ABSTRACT Title of Dissertation: THE FEELING OF PERSUASION: A COGNITIVE RHETORICAL ACCOUNT OF THE EMOTIONAL APPEAL Cameron Mozafari, Doctor of Philosophy, 2019 Dissertation co-directed by: Professor Michael Israel, Department of English Language and Literature Professor Vessela Valiavitcharska, Department of English Language and Literature Emotion often takes the back seat in contemporary rhetorical investigation, as emotions are treated as subjective reactions rather than the result of deliberate forms of argumentation. In classical antiquity, rhetorical training for emotional persuasion required students not only to learn what sorts of arguments could move their audiences but, more importantly, how that movement was composed linguistically and psychologically. Yet as history progressed and disciplines branched, the formal study of language and cognition separated from the study of rhetoric, resulting in a conceptually stunted understanding of the emotional appeal. This dissertation returns to classical questions and theories of emotional persuasion but does so with insights from contemporary emotion science and cognitive linguistics. Emotion is understood as neither purely physiological nor purely conceptual but rather is understood as embodied conceptualization grounded in culture-specific scripts. The dissertation lays out a model for understanding how non-emotive language links up to emotion activation through the introduction of the theater of the mind model, an expansion on the stage model of Cognitive Grammar. It then traces three strategies for arousing and controlling audiences? emotions from classical rhetorical theory: the enthymematic activation of emotion concepts, the enargeiac amplification of emotion events, and the mitigation of potential threats so as not to excite emotions. Analyzing discourse from politics, fundraising letters, and college student writing, this project argues, contrary to popular opinion, that emotional appeals are not antithetical to reason but instead very much dependent on reason, in that they act as grounds for arousing and guiding inferences in predictable ways for rhetorical purposes. THE FEELING OF PERSUASION: A COGNITIVE RHETORICAL ACCOUNT OF THE EMOTIONAL APPEAL by Cameron Nazer Mozafari Dissertation submitted to the Faculty of the Graduate School of the University of Maryland, College Park, in partial fulfillment of the requirements for the degree of Doctor of Philosophy 2019 Advisory Committee: Professor Michael Israel, Co-Chair Professor Vessela Valiavitcharska, Co-Chair Professor Jeanne Fahnestock Professor Jessica Enoch Professor Linda Coleman Professor Damien Smith Pfister, Dean?s Representative ? Copyright by Cameron Nazer Mozafari 2019 Dedication To Mom, Dad, Edward, and Alf. ii Acknowledgements The ideas of this dissertation sprung from two courses: Vessela Valiavitcharska?s pathos and argumentation seminar and an independent study with Michael Israel on emotion and language. Thank you, first and foremost, to my tireless dissertation directors, Michael and Vessela. I feel lucky to have been blessed with two brilliant and nurturing mentors. Michael?s (at times heavy) criticisms helped me to see my contributions, and, when paired with his genuine encouragement, helped me to become a better scholar. Vessela?s careful reading of my work raised questions I would have never thought of and connected dots I wasn?t even seeing. Her endless knowledge is rivaled only by her endless patience. I couldn?t have asked for better dissertation directors, and I am so thankful for them. I owe a great deal of thanks to my dissertation committee. Jeanne Fahnestock?s support for this dissertation project and encouragement on my scholarly work in general has been truly motivating. And the conversations about politics in the Academic Writing Office I had with Jessica Enoch the days and weeks after the 2016 election led me to investigate reason and emotion in politics with a focus on what it means to feel right. I am also in debt to her for comments on a conference paper that developed into Chapter 5, which draws on research the Academic Writing Program was conducting while I was serving as an assistant director. Linda Coleman, who graciously allowed me to sit in on her discourse analysis course, influenced the way I understood the role of emotion in contemporary American political discourse. I am also appreciative of Damien Pfister for offering to be my outside reader for this work and providing feedback during the defense. A special thanks should also go to Jane iii Donawerth. Jane taught me not to pick fights and to treat academic engagement as a conversation rather than an argument. Jane also provided words of encouragement after I had a complete hard drive crash in 2016, losing three chapters of my dissertation and having to start all over again. On that note, I would like to thank my dog, Alf, who not only broke the mold when he came into my life but also broke my laptop. The dissertation was much better the second time around as a result. Along with Alf, I would like to thank the rest of my family: my brilliant and caring parents, Brenda and Davoud, my brother Bobby, my aunt Parivash, my grandmother Mamani, and all of my aunts, uncles, and cousins who have provided me support, love, and sometimes food. This dissertation would not have been possible without the encouragement and emotional support I received inside and outside of the walls of Tawes Hall. I want to give a special thanks to Brian Davis, whose weekly accountability meetings and sporadic dinner outings offered me a space to recalibrate and recharge. I am fortunate enough to have had amazing friends during my graduate education: Andy Black, Oliver Brearey, Jesse Brooks, Martin Camper, Kim Calder, Dustin Chac?n, Chip Cobb, Brooke Feichtl, Shaun Gannon, Lew Gleich, Danielle Griffin, Maria Gigante, Sarah Bonnie Hummud, Kayla Harr Doucette, Nabila Hijazi, Joyce Hsiao, Lindsay Dunne Jacoby, Heather Lindenman, Elizabeth Ellis Miller, Lindsey E. R. O?Neil, Ruth Osorio, Domingo Ruiz, and Katie Bramlette. My sincere gratitude to Terp Wushu and my martial arts brothers and sisters: Jason Liu, Dennis Shyu, Janet Hsu, Tina Zhang, Henry Hong, Justin Benedik, Sunny Zhang, Jeff Lui, Allen Chung, Sherry Feng, and Chester Lee. I lost a good friend in Jacob Ekstein, who passed iv before this dissertation was completed. Jacob was always a source of encouragement, and I miss him dearly. I must also include an earnest thank you to Scott Eklund and Marybeth Shea. Scott is surely one of the kindest people I have met and is one of the only people at Maryland who can steal your afternoon talking about comic books. Marybeth is perhaps the hardest working person I have met in my academic career. Our discussions about pedagogy, science communication, and cognitive science were formative in the direction I have gone in my scholarship. Her generosity mentoring me in science communication will forever put me in her debt. Finally, I would like to thank my loving partner, Edward Phillips. Edward was there through the entire thing, was there to listen to my half-baked ideas, was there to read sections of this dissertation as they were strewn out over our couch, was there to distract me with music, and was there to support me when I needed it most. Throughout all the stress, doubt, and boredom, Edward was there. I couldn?t have done any of this without your love and support, Edward. Thank you. If I could play you the movie of this project, everyone listed in my acknowledgments would be listed as writers, directors, producers, and actors of the dissertation. Without them, the set would have been empty, the acting stilted, and the script nearly 650 pages. What you are reading is as much their contribution as mine. v Table of Contents Dedication ..................................................................................................................... ii Acknowledgements ...................................................................................................... iii Table of Contents ......................................................................................................... vi List of Tables ............................................................................................................. viii List of Figures .............................................................................................................. ix Chapter 1: Toward A Cognitive Rhetorical Model of Pathos ...................................... 1 1. Introduction: Rhetoric and Emotions ................................................................ 1 2. The Domain of Emotion ................................................................................. 10 3. Definitional Difficulties .................................................................................. 11 4. Emotions as Events ......................................................................................... 14 5. The Encoding and Construal of Experience ................................................... 23 6. The Aristotelian Pathetic Appeal .................................................................... 28 7. Cognitive Rhetorical Model of Emotional Appeals........................................ 35 8. Forecast of Chapters ....................................................................................... 41 Chapter 2: The Imaginative Mind ............................................................................... 46 1. Introduction: The Place of Imagination in Rhetoric ....................................... 46 2. Images and Argumentation ............................................................................. 51 3. Dimensions of Imagery ................................................................................... 55 4. The Theater of the Mind ................................................................................. 62 4.1. Level 1: Force-dynamic Relations .......................................................... 64 4.2. Level 2: Thoughts and Feelings .............................................................. 66 4.3. Stage 3: Social Interactivity and Value Alignment................................. 69 4.4. Stage 4: Persuasive Strategies ................................................................. 84 5. Conclusion: Returning to the Euphantasiotos ................................................. 94 Chapter 3: Emotive Reasoning in Political Rhetoric ................................................ 105 1. Introduction: A Reason to Feel ..................................................................... 105 2. The Pathetic Enthymeme .............................................................................. 109 3. Frame Metonymy .......................................................................................... 116 4. Figurative Reasoning in Pathetic Enthymemes ............................................ 119 5. Afraid of Breakfast: Involved Movement Through and Across Frames ...... 131 6. The Constitutive Grammar of Fear ............................................................... 142 7. Rising Homicides: Metonymy in Trump?s RNC Speech ............................. 155 8. Conclusion .................................................................................................... 158 Chapter 4: The Function of Enargeia in Emotional Appeals ................................... 163 1. Introduction: Emotion in Charity Appeal Letters ......................................... 163 2. The Appeal Letter Corpus ............................................................................. 167 3. Crafting Intimacy .......................................................................................... 170 4. Crafting Enargeia through Focus and Positioning ........................................ 187 5. Imperative Imagine, Irreality, and Enargeaic Engagement .......................... 195 6. The Identification Functions of Vividness in the Historical Present ............ 210 7. How Vivid is Vivid? ..................................................................................... 216 Chapter 5: Defensive Stance in the Self-Assessments of Student Reflective Writing ................................................................................................................................... 225 1. Introduction: Rhetoric in Reflective Writing ................................................ 225 vi 2. Corpus Construction and Methods of Analysis ............................................ 228 3. Mitigation, Hedging, and First Year Writing................................................ 232 4. Analyzing Reflection .................................................................................... 240 5. I Think As An Emotional Appeal.................................................................. 247 6. I Think Across the Semester ......................................................................... 257 7. Conclusion .................................................................................................... 260 Concluding Remarks ................................................................................................. 263 Appendices ................................................................................................................ 271 Appendix A ........................................................................................................... 271 Appendix B ........................................................................................................... 277 Bibliography ............................................................................................................. 278 vii List of Tables Keywords of the Appeal Letters Corpus ????????????????..171 Frequent p-frames that ground????????????????????..173 Top 8 markers of hedging??????????????????????.237 Frequency of Reflective Moves????????????????????.240 Keywords of Assessment Sub-Corpus Measured Against Full Corpus?????244 viii List of Figures Core affect and corresponding prototypical emotion episodes?????????20 Dynamic landmark-trajector alignment?????????????????...58 A CG diagram profiling the flooded region????????????????.59 Diagram of Gore's "never again" utterance????????????????.61 Direct and indirect construal of thoughts and feelings????????????67 The Stance Triangle?????????????????????????.74 Example of frame metonymic emotion arousal of panic???????????.86 Construals of force-dynamic events that denote potential fear and dread????...91 An Infographic from All In with Chris Hayes??????????????...120 A representation of being afraid of breakfast???????????????134 The Come frame??????????????????????????147 A conceptualization of the progressive coming??????????????.148 A schematic of the endoxic premise??????????????????..150 Image of horse tripping, provided by Four Corners Equine Rescue??????.163 The Fictive Common Ground of the Appeal Letter????????????.178 Depictions of Good Situations and Bad Situations in L1??????????183 Moving from Bad Situations in the Present to Good Situations in the Future??184 Sierra Club Wolf Experience Example?????????????????.208 Frequency of Emotion Words in the Appeal Letter Corpus?????????.221 ?I was successful in creating a summary?????????????????249 ?I think I was successful in creating a summary??????????????250 I think across the semester??????????????????????257 ix Chapter 1: Toward A Cognitive Rhetorical Model of Pathos 1. Introduction: Rhetoric and Emotions Tasked with providing a definition of rhetoric that is broad enough to capture persuasive communication across human cultures and across other species, the classical scholar George A. Kennedy (1997) settles on calling rhetoric ?a form of emotional and mental energy? (3). Certainly, emotion has been a prominent part of the Greco-Roman rhetorical tradition since Gorgias?s Encomium, wherein metered speech is said to be able to bring ?fearful shuddering,? ?tearful pity,? and ?grievous longing? to its hearers?inflicting pain on the body and soul alike with words alone (9). Nearly three hundred and twenty years later, the author of the Rhetorica ad Herennium writes, in his list of figures of speech, that ?[i]f we use Apostrophe in its proper place, sparingly, and when the importance of the subject seems to demand it, we shall instill in the hearer as much indignation as we desire? (4.15.22). But how does apostrophe arouse emotion, O anonymous author of the ad Herennium? And why is it, Gorgias, that metered words can tear at our hearts? While rhetorical theory has sought to explain the workings of figurative language and argumentative reasoning, neither classical nor contemporary theorists much bother to address the explicit roles emotion plays in persuasive discourse. This project seeks to understand how language functions to arouse emotions in others. It takes as its central object of study the verbal emotional appeal, a kind of argumentative act that seeks to move an audience from feeling one way to feeling another for some intentional purpose. Verbal emotional appeals appear everywhere, from love songs and advertisements to political speeches and bumper stickers. When 1 Lloyd Bentsen said to Dan Quayle in the 1988 Vice Presidential Debates, ?Senator, I served with Jack Kennedy. I knew Jack Kennedy. Jack Kennedy was a friend of mine. Senator, you?re no Jack Kennedy,? the audience erupted into impassioned cheers. What made Bentsen?s line so powerful? When Patricia Smith, mother of Sean Smith, one of the victims of the 2012 Benghazi attack, took the stage at the 2016 Republican National Convention, the audience in the room was visibly weeping, yet I was completely unaffected. Why does this mother?s rhetoric bring her intended audience to tears while having no emotive effect on her unintended audience? The dissertation analyzes and advances a program for studying the emotional appeal through methods borrowed from emotion science and cognitive linguistics, which, due to disciplinary divides, may be foreign to rhetoric. Rather than understanding emotions as subjective reactions to events, the emotion science demonstrates how emotions are predictive constructions of the world to be experienced, born out of imagination and mental simulation. It posits that emotions are embodied rationalizations that emerge from situated categorizations and bodily attention. While Aretha Franklin?s demand for respect and the panicked words of a student requesting an extension are very different in content, they both rhetorically aim to bring about an emotional appeal, a momentary and embodied event, brought about through intentional discourse, wherein an audience notices an affective experience. These momentary events are brought about through patterned language and can cause experiencers to change their attitudes and evaluations about some topic at hand. As such, emotional appeals were of use and of concern to classical theorists of rhetoric 2 and should be equally important for contemporary analysts of discourse and argumentation, as they can shape ideas, reinforce cultural attitudes, and encourage real world action. Insights from Cognitive Linguistics?in particular the frameworks of Cognitive Grammar and frame semantics?help to show how words and patterns of words can systematically prompt rich and embodied mental simulations of the world, ones that are subjectively construed but motivated by the search for intersubjectively shared meaning. Emotional meanings arise in discourse when there is enough shared information between the elements in the current discourse space and the component parts of an emotion?s script?its actors, actions, focus, responsibilities, sequence of events, moral evaluations. The goal of using an emotional appeal is not only to bring an emotion to mind but, by prompting audience members to play out and experience the story in their minds, to shift thoughts and change attitudes. By doing so, rhetors can move audiences to act?to march for a cause, to mentally and physically distance themselves from others, or to bring people together in communion. In this way, emotional appeals have real world consequences; they move us both figuratively and literally. Bringing a fresh understanding of emotion and a dynamic method of connecting language and embodied cognition allows for the construction of new models of studying the emotional appeal, falling under the program of the cognitive rhetorical model of pathos. Moreover, as I argue in this dissertation, this new model reflects many aspects of the classical Greek model of pathos put forth by Aristotle in the 3 fourth century BCE1 and elaborated in various ways by later Peripatetic and Roman rhetoricians. More than just a rehashing, my model of the emotional appeal adds to the classical rhetorical pursuits, providing ways to understand how the experiences, thoughts, and reasoning processes intrinsic to ordinary cognition are intrinsic to the activation of the meaningful simulations that emotion concepts conjure. The cognitive rhetorical model of pathos provides a model of conceptualization and a framework for talking about what aspects of conceptualization get encoded in linguistic constructions. This differs from so-called ?bag of words? approaches to analyzing emotion in language, popular in computational methods of sentiment analysis, which assume that emotion arousal is tied to the use of emotion words or figurative language. Furthermore, the model of the rhetorical model of pathos presented in this dissertation differs from some widely held classical and contemporary ideas of the emotional appeal?likely stemming from Stoic influence?that posit that emotion is a perturbation of reasoning. Instead of viewing words and figures as the sources of emotion, it understands lexical choices and figuration as ways of altering conceptualization, foregrounding and amplifying particular conceptual elements through patterns of language. In this way, it pays attention to the ways language systematically prompts complex conceptual processes in our imaginations in order to 1 Pathos, of course, was a concept before Aristotle, though it was not studied formally until Aristotle. Plato discusses the use of emotional appeals in the Phaedrus, citing Thrasymachus?s use of appeals to pity (267c9-d4). Thrasymachus is said to have written a handbook of commonplace appeals to pity? Eloi?for courtroom orators, but this handbook has been lost. Like Plato, Aristotle was critical of this commonplace appeal tradition, as can be inferred from his criticism of it in Rhetoric 3.1.7. 4 allow emotional meaning to emerge. Acknowledging that rhetors and audiences alike are master imaginers, euphantasiotoi, I ask what linguistic moves rhetors use to guide imaginative reasoning to arouse emotions and what imaginative cognitive operations audiences undergo to construe those emotive intentions. As rhetoricians once again broach the question of how emotional appeals work, it might be best if they be broached with some hindsight. The idea of the emotional appeal can be traced back to before Aristotle in Western rhetorical tradition. When treating language and emotion, classical rhetoricians sought to understand how certain utterances moved beyond mere emotional expression in order to arouse emotions in others, and thus move an audience. Histories of oratory overflow with examples of such emotional appeals, from speeches declaimed by charismatic kings to poems written by pining lovers?language intended to viscerally churn passions in hearers. Yet, while appeals to emotion are easily recognizable, contemporary rhetoricians do not isolate and analyze formal aspects of language or thought that play a part in arousing emotions in audiences. And while Lynn Worsham (1998), a contemporary critical emotion studies scholar, claims that rhetoricians? ?most urgent political and pedagogical task remains the fundamental reeducation of emotion? (216), a language for talking about emotion is, in her words, ?beyond the horizon of semantic availability? (223). As a result, rhetorical scholarship on emotion has abandoned describing formal aspects of emotional persuasion in order to fixate on projects that speak to critical theory, cultural studies, and revisionist historiography.2 2 For example, Gross?s (2006) work on rhetorical emotion, Crowley?s (2006) work on civil discourse, and Miller?s (2008) work on emotion and historiography. These works provide keen insights into the 5 Can rhetorical scholarship tell us anything about emotion?s role in the so-called ?hermeneutics of persuasion,?3 the inner workings of the persuasive act? I believe it can. Like Worsham and a growing number of scholars, I believe contemporary rhetoricians have neglected the role emotion plays in persuasion and that a return to emotion is a beneficial move. Our ability to register and criticize emotional appeals may have overpowered our ability to understand and craft them. Pathos has become the ugly sibling of logos, whose preference is reflected in contemporary pedagogical material in composition and argumentation. In her survey of contemporary college composition textbooks, Gretchen Flesher Moon (2003) shows that composition textbooks often give pathos less critical attention and less space than logical appeals, the emotional appeal often only acting as a straw-man presented in order to reinforce the soundness and preference of logical appeals. In a follow-up analysis, Tim Jensen (2016) has looked at 25 composition textbooks? seven of which are updated versions from Moon?s initial study?to find little change in the treatment of emotions despite emotion?s recent growing interest in composition studies. Systematically, composition textbooks fail to define what emotions are and what an emotional appeal is, overlook how different emotions require different ways we understand and can use rhetorical theory to analyze emotion but do not attempt to produce more analytic methodologies for analyzing emotional appeals. 