptions of reality cannot be separated from what we experience in our embodied interactions. The idea of the sacred does not have an autonomy of its own as a religious category, but is inevitably linked with corporeality and territoriality as the structures of knowledge that constrain human thought and behavior. The idea of the sacred as a category-boundary that at the same time contradicts and unites all the other cultural categories has been developed implicitly by Mary Douglas. I say implicitly because Douglas refused to accept that the idea of the sacred and the idea of impurity had something to do with each other. Sacrality does not only mean that all members shall confirm to the class which they belong to, so that order, unity and integrity will be maintained (see Douglas 1989; Sperber 1996). There is also another side to the idea of the sacred as exemplified by the French sociologists Roger Caillois and Georges Battaille and to which Douglas's own studies on the taxonomic status of anomalous animals also bear witness. Impurity, forbiddenness and dangerousness are also characteristics of things classified as sacred. By the term 'sacred I refer here to a more general semantic content of the term that comprises both its positive and negative, its right hand and its left hand dimensions (see Burnside 1991).

When taken at face value, the sacred seems to be an unproblematic concept. Times, places, persons, animals and objects are classified as `sacred', because they have or have had a religious or spiritual significance for people in specific historical and social contexts. Religious traditions and their systems of signification are taken to explain why people have set aside specific things and considered them qualitatively different from other things. Deeming something as sacred means that it is disconnected from the category in social life in which similar things are classified and bestowed with special meaning and value. Sacredness of an object means that it stands in direct relationship to specific power-laden super-human entity by which members in a given culture mirror their self-consciousness or some aspects of it. A sanctuary for instance is a place that is set apart from the rest of the social space, because it is valued as a point of contact between man and the super-human agent worshipped by the local community.
The category of experience has had a prominent role in explaining the sacrality things, times and places. They are consecrated because they represent and commemorate special events and experiences of personages sanctified by the religious tradition. It is generally thought that experience constitutes the sacred or the holy because religious experience transforms the cultural schema by which -as William Paden has put it -the world and its processes of categorial selection are shaped (Paden 1992: 7; for the 'reference to experience' see Dawson 1996).
There has been an insistent tendency among scholars of comparative religion to approach religion with the help of the category of the sacred and treat it as a dependent variable of religious experience. But which explains which? Does religion explain sacrality or does sacrality explain religion? Is the boundary that is set up to draw lines of demarcation between sacred things and non-sacred things a minimum criterion for the religious, as Emile Durkheim suggested? If so, then we have to know what is distinctive in the social and cognitive processes by which sacred things are perceived and represented in relation to the perception and cultural representation of non-sacred things. Are there universal attributes and properties in human cognition and behavior which govern the representation of sacred things? If so, what are they: devoutness, solemnity, awe-inspiringness, compellingness of the norm of non-violation, a perception of an entirely different order and perhaps a sensation of 'a power or force quite different from the forces of nature (see e.g. Eliade and Sullivan 1987, 313). On what premises we base our scholarly characterisations: are the scholars of comparative religion emicists who mediate the culture-dependent thought-worlds and its categories into wider academic and popular audience or are we eticists who transform emic categories into scholarly ones in order to explain their semantics? What kind of cultural category is the sacred? Can it be considered a special mode of performative genre (see Turner 1992: 100-101) the forms of which vary with culture, but which can be known after becoming acquainted with its both universal and culture-dependent characteristics? And how are the ideas of setting something apart as `sacred' and a super-human agent related to each other? Does the norm of non-violation presuppose a super-human agent as a sanctioning authority before a thing will be established and represented as sacred? Or should we follow Nathan Söderblom's advice and consider the notion of God methodologically less important than the culturedependent forms of classification by which the sacred things are set apart from the profane things (see Söderblom 1913: 731).

How to conceptualize the sacred?
In the following I shall make an attempt at clarifying the methodological choices that we have to make in order to specify the interrelatedness or separateness of the categories of 'sacred' and 'religion', and to reach some sort of theoretical understanding of the epistemological status of the sacred as a cultural category.
