In Chapter 6 of his Conceptualizing Religion: Immanent Anthropologists, Transcendent Natives, and Unbounded Categories, Benson Saler presents “a prototype approach” in relation to our use and understanding of the category religion. According to this approach, the student of religion’s frame of reference is determined by prototypes, that is to say models or best representations of the object of analysis (Saler, 198). The student of religion, informed by his/her own cultural setting, develops conceptual biases which subsequently lay the groundwork for the formation of certain prototypes, which then constitute the student’s general frame of reference with respect to religion, and inform his/her subsequent analysis of religious phenomena. In particular, Saler notes that Western students of religion employ Western prototypes in their approach to non-Western societies: “ideas about the natures and histories of religions in the West serve as what the dictionary calls prototypes – as the first or original models – guiding anthropologists in their development of models of religion among non-Western peoples (Saler, 200).”
For Saler, the universal predicates of the essentialist tradition are not applicable to religious studies. Instead, Saler proposes to view religion as “a network of predicates, criss-crossing and overlapping in their applicability to phenomena that we variously deem better and less-good exemplifications of the category religion (Saler, 201).” For Saler then, there are no scientifically observable religious traits or characteristics that should precisely define what constitutes a religion as such: “the appeal to prototypes eschews the sort of approach that pivots on what Fillmore (1976:24) calls ‘a checklist of criterial properties,’ a list of features that an object supposedly must satisfy if it is to be deemed properly labeled by some word (Saler, 205).” Rather, the goodness of fit of any particular phenomenon associated with religion is determined subjectively by the student of religion in relation to his/her established prototypes: “’By prototypes of categories,’ Rosch writes, ‘we have generally meant the clearest cases of category membership defined operationally by people’s judgements of goodness of membership in the category (1978:36) (Saler, 206).” For Saler, prototype effects, that is to say the variations in individual judgment of the degree of resemblance of any given phenomenon to the category religion, do not in and of themselves prove that “membership in the category is graded and that the structure of the category is given by the prototype effects (Saler, 205).” To the extent that researchers may use different prototypes in their approach to the study of religion, the resulting prototype effects should only be understood as the natural result of the interaction of these various prototypes. They reveal the prototypical basis of the approach, not necessarily the structure of the category itself.
What then can be known for certain about the category religion? Saler’s prototype approach seems to imply that the conceptual understanding of the category religion has very much to do with the particular predisposition of the one attempting to define it, and as such there is no single right answer to the question “what is religion?” Under Saler’s method, the variety of prototypes held by individual scholars and informing their work results in a mass of phenomena indiscriminately deemed predicable to religion, and fails to assist in narrowing down the category to anything specific: “While all of the elements that we deem to pertain to the category religion are predicable of that category, not all of them are predicable of all the phenomena that various scholars regard as instantiations of religion (Saler, 225).” While this approach may be suitable for Saler, who understands the ultimate purpose of scholars of religion as wanting “to say interesting things about human beings rather than about religions and religion (Saler, 226),” it is not especially helpful for scholars who, unlike Saler, are more interested in examining the ontological status of religion itself, and tend to be left unsatisfied with religious phenomena explained solely in anthropological terms. Prototype theory is helpful in understanding the mindset of students of religion and in uncovering whatever conceptual biases may be at play in their work, but at the same time it seems to ignore questions surrounding the ultimate status of religious phenomena considered independently from those individual biases.