In research dealing with religious affiliation, generic nonreligious categories -'no religion', 'not religious', 'nonreligious', 'nones' -are frequently used to measure secularity and secularisation processes. Analysis of these categories is, however, problematic because they have not received dedicated methodological attention. Using qualitative research conducted in the UK, this article investigates what nonreligious categories measure and, specifically, whether they indicate nonidentification or disaffiliation as assumed or an alternative form of cultural affiliation. Findings suggest that generic nonreligious categories are sometimes used to express substantive positions and public identities, and that these are diverse. These findings flatten distinctions between religious and nonreligious categories as 'positive' and 'negative' respectively and indicate problems therefore in using nonreligious identification to measure secularity and secularisation. They suggest nonreligious identification is, however, a useful indicator of the advance of nonreligious cultures and the 'nonreligionisation' of societies.KEY WORDS: nonreligion and secularity; quantitative methods; atheist identities; indifference to religion; secularisation; nonreligionisation; religion in Britain; religious identities; disaffiliation The generic nonreligious categories frequently used in survey questionnaires appear to differ fundamentally from other identification or affiliation categories. Whether phrased as 'no religion', 'not religious', 'nonreligious' or 'none', these categories are apparently negative, allowing participants to record that a form of cultural identification lacks meaning or relevance to them or is perhaps entirely absent in their lives. Consequently, social researchers and reporters often describe and interpret the use of these categories as acts of 'disaffiliation' or 'non-identification'. Generic nonreligious categories seem, therefore, distinct from more explicitly positive confessional classifications that are also offered in religious identification survey questions. Selecting a category like 'Christian' or 'Hindu' indicates that a respondent is familiar with this term as a cultural marker and that it is meaningful to them to some degree. Such categories allow respondents to share an emic representation -the one that they would Religion, 2014 Vol. 44, No. 3, 466-482, http use in daily life or a reasonable proxy for the one they might normally prefer ('Christian' instead of 'Anglican', for example). By contrast, generic nonreligious options seem to be etic categories, used to locate the remaining individuals inside an analytical rather than a phenomenological framework. As Pasquale (2007, n.p.) says of the 'none' category, the terms themselves indicate that they are 'a function of survey method rather than a self-description'.In many countries in the world, increasing and large numbers of people are today choosing generic nonreligious categories: in Britain, the empirical focus of this article, 51 percent selected '...