Conceptualizing the sacred is not an easy thing to do when it is not taken at face value as a religious category. Sacredness is a worldwide phenomenon, and there are categories of things in any culture which are qualitatively different from others and which include the norm of non-violation. But because of its transcultural distribution and its intracultural comprehension, the sacred is an important concept and needs to be evaluated as a methodological tool both in the history and anthropology of religion. In addition to Emile Durkheim and Nathan Söderblom many distinguished phenomenologists of religion such as Rudolf Otto, Gerardus van der Leeuw, Joachim Wach and Mircea Eliade held sacredness (or holiness) to be not only the hallmark of religion, but its very essence. According to these theorists cultural systems of belief and practice cannot be given the title `religion' if nothing is deemed sacred by their adherents. In the methodology developed by the afore-mentioned phenomenologists sacredness has been treated as a special quality in human consciousness. The idea of the sacred entails that things, persons, places, times etc. are separated for a ritual purpose, because they are experienced as points of contact between humans and the transcendent reality. For this school of thought the sacred is a dynamic force that is manifested on the social level of spatial divisions, where religious persons can have their share of the force, hold communion with the sacred (see Eliade 1959: 367-369). Sacred space serves as a center which gives orientation and moral direction to the religious person. Eliade wrote that "(w)hatever the historical context in which he is placed, homo religiosus always believes that there is an absolute reality, the sacred, which transcends this world but manifests itself in this world, thereby sanctifying it and making it real. He further believes that life has a sacred origin and that human existence realizes all of its potentialities in proportion as it is religious -that is, participates in reality." (Eliade 1959: 202) It is largely agreed today that the phenomenologists' notion of the sacred as a dynamic force originating in another world blurs the boundaries of religious and scientific discourses. By emphasizing an introspective understanding of emotions and numenal structures in subjective religious experience, the phenomenologists have detached the sacred from the social matrix in which all human experience, including religious experience, takes place. By keeping the notion of the sacred reality intact from the cultural and cognitive processes constraining human thought and action and socially transmitted systems of meaning, these scholars can be criticized for taking part in the very cultural process that they were supposed to study.
As William Paden points out, there is no reason to equate 'sacred' with religion. The sacred is not a uniquely religious category, although its religious meanings and history of use dominate its popular as well as scholarly discourse. As Paden emphasizes, this is due to the so-called prototype effect. (Paden 1996: 16). Hebrew, Greek and Latin terms denoting 'sacred that appear in various religio-cultural texts and scriptures of Judaism and Christianity have exercised their influence also on the scholarly discourse in comparative religion.
I also agree with Paden when he states that sacrality is a distinctive factor in the logic of human behavior (Paden 1996: 16). I would add that not only in behavior, but also in human cognition that guides behavior. Students of comparative religion should not be out hunting only for the religious sacred according to prototypes given by Jewish, Christian and Islamic religious traditions. My own interest in this category is based on the idea of creating scholarly debate that has its point of departure in the semantic analysis of emic terms denoting sacred in various languages and in an attempt to explain their cultural logic with the help of the ethnographic texts in which the terms appear (see Anttonen 1996a;Anttonen 1996b). The use of the terms as specific concepts in the theologies of different religions should be approached with the same methodological attitude and seen as instances of specific cultural logic that has directed the formation of religious ideas in the contexts of their location. In this attempt we need to go beyond the category-formation of the phenomenologists of religion.

Evidencies from the past: the Finnish case
The word pyhä denotes 'sacred' in Finnish language. Before the term was adopted into the Christian vocabulary in the 12th and 13th centuries when Christianity was established in Finland, it was used in vernacular as an attribute in conjunction with prominent and exceptional natural places such as lakes, rivers, rapids, ponds, mountains, larger hills, capes, bays and fells. There are place names all over the Baltic Sea Culture Area, especially in Finland and in Estonia where the term occurs in a compound word as an appellative designation for a place.
