43 THEOLOGICAL LIBRARIANSHIP An Online Journal of the American Theological Library Association Volume 5, Number 1 • January 2012 PEER-REVIEWED ARTICLE Daniel R. Roland, Ph.D. is an Assistant Professor in the School of Library and Information Science at Kent State University and Primary Researcher at the Center for the Study of Information and Religion. Using Sermon Text Archives to Investigate the Construction of Social Values: A Proposal for a Collaborative Research Agenda in Social Epistemology by Daniel R. Roland ABSTRACT: This article presents a detailed description of a research agenda and methodology inspired by Jesse Shera’s notion of social epistemology as the study of “the ways in which society generates new knowledge, disseminates it, and uses it to contribute to the values the society seeks.”1 The research agenda is ambitious and echoes the call that Shera put forth with Margaret Egan that librarianship be a discipline “for the effective investigation of the whole complex problem of the intellectual processes of society.”2 The research agenda focuses on sermons as a communication medium that significantly influences the social construction of knowledge. With the advent of the World Wide Web and the increasing ease by which anyone may publish messages of every sort for all the world to receive and for scholars to study, Egan and Shera’s vision of macrocosmic bibliography and documentation is coming to fruition. The article proposes a collaborative effort between theological librarians and researchers in the field of religion and information science in order to facilitate Egan and Shera’s call to analyze “the production, distribution, and utilization of intellectual products [i.e., sermons] in much the same fashion as that in which the production, distribution, and utilization of material products have long been investigated” (133-134). InTRoduCTIon M any religious leaders were early adopters of the World Wide Web as a distribution and archival system for the sermons that they prepare and deliver on a regular basis. A growing number of religious leaders blog and podcast their sermons every week. Many of these sermons date back as far as the mid-1990s so that many websites contain archives of several hundred sermons. These sermons are typically stored in the websites of the local congregations served by the religious leaders in one or more formats including text, audio, and video, and made available to the public. The congregational websites include links to the sermons with instructions for interested persons on how to download, open, read, listen to, or view the files. As such, these congregational websites contain a treasure trove of primary data for researchers in the areas of religion, communication, sociology, and more. This article proposes a systematic and potentially collaborative method for the collection, organization, and indexing of large volumes of these sermons and thus creating digital repositories of primary data for the facilitation of research. 1 Jesse Shera, Libraries and the Organization of Knowledge (Hamden, CT: Archon Books, The Shoe String Press, Inc., 1966), 66. 2 Margaret Egan and Jessa Shera, “Foundations of a Theory of Bibliography,” The Library Quarterly 22 (1952): 132. THEOLOGICAL LIBRARIANSHIP An Online Journal of the American Theological Library Association 44 Volume 5, Number 1 •January 2012 A search of the literature indicates that what little research has been done using sermons as primary data has focused on small quantities. Ethelston analyzed twenty-four sermons that spanned nearly sixty years for instances of what he called “the misguided voice” in sermons delivered by evangelical religious leaders.3 Haskell, Paradis, and Burgoyne analyzed seventy-eight sermons all delivered on Easter Sunday, April 16, 2006, in the search for references to supposed challenges to the Christian faith in discourses of popular culture.4 Witten analyzed forty- seven sermons in order to compare how religious leaders from two different denominations within the Christian faith tradition interpreted the same passage of sacred text.5 Historical studies of sermons include Garner’s analysis of eighteen posthumously published sermons by the sixteenth-century Scottish theologian Robert Rollock.6 Washington analyzed an unspecified number of sermons and layman’s speeches in three Christian congregations located in Tokyo from 1890-1917 as vehicles for social change by creating a new discursive space previously unknown in Japanese culture.7 By contrast with these small data sets, the data collection and management method presented below facilitates research projects such as these and other potential research scenarios that expand the data pool by many factors of scale. The article describes a methodological approach that enables researchers to greatly expand the scope and volume of primary data collection across many variables such as time, space, gender, particular events, passages of sacred text, and religious affiliation. The sheer volume of sermons currently archived and added to the Internet each week presents logistical problems for data collection and invites a collaborative effort by a network of librarians and researchers. Sermons are distributed over thousands of websites, in different digital formats, and lacking a systematic filing and naming scheme. Organizations such as the American Theological Library Association (ATLA) and the Center for the Study of