Semiotics based discursive communities in online book reviews

Introduction

Inquiry on information meaning is one of the central foci of LIS research. Any information activities assume an understanding of the meaning of information for the producer, the user and a processor. Information organization such as indexing or cataloguing involves a process of denoting and interpreting what it means to represent and create a surrogate for retrieval. Information users make meaning of information according to their need and use context. Meaning is embedded in information phenomena during its production, organization, retrieval, use and any intersection of these activities.

The rise of user generated content (UGC) and big data have resulted in an increased focus within LIS on machine processing, natural language recognition, and semantic textuality (Xiao and Conroy, 2017). Establishing a methodological framework for identifying communities of meaning within UGC would be beneficial to the identification of proto-communities of users in UGC contexts where formal communities and community building features may be lacking. In this paper, we will discuss semiotics as a useful methodological framework for interpretative analysis of user meaning in UGC, using UGOBR as a case study.

Semiosis and community

According to Peirce, ‘A sign is something, A, which brings something, B, its interpretant sign determined or created by it, into the same sort of correspondence with something, C, its object, as that in which itself stands to C’ (Huang and Chuang, 2009, p. 342). The process by which people negotiate between these elements to create meaning is dynamic, ongoing, and generative of further signs and semiosis. The end result of the semiosis, or the interpretant, takes the form of a new sign generated from a meaningful interpretation of the relationship between the form a sign takes (representamen) and the concepts/things it represents (object) (Mai, 2001; Thellefsen, Brier and Thellefsen, 2003; Thellefsen, Thellefsen and Sorensen, 2018). As such, any new interpretant always bares the possibility that it could in turn become a representamen for a new process of semiosis, potentially stretching into infinity. Peirce grounded these distinctions with a hierarchical phenomenology outlining the varying ways in which a sign’s triadic elements are manifested through human perception, sensation, and cognition. Based on this, he developed 10 sign classes (Mai, 2001).

He acknowledged that a single sign can mean different things to different people based upon their past experiences with the sign. Via his conception of legisign, or signs as generalizable laws, Peirce acknowledged that the only way that language can function and discourse can arise is via shared experience with signs (Peirce, 1955, p. 11). This invites the possibility of the formation of semiotic communities, or communities of individuals defined by a shared means of interpreting signs. Several within the field of ILS have begun to think of ways that semiotics might be applied to the study of communities (Brier, 1996; Friedman and Smiraglia, 2013; Sorenson, Thellefsen and Thellefsen, 2016). Schreiber (2012) adapts Irvine Goffman’s notion of frame to study of individual users of collaborative learning software employ different framing devices for the interpretation of meaning.

Thellefsen and Thellefsen (2013) supplement this conception of semiotic framing devices by distinguishing between the roles of collateral experience and discursive communities in semiotic processes. They use the concept of collateral experience or the ‘actual knowledge shared by the utterer and interpreter about concepts or phenomena in a communication process’ (Thellefsen and Thellefsen, 2013, p. 1739). The more experience that an individual has with the interpretation of a given representamen, the more limited and defined their interpretations will be. Individuals with shared collateral experiences are better able to arrive at a common understandings of the meaning of a sign, opening up a space where communities of discourse can arise. While discourse communities refer to norms for discussing a given topic, be they ad hoc or highly structured and specialized, collateral experience refers to shared experiences with a given sign that allow communities of discourse to arise.

This distinction between discourse community and communities of collateral experience is an important one to the study of ad hoc discursive communities that may arise along with UGC. While all processes of semiosis will be contextualized within some sort of discursive community, examining book reviews in terms of communities of collateral experience will help to identify common features within individuals that may lead to certain types of interpretations, even in the absence of any formalized community structures.

Semiosis and reading

Peirce’s conception of semiosis provides a methodological framework for unpacking and differentiating between the complex relationship between individual perception and emotion, informational content, and socio-cultural context involved in the process of reading books, writing about books, and understanding the multiplicity of meanings that those books can have for individuals.  Peirce allows us to begin to separate the different semiotic processes that are involved in the review writing process: (1) the semiosis of the book’s author, (2) the semiosis of the reader who writes the book review, and (3) the semiosis of researcher analysing the review.

Applied to Peirce’s triadic model of the sign, the text of the book can be viewed as a collection of signs each of which has a representamen, object, and interpretant.  However, the exact nature of each of these triadic elements changes depending upon whether it is placed within the context of the book’s author, the book’s reader, or the researcher’s analysis of UGOBR.  To the author, the text of the book, as the end result of a semiotic process, can be viewed as the interpretant. The author’s own semiosis, in writing the book, involves interpreting a host of signs and synthesizing this interpretation into collections of rhemes, dicent statements, and arguments. To the reader, the book’s text constitutes a collection of representamen, a host of signs and also an object, what is signified by the book content, the idea that is conveyed in the book by the author. For the reader, the semiotic processes through which they interpret and make sense of the book, as a representamen, involve a complex interplay between their previous collateral experiences and their emotional and cognitive experiences of reading the book. As researchers, we analyse the text of the OBR as the reader’s interpretant (or representamen, from our perspective), employing semiosis as a framework for interpretation and analysis.