Pyhä is a Germanic loan word. Its proto-Indo-European root is *ueik-denoting to separate'. In proto-Germanic the root is *vikwhich has given the adjective *whet-which the speakers of early proto-Finnic language turned into *püšä (> proto-Finnic pyhä). As a geographical term pyhä-designations in the Baltic Sea Culture area date back to the cultures of Bronze and Iron Age populations.
The question is why these places were designated as pyhä, i.e. sacred? Should we understand the prehistoric term pyhä meaning something altogether different from what it does today? Did it, per-haps, have none of the religious connotation that it has in Christian parlance and in popular discourse in today's Finland? What the sacrality of the places actually entailed?
According to my findings, the term was used only when all of the following conditions obtained: 1) The place was situated outside in an uninhabited area in the wilderness. 2) There were no previous names in this area. The attribute pyhä is the first name to be given in the place. The place or the area designated by the term pyhä was newly occupied land; the first people ever in the area had just taken the land into their possession. 3) The place had a special function for the people whose territory it belonged to and who had the right to use its natural resources. A "pyhä-place" was used as boundary marking the limits of the occupied territory and of the right of exploitation. 4) The "pyhä-place" as a boundary point was chosen from among the topographically exceptional or anomalous places in the region, or from places where routes intersected. Since the term pyhä appeared in similar places all over the geographical area where Finnish was spoken, it became an established term for marking places and boundaries in the landscape.
The adjective pyhä had a religious referent only to the extent as the category of 'religion' can be equated with the categories of the social' and the territorial'. According to methodologies of both Emile Durkheim and Arnold van Gennep religion as a category can be used in connection with popular traditions of hunting and agricultural societies in the meaning of its comparative use. It does not mean that religion actually has an autonomous ontological existence, but forms of cultural representation in local settings are theoretically conceptualized as such (see e.g. McCutcheon 1997: viii). According to this comparative methodology linguistic expressions in vernacular, oral narratives in folklore and other forms of cultural representation forming the nucleus of performances conceptualized as 'religious' are part and parcel of the overall social and spatial categories by which the members of ethnic communities comprehend and communicate the structures of meaning of their life-worlds. As Arnold van Gennep has emphasized spatial boundaries are not only legal and economic in nature, but also magico-religious. The boundaries marked by natural features such as rocks, trees, rivers and lakes or by natural objects such as stakes, portals or upright rocks are known by local people through collective tradition: as van Gennep writes the inhabitants and their neighbors know well within what territorial limits their rights and prerogatives extend (Gennep 1960: 15). The boundary points cannot be crossed or passed without the risk of supernatural dangers and sanctions. The boundary point is most often accompanied with interdictions, behavioral norms, rules of avoidance and prohibitions. Depending on the cultural value of situation when boundaries are crossed, socially prescribed rituals are considered as only proper ways to deal with the crossing (Gennep 1960: 15-17).
For the population groups of prehistoric Finland, pyhä represented a boundary between two conceptual spheres of sociocultural processes, i.e. those taken place within the inside and outside of the inhabited territory and the human body. Such a categorization is a major cognitive element on which various population groups have traditionally based their symbolic cultural behavior. The symbolism of the boundary crossing from the inside to the outside and from the outside to the inside have become manifest both in hunting and agricultural rituals, but also symbolically in so-called crisis rituals and in rites of passage.
The majority of the Finnish place-names beginning with pyhä are the product of the concepts guiding the categorization of space and the customary law tradition by which groups of settlers sought either to separate themselves from one another and to mark off the territory claimed by them from the shared inner domain or the outer domain. In place-names pyhä signified the outer border of the inhabited area. As a temporal category pyhä was used to denote times that are, as it were, on the border and Tall between temporal categories. It thus became a basic term in the reckoning of time according to the lunar calendar. Among the Baltic Finns it was used to mark off times into periods by virtue of its meaning of prohibition and non-violation. Pyhä meant forbidden, something to be avoided, dangerous, so that the behavioral norms prescribed by society had to be observed during the time marked off as sacred. In addition to territorial and temporal borders, the notion of pyhä was used as an adjective to mark off an object, a phenomenon, a time, an animal or a person that was to be avoided and held as forbidden because of its dangerousness or impurity and to separate it from the sphere of everyday social life.