Information and Religion (CSIR) are logical networking nodes for collaborative efforts to standardize and systematize a pioneering research agenda in the area of information and religion. The doCumenT ColleCTIon And mAnAgemenT SofTwARe The dimensions of the primary data under consideration include hundreds of thousands, perhaps millions, of individual documents spread over thousands, perhaps tens of thousands or more, websites. The mere collection of these documents is a daunting task. Even more so is the organization and management of these documents once collected. The current author has compiled a collection of more than twenty thousand sermon texts from approximately two hundred congregational websites over the course of just two months. These sermons were prepared and delivered between the years 1996 and 2011. However, this collection is but a very small fraction of the growing number of sermons delivered in religious congregations each week. To adequately address the magnitude of the situation, a collaborative network of librarians and researchers using a shared technology platform could divide the data collection and management process by such natural divisions as faith tradition, denominational affiliation, and geographic area. 3 Graham Ethelston, “Appraisal in evangelical sermons: the projection and functions of misguided voices,” Text & Talk, 29 (2009): 45-70. 4 David M, Haskell, Kemmeth Paradis, and Stephanie Burgoyne, “Defending the faith: Easter sermon reaction to pop culture discourses,” Review of Religious Research 50 (2008): 139-156. 5 Marcia G. Witten, All is Forgiven: The Secular Message in American Protestantism (Princeton, NJ: Princeton University Press, 1993). 6 Mark Garner, “Preaching as a Communicative Event: A Discourse Analysis of Sermons by Robert Rollock (1555-1599),” Reformation & Renaissance Review: Journal of the Society for Reformation Studies 9 (2007): 45-70. 7 Garrett Washington, “Pulpits as lecterns: Discourses of Social Change Within Tokyo’s Protestant Churches, 1890-1917,” Japanese Studies, 29 (2009): 381-399. 45 THEOLOGICAL LIBRARIANSHIP An Online Journal of the American Theological Library Association Volume 5, Number 1 • January 2012 One potential platform is the knowledge management tool DEVONthink Pro Office.8 Billed as a smart information assistant, DEVONthink Pro Office easily facilitates the collection of documents from websites, automatically converts documents for keyword searching, includes a powerful concordance tool, facilitates customized document tagging, and much more. A strong user community that includes user forums, video tutorials, FAQ pages, and much more supports the software. DEVONthink Pro Office is the software of choice for the current author, and its use is explained below. Unfortunately, DEVONthink Pro Office only works on the Apple Macintosh platform. Similar software products include EverNote, which is available for both Mac and Windows platforms as well as popular mobile devices.9 EverNote offers free downloads for computers, tablets, and smart phones and effortless synchronization between devices via its cloud-based platform. A more ambitious proposal would involve AskSam SDK,10 a web server mounted, full-text searchable database for the facilitation of collaboration among an unlimited number of users. The dATA ColleCTIon PRoCeSS The most time-consuming, labor-intensive step in the data collection process is locating congregational websites that include sermons. Denominational websites often include directories of regional and state levels of organization within each denomination, which in turn often include directories of local congregations and links to congregational websites. The current author employs a graduate research assistant to conduct the initial searching of websites and to record the findings on a cloud-based spreadsheet in Google Docs. The URLs of congregational websites found to contain sermons are recorded in the spreadsheet state by state and denomination by denomination, along with contact and demographic data regarding the author of the sermons and the location of the congregation. A simple collaborative effort might involve seminary libraries and denominational research offices creating and maintaining web-based directories of congregational websites that include sermon texts. Such an effort could divide the task of initial searching into manageable portions and be of invaluable assistance to researchers in finding sermons for data collection. For example, a seminary library located in New England might commit to creating and maintaining a directory of congregational websites with sermon text archives of all congregations, or just those of a particular denomination, located in a particular state or region. Other seminary libraries around the country might commit to do the same, and a list of these links could be accessible from the home pages of ATLA and CSIR. This and other collaborative proposals are presented in more detail in the conclusion of the paper. The current author and other researchers would save a great deal of time by using such directories. The method used by the current author involves copying and pasting the URL of a congregational website into the DEVONthink Pro Office “Import Site” feature under the File menu. The Download Manager includes options for types of file to be imported and controls how far to drill into the website in search of sermons (see Figure 1). The import function