![]()
1
A STORY ABOUT STORIES IN STRATEGIC COMMUNICATION
Michael Dahlstrom
Stories, more formally known as narratives in the scholarly literature, represent a format of communication describing the thoughts and actions of specific characters over time and the cause-and-effect relationships surrounding their choices. As narrative represents a general format of communication, the actual construction of particular stories varies greatly, ranging from short testimonials describing an individualâs success after using a certain brand of toothpaste to great cultural literature exploring universal themes of human experience. Narratives abound in our communication environment, guiding the construction of news stories, driving the majority of entertainment media and packaging content for sharing over social media.
In addition to their ubiquity, narratives are also intrinsically persuasive. Because they present a specific experience from an individualâs point of view, narratives have no need to justify the accuracy of their claimsâthe story itself demonstrates the claim. As such, narratives are able to assign normative values to real-world objects without the need to construct an argumentâmaking it difficult to counter their claims (Bruner, 1991). Strategic communicators often take advantage of these persuasive aspects of narrative, using anecdotes, case studies or testimonials as evidence, or by using product placement within existing stories to infuse their brand or product with the influence of the narrative itself.
The intrinsic value of narratives to strategic communication make it particularly surprising that the scholarly examination of narrative persuasion is relatively recent. As a growing field of study, it should come as no surprise that two chapters in this volume address narratives. In this chapter, I will use narrative persuasion (as a narrative device) to reveal the history of narrative persuasion. In addition, however, I hope to portray a larger journey within this chapterâthe shift from a focus on message effects to the larger normative questions about when such message effects are appropriate to employâa journey I argue is important for both the individual scholar as well as the field itself. In Chapter 2 in this volume, LaMarre provides a detailed examination of the fractured streams of research that currently constitute narrative persuasion. Together, these two chapters should provide readers with a good understanding of narrative persuasion past, present and future.
Exposition: The Setting of Narrative Persuasion
At the turn of the previous century, the study of narrative resided in the domain of literary studies and criticism. Literary scholars from this period focused on the fictional novel to investigate the formal and functional aspects of narrative, including its internal composition, underlying meanings and differences to other forms of discourse (Kreiswirth, 2000).
Different traditions developed within this literary domain, each with a specific focus on the concept of narrative. The earliest group was the Russian formalists who stressed a distinction between a story and its discourse. Later, the American tradition focused on differences of point of view within a narrative. The French structuralists around the same time searched for a common underlying structure to all narratives (Culler, 1981). Later literary traditions also arose in Germany, the Netherlands and Israel with their own narrative foci and the eventual merging of the American tradition and French structuralism created the field of narratology (Kreiswirth, 1992). Yet, even with the subtle differences between traditions, narratives still remained within the domain of literary theory.
It wasnât until 1981 that narrative as a serious topic of study was introduced to the larger circle of scholarly discourse through a special issue of Critical Inquiry, titled âOn Narrativeâ (Bruner, 1991; Kreiswirth, 1992). In contrast to its literary roots, âOn Narrativeâ presented views of narrative outside the fictional novel, from scholars outside of the literary field, and emphasized narrative as human activity and meaning rather than its form and function (Kreiswirth, 2000). It was a shift from âwhatâ narratives are to âhowâ they work and âwhyâ they exist (Kreiswirth, 1992).
This shift in thinking led to the ânarrative turn,â an explosion of interest in narrative across many disciplines (Bruner, 1991). Narrative was no longer just for appreciation but had potential effects on how humans organize knowledge. Areas as diverse as economic theory and medicine that were dominated by scientific and logical approaches began to be interested in how narratives could influence their field of study (Kreiswirth, 2000). The result was the study of narrative spreading across cognitive science, pedagogy, policy analysis, sociology, experimental psychology, therapy, visual arts, music and natural sciences (Kreiswirth, 2000).
Two influential scholars helped integrate narrative within the communication and psychology fields in the 1980s. Within communication, Fisher argued that the prevailing view that people interact with the world in a rational or logical manner is false and should be replaced with a narrative paradigm, stating that people are essentially storytellers and base decisions on the coherence and probability of stories that already align with their existing beliefs and values (Fisher, 1984, 1985). Within psychology, Bruner empirically showed that narratives and rationality represent contrasting forms of cognitionâa paradigmatic pathway that controls the encoding of logic-based arguments and a narrative pathway that controls the encoding of situation-based exemplars (Bruner, 1986, 1991). This division aligned with the existing distinction between semantic and episodic memory (Tulving, 1972) but began to explore the pathways that created these different forms of memory and the factors that led to the activation of one over the other.
