Semantics of 'spirituality' and 'religion' in light of corpus linguistic analysis
Logo der Einrichtung
 
Hintergrundbild
Hintergrundbild
  About  News  Research  Projects  Staff  Publications  Open Access
  Flag Deutsch
Theologie > CIRRuS > Research Disciplines > 'Biographical Studies in Contemporary Religion' > spiritualitaet >
  

The semantics of "spirituality" and "religion" in light of corpus linguistic analysis

We thank Stefan Altmeyer, University of Bonn, Germany, for his contribution to the analysis of the data of our project. These corpus analyses are his work.

In order to more closely or more deeply assess our respondents' understandings of "spirituality" and "religion," we have invited them to write down their subjective definitions in free entries in the questionnaire. And because 1039 respondents in the US and 727 in Germany have accepted this invitation, we have a large number of entries, which range from a few words to two or three sentences (and sum up to about 40,000 words in total for the US and 30,000 for the German sample). This rich data base opens a new perspective on the semantics of "spirituality" in relation to "religion." How did we analyze these data? We have taken two avenues: one is corpus linguistic analysis, a procedure to identify characteristic and outstanding words (keywords); the other is a detailed procedure of coding meaningful units in all entries for "spirituality" - and later reducing the many categories /codes by means of principal factor analysis. Here, we present only selected results of the corpus analyses.

For the corpus investigation we focused on keyword analysis, comparing a body of texts (research corpus) to another body of texts (reference corpus), which can be as large as an electronic data base of spoken or written language with millions of words. The aim is identifying the linguistic unique characteristics of the research corpus - which is indicated by the unique (deviant from the reference corpus) frequencies of specific words (keywords). The measure for such "uniqueness" is called keyness (and calculated on the basis of a log-likelihood procedure).

For a contrastive comparison of "spirituality" and "religion," it is revealing to analyze the free entries for "spirituality" (research corpus) in relation to the free entries for "religion" (reference corpus). The following figures present the nouns with highest keyness.

Keyword Analysis of Free Entries for "Spirituality," with Free Entries for "Religion" as Reference Corpus (and vice versa) in the US Sample (n=1,039)

In the corpus of the total free entries for "spirituality," the following nouns emerge as keywords with highest keyness: "feeling," "spirit," "connection," "soul" and "self;" in contrast, in the entries for "religion," a completely different set of nouns appears as characteristic: "set," "group," "rules," "beliefs," "people," "worship," "system" or "organization."

The next figure presents the same comparison of free entries for "spirituality" and "religion" for the German sample. The German nouns that are characteristic for "religion" are similar to what is presented for the US sample: it is "Regeln (rules)," "Gemeinschaft (community)," "Kirche (church; worship)" and "Dogmen (dogmatic propositions)," that are characteristic for "religion." The nouns which emerge as characteristics for "spirituality" mark - again - a clear contrast to the texts for "religion:" "Geist (spirit)," "Meditation," "Natur," "Esoterik," "Jenseits (other world)" and "Seele (soul)" are the outstanding characteristics for "spirituality."

Keyword Analysis of Free Entries for "Spirituality" with Free Entries for "Religion" as Reference Corpus (and vice versa) in the German Sample (n=727)