3 This term is first used by the linguistic pragmaticist Marcelo Dascal and the rhetorical scholar Alan G. Gross in ?The Marriage of Pragmatics and Rhetoric.? It is revisited and expanded on by the linguistic pragmaticists Yameng Liu and Chunshan Zhu in their ?Rhetoric as the Antistrophos of Pragmatics.? The term helps to distinguish one aspect of the study of rhetoric, namely the study of how persuasion works, from other aspects, such as the role of women in the history of rhetoric or the ways in which politicians are products of socially constructed discourses of power. 6 strategies for arousal, and are quick to warn readers to be weary (rather than mindful) of emotional appeals. Moreover, teaching students to avoid emotional appeals may be doing them a disservice, as Brian Jackson (2011) argues, since political discourse has shown that ?any appeal without some kind of emotional salience will fail? (475). There is pedagogical and civic exigence in studying the production of emotion-arousing reasoning processes. Jackson, who, like Moon and Jensen, surveys pedagogical material in rhetoric and composition, concludes that despite the need, rhetoric and argumentation textbooks tend to treat pathos ?in about two pages, emphasizing only that emotional strategies lend ?presence? to an argument by creating immediacy and emotional impact? (485). While Jackson?s criticism focuses on textbooks and does not address work from critical emotion studies by scholars like Lynn Worsham (1998, 2003), Shari Stenberg (2011), and Laura Micciche (2007), Jackson?s point is still relevant?that an analysis of the emotional appeal grounded in understandings of individual and social psychology, formal features of language, and patterns of argumentation has yet to be undertaken.4 Furthermore, Jackson warns that, if civically 4 In briefly addressing work done by Douglas Walton, an influential scholar in argumentation theory who has written three full-length books on emotion in argumentation (2000, 1997, 1992) at the time that Jackson wrote his piece and even one since (with Fabrizo Macagno 2014), Jackson points to an attitude that Walton and other argumentation scholars have shown when evaluating emotional arguments; namely, that even if they work in persuading, they are often times, in Walton?s words, ?informal fallacies? or even outright ?irrelevant? lines of reason (Walton 1997, 151). I agree whole heartedly with Jackson when he appraises current argumentation scholarship as missing the point of emotional arguments. The point is not how rational emotional arguments are but how effective they are, how they work, what their structures tend to be. These are the things that beg for our scholarly attention. 7 engaged pedagogues fail to understand how language is imbued with emotion, ?we will not understand public argument, nor will we know how to teach students how to make effective public arguments? (475). Pointing to politically motivated work on framing by linguists George Lakoff and Geoff Nunberg, Jackson asserts that rhetoricians must pay attention to what cognitive science is telling us about language and emotion. Though cognitive scientists and linguists rarely address the arousal of emotions for persuasive purposes?the chief interest of rhetoricians?they may provide empirical grounding that can, as Jeanne Fahnestock (2005) has suggested, offer a productive explanation of how rhetoric happens in the brain (161). To put it differently, rhetoricians can learn more about how rhetors construct emotionally persuasive arguments by addressing research findings on emotion and language from the linguistic and cognitive sciences. An approach to understanding rhetorical persuasion, cognitive rhetoric applies ideas and research findings from cognitive science and cognitive linguistics to classical issues commonly addressed by rhetorical scholars. For example, Mark Turner?s (1998) interdisciplinary recasting of the category of rhetorical schemes in terms of cognitive semantics and construction grammar attempts to demystify the effects of schemes by highlighting the mental operations schemes induce in listeners. While Turner draws mostly on classical rhetorical figurative theory, Todd Oakley (1997) addresses issues of the New Rhetoric project, and in doing so, Oakley shows how methodologies from Cognitive Linguistics can illustrate the linguistic ground for the ?contact of minds? necessary to cue attention and craft ?presence? (66). Yet, little work has been done within the cognitive rhetorical framework to address emotional 8 appeals, even with prominent Cognitive Semantic work on the conceptual structures of emotions (e.g. Lakoff 1987). And no cognitive rhetorician nor cognitive linguist has addressed the classical rhetorical strategies of crafting emotional appeals found in Aristotle, Cicero, and Quintilian, let alone pseudo-Longinus, Hermogenes, or lesser known Greek rhetorical handbooks of the Roman era, such as that of the Peripatetic author of the Anonymous Seguerianus and Apsines of Gadara.5 Needless to say, the scope of classical rhetorical material, elusive to even trained rhetoricians, still could be useful for understanding emotional persuasion, with the aid of careful cognitive rhetorical analysis. Interfacing classical rhetoric with contemporary emotion science, cognitive linguistics, and linguistic pragmatics, this book brings neighboring fields to a shared focus on the emotional appeal. By combining cognitive and classical rhetorical approaches to emotion, my project addresses questions central to contemporary rhetoric: What is the relationship between patterns of language and communicating emotion? Between communicating emotion and intentional persuasion? Writing this in a post-2016 American political climate, where emotional arguments are routinely weaponized to polarize political affiliations, these questions are more important than ever. To echo both Lynn Worsham and Brian Jackson again, our current social situation demands that we understand and teach the ways that rhetoric shapes our emotions. To do so, we will need a method mindful of both the situational linguistic and the emotive aspects of rhetoric. And that is what I hope this project provides. 5 See Two Greek Rhetorical Treatises from the Roman Empire, trans. and ed. Mervin R. Dilts and George A. Kennedy (Leiden: Brill, 1997). 9 2. The Domain of Emotion The modern treatment of emotion in language and rhetoric tends to make strong, sometimes unhelpful generalizations about the role emotion plays. Both Roman Jakobson (1960) and Kenneth Burke (1968) have expressed the idea that the act of arousing emotion is implicit to all language usage. Jakobson claiming that emotion ?flavors to some extent all of our utterances, on the phonic, grammatical, and lexical level? (354), and Burke that all rhetorical utterances come with ?emotional curves? (41). While there may be some truth to these beliefs, sweeping statements shroud more than they reveal. What do we lose conceptually from believing that emotion is in everything? Or, to put it differently, if emotion is in everything and present on every level, how does that help us understand how it works, or why it feels so different in different places and at different times? It is the general statements about emotion and language?that all language is emotive or that all rhetorical acts shape emotion?that allow us to overlook both the role of language and emotion.6 We may think that, following Plato?s example, we ought to begin by defining our terms. Rhetoricians are quick to provide definitions of rhetoric: Rhetoric is the faculty of observing in any given situation the available means of persuasion, or rhetoric is the art or talent by which discourse is adapted to its ends, or there is the 6 To this point, it is worth noting that this project does not directly engage with the affect theoretical work in contemporary rhetorical theory. I do this not out of ignorance of the works of rhetoricians like Thomas Rickert, Joshua Gunn, and Jenny Rice but because I believe we see the role of emotion and affect working differently in our projects. 10 definition that began this chapter, that rhetoric is a form of emotional and mental energy. There are plenty of well-known definitions of rhetoric, and perhaps a more detailed and precise one is in order for the twenty-first century. Nevertheless, rhetoricians don?t tend to have as definite an understanding of emotion. To answer what we would conceptually lose from believing that emotion is in all aspects of language, we must first start with a good definition of the word emotion. 3. Definitional Difficulties In the cognitive sciences and in modern psychology in general, coming to an acceptable definition of emotion has become something of a holy grail. In 1884, William James argued that emotions are mental states that arose as the result of bodily change, each bodily change constituting a new emotion event.7 Before him, Darwin (1872) argued the exact opposite?that emotions are natural kinds, inherited by evolution, that presented themselves in expressive behavior, such as facial expression, perspiration, and activity in the heart and other organs prototypically associated with certain feelings. Since Darwin and James, various theses have been presented. In 1927, Walter Cannon argued that emotions co-occur with their expressions. In the mid-twentieth century, Richard Lazarus argued that emotions are the result of conscious appraisals and cognitive activity. In 1962, Stanley Schachter 7 James?s view has a history of being misread as a reduction of emotions to bodily changes. This reading is taught as the ?James-Lange theory of emotion? and was not proposed by James but by the American philosopher John Dewey. For more discussion of James?s argument and its misreading, see Lisa Feldman Barrett?s How Emotions are Made, 161-162. 11 and Jerome Singer proposed a model that merges appraisal and biology. Since, there have been countless redefinitions of emotion, each emphasizing different aspects of the biological, behavioral, and cognitive features and leading to an internal conflict in the sciences over what the meaning of the word ?emotion? is to those who are tasked with studying it. And this is just in biology and psychology, before the issues of cultural variation were raised by anthropology. I bring up this truncated and simplified history of the definition of emotion not for narrative reasons but to highlight a point that is often unheard in the humanities: we may not be beginning on the right terms. As the historian Thomas Dixon (2012a) reminds us, the word emotion did not even appear in the English language until the seventeenth century, first penned in John Florio?s translation of Montaigne?s Essays. Florio himself apologized for the ?uncouth termes? he introduces from French, one of which being the word ?emotion,? which in this sense referred to physical disturbance and bodily movement??public emotion??and later a physical agitation in the body, tying the term back to Stoic roots (see Dixon 2012a, 376). While it may seem easy to remedy this problem of definition by simply crafting a definition and then holding claims relative to that definition?this is indeed what the emotion sciences are and have been doing for the past century or so?the problem of definition runs deeper, and this problem causes complications for studying argumentation that depends on a shared definition of emotion. Aristotle asserted that there were three questions one must address before one could understand the emotional appeal. The first question asks what an emotion is; the second asks who is feeling the emotion and in what context (?against whom?); 12 and the final question asks ?what sort of reasons? people give for feeling the way they do when in the throes of their passions (Rhetoric 2.2.9). Over two millennia later, the psychologist Caroll E. Izard (2010) conducted a widely cited and revealing interview with 34 leading emotion scholars, asking basic questions about definitional terms in the field?the first, and most important of which being the same one that Aristotle begins with: ?What is an emotion?? (364). It ought to come as no surprise that, among the 34 emotion scientists interviewed, there was little agreement on a definition of emotion. Nevertheless, Izard?s study shows that there was ?moderate to high agreement on the structures and functions of emotion? (367). In order to reconcile the seeming contradiction that most scientists don?t have a consistent definition of emotion, that they find the term too ambiguous to be useful in science (367), that most believed that emotions were motivational in human behavior, and that there is a high agreement on the structures and functions of emotion, Izard proposes that perhaps ?the idea that emotion (or a specific emotion) is motivational does not necessarily conflict with the idea that it has particular neural substrates and is informational, social, and relational, and monitors or assesses the significance of events, and may include appraisal processes and other forms of cognition? (368). In other words, various aspects of what emotion scientists study?the neural substrates, appraisals, relational and social factors and so on?could perhaps come together to comprise a unified model of emotion, even in the absence of an agreed upon definition. Many have taken Izard?s work to show little more than a problem with the semantics of the word ?emotion.? The cultural anthropologist Geoffrey White (2010) 13 suggests that Izard?s conclusion that emotions may have a unified model should not be surprising. Although emotion scientists study different phenomena and sometimes hold contradictory technical definitions of emotion, they agree that they are studying emotion, a concept that is common to non-specialists and specialist alike (376). This is because both the non-specialist and the specialist seemingly rely on the same prototypical model of emotion?i.e., that emotions are bodily and psychological phenomena that affect how we feel. Anna Wierzbicka (2010) also weighs in on Izard?s research, criticizing Izard for trying to define emotion by asking English- speaking emotion scientists questions in English. Following a line of argument from earlier work (Wierzbicka 1999), Wierzbicka claims that, without a metalanguage to discuss the meaning of any word, any definition will fall into a circular trap that only further strengthens the bond between a word and some ethnocentric folk psychology. In a publication on the uses of the term ?emotion? as a technical term in the sciences, Dixon (2012b) asserts that, since its introduction into the English lexicon in 1884 as a scientific term to be used by scientists, the word ?emotion? has been a ?keyword in crisis,? a word that hasn?t stopped being defined and redefined to capture aspects of the words that it was meant to replace: sentiment, feeling, passion, morality, and so on. From its onset, emotion has had a problem with definition. As such, the emotional appeal has had a problem with definition. 4. Emotions as Events In contemporary rhetorical stasis theory, when there is a problem in definition, one can try to resolve this problem by reexamining the phenomena being defined. Neuroscientists and 14 psychologists who subscribe to the so-called ?basic? emotion theory?a theory of emotion that believes that all emotions have unique properties that are hardwired in the brain or mind?ground their definitions on a search for physical emotion essences.8 For example, the emotion scientist Paul Ekman believes that contortions of the face are essential markers for an emotion?s biological existence. Jaak Panksepp believes that the essence of an emotion is in the neurological circuitry of the brain. To date, no such emotion essences have been found, despite research in localized parts of the brain, like the amygdala, thought to be essential for anger and fear processing. The emotion scientist Lisa Feldman Barrett (2006) has argued that emotion science has largely ignored the ?emotion paradox,? the problem that, if bodily phenomena like amygdala activation are meant to accurately predict how a person is feeling, what the body does and how a person is feeling don?t always match up. Sometimes people report feeling angry without amygdala activation, for example. If the basic emotion theory is correct and emotion is instantiated in the body through hardwiring in the brain, either people misrepresent their emotional experiences (e.g. saying they are angry without amygdala activation) or amygdala activation can?t be taken to be a sign of experiencing anger. Studying neurological phenomena, like amygdala activation, or physiological phenomena, like facial contortions taken as indicators of emotion in order to prove the existence of an emotion, may be begging the question. Grounding the definition of an emotion like anger in amygdala activation has been 8 See, for example, the work of Paul Ekman, Emotions Revealed: Understanding Faces and Feelings. In Ekman?s basic emotion theory, emotions are marked by universal facial physiological change. For a textbook treatment of the neuroanatomy of emotion from a different basic emotion perspective, see Jaak Panksepp, Affective Neuroscience: The Foundations of Human and Animal Emotions. 15 shown to be problematic from an empirical perspective as well. While it is true that many instances of anger exhibit amygdala activation, a meta-analysis of all publications on amygdala activation and anger, conducted by Kristen Lindquist and her colleagues, has shown that people who experience anger exhibit incredible variation. It is only when this variation is statistically averaged and projected onto an idealized average person that the amygdala appears to be the seat of anger. Amygdala activation, then, is not a necessary and sufficient condition for the experience of an anger episode. On top of this, meta-analyses of facial expressions have also failed to yield essential components of an emotion?s expression.9 Recent recognition that the so-called emotive regions of the brain are not essentially emotive has become apparent to some of the most prominent neuroscientists of emotion. Joseph LeDoux, whose landmark 1996 book The Emotional Brain, trumpeted the claim that fear and anger are located in the amygdala, has more recently backed away from grounding emotions in neurobiology.10 It is a mistake, LeDoux (2015) now acknowledges, to think that neuron firing in the amygdala?a neurological behavior?is the same thing as the emotion of fear. For starters, the amygdala is responsible for many behaviors that have nothing to do 9 See Juan I. Dur?n, Rainer Reisenzein and Jos?-Miguel Fern?ndez-Dols, ?Coherence Between Emotions and Facial Expressions: A Research Synthesis.? 10 LeDoux (2015) now writes: If we assume that emotions are feelings, that feelings are states of consciousness, and that states of consciousness are inner private experiences predicated on the awareness of one?s own mental activity, questions arise about the scientific study of the brain mechanisms underlying emotions in animals. (97) The terminological problem of separating feelings, which are conscious and able to be attended to, from what neuroscientists like LeDoux called emotions, which are in essence patterns of brain activation associated with certain kinds of behavior like avoidance and defense, resulted in a misunderstanding of the phenomena being studied. What LeDoux was calling an emotion was at odds with what people generally understand to be an emotion, namely a subjective feeling. 16 with fear-associated behaviors like avoidance. Furthermore, as stated in Lindquist and her colleagues? meta-analysis, reported feelings of so-called ?basic? emotions don?t seem to show consistent patterns of brain activation in localized areas of the brain. So it appears that emotion scientists as a group began the conversation in the stasis of definition, while issues in the stasis of conjecture were still unresolved. Conjectural work, like the laboratory work done by neuroscientists like LeDoux, only complicated the problem because it presupposed a definition. As a result, the study of the brain?s neurocircuitry reinforces the theory of emotion that reduces experienced emotions to essential properties of neurophysiology. Because no emotion essences have been found in the brain, more research is conducted to find them, all without questioning the premise that an emotion is properly and essentially defined by some neurophysiological essence. Cross-linguistic analysis has shown that the folk model associated with the English word ?emotion,? which has been exported to the international scientific study of emotion, is unique in that it brings together the concepts of thinking, feeling, and bodily change (Wierzbicka, 1999). Equivalent folk models of words for ?emotion? across the world?words used to represent the inner feelings of an experiencer?tend to only bring together concepts of thinking and feeling. Wierzbicka (1999) postulates that the fixation on neurophysiology in emotion science derives from the uniqueness of the semantics of the English word?s reference to the body. One solution to the definitional problem comes from the Theory of Constructed Emotion, a theory that understands emotional experiences not as expressions of essential neurocircuitry found in an already emotional brain but as emergent events, formed from lower-level general processes. Calling emotions emergent events means that there are several more basic psychological and neurological layers of activity co-occurring in the brain, none 17 of which has the sole function of producing an emotion. Lisa Feldman Barrett (2017) proposes a number of basic cognitive functions: interoception of core affect in the body (i.e., the conscious internal focus on how aroused, calm, comfortable, or uncomfortable one is at any given moment), a conceptual system that can categorize and make meaning of registered phenomena (i.e., using one?s web of beliefs, memories, and knowledge to make sense of perceptions), and a social reality where these meanings circulate (i.e., an always updating representation of a world that others have access to as well). I subscribe to the Theory of Constructed Emotion because its model of understanding meaning as emergent from more general cognitive processes coincides with the theories of language I use in this dissertation, namely Cognitive Grammar and associated theories, like Conceptual Integration Theory and frame semantics. Both the Theory of Constructed Emotion and Cognitive Grammar understand subjective meaning as being emergent from dynamic lower-level cognitive processes, not hardwired into the brain. And both recognize the importance of cultural mediation and intersubjective alignment for the acquisition and communication of concepts, emotional or not. In the case of communicating emotion concepts, if a person reports feeling fear, ?