From vernacular to scholarly construct: baby steps in category-formation
the sacred as a scholarly category in comparative religion A > B terms denoting 'sacred in the sacred as a vernacular of ethnic cultures religious concept The scholars in comparative religion have lacked a theoretical appraisal of the conceptual frames of reference suitable for comprehending and explaining the terms and concepts of the sacred in different cultures and religions, i.e. the religious and vernacular 'sacred', and the ritual systems of representation surrounding them. My attempt to create a new theoretical model for operationalizing the sacred as a methodological tool in comparative religion is founded on the mutual incompatibility of notions of the sacred in the discourse of theologically and anthropologically trained scholars of religion. The sociologists of religion are usually located somewhere in the middle of these two extremes.
Steps in this methodology are three-fold. First, one has to collect the linguistic and ethnographic evidence of the terms denoting 'sacred' in different cultural traditions. The scholar has to examine the words in vernacular denoting 'sacred' in contexts of their appearance. While taking the second step one has to pay attention to how agents in cultural systems perceive attributes and properties in places, times, persons, animals, actions and objects and categorize them in regard to attributes and properties that define members in the same category. The final and most important step from the point of view of comparative studies is to analyze the cultural and religious meanings conveyed by the context-specific usages of these terms. The scholar has to delineate the context-specific structures of knowledge and explain the cultural logic which underlies the sacred-making behavior within the symbolic system in question. This last step transforms the sacred from an emic category into an etic one.

Grounding the sacred in human cognition
In order to clarify how human disposition to sacralize is connected to human cognition, I have to call linguistics and symbolic anthropology for assistance. The concept of the sacred has been adopted into the theoretical language of various arenas of scientific specialties which cannot be ignored when theorizing it. It is basically in the Kantian tradition of the sociology of knowledge where the sacred has been treated as a fundamental category of human mind. Psychoanalysts, structuralists, students of cybernetics and cultural geographers have utilized the concept as a cognitive structure in human thought and behavior. Emile Durkheim and Marcel Mauss were first to suggest in their book Primitive Classification that the sacred should be dealt with in reference to the culturally dependent classificatory systems and theorized in connection to the social constraints that bring about collective conscience. Durkheim and Mauss treated the sacred as a collective representation that is set apart as a symbol that unites divisions, distinctions and oppositions into a meaningful whole and gives legitimacy to the behavioral norms connected with the symbol. Things set apart as 'sacred transcend individual consciousness and act as a divinely legitimated source for sentiments that bind members of the social group together. Any category within the social systems of classification which has a specific value for the local community needs to be dealt with in relation to sacred things, times and places when its taxonomical status is about to change or needs to be changed. Ritual is the only proper context for category transformations since ritual creates an in-between boundary space within the social system of categories (see Leach 1976). Since ritual is the social system of behavior that makes a difference in showing the flexibility of distinctions and oppositions between social categories, it is the prime locus that in the final analysis also creates the sacred (see e.g. Smith 1987;Bell 1992).
In my own study referred to above I have argued along the Durkheimian lines that the sacred and the ritual can be treated as corresponding scholarly categories and they can be used to analyze the logic of behavior in liminal spaces that people create in order to make a difference in their cultural systems of categorization. The semantics of both categories can be approached with ethnographic data on cultural and religious symbols that represent cognitive boundaries by which ethnic or other social groups maintain, secure and reorder their social edifice. Both categories are cognitively connected to the ideas of placeness and placelessness and corporeality and noncorporeality. Although Durkheim failed to understand human mind and its cognitive faculties separately from social organizations, his contribution has been nevertheless of prime importance in conceptualizing and operationalizing the sacred as an anthropological category (see e.g. Paden 1991;Paden 1996;Parkin 1991;Anttonen 1996a). Durkheim went wrong, however, in assuming that the opposition of the sacred and the profane stems from social sentiments. There are far greater cognitive operations behind the categories of `sacred' and 'profane and their opposition. In Rodney Needham's words if the mind is taken to be a system of cognitive faculties, it is absurd to say that the categories originate in social organization...the notion of class necessarily precedes the apprehension that social groups, in concordance with which natural phenomena are classed, are themselves classified" (Needham 1963: xxvii).