can include the entire host or subdirectories, or be limited to a single directory. A potential concern might be that of copyright regarding the sermon texts. While some clergy members include a copyright statement within their sermon texts, most do not. Many of the texts do not include the name of the clergy member, nor the date of preparation and delivery, nor the location. These attributes add to the problems of data organization and retrieval. Only sermons archived on publicly accessible websites are collected. The sermon 8 http://www.devon-technologies.com/ 9 www.evernote.com 10 www.asksam.com/products.asp THEOLOGICAL LIBRARIANSHIP An Online Journal of the American Theological Library Association 46 Volume 5, Number 1 •January 2012 fIguRe 1 contents are not used by the current author in any form that misrepresents the intellectual content as belonging to anyone other than the original author. Clergy members are never identified by name or location in any write-up of research findings. Advantages of the data collection method. The primary advantage of the data collection system described above is the ability to access a large and growing body of primary data in a timely and efficient manner. Creating an archive of sermons spanning fifteen years and representing numerous religious affiliations and geographical locations enables multiple strategies for information discovery. These strategies include such possibilities as researching sermon content on particular dates, regarding particular events and issues, or regarding particular passages of sacred text. These and additional strategies are presented in more detail below. Disadvantages of the data collection method. There are several disadvantages to the data collection system described above. First, the system collects only those sermons posted to the World Wide Web. While increasing numbers of religious leaders are adapting the practice of posting their sermons to the Web, not all do so and perhaps some never will. Second, the system does not collect sermons prepared and delivered by those who do not prepare a written text. Many religious leaders prepare only an outline of a sermon so that large portions of the spoken message are extemporaneous. This practice is common throughout particular denominations and faith traditions so that the data collection system described above is not representative of all denominations and faith traditions. Third, the system favors data collection from the websites of congregations that belong to denominations with more organizational structure. The data collection system depends on denominational directories of organizational regions, synods, presbyteries, conferences, etc., in order to locate congregational websites. Therefore, independent congregations or those only loosely affiliated with a governing body are more difficult to locate on the World Wide Web. 47 THEOLOGICAL LIBRARIANSHIP An Online Journal of the American Theological Library Association Volume 5, Number 1 • January 2012 Finally, the system does not currently collect sermons posted to the World Wide Web in an audio-visual format. More and more religious leaders and congregations are utilizing technology to create podcasts and streaming video of sermons and entire worship services. This is especially true of those denominations and faith traditions mentioned above for which a written text of a sermon is uncommon. While the data collection system can be adapted to allow for the addition of audio-visual files, the inability to search these files for keyword content discourages the expenditure of time, effort, and computer memory space to include these files at the present time. The dATA oRgAnIzATIon PRoCeSS The following section describes how the current author uses DEVONThink Pro Office to organize and analyze primary data used in research of sermons. The goal here is simply to articulate and demonstrate a particular method and thereby provide a potential starting point for discussion, collaboration, and standardization for the systematic collection, organization, and analysis of sermon documents for the sake of social epistemology. The hope is that librarians and researchers will find the following information intriguing enough to desire to enter into constructive conversations that share similarities and suggest improvements in research methodology. Organization of the data in DEVONthink Pro Office begins with creating a file directory system that begins with the largest common denominators such as “Religious Leaders,” “Denomination,” “Gender,” “Year,” and “Location.” Each of these data identifiers requires a “Group” that is easily created from the “Data/New/Group” menu or by clicking on the “Group” icon in the toolbar. The folder icons that designate Groups resemble the data organization schemes in both Windows and Apple operating systems. Within each large group, smaller groups are required for individual denominations, gender, years, and locations by country and/or state. Within the “Religious Leaders” group, a new Group folder is required for each religious leader whose sermons are collected with each group labeled by last name, first name, and the denominational affiliation of the religious leader. The sermons are imported from Finder by highlighting the appropriate Group folder and selecting “Import: Files and Folders” from the File menu. Changing the label on the Finder file from which the sermons were imported to blue