[t]he person may feel fear,? LeDoux (2015) writes, ?but this does not mean that the same brain circuits create feelings of fear? across all people or even within the same person (99). An emotion is the emergent result of complex coordinated cognitive processes? including but not reducible to activation of certain brain circuitry. It?s like the flavor of a soup, formed by many ingredients working together, none of which having a sole function in creating soups (LeDoux 2015, 104; Barrett 2009, 1284-86). Just like how the flavor of a soup can change with different chefs, in different moments, at different temperatures, in different places, or even at different times in one?s life, emotion 18 events change due to underlying processes. While there may be certain categories like chicken noodle soup and anger that our minds preserve in the conceptual system to help us to understand what our bodies are feeling (I?m hungry and this tastes good) and predict future actions (I will eat this), we often categorize very different tastes, very different emotional experiences, using the same categories. This is how someone can have very different kinds of feelings of anger?e.g., a prototypical anger with a clenched fist and a furrowed brow brought on by a conventional slight, versus a calm anger brought on by hearing an annoying song on the radio?and still categorize them as experiences of anger. The concept anger, like the concept chicken noodle soup, is also culturally bound and culturally mediated. As such, it is particular to a culture and a language. Other languages may have concepts that resemble anger to an English speaker but differ conceptually, like how the Western chicken noodle soup is much like the Chinese ji si mian (???), a noodle soup with chicken broth, but slightly different in noticeable ways. To recognize an instance of anger, one must have that concept already, and to feel angry requires the dynamic categorization of ongoing experiences using emotion concepts to make meaning of those experiences for some purpose (e.g., to excitedly retaliate against a slighter, or to change the radio station and mutter). While every person generally has the same neurobiological underpinnings, what is neurologically activated in an emotion event differs across individuals. What is necessary for an emotion event?s categorization is not some universal set of neurobiological events?though cognition requires some neurobiological activation of course ?but some learned emotion concept that provides an experiencer meaning. Contrary to the essentialist view, the Theory of Constructed Emotion views emotions not as reactions to the world but rather as emergent experiences of the world, and thus is uniquely different from ?basic emotion? theories that 19 reduce emotion to neuronal activation or physical behavior. Emotion events are emergent experiences of categorizing the world, and as such, they are categorizations and mental simulations of potential ways of understanding an experience, its causes, its consequences, and its potentialities. But this is not to say that they are disembodied phenomena, solely conceptual in nature. When the mind is going through processes of conceptual categorization?an always ongoing process that is crucial for perception and action?it is conscious of not only to the elements involved in a situation but also to the way that the body is reacting to that situation, i.e. the body?s core affect. If an intruder breaks into your house, it?s not just the element of intruder and barrier coming to conscious attention, but also, for example, the racing of the heart and flushing of the face, that help to psychologically construct the emotion event. When attention is placed on the body, an experiencer comes to understand what state of arousal they are in, as well as what the valence of this feeling is?i.e., whether this feeling is good or bad. Prototypes of emotion concepts can be mapped onto these core affective parameters, as illustrated in figure 1. Figure 1. Core affect and corresponding prototypical emotion episodes, as illustrated in Russell & Barrett (1999) 20 Recognizing the core affective aspect of emotion episodes helps to understand how interoceptive awareness?that is, awareness of how the body is feeling?works to direct the conscious mind toward categorizing an experience a certain way. Core affect alone, however, cannot be the basis for determining an emotion event. That nauseated feeling that often accompanies falling head over heels for someone, after all, more than closely resembles that nauseated feeling that accompanies food poisoning. Context counts. If a feeling of emotion is the emergent experience of a certain kind of categorization, namely categorizations that rely on emotion concepts, that feeling of light-headed nausea as recognized in an amorous context means something else entirely, even if the same kinds of bodily representations are coming to mind as when feeling ill. Those representations are being used to categorize a different experience, one that shares some physiological aspects but not the conceptual aspects that provide the emotive meaning of love. The same kinds of categorization that occur when personally feeling caught up in an emotion event are also present when recognizing emotions in others, too. That is, we only recognize that a crying baby is upset or a yelling man is angry because we cognitively construct the emotion when categorizing the experience. While the categorization of a crying baby as being upset may feel like it?s immediate, it is worth noting that walking or recognizing objects or reading words also feel immediate. One prediction on how we are able to make these inferences about others? emotion states so quickly is through conceptual integration, also called conceptual blending. In their 2002 book The Way We Think, the landmark text on conceptual blending, the linguists Gilles Fauconnier and Mark Turner argue that understanding the meaning of behaviors from facial expressions, postures, and tones in 21 voice requires unconscious blending of some perceptual stimulus with some prediction of its cause and effect. ?We even say that someone looks or sounds ?violent? or ?criminal?,? Fauconnier and Turner write, ?when what we mean is that their appearance makes us think they might cause violence or commit a criminal act? (76). From perception, humans are prone to make predictions about effects in possible worlds that don?t yet exist and may never, and we do this quickly and effortlessly. For Fauconnier and Turner?like for proponents of the Constructed Theory of Emotion, Cognitive Grammar, and strands of cognitive linguistics like Embodied Construction Grammar?humans think through effortlessly constructing counterfactual or possible worlds, comparing selective perceptions or thoughts in a perceived ?real? world with that possible world, and making inferences by blending real-world perceptions with mental models of these worlds. This is what I will be calling simulation in this dissertation. My use of the term simulation importantly refers to commonsense observation of mental motion in the theater of one?s mind; that is, I am not empirically observing how simulation is occurring.11 Emergent emotive events, as I present them, are in part the result of an ability to form mental simulations that allow the mind to create rapid predictions of how an experience will feel before experience even happens. As I raise a spoonful of chicken noodle soup to my lips, the simulation has already begun. Already, I have expectations of what the soup should taste like, based on past memories and moments of categorizing chicken noodle soup. When I sip the soup and it tastes salty, this is the result of my felt experience not aligning with my 11 While I do not provide an empirical account of simulation, many of the theories I reference?from the Theory of Constructed Emotion to Embodied Cognitive Grammar?do provide empirical grounding. 22 simulated expectations. The asymmetry between what I expected and what I experienced causes me to experience an event?in this case, a moment of disgust or a moment of surprise?which can further lead me to reason about the experience: I should request less salt, or I will never come back to this restaurant. The expectation acts as the background, on which the experienced deviations become objects to further guide judgment and reasoning. Predictions, perceptions, experiences, and reasons are all bound together in emergent emotion events. That these different kinds of cognitive phenomena are all part of constructing emotive meaning is important for understanding the emotional appeal. If emotional appeals are only recognized in terms of their ability to activate an emotion concept, the kinds of predictions and common ground assumptions that guide reasoning are overlooked. In addition, emotive meanings are grounded in perceptual and experiential knowledge. They are embodied, in other words. While rhetorical theory has provided ample ways for describing patterns of argumentation and informal logic in persuasive situations, I am unaware of any rhetorical framework that pays attention to the ways that physical, bodily experience is captured in language, let alone how these experiences come to bear on argumentative discourse. To adequately capture emotive phenomena in argumentative discourse, it helps not only to understand what emotion is and how it arises in the body and mind but also how emotion interacts with and emerges from language. 5. The Encoding and Construal of Experience To adequately account for the myriad ways that emotional appeals are constructed for 23 rhetorical purposes, a framework is necessary for the integration of two crucial components: the cognitive process that makes up an emotion, and the ways that these cognitive processes are cued discursively in patterned language usage. The framework I propose?the cognitive rhetorical model of pathos?blends two sources: the Theory of Constructed Emotion described above, and the theory of Cognitive Grammar (Langacker 1987; 1991a; 2013), one of the most prominent theories of language in the umbrella of Cognitive Linguistics. Both the Theory of Constructed Emotion and Cognitive Grammar understand meaning?emotive or linguistic?as being the result of lower-level cognitive processes like categorization, attention,12 and bodily experience. Given that the Theory of Constructed Emotion has been introduced above, this section will provide an introduction to basic ideas in Cognitive Grammar, as they will be useful for furthering the cognitive rhetorical model of pathos. Cognitive Grammar (CG) understands language as an emergent phenomenon, born out of communicative needs and social interactions, as well as the basic cognitive abilities that shape experience. These cognitive abilities take the form of basic conceptual knowledge like the perception of shapes and colors and the experience of sensation and affect, as well as basic cognitive functions like the relationship between two things or events (comparison), the understanding of one experience as an instance of another (categorization), and the ability to build higher level concepts out of observed patterns (abstraction) (Langacker 2000, 2-4). Instead 12 Of course, Cognitive Grammar is not alone in developing a story of the roles that embodiment and lower-level domains of cognitive processing play in the emergent construction of meaning. Todd Oakley, in his work on cognitive semiotics, has provided a rich description of how meaning is constructed from the cognitive attention system in his From Attention to Meaning: Explorations in Semiotics, Linguistics, and Rhetoric. 24 of understanding the meaning of a linguistic utterance in terms of its objective, truth- conditional value?a perspective common in semantics and the philosophy of language?CG takes meaning to be something that is constructed through language usage and subjectively coordinated in the minds of interlocutors. That is, language?s function is communicative in nature, not solely descriptive. CG thus understands language as providing a unique window into the way meaning emerges in the minds of language users. In CG, all language usage is meaningful, as it is composed of pairings of phonological forms and semantic meanings. The pairings of form and meaning creates a symbolic structure, the most fundamental unit of language. While the idea that lexemes or words are composed of symbolic structures is by no means novel, CG posits that every aspect of linguistic meaning?from morphology to syntax?is the result of assemblies of symbolic structures. Moreover, differences in symbolic assemblies afford different conceptualizations of meaning. Because all language is meaningful and different utterances result in different kinds of meanings, CG presents itself as a useful way of articulating differences in meaning and meaning construction. These descriptions will prove to be helpful in showing how emotive meaning arises in the conceptualization and construal of emotional appeals. Part of what makes CG a particularly helpful theoretical tool for analyzing language is its ability to capture various sorts of cognitive phenomena that language facilitates. These phenomena include a language user?s ability to make inferences by reference to encyclopedic knowledge (i.e., semantic frames) that underlies words and phrases (e.g., Fillmore 1985); attending to the ways bodily experiences and force dynamics factor into linguistic meaning (e.g., Talmy 1988); the use of mental spaces in discourse to model alternate, counterfactual, 25 conditional, or more schematic conceptions of reality (e.g., Fauconnier; Fauconnier and Turner; Dancygier and Sweetser 2005); the way that metaphor and metonymy structure ordinary cognition and reasoning (Lakoff and Johnson); and way that language users undergo processes of conceptual blending to compress, elaborate, and enrich meaning (Fauconnier and Turner). Because understanding emotive meaning means understanding how experiences, events, and evaluations can be categorized as being part of emotion concepts or emotion frames, having a theory of language that is able to explain how words and grammatical constructions correspond to the mental life of lived experience is crucial. While CG was not developed to describe the kinds of indirect emotion activation emotional appeals elicit, let alone rhetorical phenomena, it has proven to be useful in providing a window into not only the ways we reason through language but the ways we perceive and experience language in communication. Integral to the study of emotional appeals is the way that language prompts speakers and hearers to conceptualize semantic meaning from particular perspectives. In CG, this aspect of language is called construal, the ?ability to conceive and portray the same situation in alternate ways? (Langacker 2008, 43). If we were to imagine linguistic meaning construction to be a movie, what the camera man does?placing the camera to face one actor rather than other, focusing on the scenery, zooming in or out?would be to construe an event from a particular perspective with a particular focus. Language, too, works in this way, inviting listeners to take particular viewpoints or perspectives. Differences in construal be prompted by the use of vocabulary; for example, if one were to designate the point at which the land and sea meet from the vantage point of the land, it would be called the coast, whereas, if they were to designate the same location from the sea, it would be called the 26 shore (Fillmore 1982, 121). Construal can also be prompted grammatically; for example, by using the grammatical pattern [be/ get + V past participle] to fix the vantage point onto an entity undergoing an action (as in the tree was cut). Linguistic content is often presented with more or less fine-grained granularity, another aspect of construal, leading to difference in (1)a-d. (1) a. A person did something somewhere with someone b. A Roman politician stood on the ground with a woman c. Verres, praetor of the Roman people, stood on the shore with a worthless woman d. There on the shore stood the praetor of the Roman people, shod with slippers, clad in a purple cloak and tunic and leaning on this worthless woman.13 The more schematic scene in 1(a) increases in detail in complexity until it becomes the striking expression in 1(d) that provoked Quintilian to vividly imagine the Roman politician Verres from reading Cicero?s Verrine Orations. Attending to how linguistic expressions motivate language users to understand them in particular ways?taking viewpoints, zooming in, positioning items in space and time? illustrates how language affects cognition. It also illustrates the dependence on knowledge of physical and embodied experiences, like the experience of viewing an event from a particular location, in the general processing of constructing meaning from language. These experiences, CG argues, are (semantically) encoded in linguistic expressions and can cue simulations that allow for (pragmatic) meanings to emerge in and across context. Understanding how language guides experience is thus an important part of learning how 13 1(d) is used as an example of enargeia, or vivid description, in Quintilian?s Institutio 8.3.64-5 and is drawn from Cicero?s Verrine Orations 5.33.86. Enargeia, a chief strategy of emotion arousal in classical rhetoric, will be discussed in more detail in Chapter 2, as well as Chapter 4. 27 argument guides feeling. 6. The Aristotelian Pathetic Appeal It is common knowledge among emotion scientists and rhetoricians alike that Aristotle is responsible for the first serious account of the emotions. In the Rhetoric, Aristotle explicates 16 emotion words, providing cognitive sketches of how each emotion mental state is understood, as well as how it affects experiencers. That this is primarily dealt with in a book on rhetoric is both odd and interesting. In the beginning of the section on the path?, Aristotle states that those who are serious about learning how to arouse pathos in others must first be able to define that mental state. As we have just seen, this job of defining is rather difficult and confused from the onset. How did Aristotle answer his first question, then? What did Aristotle believe an adequate definition of pathos, the closest equivalent to the English word ?emotion?, was? Aristotle himself does not provide a good account. He does mention terms that have modern equivalents, such as anger (org?) and shame (aiskhyn?). But it would be anachronistic to call these emotions, as they are categorized alongside oddly specific path? such as the calming down of anger (praot?s) and the lack of shame (anaiskhyntia). He even includes unkindliness (akharistia), makes sure to differentiate this from being unfriendly (ekhthra) and, almost as if he could predict that some would mistake his list as a list of Western emotions, he tags emulation (z?los) to the end of his list of path?. Surely the set of path? included in Book II of the Rhetoric is far from complete. To think that Aristotle is presenting a complete list of emotions would be strange, yet this is what many classical philosophers (especially Stoics) 28 and contemporary psychologists believed. What is often overlooked or forgotten is that this list of emotions is presented in his book on rhetoric, not his book on psychology, De Anima, or his book on social morality, The Nicomachean Ethics. The reason, a rhetorician would infer, that these emotions are present here is not because emotions are relegated to the discipline of rhetoric but that rhetoricians can use knowledge of the emotions to construct persuasive arguments. The emotion concepts that Aristotle writes on are not meant to represent a complete list of emotions but a list of useful emotions for oratorical purposes. I am not alone in reading Aristotle?s list of path? as ways of constructing arguments. Jeffrey Walker (1992, 2000) has argued that Aristotle?s path? are topoi from which enthymemes?partially-filled deductive arguments14?can be constructed. Walker argues that enthymemes of an emotion are constructed when various premises lead an audience to conclude that they should be feeling a certain way as a result of accepting the truth of these premises and reasoning to their conclusion. The emerging emotion, Walker (1992) argues, is 14 Aristotle defines the enthymeme as a ?rhetorical syllogism? in Rhetoric 1.2.7. This definition, however, has led to many arguments. Thomas M. Conley?s ?The Enthymeme in Perspective? argues that modern interpretations of enthymeme are simplistic, though the jury is still out on what the ?actual? meaning of the term is. The term changed over time and over systems of reasoning. The system that Aristotle introduces in the Prior Analytics and in the Topics was probably not consulted in the Rhetoric, meaning that the term ?syllogism? that moderns use to mean an argument with a major and minor premise and a conclusion was not part of what Aristotle meant by enthymeme being a rhetorical syllogism. For more on the history of the term, see Thomas M. Conley?s ?The Enthymeme in Perspective.? See also Jeffrey Walker?s ?The Body of Persuasion: The Enthymeme,? and also Carol Poster?s ?A Historical Recontextualization of the Enthymeme.? The idea that the word enthymeme is constructed around the root for thymos, the part of the body that Plato claimed was the seat of the emotions, is probably not true, a point made by David Mirhady in his essay ?Aristotle?s Enthymeme, Thymos, and Plato? in Influences on Peripatetic Rhetoric: Essays in Honor of William W. Fortenbaugh. 29 a ?diffuse arousal-state? that helps the body both ?prepare for bodily action? as well as selectively find reasons to justify that plan for action (359). Following from this, the definition of path? are not necessarily (or not only) descriptions of path? but blueprints for constructing arguments. Because emotional appeals depend on emphasizing the good reason for an audience?s activation of an emotion script, they can be understood as requiring both pathos and logos, constituting a ?sort of rationality? grounded in practical reasoning, as Walker (2000) has suggested (81). The arousal of a pathos, in Walker?s terms, ?will strongly determine how [an audience] perceives and interprets any ?premises? presented to it? (81). In other words, the experience of an emotion event can guide what sorts of worldly phenomena are attended to. Arousing emotions from a rhetorical perspective can help a rhetor to focus on specific actors, arouse selective identifications, prompt certain lines of reasoning, and otherwise control attention to the cause and consequence of registering an emotion. It is this point about how to use the arousal of emotion for argumentative purposes that found its way into Aristotle?s theory of rhetoric. When Aristotle was writing his list of path?, he was sure to pair each emotion word with another: anger (org?) with calmness (praot?s), kindness (kharis) with unkindness (akharistia), and so on. While some words in these pairs are opposites?one could say that the opposite of fear (phobos) is confidence (tharsos)?it would be strange to think that the opposite of anger is calmness. Rather than reading this list as pairings of opposite emotions, it may be more fruitful to think of them as topoi for constructing arguments in contrasting situations. In a situation where a rhetor finds an angry audience, it may 30 be helpful to produce arguments that calm them. In a situation where a rhetor finds an audience quivering in fear at the thought of combat, bestowing confidence may help the audience to pick up their swords. The trick to constructing effective emotional appeals, given Aristotle?s descriptions and pairings of path?, is knowing what strategies and conceptual content can guide audiences to effortlessly reason themselves into an emotion state. While from a rhetor?s perspective, achieving the desired effect from an emotional appeal can be something to labor over, from an audience?s perspective, emotion arousal does not (and must not) feel effortful at all. Audiences must feel as if they came into the rhetor?s desired emotion state on their own and crucially must not be aware of a rhetor?s suasive strategies for arousing that emotion. Many classical and contemporary critics of rhetoric have argued that emotional appeals pervert reasoning and cause deviations of thought. On the contrary, emotional appeals work through the everyday reasoning practices that occupy most of our conscious mental activity. If the twenty- and twenty-first century study of reasoning has taught us anything, it is that deliberate rational thought is the exception rather than the rule when it comes to how we think (Johnson-Laird; Kahneman; Clark). Instead of rationally weighing the risks and benefits of any given situation, we more frequently use quick heuristics to guide our reasons, as contemporary scholars of cognition have shown. These heuristics don?t exhaust our mental energy and allow us to make reasonable predictions. Rather than undergoing exhaustive and deliberative reasoning, our minds default tor fast-paced reasoning?sometimes called ?biased? reasoning? that will allow us to make quick decisions and do more with our time. In moments 31 where we need to make decisions quickly and we don?t perceive risk in our decisions, slow and deliberate reasoning put as at a disadvantage. Similarly, at moments when we feel that the outcome of a decision is crucial and warrants careful attention, the potential errors of fast-paced reasoning can lead to risk and unfavorable consequence. Both deliberate reasoning and fast-paced reasoning have their advantages and disadvantages, and that they are both useful is why they both continue to persist in the epoch of our human psychology. It is for this reason that I am wary of calling fast- paced reasoning a ?cognitive bias,? as the term ?bias? has a negative connotation. This sort of reasoning is grounded in lived experiences, making ample use of default and learned assumptions (prototypes and stereotypes). It is reasoning that is integral to everyday cognition, and it more often than not goes unnoticed. For this reason, I call this fast-paced reasoning ?everyday rationality?. This being said, everyday rationality can certainly be influenced by a rhetor, and conventions of reasoning through everyday rationality can also result that in ways one might not expect when viewing reason only through the lens of deliberate rationality. Adam Smith writes in his Theory of Moral Sentiments that he would not expect a European person to show any less sympathy for someone whose pocket watch has just broken than for hearing about a devastating earthquake in China (3.1.46). In fact, more sentiment may be shown for the person with the broken watch?the earthquake is far away, but the broken watch is right here and causing immediate distress. The Yale psychologist Paul Bloom (2017) calls this phenomenon the innumerate bias and claims that it is central to all appeals to empathy: 32 To get a sense of the innumerate nature of our feelings, imagine reading that two hundred people just died in an earthquake in a remote country. How do you feel? Now imagine that you just discovered that the actual number of deaths was two thousand. Do you feel ten times worse? Do you feel any worse? (89) The point is that emotional responses grounded in empathy or pity or anger are innumerable?they don?t vary according to amount. He, like Smith, offers this as a point of concern. Can we ever think reasonably when it comes to issues that move us? Really, this is a moot point that misunderstands reasonableness by conflating everyday rationality, which has its own grounds for reasonableness, with the utilitarian deliberate rationality that was celebrated by Enlightenment philosophers. It?s not that we don?t want to care, it?s that judgment that stems from caring is biased to find prototypical examples of suffering from which we can generate our evaluations of the situation and decisions as to what to do. That is, our use of everyday rationality guides the arousal of our emotions. In a world modeled by deliberate rationality, the number of sufferers would affect the degree of our empathy, but, in the arousal of empathy, as with the arousal of many emotions, the value we place in the thing affected is the determining factor. This is just how we normally think.15 15 While there is no ideal way of thinking, experiences of empathy depend crucially on our attention to, investment in identification with someone we value. Importantly, experiences of empathy vary across individuals, cultures, and neurodiversities. For example, neurodivergent people and neurotypical people may experience empathy differently, but those experiences, however they are configured, still depend on identification and valuation. I regret that this dissertation does not spend much time analyzing or accounting for neurodivergent experiences of emotion, but I have no doubt that the arguments made here and throughout this work are not restricted solely to neurotypical experiences. 