A Measured Faith: Edwin Starbuck, William James, and the Scientific Reform of Religious Experience

A number of recent studies have drawn attention to how the study of religion and religious seeking were intertwined in European and American cultures in the nineteenth and twentieth centuries. Ann Taves, Leigh Schmidt and Hans Kippenberg, for example, have pointed to ways that particularly Protestant anxieties and dilemmas shaped scholarly thinking about categories such as experience and “mysticism.” Scholars have been less interested, however, in the other side of the exchange—less interested, in other words, in how scholarship has reshaped religious belief and practice. The first Americans to study religion scientifically, American psychologists of religion, serve as a particularly useful illustration of how scholarly methods influenced modern ways of believing, but there is still little historical scholarship on the key figures involved. There remain few critical works, for example, on the pioneer psychologists of religion—Edwin Starbuck (1866–1947), George Coe (1862–1951), James Bissett Pratt (1875–1944), and G. Stanley Hall (1844–1924)—and their ways of studying and attempting to reform religion. The notable exception is, of course, the literature on William James, which includes an enormous number of dissertations and monographs, including several important studies examining the Varieties of Religious Experience and James's other efforts to help fashion a science of religion. But even the scholarship on James does not consider how he and others used the sciences to reform religious belief and revitalize American culture. Given the fact that James identified himself as a psychologist, engaged a wide range of neurological, physiological and psychological thinkers in his work, and drew extensively on psychologists like George Coe and Edwin Starbuck, it is remarkable that these contexts have been overlooked. His debt to the psychologist Edwin Starbuck is particularly striking. In his Varieties, he uses or refers to Starbuck's empirical work twenty-six times, he draws from Starbuck's questionnaire data thirty-seven times, and he mentions Starbuck by name a total of forty-six times, which is roughly the equivalent of once in every six pages of text.