The sacred stems from the systemic character of human thought
Claude Lévi-Strauss, who took structural linguistics as his point of departure, developed Durkheimian ideas into a more general theory of the human mind. While Durkheim had a social deterministic conception of the opposition between the sacred and the profane, Lévi-Strauss converted the idea of oppositions into a more semiological and symbolic approach. Cultural symbolic structures and models are not grounded in specific forms of social organization, but vice versa: all social categories have a symbolic origin. Lévi-Strauss thought that human beings process information on three categorial levels: the real, the symbolic and the imaginary. He treated culture as a system of communication in which language carries thought back and forth across these three structural levels (see Morris 1987: 266;Sullivan 1984: 152-153). But as Brian Morris writes, one is never too sure as to which level of reality or experience the symbolic systems are based upon (Morris 1987: 204) and what is the cognitive mechanism by which they are fabricated. In analyzing the work of Lévi-Strauss Lawrence E. Sullivan says that processes of thought transform elementary structures of the mind by building symbolic bridges between contradictions. These symbolic bridges become in turn the focus of the same unceasing formal processes and are recycled as images which, in their turn, become object (or victim) of processes which reorder their relations in the attempt to give them meaning" (Sullivan 1984, 152-153). Things, animals, persons, times and spaces set apart as sacred are in Lévi-Straussian terms symbolic bridges that carry thought back and forth on these three structural levels and become represented not only in ritual, but also in myth, epic and fiction.
In Lévi-Straussian terms the idea of the sacred is like the numerical value zero. In itself it signifies nothing, but when joined to another number it is filled with differential significance (see Smith 1987: 108). In religious systems the idea of the sacred as the numerical value zero becomes evident when we for example think of the symbolism in Christian rituals. Jesus Christ can be paralleled to a numerical value zero: in itself it signifies nothing, but acquires meaning and acts as a source of meaning when joined to different aspects of value in the category systems of Christian individuals and communities. We only have to think of Christian rites of passage. The idea of Jesus as an embodiment of sacrality becomes represented in liminal boundary states such as in birth and the baptizing ritual, in the ritual of confirmation and also in rituals of marriage and death. Jesus is a culturally established symbolic bridge by which oppositions such as male/female, life/death, pure/impure, inside and outside sanctuary, inside and outside of the human body are brought into differential relationships. Let me here quote how Jonathan Z. Smith has described the logic of the sacred: "Here (in the world) blood is a major source for impurity; there (in the ritual) blood removes impurity. Here (in the world) water is the central agent by which impurity is transmitted; there (in the ritual) washing with water carries away impurity. Neither the blood nor the water has changed; what has changed is their location" (Smith 1987: 110) Ritual exhibits the religious system and its differences by focusing attention on one or several aspects of the systemic elements. Arnold van Gennep had a special expression for this: he called it the pivoting of the sacred.
Unlike Eliade Lévi-Strauss did not think that the sacred is a structure in human consciousness which refers to specific symbols in religious narratives in order to display their divine origin. Just like the concept of value, the sacred is a differentiating device that emerge from the systemic character of human thought. For Lévi-Strauss the sacred is at the same time an order of universe and a transformational situation when things are removed from places allocated to them; as Lévi-Strauss writes "being in their place is what makes them sacred for if they were taken out of their place, even in thought, the entire order of the universe would be destroyed" (Lévi-Strauss 1968: 10).
Along the Lévi-Straussian lines it can be argued that the idea of the sacred as it appears in the reports written by ethnologists and anthropologists and also in literary sources of established religions, displays a primordial structure of human cognition that takes different forms and contents according to the master narratives in the symbolic-cultural systems. The sacred is a socially constructed metacategory by which metaphoric and metonymic relations between other cultural categories are established and mediated, e.g. between categories of person, gender, kinship, marriage, nation, or between moral categories such as justice, liberty, purity, propriety.