indicates that the files are within the DEVONthink Pro Office database. The second level of organization of the sermon documents requires the replication of the documents to appropriate sub-groups. The replication creates an alias of the document rather than a duplicate in order to save computer space, but any change to the original document also applies to every replication. Any documents within a group for an individual religious leader that properly belong to another group should be moved before the replication process. For example, if the files imported from a congregational website are found to include sermons from more than one religious leader, you must create a separate group for each religious leader and move the appropriate documents to the proper group. The replication process begins with the selection of document files that share the least common denominator such as Year, as demonstrated in Figure 2. An archive of sermons may contain documents that span several years in the career of a religious leader. Organizing these files by date creates a vital search feature for the database. This is also another challenging and time-consuming step in the data organization process since the various document filenames lack a universal code, but rather reflect the preferences of the document author. While many file names include a six-digit code to indicate month, day, and year or even the full spelling of the date on which the sermon was delivered, many more file names include only a sermon title or perhaps just a linear numbering system. THEOLOGICAL LIBRARIANSHIP An Online Journal of the American Theological Library Association 48 Volume 5, Number 1 •January 2012 fIguRe 2 DEVONthink Pro Office sorts documents within groups by file name, modification date, document type, and URL; this feature can help in determining the delivery date of a sermon. It is also possible to search the documents for a reference to a particular delivery date or year. The next step is to select all the documents that fall within a particular date range and replicate them to the appropriate year group. This process can be repeated for as many different years as are represented by the documents in the particular religious leader group. Next, all of the documents within the particular religious leader group can be selected and replicated to the appropriate groups for denominational affiliation, gender, and geographical location. Once this organizational step is completed and after you select a single file, different tags or labels will appear along the bottom of the file window. These Group tags facilitate Boolean searching within the database in order to quickly and easily expand or narrow a search for a custom set of document files. When all the documents within a particular religious leader group are replicated to the appropriate sub-groups, the group name can be highlighted “Flagged” to indicate that the replication process is complete. The InfoRmATIon dISCoveRy PRoCeSS Data become information when it is viewed from a particular context. The database of sermons described here allows the researcher a variety of contextual perspectives from which to view the data. The organizational scheme detailed above allows the researcher to easily view the data from the contextual perspectives of denominational affiliation, gender, geography, individual religious leader, and time. The researcher can easily narrow these rather broad perspectives by using the Boolean search logic built into DEVONthink Pro Office. These contextual perspectives are presented below along with examples that illustrate how to use the various search and discovery tools within DEVONthink Pro Office. 49 THEOLOGICAL LIBRARIANSHIP An Online Journal of the American Theological Library Association Volume 5, Number 1 • January 2012 Denominational or Faith affiliation. By highlighting one of the Group folders labeled by denomination or faith affiliation, the researcher may search just the sermon files within that particular group by selecting the appropriate folder and entering a search term. For example, Figure 3 displays findings for a search of sermons prepared and delivered by United Methodist religious leaders in which the word “abortion” appears. DEVONthink Pro Office returns a weighted list of documents that contain the search term based on the number of occurrences of the word and the perceived importance of the word within the particular document. By clicking on a file name within the search return list, the document contents appear in a windowpane below the list of documents, and you can review the individual document. At the bottom of this pane is a bar that includes all of the Group tags applied to the particular document. At this point, the researcher may wish to add a tag that applies just to the individual file, e.g., the sacred text passage from which the sermon is based, or a tag that designates that the sermon takes a “Pro-Choice position,” or a “Pro-Life position.” To do so, the researcher can simply click on the bar with the group tags and type the individual document tag. This tag will then appear in a list fIguRe 3 on the left side of the computer screen in order to facilitate the easy application of the tag on additional documents by simply clicking and dragging documents to the tag. Geography. Does the content of sermons differ from one geographical area to another? Do religious leaders in Arizona speak about illegal immigration