33 If we hold that there is some form of reasoning called ?rational reasoning? and then mark off all of the sorts of perversions?biased reasoning, reasoning from feelings, reasoning from prototypical concepts, reasoning from experiences, and so on?we quickly find that ?rational reasoning? captures very little of our human experiences. Far from showing us ?perversions? of thought, an analysis of the emotional appeals illustrates the less laborious and more everyday ways that we normally think. Because emotional appeals are intentional argumentative moves, what Aristotle calls artistic appeals of invention, studying them also provides insights into the hidden ways that rhetors guide thinking through everyday rationality. What the Aristotelian pathetic appeal reveals are patterns of inference from value-laden scripts to the activation of emotion states that influence everyday rationality. As I will go on to demonstrate in this dissertation, rhetors, using only a few words, are able to intentionally activate rich and complex value-laden scripts, and this is due to the metonymic nature of emotional appeals. The Aristotelian pathetic appeal carves out a sort of enthymematic reasoning, a particularly common form of thought that is surprisingly under-appreciated and under-theorized, where, given a few premises, audiences can metonymically activate whole evaluative scenes, quickly infer where audiences stand on issues, and effortlessly align their own stances to match. The cognitive rhetorical model of pathos brings the study of the pathetic appeal into the twenty-first century by reviving Aristotle?s original line of inquiry into pathos and putting that inquiry into conversation with contemporary theories of mind, meaning and language in Linguistics, Cognitive-Neuroscience, Anthropology, Rhetoric, and Composition. 34 7. Cognitive Rhetorical Model of Emotional Appeals Everyday rationality is affected by emotion, and emotional appeals are everywhere, from the fuming tweets of politicians to clever couplets in protest chants to the doting notes in children?s lunch boxes. These appeals ask us to pay attention to causes, to reconsider how we feel about politicians, to empathize with those who are less fortunate, to imagine events unfolding differently. Understanding how they function may provide insight into how we reason, the chief goal of the cognitive rhetorical model of pathos. To offer an example of how the cognitive rhetorical model of pathos can be useful in describing the links between language, emotion, and argumentation in everyday life, consider how the emotive meaning of the following sentence, seen on a bumper sticker, can be described. (2) War doesn?t show who?s right, just who?s left The sentence is aphoristic, just eight words (eleven if you separate the contractions). It may come off as striking or it may not, yet the pathetic call for a recognition of anger and sadness cannot be ignored. A cursory reading would probably locate the main spots of rhetorical force: the negation of who?s right juxtaposed with the assertion of who?s left, the antithesis between right and left, and the syllepsis in the competing senses of left. Perhaps the quasi- personification, via metaphor, of war as it shows something would be part of this description, too. Yet, none of this explains how the utterance moves its audience to an emotional conclusion. The cognitive rhetorical model of pathos argues that emotions arise as the result of indirect activations of emotion concepts and are intensified through the use of linguistic 35 constructions which shift attention on key elements in value-laden emotional scripts. Many of these linguistic choices operate under the register of conscious awareness but are important to the overall construal of the utterance. When beginning to analyze the bumper sticker sentence from above from a cognitive rhetorical perspective, it is not that the rhetorical stylistic elements are not important factors. They certainly are. The quasi- zuegmatic figuration of left is one way in which focus is shifted between cognitive entities. Yet, there is much working at the level of cognitive conceptualization that is not present in the words alone. Of course the utterance in (2) is already argumentative, though implicitly so through its negation of the proposition that war shows who is right. As Langacker (2008) notes, ?negation evokes as its background positive conceptions of what is being denied? (59). In other words, the negation of war showing who is right requires a shared understanding between the rhetor and audience that some people?and perhaps the audience?believes that might makes right. It may also be worth noting that what is negated (war shows who is right) and what is asserted (war shows who is left) are not mutually exclusive. Nevertheless, when the propositions are placed in the not x, just y construction, the value of y appears as antithetical to the value of x, such that, to accept y is to reject x. This may be fallacious reasoning in the domain of deliberate rationality, but it is rhetorically effective at the level of construal in everyday rationality, as it invites audiences to take a stance on what war just shows, exclusively. In persuading the audience to accept that the only thing war can show is who?s left, there is, of course, the play on the sense of right and left as ego-centric relative directions and opposing political views. There is also a contrast of semantic frames. When war shows who?s 36 right, it is showing who has the correct perspective on the irreconcilable difference in opinion that resulted in war. Right here brings on stage a kind of Reasoning frame.16 The person who is right is the person who reasoned correctly, in contrast to those who are wrong. When war shows who is left, the frame being evoked is not the Reasoning frame but the Remainder frame. In the Remainder frame, there is some amount of some resource, and then some of that amount is taken away, resulting in a remainder or what is left of that initial resource. The two frames focus on different qualities of war. When war is viewed through the Reasoning frame, the results of war show the correct reasons. When war is viewed through the Remainder frame, the results of war show the remaining lives after war, the survivors. The difference between the invocation of Reasoning and Remainder is also a difference in feeling. When considering who is right, at least two stances must be considered, and the correct stance is epistemically unknown, as indicated by the interrogative pronoun who. The frame thus invokes a sense of tension17 between agonistic positions and an uncertainty over which position will prevail. This is contrasted with the feeling of who is left, which, while it also invokes epistemic uncertainty, does not present an agonistic relationship but instead the results of an action. Who is left implies that not everyone is left, and, by 16 Following conventions in Cognitive Linguistics, the names of frames will have capitalized first letters and underscores between multi-word names. 17 Rhetoricians and non-cognitive grammarians alike rarely pay attention to the build-up and release of tension implicit in basic linguistic utterances. Work in Cognitive Linguistics by Leonard Talmy (1988) has argued that language captures basic senses of force interactions between elements in utterances, called force-dynamic interactions. These interactions, I argue, have an affective role in how language users make meaning. 37 extension, entails that some have died. Instead of increasing tension and affective arousal, the Remainder frame in this instance relaxes tension, thus lowering arousal and granting greater cognitive accessibility to emotion concepts that are less excited, like the negatively valenced sadness and disappointment or the more positively valenced calmness and content.18 What the bumper sticker does, then, is persuade the reader to reject the view that the results of war reveal correctness of tense, agonistic reasoning and to accept the view that the results of war only reveal those who have not died, a relaxed but somber stance. How this stance is somber may be obvious but is worth articulating. In Western traditions, as well as many non-Western traditions, life is considered a precious resource, something valued, such that the loss of life is undesirable and thus negatively valenced. While the bumper sticker invites a construal of survivors of war, that there are survivors implies that some people may have also died.19 The rhetoric of the bumper sticker?s assertion, thus, begins with an explicit foregrounding of the survivors in who is left, invites audiences to recognize the bleak background in those who have died at war, and focus attention on that death until an optimally satisfying interpretation has been reached. Furthermore, because death involves the loss of life, a presupposed thing of value to the audience, it shares conceptual structure with other negatively valenced emotion concepts that involve loss, such as sadness. Focusing in on a scene of loss, as is done by emphasizing who is left after war, thus allows the scene to 18 See figure 1 for a chart indicating core affective states, measured by valence and arousal, and the prototypical emotion concepts that correspond with these states. 19 More precisely, to say in a neutral context ?Some people survived? licenses a scalar implicature that ?Not everyone survived,? because if they had, the speaker would have said that. This is an example of a Gricean quantity implicature (see Grice, Levinson 138-146). 38 become categorized as a sad event.20 How the bumper sticker works to construct emotive meaning, in conclusion, is as follows. It narrows the focus of the scene to exclude the speaker and hearer, thus focusing on the proposition. The proposition is composed of a rejection and assertion that are written in such a way as to imply that the rejected premise is in opposition to the asserted premise. The scene rejected invokes a swelling force of uncertainty about agonistic viewpoints in wartime, and the asserted premise invokes a relaxation that helps to both relax and focus conceptualization on the results of the act of war. The asserted premise, finally, indirectly invokes the loss of life and thus shares enough conceptual structure with the emotion concept of sadness that it can be categorized as an instance of it. The use of negating a premise that presents war through a different frame helps to bring argumentative force to the asserted frame, which argues in a stance (not x, just y) that the bumper sticker?s audience is invited to accept exclusively. The words function to bring entities on stage, shift focus, frame, reject, reframe, and infer. Emotive meaning is achieved as the result of both language and argumentation, style and content, moments of thinking and moments of feeling. A cognitive rhetorical account of pathos does not merely locate where emotive arguments are but shows how they are composed and how they function. The composition of emotional appeals requires a considerable amount of cognitive work, most of which occurs beneath the level of conscious awareness, at the level of frame activation and categorization, 20 The enthymematic process by which this categorization is done is called frame metonymy, which is the focus of Chapter 3. 39 all of which is implicit in the rhetorical styling of the utterance. Even if they account for them, popular and expert models of emotion rarely explicitly attend to the cognitive dimension that rhetorical and linguistic acts play in constructing emotion events. My rhetorical and discourse analytic methodology allows for careful attention to the construction of meaning in linguistic discourse, and, importantly, this methodology reveals the ways that both pathos and logos get co-constructed dynamically and dialogically in discourse. With this, I am able to both describe the pathetic texture of texts in fine detail, as well as to see the ways emotional appeals get constructed out of apparently neutral words and expressions. One claim, which is central to this project, is that emotional appeals hinge on valuation and presupposed beliefs in common ground. In the above example, the evaluation that the loss of life is bad is (culturally or pragmatically) presupposed, as is the evaluation that some people think that might makes right. Objects of evaluation are rarely introduced into discourse as such, and, as George Lakoff (1996, 2004) has argued, hierarchies of moral evaluations underlie social, political, and ideological modes of thought. Like those entities that were hidden in the background of the bumper sticker?the speaker making the assertion, the semantic frames of Reasoning and Remainder, who war is showing something to, etc.? presupposed beliefs and guided evaluations, while unstated, are vital components of an audience?s affective construal of an emotional appeal. They entice us to act by threatening or protecting, demonizing or defending those social orders likeminded communities find to be important. What analysis reveals is how these elements of evaluation are brought on stage or taken off, put in danger or protected from harm, referenced indirectly or amplified through zooming in?all to craft identification, buttress persuasive acts, and motivate changes in feelings and actions. 40 While a bumper sticker may not change minds, it resonates within everyday rationality. Through its reverberation and through its part in constructing collective attitudes toward disputable evaluative elements, the bumper sticker does its job. In processing its meaning, readers come to understand the stance of its intended reader, the presumed stance of its owner, and the general political and social stance of others who hold similar beliefs. It further reinforces these beliefs to enacting habits of thinking, feeling, and evaluating, all by reading the sticker. Moreover, it does all of this without announcing its effect; it doesn?t say what to believe or feel. And herein is the strength of the emotional appeal: its effects depend on audience reasoning. Reasoning is integral to the emotional appeal. 8. Forecast of Chapters The Western rhetorical tradition has had a long history of rethinking cornerstones of rhetorical theory when confronted with changes and innovations in philosophy and psychology. Conceptions of the emotional appeal changed when Greek thought was adopted into Roman rhetoric, again when emotion concepts changed in the Renaissance, and yet again when Enlightenment reason and faculty psychology dominated rhetorical theory during the eighteenth and nineteenth centuries. While this project contributes to this history, it also gestures to the earlier insights and thoughts that asked questions about how rhetors get audiences to change their thoughts and feelings using language. This question is pertinent not only to rhetorical history but also to the contemporary study of emotion and linguistic communication. In the next four chapters, the method of the emotional appeal is analyzed from different perspectives, each chapter focusing in on a different facet of the intentional 41 construction of an emotive argument. Together, they layout a framework, the cognitive rhetorical model of pathos. Using cognitive rhetorical and corpus linguistic methods, I show reoccurring stylistic strategies at the level of grammatical construction that motivate imaginative and argumentative strategies. Chapter 2 focuses on the mind of Quintilian?s fabled euphantasiotos, the skillful imaginer. Drawing on work from Cognitive Grammar, the chapter lays out basic ideas of the cognitive rhetorical model of pathos with regards to how language, imagination, and mental visualization interact in ordinary and persuasive discourse. Through the introduction of the theater of the mind model?a model of meaning construction based on Langacker?s (1992) stage model of conceptualization?the chapter shows how linguistically cued simulated experiences can activate frames and emotions. While the chapter focuses primarily on the subjective experience of an audience, it concludes with a reminder that knowledge of how emotional appeals are constructed provides a way into thinking critically about emotional persuasion. Chapter 3 focuses on the role of enthymemes in emotion arousal. Assuming that an enthymeme is an argument from underspecified knowledge whose inferences are motivated by unstated common beliefs, I argue that the enthymeme is the primary argumentative strategy used to construct emotional appeals. These pathetic enthymemes, furthermore, operate via a part-for-whole frame metonymy, where description of a salient event provides conceptual access to a whole emotion script. I demonstrate how emotions can be aroused through pathetic enthymemes by analyzing contemporary American political discourse surrounding Donald Trump, focusing on reactions to Trump?s election, as well as the fear appeals Trump constructs in his 42 rhetoric on immigration. In addition to describing the composition of enthymemes of fear in Trump?s discourse, my analyses illustrate the subconscious transformational power of emotional appeals, for example, turning asylum seekers into national and bodily threats. Chapter 4 analyzes the pathetic appeal from the perspective of the technique of constructing vivid descriptions, or enargeia. In classical rhetorical theory, enargeia is asserted to be a powerful rhetorical tool that turns listeners into eyewitnesses, producing emotive effects indirectly. Using the theater of the mind model, I demonstrate this principle by showing how emotion scripts emerge from discourse that produces vivid images of interactions among valued objects. I use a 120,000- word corpus of environmental, humanitarian, and political fundraising letters to continue my argument that emotion arousal is dependent on reasoned use of language that influence everyday rationality. My analysis shows that the emotions rhetors seek to arouse are very rarely announced in fundraising discourse but instead emerge as the result of the effects of enargeiac linguistic choices, such as verbs of thought, tense and aspect constructions, and first person plural pronoun usage. By pairing cognitive rhetorical and corpus linguistic methods, I show reoccurring stylistic strategies at the level of grammatical construction that prompt imagination and craft intimacy between subject matter, rhetor, and audiences. Enargeiac techniques constitute a second strategy for emotional appeal, though this strategy does not conflict with enthymematic emotional appeals and usually work in service to them. The final chapter analyzes the relationship between mitigation and emotive expression. The Rhetorica ad Herennium notes the role that mitigation plays in 43 managing audiences? reactions and reasoning within emotive events, calming them when they are angry or avoiding unwanted attention when wanting to appear humble (4.38.50). This chapter uses corpora of first-year student reflective writing to examine the ways and reasons why writers use epistemic and affect hedges to mitigate moments of self-assessment in ungraded reflective writing assignments. What this final chapter does is move the discussion of emotional appeals from places that have traditionally been examined for emotion in rhetoric and discourse analysis, such as political discourse, to places regularly associated with ?reasoning? and ?metacognition,? such as student reflective writing. I argue that lexico-rhetorical constructions not only help students to mitigate positive self-evaluations, thus coming across as more humble, but also work to redirect their audiences? attention away from assessed content and onto their personal feelings about their assessments. I understand these rhetorical moves that mediate emotive reactions in the same way Aristotle understood the calming down of anger: as a vital component of pathos a rhetor needed to learn. While the material analyzed may seem eclectic, the three body chapters present how emotional appeals are constructed in three very different discursive fields: political oratory, professional charity writing, and student reflective writing. Certainly, these fields consist of different language patterns and different rhetorical strategies, yet, in all three fields, I argue that speaker and writers intentionally use language to prompt vivid and emotive construals of events in their audience?s minds, a shared rhetorical objective. What this argument does is ground emotional appeals in shared cognitive and linguistic processes. It makes no special concessions to rational 44 appeals as it shows that the emotional appeal works via the very same cognitive and linguistic processes for everyday rationality. In doing so, I argue that we are working toward better understanding how rhetorical persuasion occurs and what aspects of human cognition are involved in producing rhetorical discourse in any given context. The cognitive rhetorical model of pathos developed in this book adds to a growing body of work on rhetoric, linguistics, and argumentation that challenges the separation between emotion and reason, pathos and logos, style and content. In doing so, this project demonstrates how paying close attention to small lexical and grammatical cues, when combined with careful attention to habits of conversational inferencing and pattern completion, can explain how we systematically get our audiences to imagine things they may not want to in order to get them to feel in ways we intend them to. 45 Chapter 2: The Imaginative Mind 1. Introduction: The Place of Imagination in Rhetoric In his outline of rhetorical education, Quintilian classifies declamation as a necessary practice for training an orator but saves it as a practice that should come after a student has studied the full rhetorical curriculum, as declamation draws on all aspects of rhetorical theory. Quintilian?s inclusion of declamation could have been seen as controversial, as arguing fictive cases was not immediately practical, but Quintilian defends the practice by arguing that the ?imitations of battle? work to prepare orators for the field (2.10.8). Even though declamation is fictitious, it provides an arena for training. In particular, it provides imaginative training, ways of imagining cases as if they were real, and producing arguments to fit them. Imagination is an often-overlooked aspect of rhetorical training, but it is an essential part, nonetheless. Without imagination, a rhetor loses the ability to invent arguments, as well as amplify, mitigate, and otherwise figuratively develop arguments for rhetorical effects. And imagination is central not just to acts of producing persuasive discourse but, on a more fundamental level, to understanding an audience, imagining what they know or what they don?t, imagining what propositions they would find ordinary or extraordinary. For Quintilian, imagination was also a necessary component of feeling emotions and, in turn, of crafting emotionally persuasive discourse (6.2.30). Without imagination, there is no art of rhetoric. Given the relative importance of imagination in rhetorical training, one might expect for rhetorical theorists to spend time discussing techniques and methods for improving imagination. Yet discussion of imagination is surprisingly narrow, often 46 focusing on the conjuring of vivid impressions (phantasia) and ignoring the role of underlying cognitive processes, both universal and culture specific. Universally, to arouse an audience?s imagination, a rhetor must rely upon shared conceptual system with the audience, must know lexical and grammatical constructions that can cue these concepts, must assume that the audience can abstract new concepts from known concepts, and must predict that the ways in which the audience is combining and simulating concepts is similar to the ways that the rhetor expects. That is, the rhetor must assume that they and their audience are jointly attending to the same or functionally similar imaginative simulations. 