As narratives began to gain attention from scholars in these and other disciplines, other scholars resisted the inclusion of narrativeânarratives were deemed by resisters as distortions or even falsifications of the truth and thus not to be trusted as representative of anything meaningful. These anti-narrativists pointed to the fact that there are no empirical procedures of verification for narrativesâa narrative based on truth does not differ in appearance or structure from one that contains no truth (Kreiswirth, 2000). Such sentiments are still voiced in certain contexts (Katz, 2013).
As a result of resistance, narratives were sometimes slow to be permitted into the canon of accepted literature across disciplines. Brock, Strange and Green note that a prominent graduate-level textbook about the psychology of attitude change published in 1993 had no mention of narrative influence (Brock, Strange, & Green, 2002; Eagly & Chaiken, 1993). Furthermore, as much of persuasion and strategic communication borrow constructs and theories from psychology, it is not surprising that persuasion was equally slow to incorporate narrative into its textbooksâas late as 2002, there was still no mention of narrative as a persuasive influence in OâKeefeâs textbook of persuasion (OâKeefe, 2002).
It wasnât until Green and Brock published a seminal article in 2000 introducing the transportation-imagery model that narratives began to gain attention within persuasion (Green & Brock, 2000). The model, which is explained in more detail in LaMarreâs chapter, states that the primary factor controlling narrative persuasion is the degree to which an individual becomes absorbed into the world of the narrative; the authors labeled this phenomenon âtransportation.â They argued that the greater the degree of transportation, the more likely the individual will accept the persuasive message embedded in the narrative. This resonated with the earlier work of Gerrig, who asserted that, when highly absorbed in the narrative (transported), individuals become so focused that they lose access to certain types of real world knowledge (Gerrig, 1993) and are thus limited in their the ability to generate counterarguments to the content being consumed. Acceptance of persuasive information was therefore described as a default state of narrative exposure, where rejection was only possible with added scrutiny afterwards (Gerrig, 1993).
These studies and others that followed supported the idea that narrative persuasion is categorically different from traditional persuasion, as many important variables of traditional persuasion models, such as motivation or personal relevance, seem to have limited significance to narrative persuasion. For example, narratives with settings or characters high in personal relevance are rarely more persuasive than narratives without (Oatley, 2002; Wheeler, Green, & Brock, 1999). This is in stark contrast to the dual processing Elaboration Likelihood Model from traditional persuasion, wherein perceived relevance is a key determinant of deeper message processing. Even need for cognition, a trait-based measure with a long history of importance in traditional persuasion, has been found to have little to no role in narrative persuasion (Appel & Richter, 2007; Green & Brock, 2000; Wheeler et al., 1999).
Instead, new variables were conceptualized to account for differences in narrative persuasion, such as identificationâwhich is emotionally connecting to a character or situation because the narrative is perceived to be aligned with some aspect of the audienceâs personal history (Cohen, 2001)âand narrative realismâwhich is how tightly the narrative conforms to the internal rules defined by the world of the narrative itself (Busselle & Bilandzic, 2008).
One of the few situations that have been found to routinely disrupt transportation is when the persuasive intent of the narrative becomes salient (Moyer-Guse, 2008; Slater & Rouner, 2002). Such a realization draws the mind away from the narrative world causing the persuadee to begin scrutinizing the narrative information with more cognitive awareness. Many persuasive narratives, therefore, shroud their persuasive intent within the logic of the narrative world to avoid rejection based on a reaction to the intent rather than the narrative information itself.
In the last decade, the field of narrative persuasion has expanded rapidly. For example, health communication researchers began examining if narrative formats would be more persuasive in convincing individuals to engage in beneficial health practices (de Wit, Das, & Vet, 2008; Mazor et al., 2007). However, as narrative-focused research took off, inconsistencies within narrative influence soon became apparentâsometimes the research findings pointed to increased persuasion, but other times not. A meta-analysis of narrative studies within health communication specifically found no evidence that narratives had any systematic influence when compared to statistical communication, and the authors warned against the use of narrative persuasion until the phenomenon was better understood (Winterbottom, Bekker, Conner, & Mooney, 2008).
Explanations of these inconsistencies were often attributed, albeit vaguely, to a lack of âqualityâ in the narrative used (Green & Brock, 2000; Slater, Rouner, & Long, 2006); the argument was that well-written narratives should be more effective than poorly written ones. However, while that is likely true, there was little discussion as to what potential variables underlie this construct of quality (Kreuter et al., 2007) and no empirical results to back up such a claim. One of the few studies that empirically examined narrative quality came from an investigation of jury deliberation that found a break in coherence or chronology of a prosecuting summary resulted in fewer guilty verdicts (Voss, Wiley, & Sandak, 1999).