with the same frequency and message as religious leaders in Wisconsin? Do religious leaders in California speak about same-sex marriage with the same frequency and message as religious leaders in Kansas? Did religious leaders serving congregations along the Gulf of Mexico have more to say about THEOLOGICAL LIBRARIANSHIP An Online Journal of the American Theological Library Association 50 Volume 5, Number 1 •January 2012 the Deepwater Horizon oil spill than did religious leaders in the Pacific Northwest? Such queries of this data management system lead to both quantitative and qualitative results. Patterns emerge from the data that lead to the construction of theory and call for additional data to check for theory consistency. By combining the DEVONthink Pro Office database with an online survey tool, the researcher is able to quickly identify potential survey informants and take the discovery process to the next level. Individual religious leader. As mentioned above, many of the sermon archives located and downloaded from congregational websites span several years in the career of a particular religious leader. This affords the researcher an opportunity to take a longitudinal view of sermon content. What patterns emerge over time in the interpretation and application of sacred texts by a single religious leader? Do positions on certain issues change over time or remain the same? With such data as a starting point, the researcher could seek to explore deeper into the life of the religious leader and the congregation served by use of a case study method. What impact have the sermons of this religious leader had on the values and beliefs of congregation members? What has happened in the life and understanding of this religious leader to effect change or the lack thereof in the interpretation and application of sacred text and current issues/events? Current Issues or Events and the Smart Group Tool. A very helpful and time-saving tool in the DEVONthink Pro Office database is the Smart Group. The Smart Group allows the researcher to create and save a query that automatically searches the database of existing documents and all new documents during the import process to create and continually update a list of documents containing the search terms. This tool effectively saves any search string found to be useful for a particular research project and eliminates the need to repeat the search process manually and on a regular basis. Perhaps the researcher is interested in references to marriage within sermons or to healthcare, the environment, 9/11, or any number of issues or events. This research method allows the researcher to quickly scan the primary data and then to drill deeper by organizing the data by date, location, denominational affiliation, and so on, in the search for patterns and concepts for theory development. Sacred text passage or doctrine. The research method facilitates searching the primary data for particular passages or phrases from sacred text or in reference to particular doctrines. The researcher can use the Smart Group tool to create and save queries for “John 3:16,” “an eye for an eye,” “take up your cross,” “original sin,” and so on. What patterns emerge in the use of particular passages of sacred text or doctrine in relation to gender and/or denominational affiliation or over time? Recurring Events. Religious congregation members expect their leaders to create and deliver sermons that address particular and recurring events important to the particular faith and social institution. These include but are not limited to Christmas, Hanukah, Easter, Passover, Thanksgiving, Mother’s Day, and the Fourth of July. Again, the research method facilitates the comparison of sermons addressing these events by a number of variables relating to religious leaders and time. Time. What did religious leaders speak about in a particular year? What was the content of sermons following a major event such as 9/11 or a particular natural disaster? When did a particular phrase such as “purpose filled life” begin to appear in sermons and spread geographically or increase/decrease in frequency over time? Which resources did religious leaders cite and refer to in a particular year and what patterns emerge in citations and 51 THEOLOGICAL LIBRARIANSHIP An Online Journal of the American Theological Library Association Volume 5, Number 1 • January 2012 references? What events or issues during a particular period received the most or least attention from religious leaders as reflected in the content of sermons? The research method facilitates the exploration of such questions and more. An example in the use of the research method from this particular perspective is described below. An exAmPle of The InfoRmATIon dISCoveRy PRoCeSS As of April 6, 2011, the primary research database contained 22,006 sermons that spanned the time period from 1996 through March 2011. In order to search only those sermons prepared and delivered in 2010, the researcher first prepared a new database labeled as “Sermons 2010.” Within this database, he created separate Groups for each of the denominations and faith traditions represented within the database. He labeled each Group with the name of the denomination or faith tradition and the year 2010. The researcher then returned to the primary database and selected the Tags view, clicked on the Tag “2010,” and then command-clicked on the tag for a denomination or faith