1 Given these universals, much of what is imagined and the much of the ways in which mental images are simulated relies on culture-specific?and often deeply stylized and ritualized?scripts used to categorize and reason about our personal and social affective experiences. What the audience is presumed to know, what the audience is presumed to like, what is disgusting, what is anger-inducing, what an audience wouldn?t ever guess: these, of course, differ across 1 Recent arguments have posited that the development of the sense of ?we? that comes with a community of people who know the same routines and perform the same social actions is grounded in the structured development of imagination. Work by Hannes Rakoczy, Michael Tomasello, and Tricia Striano (2004) at the Max Plank Institute for Evolutionary Anthropology, for example, notes the ways two-year-old children?s sense of self is altered in the imaginative acts of pretend play (pretense), where children make up rules and change their imaginative behavior as a result. They find that play not only aids in the development of communication but places this development in the realm of cultural learning?learning to cooperate, to share intentions, and most importantly, to build common grounds of knowledge, which requires treating social constructs as real. Pretense, thus, acts as a catalyst for developing a sense of a collective intentionality?i.e., shared conceptual knowledge and actions? necessary for skillful language usage and enculturation through culture-specific concept learning. Elizabeth Picciuto and Peter Carruthers (2016) have similarly emphasized the embodied social nature of pretense and have argued that pretense is, at its core, embodied imagination, and that pretense?and by extension, imagination?serves a much overlooked but crucial role in child and adult creativity and learning. 47 individuals, cultures, and times. Surprisingly, neither the universal nor the culture- specific aspects of imagination are touched upon much in the extant books of rhetorical theory from classical antiquity. Imagination is treated as a prerequisite to rhetorical training or as a basic cognitive function. Where imagination does show up, however, is in discussions of emotional persuasion, where good rhetors are thought to be good imaginers. Toward the end of Book 6, chapter 2, Quintilian states that the person most skilled at producing emotional appeals is the person most skilled at imagining, the euphantasiotos (6.2.30). The euphantasiotos is good at producing emotional appeals precisely because the production of emotional appeals requires a good imagination. The good imaginer can produce a vivid impression (phantasia) of some act or scene, and through this mental simulation, activate associated emotion concepts in himself or infer that such a mental simulation, when articulated with vivid language, would activate an emotion concept in his audience. And while he did not praise (or even discuss) the euphantasiotos, choosing instead to focus on the pathetic topoi, Aristotle, too, believed that at the heart of the emotional appeal was a vivid mental simulation of an act or scene (see e.g., Rhetoric 2.2.2; 2.6.14). The Greek and Roman work on the emotional appeal recognize that it?s not a matter of putting the right words in the right order, but a matter of activating imaginations that are useful for guiding inferences and arousing emotion concepts. The cognitive rhetorical model of pathos, as I will show, explains how the imaginative work of language usage and comprehension makes the emotional appeal possible. It allows us to understand how utterances about physical movements, descriptions of physical objects, and moral 48 categorizations of actions can be ?moving? and persuasive. That is, it allows us to link the sensorial aspect of linguistic conceptualization to the emotional aspect of rhetorical suasion. Emotional appeals typically focus on physical objects or movements, which raises the question of why this should be the case, while providing an opportunity for crafting a model that can answer this question. The bulk of the chapter focuses on what I am calling the ?theater of the mind model,? a metadiscursive model of interconnected levels of semantic space. Having such a model is imperative for a systematic analysis of the emotional appeal because it is able to represents the imaginative, discourse-processing mind in the process of making sense of language in context. Because the model is an extension of Ronald Langacker?s (1991a) stage model of language, which is a foundational concept of Langacker?s Cognitive Grammar and a central concept of Cognitive Linguistics more broadly, the chapter begins with a review of relevant concepts in Cognitive Grammar: the construal of mental images, grounding, and intersubjective alignment. These concepts are introduced with an example of a fear appeal from Davis Guggenheim?s documentary, written and starring Al Gore, An Inconvenient Truth. From this review, larger cognitive, argumentative, and rhetorical phenomena are addressed via the theater of the mind model. The theater of the mind model takes a multi-leveled approach to mental conceptualization. It demonstrates how meanings emerge from lower level processes?for example, how the description of physical action at one level can have an emergent emotive meaning on another level and a rhetorical effect on yet another level. 49 From the theater of the mind model, I trace out the ways that movements in the mind map onto rhetorical intentions. Classical conceptions of pathos point to specific kinds of moves that are frequently considered to be emotive in some way, for example the use of enargeia or the enthymeme. These two rhetorical strategies have often epitomized the emotional appeal in classical rhetorical theory, particularly in the Peripatetic model of argumentative pathos construction2 and the Roman model of indirect emotion arousal through enargeia. These strategies correspond to patterns of imaginative thinking that arouse emotion concepts from different levels of the theater of the mind. The chapter concludes with a discussion of the emotional appeal as a kind of embodied communicative act, contingent not only on cultural and experiential knowledge but also on conventionalized patterns of imaginative mental simulation that speak to communities of people who hold certain value systems. The chief goal, however, remains to explain the kind of work the euphantasiotos does when crafting emotionally persuasive discourse and to make that work available for a more fine- grained analysis of the emotional appeal. 2 The Peripatetic tradition following Aristotle seemed to understand the enthymeme in terms of a controlled thought-provoking construction of a controversy or dilemma, with a strong implication of what the orator believed the conclusion to this problem ought to be. See, for example, Demetrius, On Style (esp. 1.30-31); Apsines (8.1); and Anon. Seguerianus (160). 50 2. Images and Argumentation In an often-criticized scene in the 2006 climate change documentary An Inconvenient Truth, Al Gore constructs an appeal to fear when he articulates what displaced water from Greenland would do to the United States: 3 (1) This is the World Trade Center Memorial Site. After the horrible events of 9/11, we said never again. [An animation depicts flooding of lower Manhattan.] But this is what would happen to Manhattan. They can measure this precisely, just as the scientists could predict precisely how much water would breach the levy in New Orleans. The area where the World Trade Center Memorial is to be located would be under water. Is it possible that we should prepare against other threats besides terrorism? Maybe we should be concerned with other problems as well. (Guggenheim) Implicitly, Gore?s argument argues for the relevance of addressing climate change by amplifying its potential material and symbolic consequences. An undesirable consequence?flooding of the World Trade Center memorial?is meant to induce fear in proud Americans, ones who hold the World Trade Center memorial in high regard because of its symbolic meaning as a memorial to the victims of terrorism. 3 The scene was thought to be, as Michael Dobbs of the Washington Post called it, ?alarmist? in its characterization of the effects of flooding on Manhattan. Film critic Jonathan Greenland of The Guardian criticized the scene as being a deliberate emotional appeal??the range of emotions this prompts begins with shock, then anger.? It?s worth noting that Dobbs, Greenland, and others like them act as if the recognition of the appeal abates the legitimate risks the appeal is meant to amplify through dramatization. 51 The sort of indirect argument that Gore employs here is the sort that Douglas Walton (2000) has argued to be nearly entirely subjective rather than grounded in reason (146). These arguments rely more on psychological aspects and the relevance of threatened valued objects than the evaluation of premises and argument forms. As such, these kinds of arguments have been, as Walton claims, ignored or brushed aside in the informal logic and argumentation literature on fear appeals. Because argumentation is grounded in the assessment of reason, these sorts of fear appeals, which work more through imaginative arousal rather than reason, are said to not qualify. Yet there is a kind of reasoning occurring when audiences are moved to fear as the result of mental visualization. This reasoning is how words in a discourse cue and constrain the conceptualization of scenes in fearful frames; it?s a kind of reasoning that depends on activating imagination and presupposing shared values through language. While the skeletal structure of Gore?s argument may not be interesting to someone who studies argumentation structures, the ways he attempts to amplify and ground fear in his audience is relevant to cognitive rhetoricians who want to understand the patterned relationships between language and cognitive arousal. By using the framework of Cognitive Grammar, coupled with classical rhetorical theory and contemporary emotion science, cognitive rhetoric can demonstrate the logic of image formation, grounded in grammar, cognition, and cultural knowledge. To better understand my proposed perspective, it is worth reviewing the relationship between embodied imagination and language usage. Within Cognitive Linguistics, language is commonly thought of as an emergent system, emerging from 52 embodied schematizations of experiences and usage. The embodied cognition hypothesis posits that knowing and using a language is more than just encoding and decoding signs; language is experiential and draws on basic concepts that are learned as a result of being in the world. Instead of separating thought and the body, proponents have argued that we think with our bodies, utilizing experiential knowledge to solve problems and make meaning in other cognitive domains (see e.g. Lakoff 1987, Lakoff & Johnson). There is now a good amount of experimental evidence from cognitive neuroscience strengthening the fidelity of embodiment models of cognition. Work by Raymond Gibbs (2011) and Marlene Johansson Flack and Gibbs (2012) has shown that manipulating a conceptualizer?s environment affects language processing. Work by Rolf Zwann and his colleagues (2004) has shown linguistic processing is a dynamic process that can be affected by visualization rather than a static and propositional process. Teenie Matlock (2004) has shown that processing time of fictive motion sentences like The river runs down the mountain vary based on the kind of mental scanning licensed by the motion verb (e.g. runs vs snakes). And Ben Bergen and Kathryn Wheeler (2010) have shown that grammatical constructions like tense and aspect modulate the ways in which the brain simulates a scene. The groundwork for embodied cognition and the basic tenants of simulation, however, were established well before cognitive and psycholinguistic findings were available, as these groundworks were built into the domain-general beliefs of CG. Instead of viewing language as series of rules and transformations, CG and similar Cognitive Linguistic approaches to grammar posit that the meaning of an 53 utterance comes from a dynamic process of conceptualization?or mental simulation?where events play out on the stages of theater of the mind. This process of image-making is very similar to Aristotle?s phantasia, where mental visualizations derived from language to produce rich experiences. To understand a simple sentence like I took my dog for a walk, a hearer must know much more than just the sum of the words I, took, my, dog, for, a, and walk with the syntax that connects them. That is, the folk model that linguistic meaning is calculated by adding up the associations of a bag of words misses much of even the basic meaning of the sentence. In the theater of the mind, a hearer can see that the speaker was walking with a dog, that the dog and speaker were walking on the ground, that they moved along a path forwards, that the dog presumably had a leash that the speaker held, that there is a reason for walking the dog (e.g. to give the dog exercise, to let the dog play, so that the dog can poop), that there are routine times for walking the dog, that there are routine routes that the dog likes to wander, and so on. The Dog Walking frame that is put on stage is conceptually rich, even if many of the details in the mental simulation are not explicit in the sentence. This account proposes that meaning making is dynamic and highly imaginative. ?It?s not about activating the right symbol,? as the cognitive scientist Benjamin Bergen comments, ?it?s about dynamically constructing the right mental experience of the scene? (Bergen 16). While mental simulation involves purely subjective experience, as Walton and other argumentation theorists are right to point out, systematically cued mental simulation must be coherent enough across interlocutors so that communication can occur and produce coherence between interlocutors. Mental simulation may have 54 subjective qualities, but no one reads I took my dog for a walk and imagines a candlelit dinner. The dog may be different, and maybe the walking terrain, but the simulation is diffusely shared across readers. As such, the grammar of a language is, if properly understood, precisely a toolkit of constructions for cueing and activating mental simulations. Understanding language comprehension in discourse in this way affords a very valuable means for understanding and describing the emotional appeal. Effective emotional appeals often engage audiences beyond what is literally said, drawing them to imagine bad things happening to good people without explicitly stating as much. Because so much detail is often left unsaid in emotionally persuasive discourse, having a methodology for understanding how particular linguistic assemblages link up to conceptual and emotional experiences is invaluable. 3. Dimensions of Imagery A large part of the methodology in this dissertation for describing embodied mental visualization is built out of Langacker?s (1987, 1991a) work on construal in CG. In the CG literature, the construal of a mental image is broken down into ?dimensions of imagery? (Langacker 1986, 6-13).4 While there are no definitive taxonomies of dimensions, the central dimensions of relevance are specificity (pertaining to granularity and schematicity), prominence (pertaining to what entities ?stand out? 4 The term ?dimensions of imagery? seemed to have fallen out of favor for ?construal operations? (e.g. Croft & Cruse 2004). I choose to use the former term because I find it to be a more informative. Also worth noting is that the term ?image? and related visual terminology in CG (viewing, on-stage, etc.) are meant not only to designate visual sensations but all vivid multisensory mental representations. 55 against a conceptual base), and perspective (pertaining to the vantage point from which a situation is viewed). In addition to these three dimensions of imagery, two additional aspects of imagination are considered: force dynamics and intersubjective alignment. Force-dynamic states?such as the simulated build up and release of tension?are helpful for describing how language can cue affective response, and intersubjective alignment reminds us that language usage is always an act of coordination of meaning and knowledge between language users. The concept of specificity was covered in section 1.2.3 when discussing the degrees of detail that Cicero used to describe Verres, the regal praetor standing on the beach with a worthless woman. Specificity is more than just the details, though. It guides meaning-making. For example, imagine if Al Gore were to make any of these utterances rather than the one in the first sentence of (1): (2) a. This is a place. b. This is a memorial. c. The World Trade Center Memorial is a memorial in lower Manhattan on Greenwich Street, a few blocks from the Cortlandt Street subway station. In (2a) and (2b), the noun phrase predicates are said to be schematic with respect to the original utterance: the conceptualizable entity is abstracted to a point where it loses some of the meaning that comes with its lexical designation. This can be illustrated with the following degrees of specificity: place ? memorial ? World Trade Center Memorial. With each new expression, the conceptual entity becomes less abstract and becomes more granular. As it becomes more granular, more fine- 56 grained visual detail is made available for mental visualization. The name of the memorial designates a specific entity, which Gore?s interlocutors can quickly access and register the symbolic weight of. This kind of symbolic meaning may be accessible in (2c), but because (2c) has been specified so granularly with regards to its location, it may lose symbolic weight as a memorial. As such, the degree of granularity with which a conceptual entity is put ?on stage? in the theater of the mind affects the kind of meaning it can produce. While it is tempting to think that the dimension of granularity carries with it an intrinsic emotive quality, as it can be used to construe a valued entity in vivid detail, it is not necessarily the case that the more granular a construal, the more emotive impact that construal has, as evidenced by (2c). It is often the case that different dimensions of imagery work in harmony to construe a conceptual entity, enhancing the weight of the entity?s symbolic value. The second dimension of imagery pertains to the prominence of an image, or what ?stands out? as salient in a mental visualization. Langacker identifies two phenomena responsible for prominence: profiling and landmark-trajector alignment. For any construable conceptualization, part of that scene is profiled against a base or background. In the scene in (3), the entity that is profiled is the asserted potential relation between the area in which the World Trade Center is located and the state of being under water. (3) The area where the World Trade Center Memorial is to be located would be under water. Implicit within the simulation of the World Trade Center Memorial becoming flooded is the image of water rising above the memorial. In this image of water rising above 57 the memorial, as depicted in Figure 1, the World Trade Center acts as a landmark and the water, which changes over time to eventually rise above the memorial, is a trajector. Figure 2. Dynamic landmark-trajector alignment of water (tr) rising above the World Trade Center Memorial (lm) over time In CG, the trajector is understood to provide a primary focus in mental simulations, whereas a landmark provides a secondary focus. While it?s the World Trade Center that is being flooded, it?s the act of flooding that is primarily on stage in the theater of the mind. The very term ?flooded? depicts a scene like the one in Figure 1, where a landmark which was not under water, over time, becomes under water (at some time prior to the moment of the utterance). The relationship between landmark and trajector?a relationship that is often profiled?is one that is systematized in CG and has great explanatory power. In a word like ?flooded,? what is profiled is the part of the landmark that is under the trajector, as can be seen in Figure 2. 58 Figure 3. A CG diagram profiling the flooded region, in bold In CG diagrams profiled entities are bolded. Diagrams are meant to provide visual depictions of mental visualizations, and they are meant to do so in a somewhat systematic fashion.5 Profiling and landmark-trajector alignment are two aspects of meaning making that are often overlooked in linguistic analysis, yet they explain what elements on stage in the mind?s eye are in immediate focus. Elements on stage that are not in immediate focus, like the rest of Manhattan, the Hudson River, or the East River in the case of (3), are nevertheless essential for providing a detailed account for how meanings and images are constructed from just a few words strewn in a particular order. Without having the background of the rest of Manhattan and maybe even the rest of New York City, the World Trade Center fails to be a salient signifier. Part of our emotive reaction to the World Trade Center flooding is precisely because we understand it as an iconic and important symbol located in New York City. The dimension of prominence need not only tend to words or sentences. In discourse, utterances can be prominent if they stand out against the background of 5 While diagrams are not necessary for understanding CG descriptions, they can be helpful for guiding attention. Some find Langacker?s diagrams confusing and difficult to interpret. I sympathize and will try my best to describe all diagrams I produce. 