Using Semantic Technologies to Analyze the Semantic Orientation of Religious Sermons - A Validation of the Early Work of McLaughlin

This research derives from a growing awareness in the knowledge management community of three factors: the value of language to knowledge management, the value of knowledge in all economic sectors and all aspects of human endeavour, and the “knowledge-richness” of belief systems and religion. Three research questions are addressed: What is the nature of language found in sermons? Is the use of semantic analysis technologies a feasible method for increasing our understanding of language patterns and characteristics? And, finally, Are there different approaches to the use of language in sermons across Christian religious communities? The research leverages semantic criteria defined by the early work of Raymond McLaughlin on the use of intentional and extensional language. McLaughlin’s research was necessarily limited in scale and scope and performed manually in 1940. In 2011, this research leverages semantic technologies to apply his well-formed semantic criteria to a larger scale (300 sermons) and broader scope (nine religious denominations). The research results suggest that McLaughlin’s criteria retain their value to language analysis today, that semantic technologies are a practical approach to applying these criteria to the use of language in religious communities, but that there are variations in the conclusion drawn by McLaughlin 70 years ago. The primary result suggests a high degree of balance of intentional and extensional language in modern day sermons. Language, Communication and Knowledge Management This research derives from a growing awareness in the knowledge management community of three factors: the value of language to knowledge management, the value of knowledge in all economic sectors and all aspects of human endeavour, and the “knowledge-richness” of belief systems and religion. Communication and language are essential elements of all stages of a knowledge life cycle. By the knowledge life cycle we mean all stages – from the discovery of a knowledge gap, to idea generation, to knowledge creation, mobilization, propagation, capture, management and preservation. In the knowledge management literature and dialog, communication is generally recognized as a key factor. Language has typically received less attention from the knowledge management community. Language, in the form of semantics, though, is assuming a more prominent role in the knowledge management domain with the popularization of semantic analysis technologies and the increasing use of computational intelligence methods. Traditionally, the field of knowledge management has focused on knowledge that is created in workrelated settings, in high profile organizations, and sectors which have high economic value. As the knowledge economy expands into all aspects of society, there is increased awareness of the value of understanding how people transfer knowledge in everyday contexts, beyond the formal work or business environment. Some important knowledge management lessons are resulting from the study of knowledge practices in the skilled trades, manufacturing and farming. Knowledge management is relevant to all aspects of our lives and our economy in the 21st century. It is important for knowledge management professionals to extend their understanding beyond the Fortune 500 company or the high profile public sector organization. Religious communities are “knowledge rich” in both foundational knowledge and beliefs and knowledge practices. A large portion of the population in any society participates in a religious community. Most religious or belief systems today have an explicit foundational knowledge and belief system, and a set of established practices for mobilizing and propagating (i.e., transferring or handing down) that knowledge across communities and over time. To date, though, the knowledge management literature has not considered religion as an area of study or research. What research exists on knowledge management in religion derives from the work in the theological community. This prevents both a gap and an opportunity. The goal of this research is to take a small step forward in bridging that gap, to establish deeper research collaboration between the theological and knowledge management communities, and to provide an opportunity for knowledge management professionals to gain a deeper understanding of the role of language in knowledge Research Context and Questions There are many belief systems across the globe. While there are common elements to all belief systems, we believe that common knowledge practices are more likely to be found within a single system. The Christian religion provides us with a well defined knowledge foundation and set of beliefs. It also provides us with a common knowledge practice – use of weekly sermons – as a method of knowledge mobilization and propagation from religious leaders to religious community leaders. A sermon is oration, a form of discourse generally delivered as a monologue by clergy or other recognized religious expertsi. Sermons are preached – using language which is chosen to effectively instruct the religious community, and to convey theological, religious or moral knowledge. The purpose of a sermon may address a wide range of issues, including general religious issues and topics, conversion efforts, narratives which tell stories and convey morals, and interpretations of current events. Some of the most effective sermons are drafted as stories, engaging the listener in the narrative. In the early 21st century, sermons are available not only interactively in a place of worship such as a church, but also via broadcast communications, as transcripts and podcasts downloadable from the World Wide Web. The widespread available provides an opportunity to expand our understanding of the use of language in knowledge mobilization and propagation within religious communities. Until recently any analysis of the language patterns has been a laborious and manual process. Humanbased, manual analysis also carries a risk of high subjectivity and variability across individual researchers. The labor-intensive nature of the work may tend to limit the scale and scope of the analysis. Subjectivity, limited scale and scope are all factors which limit our ability to generalize. The availability of userdefinable semantic analysis technologies addresses some of these risks. Tools available today allow us to encode analytical parameters and apply them consistently to large bodies of text. This provides data which can be objectively reviewed and interpreted by several researchers. Understanding the use of language in sermons may increase our understanding of effective use of language to support knowledge management in other domains. If semantic technologies can be used to analyze language automatically, we can increase both the scale and scope of our analysis – to produce generalizable and reliable results. Given the fact that all belief systems have variations in practice – and this practice may be reflected in the use of language – we explore whether variations in practice are reflect in language. This research strives to answer three essential questions: • Question 1: What is the nature of language found in sermons? • Question 2: Is the use of semantic analysis technologies a feasible method for increasing our understanding of language patterns and characteristics? • Question 3: Are there different approaches to the use of language in sermons across Christian religious denominations? Research Methodology The research methodology for each of the three questions is described below. Question 1: What is the Nature o Language Found in Sermons? There are multitudes of ways to describe the nature of language. This research focuses on the use of intentional and extensional language in sermons as an indication of different structures and approaches to conveying knowledge and ideasii. Generally speaking, semanticists characterize intentional language as more abstract and conceptual, whereas extensional language is more concrete and enumerative of real examples and properties. In the context of religion, intentional language may tend to be more dogmatic but also have a closer resemblance to narrative and storytelling. Conversely, extensional language may tend to be more balanced and qualified and more characteristic of logical arguments and rationality. Two commonly held assumptions are important to this research question, specifically: • From the field of theology – that sermons from fundamentalist preachers will tend to use more intentional language, and sermons from liberal preachers will tend to use more extensional language; • From the field of knowledge management -that narrative language style is more effective for engaging listeners and transferring knowledge than are argumentand rational-based language styles. An example of early work in semantics came in 1940 when Raymond McLaughliniii translated general ideas of intentional and extensional into concrete, observable linguistic criteria. In an era many years before the widespread availability of computers, McLaughlin applied his characteristics manually to sermons from two preachers: Harry Emerson Fosdick, pastor of the Riverside Church of New York City, a liberal pastor of what is even today considered to be a liberal religious denomination, and the well-known fundamentalist mass evangelist preacher, Oral Roberts. McLaughlin found 15 sermons from each pastor and analyzed their language for frequency of occurrence of intentional and extensional characteristics. McLaughlin’s criteria are described below. Table 1. McLaughlin’s Semantic Criteria for Intentional and Extensional Language Semantic Type Semantic Criteria Intentional Criteria Allness terms as represented by such terms as all, every, entire, whole, none.... Superlative terms such as best , worst, most, least, only, matchless, .... Two-valued terms such as either:or, if: when, if: not, “Is” of st