tradition, such as United Methodist. He selected all of the document files in the list of search results and duplicated them to the Sermons 2010 database and the United Methodist 2010 group. After repeating this process for each of the fifteen denominations and faith traditions represented in the database, the Sermons 2010 database contained 3,150 sermons, prepared by 123 different religious leaders representing 15 denominations and faith traditions. The breakdown of these documents by denomination/faith tradition and the number of religious leaders per denomination/faith tradition is found in Appendixes A and B. It is important to note that the document sample is only for purposes of demonstrating the information discovery process. It is not presented as representative of all sermons prepared and delivered in 2010. Browsing. By selecting Concordance from the Tools menu and then selecting the Sermons 2010 database, the researcher is presented with a list of all the words contained in all of the sermons within the database that are at least three characters in length. The word length is easily adjusted at the bottom of the Concordance window in order to focus on words longer than three characters. The Concordance tool sorts the words by frequency, the number of groups in which each word appears, word length, and weight, each of which are helpful for information discovery and also for quality control. The Groups column allows the researcher to look for occurrences of words that are unique to only one or just a few groups. For example, the word “Achicha,” a Hebrew word for “needy kinsmen,” occurs four times in three different documents in the database, but in only one group. Somewhat predictably, the single group is that of the Jewish faith tradition. By contrast, the word “justice” occurs 1,805 times in thirty-three groups (see Figure 4). The Concordance tool also provides a list of the sermons in which the word occurs. Area Searching. Browsing the word list of the Concordance can be rather time consuming so the software features a “Similar” tool to assist the researcher. As demonstrated in Figure 5, highlighting a word in the word list such as “forgiveness” and then clicking on the “Similar” button gives the researcher a list of words weighted by their relevance to the search term, “forgiveness.” In this example, the similar words include “sin(s),” “forgive,” “forgiven,” “Christ,” and others. The researcher may highlight any of the similar words, click the “Search” button, and see a list of documents containing the new word. Keyword Searching. Entering key words in the search window of the Concordance accelerates the information discovery process. For example, a search for the word “Clementi” reveals that only five sermons in the database mention by name the young student from Rutgers University who committed suicide in 2010 after his roommate published a video on the Internet showing Clementi in a romantic encounter with another male student. By THEOLOGICAL LIBRARIANSHIP An Online Journal of the American Theological Library Association 52 Volume 5, Number 1 •January 2012 fIguRe 4 fIguRe 5 53 THEOLOGICAL LIBRARIANSHIP An Online Journal of the American Theological Library Association Volume 5, Number 1 • January 2012 highlighting “Clementi” and clicking the “Similar” button, the researcher discovers that the database associated many other names and words such as “suicides,” “bullying,” “webcam,” and the names of several other young men who also committed suicide because of persecution for their sexual orientation. By using the “Similar” tool, the researcher discovers four additional sermons that spoke to the situation without mentioning Tyler Clementi by name, but rather spoke to the “It Gets Better Project,” a media campaign that targets troubled youth, especially gay youth suffering from bullying. By double clicking on the sermon file links to open the documents, the researcher is able to quickly see how different religious leaders spoke about this event, how immediate to or removed from the suicide occurrence they spoke about it, and to what extent and in what context. For example, in their sermons addressing this event, three religious leaders from three different denominations used the same passage of sacred text from the Gospel of Luke 17:11-19. This passage relates a story of Jesus healing ten lepers, social outcasts because of their disease. The religious leaders speak to the tendency of many to think of homosexuality as a disease and to use that as an excuse to separate themselves from such persons as Tyler Clementi. However, the religious leaders compare Tyler Clementi to the single leper who returned to give thanks to Jesus for the healing: a Samaritan and therefore a man who is not just shunned because of his disease but hated by society because of who he is as a person. In contrast to the Tyler Clementi story, the devastating earthquakes that struck Haiti and Chile in 2010 received mentions in eighty-two and twenty-three sermons, respectively. The Haiti earthquake in particular received sustained mentions throughout 2010 due to the subsequent outbreaks of cholera and the hurricane that struck later in the same year. While using the keyword “Chile” to search the database for documents mentioning the earthquake in Chile, the researcher also discovered fourteen documents referring to the Chilean miners stuck underground for more than thirty days. Moving from discovery to research