59 shared beliefs or established information. In CG, the term ground is used to refer to the physical and conceptual space where a speaker and a hearer jointly attend to some conceptualization within a speech act event. When nominals are introduced into discourse, they become grounded, meaning that they become accessible to both speaker and hearer within the ground. From the ground, a speaker can point out a discourse item, like an image on a projection, and point to it: This is the World Trade Center Memorial. Both speaker and hearer have access to the image because it is within their ground. The demonstrative this directs the attention of the hearer to some proximal entity within the ground?close physically or temporally in discourse. Determinatives like this, that, the, and a thus profile nominals in relation to their status in the ground. As such, the way that nominals are grounded?e.g., the World Trade Center memorial vs. a World Trade Center memorial?identifies the speaker?s beliefs about the hearer?s prior knowledge of an entity and it prominent status in the ground. Perspective, the final dimension of imagery, also pertains to the ground. When conceptual content is construed in the theater of the mind, it is construed from a particular place with a particular point of view. When Gore says (4) After the horrible events of 9/11, we said never again He grounds himself and his audience as people who witnessed or otherwise were affected by 9/11. Moreover, 9/11 was an event that occurred in the past, as indicated by the use of the preterite verb said?rather than the futurative will say, for example. And further, 9/11 had a similar kind of effect on Gore and his audience, such that he can assume that he and his audience share in figuratively saying ?never again.? The 60 complexity of this utterance and the kinds of vantage points it imposes can be seen in Figure 3. Figure 3. Diagram of Gore's "never again" utterance The G in the figure indicates the ground, where Gore and the audience are collectively situated in space and common knowledge. When Gore asserts that we said never again, he puts on stage a series of events: 9/11, a moment of a we in a speaking event, and, within that speaking event, an utterance never again. The events that Gore puts on stage happen at some time before the moment of the utterance, even though the intention of the reported speech (never again) presumably extends up to and potentially beyond the moment of the utterance. The scene that Gore puts on stage is distal with proximity to the ground: it?s a moment in the past, away from the here and now of the ground. His utterance positions himself and his audience in a way so that they are looking back, observing events from a distance. The emphatic temporality of the adverb never, coupled with the illocutionary act depicted in the verb said further acts to align Gore and his audience to take a strong and absolute stance against the imagined ?horrible events.? Perspective, in this way, does more than just position audiences: it organizes vantage points for viewing and, by 61 extension, evaluating imagined scenes, cuing audiences to position themselves in the theater of the mind in accordance to a rhetor?s direction. What these dimensions of imagination provide is a way for describing how grammatical constructions and lexical choices cue the mind to construe scenes in the ways that they do. From this, one can analyze the relationship between grammar and mental movement, understand how words can focus attention on specific conceptual elements, and describe the sorts of background processes that are not explicit in the words in a sentence but are explicit in shared mental visualizations. This attention to language and cognition is crucial for a cognitive rhetorical account of the emotional appeal because it is able to describe with detail how language moves the mind. 4. The Theater of the Mind In the traditional CG stage model presented above, the rich understandings we experience in response to language emerges as the result of placing mental images on stage. The role of the stage holds a special place in the model, as the stage is where the actions play out. Yet, the stage is situated within a larger structure. The audience is stationed in the ground, not on the stage. The sorts of performed drama on stage differs in kind from the kind of reactions that audiences experience in their seats. To further the drama analogy, I propose expanding the model to encompass more of the elements that surround the stage. There is much to say about the experience of watching a play: perceiving actions on stage, evaluating the significance of an action, whispering comments to friends in neighboring seats, and directing attention to specific elements on-stage for larger argumentative purposes. A model of discourse 62 that accounts not only for the stage and the ground, but for the whole theater, also helps to show where patterns of intentional meaning making?including emotive meaning making?stem from experientially. Expanding on the stage model, I propose what I am calling the theater of the mind model. Whereas the stage model placed emphasis on actions playing out on stage and audiences viewing those actions from the ground, the theater of the mind model further carves out semantic space by attending to different logically distinct but recursively nested aspects of on-stage events. As a heuristic for analyzing linguistically cued simulated experience, the theater of the mind model can be used for metadiscursive purposes to analyze how language stages scenes in the mind. The model consists of four different ?levels,? four different kinds of (meta)events that occur and co-occur in mental simulations. These events are ?nested? in one another, moving from physical representations, to mental (meta-physical) ones, to communicative (meta-mental) ones, and finally to the rhetorical (meta-communicative) ones. At the first level (L1), the object of focus is the world of force-dynamic motions. At this level, objects appear on stage and interact with other objects in events. Zooming out to the second level (L2), conceptualizers are brought into the scope, and with them, their thoughts and feelings about the actions unfolding in L1. In the third level (L3), speakers and hearers share their thoughts and feelings communicatively, intersubjectively aligning their evaluations of the events occurring on stage. Finally, at the fourth level (L4), rhetors direct attention to events, thoughts, and attitudes to get their audiences to make inferences in accordance to shared cultural scripts of reasoning. While these 63 events appear as if they are teleological, moving from the physical to the rhetorical, it is important to keep in mind that, because of the nested nature of the theater of the mind, any given simulated event can play out?explicitly or implicitly?on multiple levels of the theater. That is, lexical and grammatical construction that encode force- dynamic interactions at L1 (e.g. the gun fired into the air) can cue affective responses to the event at L2 that communicatively align audiences in evaluating the act as shocking and undesirable at L3. In this example, what is cued at L2 and L3 is not encodes in the semantics of the sentence, but, given the proper context, these cued inferences are intersubjectively available. The theater of the mind model should not be taken as a model in competition with the stage model, it should be noted. Instead, the model is meant to draw out the patterned ways that meaning emerges, and thus, to organize linguistic moves in terms of the role they serve in the theater of the mind. From observing the stage, images are construed?with various degrees of specificity, prominence, and perspective. The stage is thus a vitally important place of meaning making. But the stage is not the only important place of meaning making. The movement and force between objects that are construed at L1?like memorials or floods?can lead to subjective thoughts and feelings at L2, intersubjective alignment at L3, and rhetorical inferences at L4. To illustrate how this model can shine light into the workings of emotional appeals, I provide brief descriptions of each level of the model below. 4.1. Level 1: Force-dynamic Relations 64 The first level of the theater of the mind is the most basic; it is the level of perceivable and conceivable things on all levels. The sorts of meaning that arise from this level are the kinds that arise as the result of running mental simulations of events. What is in focus is the existence of things (actors, objects, events), as well as their interactions across real and possible time. Utterances at this level constitute conceptions of reality?that is, encoded beliefs of things that are thought to exist by speakers and presumed to exist by hearers, so long as they do not conflict with their own conceptions of reality.6 The workings of this level of the theater of the mind corresponds most directly to what Langacker (1991a) has called the ?billiard ball? conception of reality, where physical or conceptual things exist and interact in accordance to folk models of physical interaction, or force dynamics. Things exist, move, and interact with one another, causing changes in a scene. These changes can be the result of physical force, as in the sentence The wind blew open the door, where the force of the wind causes the door to change position. Changes can also be the result of a kind of non- 6 The concept of a reality conception is borrowed from Cognitive Grammar, where it is used to explain, among things, finite clause complementation (Langacker 2009, 290-326). In Investigations in Cognitive Grammar, Langacker stresses that any utterance makes a claim on reality from a particular speaker?s knowledge of it, rather than acting as a representation of reality. My use of the term is congruous with Langacker?s and posits that reality exists independent of a conceptualizer?s perception and knowledge of it, and reality conceptions may deviate across conceptualizers or capture different aspects of reality from different perspectives or at different degrees of specificity. I do not aim to make conjectural claims here but to show how a speaker?s language can impose conceptions of reality that can often times be competing with conceptions held by audiences (You may think X, but Y) or be constructed in such a way to be in contrast with some other conceptualizer?s (While scientists think X, I think Y). 65 literal causitive force, as with the sentence The film made everyone cry, where it?s not the film itself doing the action but the act of watching the film that causes the resulting state of crying. In both cases, the focus is on the transfer of energy from one forceful entity to a receiving entity. If we were to go through Gore?s emotive argument at (1) and look for constructions that construe images at this level, we would find that virtually every sentence presents some aspect of L1 phenomena. This is because virtually every sentence has some propositional content that is composed of things interacting in force-dynamic relationships. It may appear, then, that this level is not very useful for analysis. Nevertheless, this level is necessary for conducting analyses of emotional appeals. Indeed, what L1 presents is events on stage, rather than implicit or explicit evaluations of those events. Those events, of course, profile particular elements and are construed from particular vantage points with varying degrees of specificity. As such, each expression in L1 presents a speaker?s conception of reality, a starting point for the more complex conceptual phenomena that arise in emotional appeals. The role of L1 is useful for explaining emotion arousal only via its relationship to levels higher up in the theater of the mind. The theater is multi-leveled and it is it is this multiple embedded structure of the theater that provides explanatory value of the model. 4.2. Level 2: Thoughts and Feelings Moving from L1 to L2 involves expanding the scope of focus outward beyond the stage. L2 includes not only the objective events occurring on stage but also the subjective thoughts and feelings of a conceptualizer watching those events. Figure 66 4(a) provides a visual depiction of the relationship between L1 and L2 in instances where the conceptualizer stages their thoughts or feelings, e.g., in first person complement clause constructions like I think that or I feel that, and figure 4(b) provides a depiction between L1 and L2 where the conceptualizer expresses thoughts and feelings indirectly, such as through the use of emotion words or evaluative language. (a) (b) Figure 4. Direct and indirect construal of thoughts and feelings In Figure 4(a), thoughts and feelings are expressed explicitly by a conceptualizer. Utterances like I think this is what would happen to Manhattan or I am upset that climate change is not being taken seriously would be captured by 4(a). A dashed arrow, representing a mental action, leads from the conceptualizer (C) to the event under assessment. A dotted line connects the conceptualizer in the ground to the conceptualizer on stage, the I in the utterance. When the conceptualizer asserts their thoughts and feelings using first person complement clauses licensed by verbs of 67 thoughts or feelings?e.g., I think or I feel?they explicitly mark their assessments of events on stage by putting themselves on stage as thinkers and feelers. They, in effect, make explicit their subjective conception of reality. Often thoughts and feelings are expressed indirectly, though, through the use of intrinsically evaluative words that categorize events as bad, good, a disaster, a real head scratcher, and so on. These indirect means of expressing thoughts and feelings are captured in 4(b), where a profiled categorization, specifically an affective coloring (A), is made of the event. As before in 4(a), a dotted arrow points from the conceptualizer (C) in the ground to the on-stage region. The dotted arrow is meant to represent the origin of the on-stage categorization as the conceptualizer?s in the ground. Rather than commenting on the event explicitly, emotion words and words of thoughts and feelings often categorize those events in terms of their valence (i.e., whether they are positively or negatively evaluated). When Al Gore says the horrible events of 9/11, for example, he categorizes his feelings about 9/11 without directly expressing that he is the origin of these feelings. These expressions are implicit. Much of the work that is done at L2 shows up in other levels as well. Conceptualizers voice their affective colorings, thoughts, and feelings usually to communicate with an interlocutor, to inform them of what they are thinking or to express themselves so that their interlocutor may better understand where they stand in relation to the event on stage. L2, however, only deals with recognizing subjective mental experiences on stage by an individual conceptualizer. These communicative functions of language, which require at least two conceptualizers, are captured in L3, where subjective feelings guide audiences to align affective assessments. 68 4.3. Stage 3: Social Interactivity and Value Alignment Whereas L2 attends to the subjective thoughts and feelings of a conceptualizer, L3 is the intersubjective dimension between two or more conceptualizers who are cooperating in an effort to communicate about events unfolding on stage. It is at this level that interlocuters experience ?common ground? (Stalnaker 1974, 1977; Clark). What the common ground consists of is shared situational and cultural knowledge mutually manifest and presumed to be known by conversation participants. Within discourse, conversational moves signal proposals to interlocutors to update and negotiate the common ground through interaction. While the process of updating the common ground may feel easy and effortless, subtle differences in the effort it takes to process a sentence and update the common ground can have significant pragmatic and rhetorical effects (Sperber & Wilson). Furthermore, that the experience of updating the common ground feels easy further speaks to the underlying socio- cognitive groundworks of communication. Principally, all conversational language usage occurs at this joint attentional level and requires interlocutors to follow conventions to systematically establish and conform to the discourse?s common ground. More than just conventions for using language, the common ground, as Tomasello (2014) has argued, carries with it conventions of cultural valuation, too. Tomasello writes, 69 Conventional cultural practices are things that "we" do, that we all know in cultural common ground that we do, and that we all expect one another in cultural common ground to do in appropriate circumstances. Thus in the open barter food market in which a conventionalized set of measurements for coordination is in place, if I show up with my honey in unconventional containers, no other traders will know what to do with me and my undetermined quantities of honey. With conventional cultural practices, deviations are not punished per se; they are simply left on the outside looking in. And there are some conventions that one cannot opt out of: one can wear this clothing or that clothing, or nothing at all, but whatever one wears, it is a cultural choice that will either conform to or violate the expectations of the group. (85) In this explanation, Tomasello looks at the cultural practices of trade and dress and shows how they, like all human symbolic interaction, depend on a shared common ground. Symbolic interaction, be it linguistic or trade convention, requires some assumptions about mutual knowledge between cultural participants. Most of this knowledge is unspoken, yet it has a big effect on how symbolic actions will be received. While praising a leader in one culture may be seen as exhibiting a positive and shared value system within the cultural common ground, in another culture, that leader may be hated and the same act of praise can distance a speaker from her audience, as it breaks from the cultural conventions of the audience?s common ground. 70 As with the relationship between L1 and L2, where an L2 phenomenon builds on an aspect of meaning in L1, an L3 phenomenon builds on the thoughts and feelings presented in L2. In L3, a subjective conceptualizer?s thoughts and feelings become a starting point for sharing and aligning attitudes toward jointly attended objects and events. This stage depicts the kinds of phenomena that occur in ordinary communication between speakers and hearers, including the names of culturally shared concepts (like emotion concepts), speech act verbs, conversational implicatures, and expressions of stance. For example, in Gore?s argument, utterances like we said never again and is it possible that we should prepare against other threats besides terrorism evoke a communicative frame that involves a speaker and a hearer. This frame is lexically anchored via speech act verbs (said), first person plural pronouns (we), and even non-canonical clauses such as interrogative clauses (is it possible that?) that indirectly represent a suggestion, a demand, or a request the hearer. Declarative expressions at L2?for example, an utterance like that?s not nice?often have implicit communicative intentions, too; in this example, the speaker?s expression of a negative evaluation about some event that was jointly attended to is used to update the common ground so that discourse participants agree on the negative evaluation of that event. Speech act verbs like demand in I demand you give me a refund or complain in The customer complained to the manager express ways that speakers, hearers, and conceptualizers align themselves in L3 with respect to their L2 feelings or L1 objects of evaluation. Given that the sorts of phenomena L3 covers are vast, I will only focus on the sorts of phenomena that recurrently appear in the construction of emotional appeals. 71 In particular, these phenomena are the alignment of speakers and hearers in the production of evaluative stance, the activation and reinforcement of value systems to produce coherence, the relationship between endoxa and value expectations in discourse, and the complex role of emotion scripts, wherein patterns of feeling are named and shared in a culture. A chief goal of acts in L3 is to establish evaluative stances and ultimately to align stances and meanings between a speaker and a hearer. Much of arriving at the intended meaning of an utterance requires audiences to express charity and coordinate their beliefs with a speaker?s. Sometimes linguistic constructions can guide this coordination. For example, when Gore makes the claim that After the horrible events of 9/11, we said never again, he places himself and the audience in a collective we, as was shown in Figure 3 above. This move creates a fictive shared identity between the conceptualizers in the ground and thus presupposes a shared belief about the desire for making sure an event like 9/11 never occurs. Even if this sentiment is Gore?s alone, for the audience to understand the meaning of this utterance, they must entertain Gore?s feelings toward 9/11 as ones that are shared. Moreover, Gore never explicitly says what his views toward 9/11 are, nor does he state what he believes the significance of the World Trade Center memorial is, nor does he state that he believes a flooded Manhattan is a bad thing. The audience must come to this understanding through conceptualizing L1 phenomena (the World Trade Center memorial, 9/11, floods, Manhattan, etc.) and L2 phenomena (9/11 is categorized as a horrible event, Manhattan flooding would be bad, fear of flooding is a plausible reaction, etc.), as well as presuppositions within the common ground. Regardless of whether or not the 72 audience themselves believe that 9/11 was a horrible event, they know that Gore believes it was. And regardless of whether or not the audience cares that a memorial in Manhattan could find itself drowned, the audience knows that Gore cares. And not only this: the audience knows that Gore wants his audience to share his evaluations of that memorial and that act of flooding, or at least recognize that, to some degree, the speaking event requires them to entertain that evaluation for coherence?s sake. When hearers coordinate their belief systems to construct coherence, they achieve intersubjective alignment, a principal mechanism by which emotional appeals recruit audiences to reinforce specific acts of evaluation. These acts of evaluation derive from a certain value system, and patterns of evaluation from that value system go on to reinforce that value system. Holding the World Trade Center memorial in high esteem, for example, speaks to a nationalistic value system where the lives of innocent Americans who died as the result of terrorist acts reinforce the importance of stopping terrorism and supporting national security. Speaking about the World Trade Center memorial in the context of 9/11, then, positions the memorial as an object of evaluation and aligns speaker and hearer in accordance to the value system that guides that evaluation. Contrary to the belief that language is merely used to express and share ideas, what the intersubjective dimension of L3 shows is that language is just as important for shifting and aligning attitudes to construct coherent and cooperative communicative joint intentions. At L3, propositional utterances that arise in L1 communicative acts are simultaneously the basis for aligning attention and evaluation. Construing an on-stage event in L3 requires speakers and hearers to attend not only to the events as they 73 unfold but also the beliefs, feelings, knowledge, and intentions of their interlocutors. Figure 6 illustrates a schematic scene of evaluation, modeled on the linguist John DuBois?s (2007) ?stance triangle? (163). Figure 6. The Stance Triangle The stance triangle fits into the theater of the mind model neatly. The element of the valued object arises as the result of thoughts and feelings from L2 of conceptions of reality at L1. The mention of the World Trade Center memorial, followed by the utterance about the horrible events of 9/11, causes conceptualizers to understand the memorial as something that the speaker believes is of value in the discourse. If the hearer did not value the object, they would need to reposition themselves in accordance to the object under evaluation in order to achieve alignment.7 Because speakers hold the floor and are the ones who are able to provide L2 categorizations 7 DuBois?s model of the stance triangle is primarily used as a model for analyzing conversation, and, thus, the roles of Speaker and Hearer in his model are replaced by Subject. What DuBois is interested in is how stance is marked grammatically, e.g. via repetition in consecutive lines of conversation followed by lexical markings like either or also, indicating intersubjective alignment in evaluation. When alignment is not achieved between subjects, one subject will reposition themself in a stance following move that, for example, hedges or negates an initial stance. 