UGOBR text analysis

The primary objective of this research was to explore using semiotics as not only a theoretical but also a methodological framework for empirically identifying the underlying experiential basis for shared semiotic frames that result in similar interpretation of signs among sets of individuals that do not necessarily self-identify with a common community. We employed it in the analysis of OBR text for examining a shared collateral experience of book reading. In this paper, we discuss how we used the triadic sign model as a guiding ground for coding and analysis of review text in order to see if and to what degree the collateral experience can be identified in user generated book reviews.

A corpus of 200 reviews was purposively gathered from Amazon (amazon.com) by selecting the top 20 most helpful user reviews for 10 popular science books. The body of the reviews along with relevant metadata were archived coded and analysed.

Coding and analysis were done inductively and iteratively with an interpretative approach, allowing categories to evolve through the iterative process. The analysis started with an individual review to generate a researcher’s re-creation (i.e., researcher’s interpretant) of the meaning combining with the multiple layers of coding categories, then moved across all reviews for a given book, grouping them together by their similarities in meaning and sentiment.

The initial phase focused on categorizing claims by topic to capture the researcher’s meaning of the content as a relationship between the sign vehicle (i.e., review text) and the sign object (i.e., reviewer’s meaning) of Peirce’s triadic sign. A claim was defined as any chunk of a text representing a meaningful assertion and constituting a core unit of linguistic meaning (Yoon, 2012; Yoon and Bezdicek, 2015). It has yielded the following four categories; (1) the book itself and its author pertaining the book value; (2) the book content such as subject topics, theses, themes and summaries; (3) the reviewer’s own personal experience, knowledge, preferences, expectations, and other aspects of their personal lives (Bezdicek and Yoon, 2016); and (4) the recommendation.

The second phase of coding and analysis involved shifting focus from our own semiotic processes as researchers to that of the reviewers. At this stage analysis focused on the reviews as interpretants, or the end result of the reviewer’s semiotic processes. Within each of the 4 categories identified in stage 1, subcategories were developed using Peirce’s typology of symbolic legisigns (rheme, dicent, and argument). Where appropriate, sentiment was coded using a combination of star ratings and, at the level of argument, whether the specific claim expressed a positive or negative disposition toward the book

The third phase analysed the reviewer’s meanings across the reviews within a book, where the grouping of reviewers emerged based on the different meanings they created. Reviewer’s personal statements were analysed across reviews to identify shared experiences constitutive of groupings of shared collateral experiences. Then, content statements and critiques were compared against personal statements to identify common interpretative frames among the reviewers. Pairings of similar content and personal statements were analysed and coded one last time to identify common interpretive frames comprised of similar expressed experiences that tended to appear with similar statements about a book’s meaning. Frames were categorized at the level of content/personal pairings rather than at the level of the reviewer to detect instances where multiple frames might be employed within a single review.

Analysis of coding at multiple levels yielded a rich understanding of user meaning within the individual frame. It produced grouping of reviewers among which the collateral experience seems to play a critical role. Table 1 shows a breakdown of different sources of framing that were identified within the reviewer/personal category, along with descriptions and examples of the types of arguments deployed within each frame.

Discussion and conclusion

This paper explored using semiotics as a methodology for identifying shared interpretative framing devices deployed by writers of UGOBRs who expressed similar collateral experiences. Semiotics provided a methodological basis for bridging the conceptual relations of signs with empirical data analysis and presented semiotics as a useful methodological framework for untangling the distinct semiotic processes of the authors of the books being reviewed, the reviewers writing reviews of the book, and the researchers’ own processes of interpreting the OGBRs. The analysis suggested that reviewer’s meaning created from reading, more specifically based on his or her own knowledge and experience, can be identified as a frame to interpret the sign object. These frames are grounded in similar collateral experiences and may be deployed to identify proto-communities of users in UGC contexts where formal communities and community building features may be lacking.

About the authors

Kyunghye Yoon Associate Professor, in the MLIS Program at St. Catherine University, 2004 Randolph Ave. St. Paul, MN 55105. She received her PhD from Syracuse University. Her research interests include information seeking behavior, human information interaction, accessibility and user-experience design. She can be contacted at kyoon@stkate.edu.
Adam Bezdicek Librarian at Pine Technical and Community College, 400 4^th St. SE, Pine City, MN 55063. He received his MA in English Literature from the University of St. Thomas and MLIS from St. Catherine University.  His research interests include data librarianship, the digital humanities, and information literacy instruction. He can be contacted at adam.bezdicek@pine.edu.

How to cite this paper