project. The juxtaposition of these events and the disparity in the number of references in the sermons raises potential research questions: why is there such a disparity in the number of references in sermons between natural disasters and the social ostracizing of people based on sexual orientation? What is the respective level of awareness by religious leaders regarding these two situations? What are the information resources used by religious leaders that affect awareness, or the lack thereof, regarding these situations? Are there extenuating circumstances that affect the decision-making process of religious leaders whether or not to address particular situations? The potential research questions naturally lead to speculation on potential answers and the awareness of need to design a survey for additional data collection in order to move beyond speculation and towards valid theory construction. ConCluSIon And A PRoPoSAl foR CollABoRATIon The importance of the research method presented above is reflected in the fact that the vast number of sermons made public on numerous congregational websites represent a large body of knowledge that is largely untapped and unexamined by the academic community. These documents represent an important element in a process of knowledge construction that occurs on a regular basis throughout a rather large portion of society. From the perspective of information science, the religious leaders access data in the form of sacred text and put the data within a particular context that reflects their individual and corporate belief system according to THEOLOGICAL LIBRARIANSHIP An Online Journal of the American Theological Library Association 54 Volume 5, Number 1 •January 2012 denominational affiliation and faith tradition. This contextualization turns the sacred text data into information, which is then packaged and presented as an information product in the form of a sermon. These sermons are presented to members of religious congregations, each of whom incorporate the messages into their own personal context that affects how they understand and interpret the world and their daily existence. In other words, the sermons become a part of the individual and social knowledge base. By exploring the content of sermons across a number of variables, researchers may hope to realize Shera’s dream of understanding the generation, dissemination, and use of knowledge in the construction of social values. However, even with the advent of the World Wide Web and computer software capable of locating and retrieving vast numbers of sermons from congregational websites, the task is too much for any one person to manage. Any hope for a comprehensive and thorough project reflective of the traditional commitment and quality of the library and information science discipline requires a collaborative and networked approach. The current author envisions the creation of a Special Interest Group (SIG) focused on social epistemology and religious knowledge that would meet each year at the ATLA conference starting in the summer of 2013. To that end, the following steps might work as a means towards implementation: 1) Interested parties contact the current author by e-mail at droland1@kent.edu. 2) The use of an online real time conferencing platform is implemented for regular meetings open to all interested parties. 3) The current author establishes and maintains a wiki site open to all interested parties for the recording and sharing of ideas, questions, assistance, etc., in the effort to lay the groundwork for creating the SIG. 4) The ATLANTIS listserv is used to announce additions to the wiki site, online meetings, etc., in order to recruit additional persons, and 5) If a critical mass of interested persons is reached, the current author submits a proposal to ATLA for the initial meeting and organization of the SIG at the 2013 annual conference. Once created and organized, the SIG membership would set goals and objectives, divide and assign various tasks, and assess the collaborative efforts on an annual basis. Once established at ATLA, the SIG might also hold regular meetings at the annual conference of CSIR in order to expand its membership to include researchers from other fields relating to information and religion. 55 THEOLOGICAL LIBRARIANSHIP An Online Journal of the American Theological Library Association Volume 5, Number 1 • January 2012 APPendIx A: The numBeR of SeRmonS By denomInATIon/fAITh TRAdITIon 1) American Baptist - 181 2) Disciples of Christ - 136 3) Evangelical Lutheran Church in America – 1,026 4) Episcopal – 186 5) Rabbinical – 54 6) Lutheran Church Missouri Synod – 495 7) Mennonite – 84 8) North American Lutheran Church – 103 9) Presbyterian Church USA – 269 10) Southern Baptist – 42 11) United Church of Christ – 258 12) United Methodist Church – 138 13) United Presbyterian Church – 24 14) Unitarian – 90 15) Wisconsin Evangelical Lutheran Synod– 64 APPendIx B: The numBeR of RelIgIouS leAdeRS PeR denomInATIon/fAITh TRAdITIon 1) American Baptist - 5 2) Disciples of Christ - 7 3) Evangelical Lutheran Church in America – 36 4) Episcopal – 10 5) Rabbinical – 5 6) Lutheran Church Missouri Synod – 16 7) Mennonite – 5 8) North American Lutheran Church – 4 9) Presbyterian Church USA – 8 10) Southern Baptist – 2 11) United Church of Christ – 10 12) United Methodist Church – 4 13) United Presbyterian Church – 1 14) Unitarian – 9 15) Wisconsin Evangelical Lutheran Synod – 1