74 (horrible) of the display of the L1 phenomena on stage (the events of 9/11), hearers follow reasoning cooperatively or they assert protest?by interjecting, turning away from the displayed valued object, or ceasing to pay attention altogether. While the stance triangle is helpful for analyzing how moments in discourse are able to align speaker and audience through evaluation, it doesn?t tell the whole story. What DuBois does not represent in his model is a backgrounded logic of evaluation, grounded in a value system. The role of value systems in evaluative and even argumentative discourse has been studied linguistically elsewhere. George Lakoff (1996), in Moral Politics, his work on political argumentation, posits that American progressives and conservatives operate by two separate value systems. These value systems presuppose that objects and events present meaning in concordance to a specific hierarchy of values. While Lakoff is focused primarily on American political value systems and political reasoning, it goes without saying that competing value systems arise in the reasoning of other cultures and areas of social interaction as well. Through repetition of evaluation in accordance to these value systems, the logics of evaluation become entrenched and shared across members of communities. Lakoff?s point, it seems, is shared among classical rhetorical theorists, who believe that rhetorical reasoning often arises from endoxa, or widely shared beliefs among members of a community. In crafting coherence at L3, value systems which reside in endoxa can help to provide lines of reasoning that connect leaps in reasoning in discourse. Take, for example, the implicit leap in reasoning between the first and second sentences in (5): 75 (5) The area where the World Trade Center Memorial is to be located would be under water. Is it possible that we should prepare against other threats besides terrorism? Maybe we should be concerned with other problems as well. The rhetorical question in the second sentence, along with the suggestion in the sentence that follows, presupposes that the hypothetical flooding in the first sentence is a threat and a problem. But how can the flooding of a memorial be thought of as a national threat, let alone on the same level as a terrorist attack? For a hearer to make sense of this, they must recruit an endoxic premise, a line of reasoning that emerges from endoxa to provide evaluative coherence. 8 The kind of endoxic premise that is 8 I use the term endoxic premise rather than other terms to emphasize the general line of shared reason underlying an argument and inviting inference. My term is similar the concept of topos introduced into pragmatics and semantics by the French linguists Jean-Claude Anscombre and Oswald Ducrot (1983) in their Theory of Argumentation in Language, and to Toulmin?s (1958) concept of warrant in argumentation theory and informal logic. It better connects to the role of common ground, however, by emphasizing endoxa, or propositions generally assumed to be shared between interlocutors in a culture. I believe that the concept of endoxa in enthymematic arguments connects classical ideas of invited inference to contemporary models in pragmatics and semantics. In the Theory of Argumentation in Language, Anscombre and Ducrot trace the relationship between words and the patterns of reasoning they evoked; they unfortunately call these patterns of reasoning topoi. This term may be confusing for rhetorical scholars, as Aristotle calls a topos a ?heading under which many enthymemes fall? (2.26.1). As such, rhetorical topoi are not propositional in and of themselves but organizational or classificatory schemes. Because the propositions of Aristotelian rhetorical arguments stem from premises that are widely held to be true (i.e., endoxa) or at least plausibly true (see Renon [1998] for more on the point of plausibility and endoxa), I choose to call the general and shared premises under consideration endoxic premises rather than topoi. Endoxic premises, of course, can function within kinds of topoi, as they are recruited when constructing coherent meaning in underspecified contexts, but they are not topoi in and of themselves. 76 recruited amounts to something like the more damage done to a symbolic structure, the greater the threat to the real institutions it represents. When the ground of the World Trade Center is attacked by terrorist, this attack is understood as an attack on the United States, its symbolic institution. Similarly, when a flood damages the World Trade Center memorial, the damage done to it can be conceived of as threatening the symbolic institutions that it stands for: perseverance against terrorism, freedom, equality, the American way, and so on. The consequences of the argument are clear. As the environmental correspondent Richard Black (2006) put it in his review of An Inconvenient Truth for the BBC, ?what we have done for victims of terrorism will be compromised by what we have not done on climate.? While this line of reasoning is far from logical in the strict sense, it makes sense in accordance to everyday rationality, with the help of endoxa. In addition to helping to craft coherence in reasoning, endoxa is often used to align or contrast evaluations. In the Gore example, the value of the World Trade Center memorial is aligned in order to craft coherence, and in this act, its value is reinforced. When objects that a culture typically evaluates one way are evaluated Also worth noting is the difference between an endoxic premise and the ?suppressed major premise? of the neo-Aristotelian enthymeme as well as in the Toulminian warrant. Endoxic premises are presumed to be plausibly true and recruited in retrospect; that is, one hears one utterance and then another and finds a way to make sense of the relationship between the two. What writing out the endoxic premise is meant to represent is the kind of idea that would allow for reasoning between the two; it does not represent an intentional and sly rhetorical strategy for causing audiences to produce arguments in themselves, which is the most common reading of the role of the suppressed major premise in the neo-Aristotelian model of the enthymeme. In other words, endoxic premises are ad hoc but model an example kind of reasoning, especially as that reasoning relates to maintaining coherent valuation. 77 differently, the act of expressing a different kind of evaluation can be shocking for hearers. Take the following case, for example. In a press conference at a music festival in Hamburg six days after September 11th, the German composer Karlheinz Stockhausen was reported to have referred to the attack on the Twin Towers as the ?greatest piece of art that has ever been.?9 Much of Germany, let alone the world, was still mourning the lives of those who died from the terrorist attacks, so when Stockhausen categorizes the attack on the Twin Towers as performance art?an L2 move?what he attempts to ground is in contrast to the backgrounded endoxa that 9/11 was a tragedy. The clash between Stockhausen?s assertion and the presumed shared knowledge constituted a kind of discourse violation; Stockhausen introduced a new way of viewing the attacks that was not shared by his interlocutors? value system. As a result, Stockhausen?s audience may question what common ground for reasoning they share with Stockhausen. Stockhausen?s concerts were cancelled in protest. Stockhausen?s own daughter went as far as to publicly disown him and 9 Stockhausen?s comment was transcribed and published in various news outlets, including the September 27, 2001 Die Zeit article, ?Irre Funkspr?che aus dem All? by Claus Spahn. The original German reads: ?Was da geschehen ist, ist - jetzt m?ssen Sie alle Ihr Gehirn umstellen - das gr??te Kunstwerk, das es je gegeben hat. Dass Geister in einem Akt etwas vollbringen, was wir in der Musik nicht tr?umen k?nnten, dass Leute zehn Jahre ?ben wie verr?ckt, total fanatisch f?r ein Konzert, und dann sterben.? [?What has happened is?now you have to change all of your brain?the greatest work of art ever. That ghosts in one act accomplish something we could not dream in music, that people practice for ten years like crazy, totally fanatical for a concert, and then die.?] Shortly after his concerts were canceled, Stockhausen claimed that the quote was taken out of context and published for shock value by the newspaper. Anthony Tommasini (2001) of the New York Times documented reactions to Stockhausen?s comments in his article. 78 question his sanity. Sometimes, foregrounding evaluations that contrast expected endoxa can have visceral social and emotive effects at L3. The very concept of emotion is also something that comes relevant at L3. Like endoxa, emotions are things that are shared in a culture. As the result of communicative needs, patterns of subjective affective experiences, like the kinds that are profiled in L2, are given names at L3. As discussed early in the discussion of L2 phenomena, many names for emotions, like sadness, anger, and panic, are categorizations of complex events and experiences. For example, the kinds of events that can be construed as signs of anger may include nostril flaring, a customer yelling, or a man punching a wall. The meanings of these categorizations, however, do not emerge solely from the content of their L1 events. There is nothing inherently emotive about the trajectory of a fist onto a wall. The events are recognized as being emotive insofar as they fit into larger emotion scripts. Nostrils flair, customers yell, and men punch walls as the result of some perceived slight that they believed to have experienced. The experience of the slight comes before the behavior reaction, at least according to the script of anger. This insight that emotions are scripted concepts was at the crux of Aristotle?s theory of pathos. Aristotle believed that a scenario that could be recognized as one that induces anger involved a specific cast: a slighter, someone slighted, an offending act, and a wish for revenge (2.2.1). While Aristotle leaves the script of that event unstated, common sense tells us that, for example, a perception of a slight occurs after a slight occurs, not before it, and it is perceived by someone slighted, not by a 79 slighter. Aristotle leaves this script unstated because we all know the script of anger.10 It?s part of our cultural knowledge. Contemporary work in cognitive science and emotion science has likewise recognized that emotion events are ordered into scripts. George Lakoff (1987) calls these scripts ?Idealized Cognitive Models,? or ICMs for short. In Women, Fire, and Dangerous Things, Lakoff posits that the ICM of anger is comprised of five key roles: a victim, who is a person S; an agent of retribution who is usually that same person S; the target of anger, who is a wrongdoer W; a cause of anger, which is an offending event; and some kind of angry behavior, typically an act of retribution. These roles organize themselves in five different stages: 10 This is not to say, however, that all of what Aristotle says about org?, the Ancient Greek equivalent to the English word anger, translates across cultures. Furthermore, recent criticisms, such as that by Daniel Gross (2006), argue that Aristotle had an ?elitist? view of the emotions, supposing that the king has more right to feeling than his people, workers, or slaves (5). This contemporary reading of Aristotle misses that Aristotle?s description of org??specifically that kings tend to get angrier because they can more readily feel the weight of a slight (2.2.27)? was not meant to be read as normative. Aristotle is simply observing the expectations within the social order he lived under. Grimaldi (1988), in his commentary on book II, attempts to remedy this reading by arguing that Aristotle?s intention was to say that, in a moment of anger, the slighted believes their needs to be more important than those of their slighter. Grimaldi writes, ?If one examines the instances given by A. in this chapter of those with whom one becomes angry, it becomes clear that the one who causes anger is not necessarily an ?inferior,? but rather someone who should not by all that is right?show disdain to the other? (23-24). I agree with this and stress the importance of recognizing how emotion concepts like anger and org? differ across cultures. 80 ?Stage 1: Offending event Wrongdoer offends S. Wrongdoer is at fault. The offending event displeases S. The intensity of the offense outweighs the intensity of the retribution (which equals zero at this point), thus creating an imbalance. The offense causes anger to come into existence. Stage 2: Anger Anger exists. S experiences physiological effects (heat, pressure, agitation). Anger exerts force on S to attempt an act of retribution. Stage 3: Attempt to control anger S exerts a counterforce in an attempt to control anger. Stage 4: Loss of control The intensity of anger goes above the limit. Anger takes control of S. S exhibits angry behavior (loss of judgment, aggressive actions). There is damage to S. There is danger to the target of anger, in this case, the wrongdoer. Stage 5: Retribution S performs retributive act against W (this is usually angry behavior directed at W). The intensity of retribution balances the intensity of offense. The intensity of anger drops.? (Lakoff 1987, 400-401) What Lakoff?s ICM of anger shows is a prototypical way of understanding the ordering of events and the ordering of affective experiences in the larger conceptualization of anger, indeed quite a social phenomenon. Moreover, these conceptualizations are culturally shared and thus able to be evoked through the use of 81 linguistic expressions, like the use of emotion words and metaphors that construe internal experiences in terms of pressure, force, and heat. Not all utterances or even strings of discourse that evoke anger need to speak to all aspects of the ICM. Given some description of an event or expression of feelings, so long as it is recognizable as a stage within the ICM, the entire concept of anger comes to mind, along with access to who is feeling what and at what point in the script of anger. In her 1999 work Emotions Across Languages and Cultures, the semanticist Anna Wierzbicka also articulated the relationship between emotion words and the larger mental scripts that they bring to mind. Similar to Lakoff?s ICM, Wierzbicka organizes events into kinds of mental scripts, only Wierzbicka?s scripts, unlike Lakoff?s, are ego-centric, meaning they take the perspective of a conceptualizer. They also use a limited vocabulary?Wierzbicka?s Natural Semantic Metalanguage?that focuses particularly on acts of thinking and feeling. Here is an example of the script for the English emotion concept panic: Panic (X felt panic) (a) X felt something because X thought something (b) sometimes a person thinks: (c) "something is happening now (d) if I don't do something now something very bad will happen to me because of this (e) I don't want this to happen (f) because of this I have to do something now (g) I don't know what I can do (h) I can't think now" (i) when this person thinks this this person feels something bad (j) X felt something like this (k) because X thought something like this (82) Like Lakoff?s ICM, the script provides an ordered sequence of events (thoughts, 82 feelings, and action), resulting in the registering of a meaningful and shared feeling event (what the emotion word profiles). Parts of the script involve L1 actions (something is happening now, something?will happen, I have to do something), L2 conceptualizations (X felt something because X thought something, something bad will happen, I don?t know what I can do, I can?t think now, X felt something like this), and the whole script is couched within shared knowledge between speakers and hearers in L3 (sometimes a person thinks), as well as the specific sequencing of the script, given cultural practices. Emotion concepts, like any other concept, bring together different kinds of physical and mental events so that they may be shared intersubjectively. Core affective experiences may be felt and registered at L2, but the abstracted generalization that a word for an affective experience is not the experience in itself but a naming of that experience can only be captured once there is a shared ground of knowledge, provided by L3. As such, Wierzbicka?s script also illustrates the importance of L1, L2, and L3 phenomena in the conceptualization of emotions. What L3 of the Theater of the Mind model presents is a shift from the I in L2 to a you and a we. With the shift comes the ability to express, request, command, or otherwise communicate intentions to another conceptualizer. Speakers can express their evaluations at L3 to align attitudes between interlocutors toward events occurring in L1 or to recognize subjective experiences in L2. They can try to align stance with audiences and draw on presupposed shared knowledge and shared evaluations (endoxa). Coming to intersubjective alignment can ally speakers and hearers in value systems and, thus, create a sense of identification. And thinking and 83 evaluating similarly may even lead to feelings of trust or comradery, two emotion concepts that require alignment with another conceptualizer. Words for affective experiences also arise in L3, as these words stand in for the experiences of prototypical scripts shared between speakers, hearers, and their cultures. These scripts prototypically focus attention on evaluative scenes unfolding, at once providing access to a whole simulated scene and aligning evaluations within the common ground. What emotion words provide is names for the embodied and simulated events occurring in the minds of audiences. This is to say that intersubjective alignment is not just a passive process of connecting dots but an event that can discursively arise from imagination prompted by language. 4.4. Stage 4: Persuasive Strategies So far, the theater of the mind has attended to actions on stage (L1); the thoughts and feelings that those actions subjectively produce (L2); and the performative, intersubjective ways speaker and hears can speak about those actions, categorize patterns of affective experience that arise from those actions, and align attitudes toward those actions (L3). All of these levels of the theater of the mind are vital to L4, which is the level that tends to the rhetorical use of persuasive strategies in discourse to shift position and attention for argumentative effect. While events are played out on stage in L1 (The Hudson Bay will flood Manhattan), categorized at L2 (This will be a disaster!), and intersubjectively shared at L3 (We should be scared!), the intentional aspect of persuasive argument construction has yet to be worked into the model. What L4 brings into focus are the moves a rhetor makes to construct and 84 arrange scenarios, reframe conventional understandings, direct reasoning, and spotlight attention for argumentative effect. L4 strategies include elaborate attempts at rhetorical persuasion, where audiences are led to change their beliefs through argumentative techniques, as well as more culturally situated discursive practices, such as ritualized politeness, irony, make believe, and formal debate. Crucial to L4 rhetorical intentions are the construction of scenarios. A scenario is an event or series of events that resemble a series of related lines in an emotion script. For example, an utterance like That red car just cut me off resembles an event within the anger script. The event that the script resembles is the initial event, where a conceptualizer experiences someone doing something bad to them, and they feel bad as a result. As a result of this action, the conceptualizer wants to do something bad to the wrongdoer. The scenario captured in the utterance That red car just cut me off can stand in for the first event in the anger script but not the resulting desire for retribution. Yet, any hearer of the utterance can infer that the speaker of the utterance may have a desire for retribution. Hearers can infer this because the construal of a scenario as a part of an emotion script provides access to the entirety of that emotion script. Reasoning from part allows audiences to infer conclusion via reference to the whole. By producing discourse scenarios which bring to mind larger cultural scripts, rhetors at L4 can construe L1 events for L2 and L3 effects. What the rhetor at L4 deals with, then, is the intentional guiding of attention for the purpose of activating cultural scripts from which audiences can make inferences. In this sense, a rhetor at L4 is like a director. How L4 provides a rhetorical?rather than an experiential or 85 social?aspect to the theater of the mind is through its focus on selecting scenarios to put on stage. Attending to scenarios, in turn, allows the audience to make predictions about the arguments under discussion via attending to the coherence between events in scenarios on stage and the larger cultural scripts they bring to mind. Audiences, in effect, simulate the scenarios and play out the stories themselves to make inferences about what happened, who?s at fault, how specific actors ought to feel, and what ought to be done, if anything at all. The audience, thus, plays a crucial role in completing the scenario, given partial instruction, and each audience member, likewise, may complete the scenario in accordance to their know individual knowledge, experience, and imagination. In Cognitive Linguistics, the process by which a part of an event can bring to mind the entirety of the event is called frame metonymy, and it has been observed to be a chief mechanism for constructing inferences given partial information (Dancygier and Sweetser 2005, 2014; Barcelona 2003; Littlemore 2015). The claim is that emotions are discursively aroused in audiences by rhetors when rhetors put on stage scenarios that resemble lines within emotion scripts and thus metonymically recall the emotion script. Take, for example, the discursive arousal of panic in Figure 7. 86 Figure 7. Example of frame metonymic emotion arousal of panic The figure represents the utterances that Gore selects to construct the scenario he stages in his argument. Much of the scenario he constructs resembles a specific line (d) in the emotion script for panic. This resemblance, of course, requires audiences to have access to that emotion concept already. The recognition of resemblance, then, allows for the scenario to bring the entirety of the emotion script to mind. From this, audiences are able to make inferences about the scenario Gore is articulating by playing out the rest of the emotion script. Understanding the component parts of the emotion script allows rhetors to make predictions about how to arouse panic in an audience, given ordered descriptions of scenarios involving actions, thoughts, and feelings. For example, rhetors can craft scenes of panic by elaborating on lines (c)-(f) of the panic script?climate change is causing the ice in Greenland to become displaced (c), and that displacement will cause a flood that will submerge the World Trade Center memorial (d); We want the World Trade Center memorial to remain a symbol of America and not be flooded (e); We must act now (f). Rhetors can, in their construction of scenarios, also provide 87 potential solutions to the events that cause panic in order to alleviate the feeling of panic and transform it into something productive. When they do this, they take advantage of predictions of conceptualization from lines (g) I don?t know what to do and (h) I can?t think now. Because the audience doesn't know what to do after feeling panic, rhetors can provide the audience with an action?for example, to sign a petition. This proposed action may potentially solve the audience?s problem of not knowing what to do by providing them something to do, yet this proposed action is also potentially favorable to the rhetor. Because the audience can't think now, the rhetor can diminish the need for them to think to by explaining how others are thinking about this problem (h), a rhetor can think for them. All they need to do is provide money, become a member, support a politician, pledge allegiance to a social organization and all their problems are solved. Performing these actions is, in fact, preferable. In this way, the construction of scenarios becomes the construction of emotional appeals. The emotion script provides a skeleton from which strategies for constructing emotional appeals may be generated. It provides events, concepts, actors, actions, and contexts to fill in for words like ?something? in the lines I must do something or Something happened. It can help focus attention through controlling construal of the event using epistemic modality to influence possibility, thus providing a strategy for actualizing line (d). It can bring attention to the exigence of the situation and the need for immediate response through temporal deixis, thus foregrounding some event in (c) and amplifying the need to act in (f). It can use counterfactual constructions and perhaps evaluative adjectives that emphasize the negative appraisal of the 88 counterfactual future situation in which the reader doesn't act. The chief discursive strategies for arousing panic, if use practically by a rhetor, ought to consists of discursive moves that 1) appeal to panic metonymically via scenario selection, and 2) attempt to relieve panic with the proposal of an action. While the construal of the force-dynamic events in L1, the registering of thoughts and feelings in L2, and communicative actions in L3 are often tied to specific kinds of linguistic constructions, the construction of emotional appeals in L4 are necessarily unannounced. As Liu and Zhu (2011) note, rhetorical acts are non- cooperative