In summary, narratives have expanded as a scholarly inquiry from a focus on appreciation in literary theory to effects and influences in a wide range of otherwise divergent disciplines. The field of persuasion specifically has turned to narratives as a potentially influential format for strategic communication, leading to attitude and behavior change through mechanisms that seem to circumvent many of the traditional barriers to persuasion. Yet, this enthusiasm for understanding the effects of narrative formats of communication has only emphasized their underlying complexity and current lack of predictability. In the following section, I explore what may lie behind the apparent inconsistencies in narrative effects on persuasion.
Rising Action: Addressing Internal Variance Within Narratives
One potential source for the inconsistencies in the aforementioned research is the seeming expectation of a standard narrative effect, even though there is great variety in the possible stories that can be told and how they can be constructed. Searching for a standard narrative effect seems, in hindsight, analogous to the 1930s magic-bullet conceptualization of media effects. That original conceptualization portrayed the media as a monolith that had a single effect upon all audiences; it has since been rejected in lieu of much more nuanced models of media effects (McQuail, 2010). The same pathway to understanding beckoned for narrative research. What if variables internal to narratives can differ in their persuasive impact? If so, then ignoring within-narrative differences would result in a scattershot of potential results, similar to what was being found in narrative persuasion research.
It was this realization that led me to pursue a research agenda focused on exploring the heterogeneous nature of narrative structure and identifying variables involved in the construction of particular narratives that may influence their persuasive impact. In other words, I agreed with Greasser and Ottati that âit is a good time for psychologists to abandon some of the weary paradigm-ridden research projects, and investigate the properties of good storiesâ (Graesser & Ottati, 1995). To do so, I turned to the literature of discourse psychology.
Discourse psychology explores how the mind comprehends information and creates representations of that information in memory. The underlying premise is that the mental representation of information goes beyond the information itself and, as a result, individuals do not react directly to communicated information but, rather, indirectly through the meanings that they extract from that information (Bower & Rinck, 1999). Thus it is important to understand how specific variables within communicated information influence the creation of memory representations. While the form of information examined in discourse psychology can range from product instructions (Mills, Diehl, Birkmire, & Mou, 1993) to mathematical proofs (Fletcher, Lucas, & Baron, 1999), the majority of studies examine the comprehension of narrative. This focus exists because of the assumed correspondence between narrative processing and the processing of everyday experiencesâthis correspondence makes the comprehension of narratives more natural than other forms of communication (Graesser, Olde, & Klettke, 2002).
Once immersed in the literature, one structural narrative variable that caught my attention was causality. The causal network model (Trabasso & Sperry, 1985) claims that causal relations represent the glue that holds narratives together (Magliano, 1999); the model has served as a basis for many of the later models of discourse psychology (Magliano, 1999; vandenBroek & Gustafson, 1999). Studies using the causal network model identify causal relations using the logical criterion of necessity (Mackie, 1980), which states that element A is considered causally related to element B if element B could not occur in the narrative without element A. It has been found that narrative statements with a greater number of causal connections are easier to recall, are recalled faster, receive higher ratings of importance to the narrative (Trabasso & Sperry, 1985) and result in faster reading times (Trabasso & Suh, 1993) than narrative statements with fewer causal connections. These causal connections can also form a causal chain, which is a linked chain of statements connected by their cause-and-effect relationships that spans the entirety of the narrative. Statements on the causal chain have similarly been found to result in greater recall and importance ratings, even after accounting for the effects of absolute number of causal connections (Trabasso & Sperry, 1985).
My growing interest in causality led me to examine if the within-narrative variance of causality (borrowed from discourse psychology) would influence persuasive outcomes and possibly improve the predictability of narrative persuasion. Because this factor had not yet been a focus within the field, I wanted to see how previous narrative stimuli had been structured with regard to causality. I therefore contacted multiple authors who had conducted research into narrative persuasion studies and asked to see their narrative stimulus materials. What I discovered was that persuasive information had been generally placed within narratives at breaks in the action, where characters could pause to discuss unrelated (and therefore easy to manipulate) persuasive statements as an aside before returning to the meaningful action of the narrative. This meant the persuasive content was often not causally relevant to the full narrative. Because the influence of the persuasive elements at these unrelated locations in the narrative had not yet been compared to their influence if they were causally relevant, I conducted and published three effects studies exploring these questions.
Study 1
The first study aimed to determine if persuasive information embedded into a narrative in such a way as ...