Cognitive categories and their boundaries
In the cultural anthropology of Mary Douglas, just as well as in recent scholarship in cognitive anthropology, linguistics and philosophy the operation of human mind has received more serious attention (see Boyer 1993;Boyer 1994;Lawson 1993;Lawson and McCauley 1990;Lakoff 1987;Lakoff 1989;Johnson 1987;Johnson 1991;Lakoff and Johnson 1980). Religious and other social concepts and categories that comprise and organize knowledge do not float in the air as abstract entities, but are inseparably connected to the corporeality and territoriality of human beings. For example George Lakoff and Mark Johnson argue that the structures of human understanding have their origin in the human body. Our conceptions of reality cannot be separated from what we experience in our embodied interactions. The idea of the sacred does not have an autonomy of its own as a religious category, but is inevitably linked with corporeality and territoriality as the structures of knowledge that constrain human thought and behavior.
The idea of the sacred as a category-boundary that at the same time contradicts and unites all the other cultural categories has been developed implicitly by Mary Douglas. I say implicitly because Douglas refused to accept that the idea of the sacred and the idea of impurity had something to do with each other. Sacrality does not only mean that all members shall confirm to the class which they belong to, so that order, unity and integrity will be maintained (see Douglas 1989;Sperber 1996). There is also another side to the idea of the sacred as exemplified by the French sociologists Roger Caillois and Georges Battaille and to which Douglas's own studies on the taxonomic status of anomalous animals also bear witness. Impurity, forbiddenness and dangerousness are also characteristics of things classified as sacred. By the term 'sacred I refer here to a more general semantic content of the term that comprises both its positive and negative, its right hand and its left hand dimensions (see Burnside 1991).
The analysis by Douglas of the pangolin cult of the Lele can be taken as an example which demonstrates how taxonomic anomaly is one of the characteristics on the basis of which animals are classified as sacred. This can be done in spite of the fact that her theory of animal symbolism has been criticized e.g. by Dan Sperber in his recently published article (1996). According to Douglas it is probable that species of animals will have symbolic value if they are perceived anomalous in regard to the attributes and properties defining the animals in the same category. With his findings concerning the taxonomic status of Cassowary among the Karams in New Guinea, Ralph Bulmer corrected Douglas's imprecise formulation. Bulmer showed that it is not only the taxonomic anomaly that makes the animal `sacred', but its relation to human beings (Bulmer 1973). Anomalous character is not a sufficient criterion by which sacredness is defined, but its ability to serve as a vehicle for negotiating and reconceptualizing categorial boundaries by which difference is made e.g. between male and female and between kinship and territorial divisions. Persons, animals and objects that are chosen as sacred symbols do not only reflect the idea of making a difference, but also the idea of transcending the difference in order to produce growth of social values across the boundaries that differentiate categories. The sacred as a categorial boundary is universal cognitive property which is represented in various sorts of symbolism, not only in the symbolism of socalled folk religions, but also in the symbolism of major world religions such as Hinduism, Buddhism, Judaism, Christianity and Islam (see Pyysiäinen 1996). As long as there are categorial divisions in the world, there will always be some sort of symbolism of the sacred according to which people make differences in thought and action and create boundaries that at the same time separate and unite.

Recipe for scholars of religion to differentiate sacred-making characteristics
Pay attention to how agents in cultural systems • perceive attributes and properties in times, places, persons, animals, actions and categorize them in regard to attributes and properties that define members in the same category (possible options: taxonomic anomaly/ taxonomic completeness or wholeness) ® signify properties in relation to systems of value in culture, society and in personal lives of individuals (by creating metaphoric and metonymic linkages) • set them apart • idealize them as cultural models for guiding behavior and establish normative boundaries against transgression and violation (pure/impure and licit/forbidden) e create systems of symbols and forms of cultural representation (embodiment and spatiality as central structures of knowledge in the process of symbolization and ritualization)