in nature, because to announce the intended effect?e.g., a lawyer announcing you are now persuaded?more often has the opposite effect. Liu and Zhu argue that, far from being cooperative, no courtroom lawyer worth their weight in salt would walk into a court room assuming that they can rely on their audiences to produce the effect that they desire to produce. If this was the case, rhetorical argumentation would be unnecessary (3409). Imagine someone approaching you and telling you you now feel panic. That would probably make you feel annoyed, not panicked. Instead of announcing their rhetorical intentions, rhetors construct scenarios that, when simulated, arouse the thoughts and feelings they intend to arouse by getting audiences to activate emotion scripts. For the audience of an L4 rhetorical strategy, effects are felt in the presentation of events in L1, the thoughts and feelings they arouse in L2, and the social concepts, values systems, and positionings that follow in L3. The cumulative conceptual, affective, and evaluative phenomena constructs what I call affective force dynamics in audiences of emotional appeals. Much like the 89 physical force dynamics of L1, affective force dynamics present things enacting force on other things. The relationship between a hungry wolf and an approaching innocent sheep, for example, would express not a physical tension but an affective one. This affective tension can go on and, for example, be construed as a grounds for metonymically arousing fear. An example of how force dynamic experiences can help explain argumentative affective arousal was given in the example in Chapter 1 (1.2.3) when analyzing the bumper sticker that read War doesn?t show who?s right just who?s left. The observation there was that the relationship between tension in who?s right and the relief in who?s left amplified the potential arousal of somber feelings. The tension in who?s right increases core affective arousal, and the relief of who?s right decreases that arousal, motivating the recruitment of emotion concepts prototypically associated with high and low arousal with negative valence?for example, anger and sadness, respectively. This analysis did not attend to the ways that L1 expressions motivate L2 affective phenomena, however. The relationship between L1 and L2 is made clearer by looking at Gore?s argument. The use of L1 stative constructions like the would in would happen [to Manhattan] or in would be [underwater] construe the event of flooding as a bounded potential action, which is not one of high tension but one that indicates a forceful change taking place. This potential action is also viewed from a vantage point both distal in time and place, as the flood would happen in a possible future. As a result, the potential affective force dynamic scene is not one of particularly high arousal, since the event of the flooding is in a hypothetical future mental space and not the 90 actual or immediately certain world. Still, there is a potentiality of producing a high arousal. If Gore succeeded in framing the event of flooding as inevitable, his audience might well construe that inevitable, negatively valenced, high arousal as approaching the here-and-now of the ground. (a) (b) Figure 5. Construals of force-dynamic events that denote potential fear (a) and dread (b) The illustrations in Figure 5(a) and 5(b), don?t represent any particular utterance but the affective construal of Gore?s animation of the World Trade Center memorial flooding, accompanying the utterance This is what would happen to Manhattan. The red surrounding the on-stage region is meant to denote that the element in question carries a negative valence. The double-lined arrow inside the on-stage region and the 91 one in the L2 region, too, are red. These arrows denote force dynamics evaluated as having negative valence. In the on-stage region, the physical event the arrow represents is the flooding of Manhattan, an L1 phenomenon. The arrow in the L2 region, however, denotes an affective force dynamic relationship, one in which a negatively construed entity temporally approaches the present. There are various ways of categorizing the thoughts and feeling associated with construing the approaching flooding of Manhattan, a negatively valenced high- arousal event. A conceptualizer could focus on the flooding of Manhattan alone and think it was a terrible event, and thus, as it approaches, it is something to fear. This construal is depicted in 5(a). When a conceptualizer recognizes the negative event approaching but, instead of fixating on the event, fixates on the inevitability of it occurring, the conceptualizer pays closer attention to the temporal aspect of the event approaching. This could lead to feelings of fear, as with the construal in 5(a), but it may also present feelings of dread, anxiety, or even panic if audiences construed the approaching event as one that they believe they cannot do anything about. What this goes to show is that understanding the valence and arousal of the elements on stage can only go so far to definitively predict the kinds of affective experiences audiences can have. The same mental simulation in different audiences can produce varying subjective experiences of fear, dread, anxiety, and panic. It should be noted, however, that these are all within the range of intended emotions that an audience can presuppose Gore is trying to get them to feel, given the context. It would be uncooperative and incoherent, then, for an audience to feel happiness, believing that Gore is excited for warmer weather and a new shoreline. This is to say that, while the 92 experience of core affect and emotion can differ across audiences, this difference is still able to be predicted; the evoked emotion of an emotional appeal is not wholly relative. It is the conclusion to a shared emotion script. So far, the discussion of L4 has only looked at what the rhetor does and has ignored the role of the audience. Because the rhetor?s audience experiences the rhetor?s argument through L1, L2, and L3 phenomena, this may lead one to think that the audience has a completely passive role in L4. This is not entirely true. I posit that recognition of a rhetor?s persuasive intentions can become registered at L4 in audiences. While the structure of the sorts of emotional appeals discussed so far may require audiences to perform mental simulations in order to come to the effects a rhetor intends to produce, not all emotional appeals succeed in producing emotions. Importantly, the theater of the mind model also provides an avenue into analyzing how failed attempts at emotional appeals also demonstrate skillful acts of imaging on the part of the audience. As mentioned before, the Gore example that is used throughout this chapter has been criticized as misleading and emotionally manipulative. Katherine Mieszkowski (2006) wrote for Salon that ?[a]udiences might be left with the impression that the deluge is just around the corner, lapping at our feet.? In a 2007 court case that appeared before the High Court of England and Wales, a man named Stuart Dimmock brought a suite against the Secretary of State of Education and Skills, who refused to include Gore?s Inconvenient Truth as part of the standard curriculum for secondary school education. The presiding judge, Michael Burton, commented on the scene of the flooding of Manhattan, calling it ?distinctly alarmist,? arguing that the effects that Gore presupposes would not happen for a 93 hundred years (Dimmock v Secretary of State for Education & Skills). Where did Gore?s argument fail? With certainty, we can say that it was not in the construal of the scenario, nor was it in the registering of the emotions. The failure actually comes about as a result of external circumstances, namely the audience?s own understanding of climate change. Gore?s appeal to panic depends on audiences not having a proper timescale for the flooding of Manhattan, and Gore himself doesn?t offer a timescale. What Gore?s critics voice is a recognition that admission of aspects in L1 helps further a rhetorical objective in L4. It isn?t that Gore?s critics were not able to feel the intended effects of his emotional appeal; it?s that they were able recognize where Gore?s attempts to guide subjective construal misrepresented the objective conceptualization. To be able to recognize the lack of a timescale takes some skillful mental visualization. The theater of the mind model allows analysts to understand where that visualization is taking place. 5. Conclusion: Returning to the Euphantasiotos The metaphor of the stage is rather apt for thinking of simulation. Kenneth Burke (1942/ 1962) used it for understanding complex intentional actions in his pentadic methodology in his Grammar of Motives. Similarly, Erving Goffman (1956) used the stage metaphor to model complex social behavior when developing his dramaturgical theory. Drawing on Goffman and Burke, Chris Holcomb and M. Jimmie Killingsworth (2010) use the stage metaphor to model how stylistic linguistic and rhetorical behavior plays on and subverts the expectations of audiences. And Ronald 94 Langacker (1991a) uses the stage to offer a way of modeling how conceptualization occurs in discourse. Stages, coincidentally, are also not too far from Tomasello?s (2009) joint attentional scene, where a speaker uses communicative constructions to influence a hearer to attend to a conceptualization. Arie Verhagen (2005) has argued, following Anscombre & Ducrot?s Theory of Argumentation in Language and Langacker?s usage-based principals of Cognitive Grammar, that linguistic constructions work to coordinate minds in the joint attentional scene toward specific intersubjective construals. And the conversation analyst and grammarian John Du Bois (2007) has extended this joint attentional and intersubjective scene to account for the ways that discourse participants evaluate joint conceptual objects, position themselves toward these objects, and, in doing so, align themselves with or against their interlocutors. This is all to say that the idea of a theater of the mind is not altogether original. What is unique is the system developed for understanding how words relate to the activation of images in the mind, as well as how images in the mind construct rich meanings across various levels of physical, mental, and emotive experience. The metadiscursive language that the theater of the mind provides is unique in that it provides an insight into the ways that language makes us think. Importantly, it provides a way for articulating the invention of emotional appeals in accordance to the project of the cognitive rhetorical model of pathos. Also of importance is the epistemological foundation that the theater of the mind model provides. My model is grounded on established empirical results and takes into consideration the ideas of attention, social cognition, imagination, and 95 emotion from several distinct fields. The result is a model that provides a shift in epistemology, providing a different way of modeling how we think about the reality of imagined entities. Levels 2, 3, and 4 of the theater of the mind model importantly do not highlight physical entities but imagined, emotive, social, and rhetorical ones. Being able to show how these entities are brought into focus in the theater of the mind and being able to ground these observations not in speculation or intuition but in the current empirical cognitive science is an important contribution to the analysis of symbolic actions that rhetoricians analyze. Such a framework is sorely needed, too. Heinrich Plett, in his 2012 Enargeia in Classical Antiquity and the Early Modern Age, emphasizes how the role of imagination has been weened out of rhetorical language processing in the shift to modernity, making very clear that, in antiquity, the average listener or reader of rhetorical discourse was assumed to be a kind of cognitive ?artist,? constructing vivid impressions of the speech in real time and responding to the demands that their mental pictures entailed (4). Moreover, Plett skillfully articulates the assumption in Quintilian?s pedagogical advice for constructing moving tragic scenes (Institutio Oratoria 8.3.68-72), which is not to produce vivid impressions of a unified whole of a scene (tortum) but a multiplicity of the parts of a scene (omnia) (9). Plett?s argument is that the classical attention to stylistic detail and sensorial amplification makes attention and, by extension, engagement more ?palpable? to even the most critical audiences, thus constituting a rhetorical stylistic strategy that was well articulated in antiquity and in the Early Modern era but that has fallen out of popularity. What once was a ?fundamental concept? in antiquity and the Early 96 Modern era has become one of many as classical conceptions of imagination fell out of favor (Plett 195-196). Perhaps a reason for it falling out of favor is due to a lack of a robust methodology for linking the imagination to the verbal arts. Missing from the classical discussion of emotion-construction, as well as from its recent renaissance, is an understanding of why we feel, what we feel, and how we guide feeling with words. The concept of granularity of construal, to provide just one example, helps to understand word choice not just in terms of specificity but in terms of zooming in to the right level of detail and specificity for emotive effect. When Gore makes his argument, he asks us to imagine the World Trade Center being flooded, not lower Manhattan, not New York City, not part of America, not roughly 380,000 lives. He asks us to zoom in on an image that is easily construed and loaded with cultural value. We care about valued things, and when we imagine those valued things in bad situations, we are liable to feel moved from a good state to a bad one. What the theater of the mind model provides is a framework for connecting moves in discourse to the imaginative and emotive phenomena they conjure in the minds of language users. Too often emotive phenomena are thought to be outside of the realm of rhetoric and linguistics, yet what the model points to are clear areas where language and emotion systematically overlap. Using the model, strategies of argumentation can also be analyzed, offering new ways to look at the construction and interpretation of emotional appeals. What the model emphasizes are the areas of imagination that join together to construct the emergent phenomena of emotional persuasion. Mental visualizations at L1 may arouse feelings in L2 that go on to become categorized at L3 in service to argumentative intentions at L4. Some 97 discursive moves may focus on specific levels of the theater of the mind more prominently than others, but all emphasize the rich and dynamic ways that language systematically guides imaginative reasoning?including reasoning within emotion scripts. It would not be an exaggeration to say that the theater of the of the mind emphasizes the ways that language guides imagination and reason is the same as how a topographical map emphasizes the landscape of a place; it represents the basic contours of semantic space. The fact that language can have such a strong effect on imagination suggests that we are all, in some way euphantasiotoi, good imaginers. We all are able to spontaneously experience complex mental experiences unfolding in response to only a few words, drawing conclusions from evaluating perceived actions that we only see in our mind?s eye. Some rhetors can, admittedly, find more succinct words to construct even richer, even more emotionally moving scenes in our minds, demanding more immediate action or more visceral reaction. But even these rhetors rely on a shared system of language usage and cognitive likeness with their audiences?on the kinds of values they share, the sorts of emotion scripts their culture uses, and the sorts of actions their audience can be expected to perform. These euphantasiotoi are particularly in tune with their audiences, able to draw on or slip in presupposed beliefs, able to reframe scenes to entertain specific strategic vantage points, or able to develop scenarios that bring to mind vivid kinds of mental simulations of emotive scenes. Perhaps there exists a fabled euphantasiotos who is always emotionally persuasive in all situations, but it may be more advantageous to 98 argue that any rhetor can become a euphantasiotos, given the right words, the right audience, and the right argument. The theater of the mind model provides a rich framework for understanding the ways enargeia, enthymemes, and other rhetorical strategies contribute to arousal. Along with CG, the theater of the mind model views semantic interpretation as a process of enacting simulation, placing an emphasis on how imagination constructs meaning. This allows analysts to understand rhetorical strategies as patterns in language and imagination occurring at different levels of the theater of the mind. For example, enargeia requires a connection between more specific (or granular) depictions of L1 events, coupled with L2 affective force. Moments of enargeia can be broken down to the relationship between the specificity and granularity of construal in L1 and the kinds of subjective feelings in L2. The grammar of enargeia then requires special attention to the ways rhetors prompt their audiences to think, feel, reposition themselves in relation to the events unfolding on stage, and zoom in closer to observe those events in detail. In the background are the emotion scripts and evaluative stances that are activated in L3 or the larger L4 rhetorical strategies that the enargeiac scenarios are a part of. Similarly, the theater of the mind explains the workings of pathetic enthymemes. When analyzing the so-called pathetic enthymeme, made popular by Jeffrey Walker?s (1992) reading of Aristotle on the emotions, what is in focus are the various ways that various L1 phenomena can bring to mind emotion scripts at L3, backgrounding the subjective thoughts and feelings at L2 and the larger rhetorical intentions that emotion arousal serves at L4. 99 In ancient rhetorical theory, especially in discussions of striking or moving discourse, rhetoricians have placed a special emphasis on the imagination and on what Aristotle calls bringing-before-the-eyes (pro ommat?n) (Rhetoric 3.10-11). Aristotle believed that understanding shame meant, to some extent, simulating the shameful scene, and that to do so would also bring recognition of the sufferings and evil that comprise the simulation (2.9.14). Aristotle defines shame as a phantasia of a loss of reputation among those a person holds in high esteem (2.6-14-15). Aristotle?s emphasis that shame is a mental simulation or phantasia aligns it with an L2, L3, and L4 event in the theater of the mind. What acts constitute as shameful depend on a belief about what the observers believe to be seemly and shameful?an L3 intersubjective alignment that situates a value system?as well as a strategic activation of an L4 cultural narrative. Crucially, this means that the shame emotion script involves a conceptualizer imagining how other people would think about them if those other people knew that the conceptualizer has performed some action that the conceptualizer believes the other people would feel was wrong. To construct a shame appeal, a rhetor needs to attend to their interlocutor?s sense of values and beliefs, as well as present selective scenes that can produce a repelling affective force. The theater of the mind model?s emphasis on cultivating imagination to produce selective mental visualizations to evoke larger cultural narratives parallels Aristotle?s conception of the pathetic enthymeme, but it does so in a way that organizes thoughts, feeling, social cognition, and narrative evocations for a richer analysis. By describing the structure of the appeal, the theater of the mind model also reflects some general socio-cognitive and linguistic facts about the structure of thought. 100 I argue that this strategy is not altogether unique and is somewhat cognitively intuitive, and I predict that rhetorical actions that appeal to panic follow this form. Climate change rhetoric, partisan political rhetoric, gun rights rhetoric, animal conservation rhetoric, and environmental conservation rhetoric, in the form of fundraising letters that ask patrons to provide money, should follow this strategy. If there is a variation, the variation should not be an attempt to appeal to or to relieve panic in a patron but to do something else, such as to produce trust in the appealing organization and thus develop ethos. Hypothetically, there is a set number of constructions and a set number of events that can be construed in a set number of ways that comprise the rhetorical appeal to panic. There are, nevertheless, a vast number of ways in which events within scenarios can be construed, and herein lies the genius of a skilled rhetor at L4. Aristotle provided a list of emotion words in chapter two of the Rhetoric but we ought to assume that this list is incomplete, that, if emotions are conceptual constructions?built out of experiences, culturally learned scripts, and interoceptive feelings?rather than natural kinds floating around in the world or neural fingerprints in the brain there are innumerable emotion events that can be construed. Panic, for example, is not an emotion word that he numbers in his list, perhaps because it was not as helpful of a concept in the more debate-focused argumentative culture of Ancient Athens, given that inciting panic is more of a political or deliberative strategy. Panic is only but one of many emotion concepts for which rhetorical theory may want to consider developing strategies for invention. 101 Perhaps one reason for the shortcoming of terms in Aristotle?s list of emotion words and the outline of the emotion scripts Aristotle glosses is that Aristotle was too focused on finding positive and negative pairs of emotions in order to train rhetors to control crowds,11 to get them angry when they are calm or to make them happy when they are sad. Instead of thinking of emotion arousal in terms of learning scripts in relation to the polarity of their valence (e.g. happy on one side and sad on the other), it may be more practical to invent emotional appeals based on the nuanced goals of the situation. This would require opening up the emotional appeal to more than the handful of emotion words that Aristotle provides, and would also require becoming steeped in the conceptual differences between, for example, sad, depressed, upset, dejected, and miserable. What are the lines of the emotion script for each word? How can a rhetor, in any given situation, highlight the conceptual scenes in each line of the emotion script? How is feeling dejected different from feeling the more general sad? Learning to go in depth when conceptualizing emotions should prove to give more dexterity when constructing emotional appeals. In other words, it?s time that rhetoricians begin focusing not just on the repertoire of appeals at hand but on the conceptual specificity of the construal of any potential appeal. Learning granularity of a series of related emotion concepts also places a focus on the detailed ways that a rhetor can verbally construct emotion events. True to the classical idea of a rhetor as a 11 This, of course, assumes that the intention behind the Rhetoric was to train rhetoricians. Some, like Carol Poster (1997) have argued quite convincingly that Aristotle?s work on rhetoric was less about teaching rhetoric than it was about teaching philosophers to observe rhetorical methods of argumentation. 102 painter and an audience as a viewer, this perspective of the rhetorical appeal pays careful attention to the details of the picture being presented. The theater of the mind model provides a way of understanding the mind of the euphantasiotos. While having the language to articulate a cognitive rhetorical model of pathos is not necessary for an audience to experience the emotional appeal, rhetorical argumentation has always operated effectively without the need for analyzing argumentation. Rhetorical theory aims to articulate what we do, and by doing so, it allows for practitioners to train and analysts to criticize. The theater of the mind model offers a way of making explicit the unstated and often intuitive kinds of systems of thinking and feeling that accompany emotional appeals. It offers a system for analyzing the euphantasiotos?s arguments?which is to say, it offers a system for analyzing the linguistic and argumentative strategies that we draw in persuasive contexts to move our audiences to think and feel differently. And this is, at its heart, the project that Aristotle lies out in his rhetoric: to make apparent anyone?s ability, in any given situation, to understand how a particular means of persuasion functions to produce an intended effect. 103 104 Chapter 3: Emotive Reasoning in Political Rhetoric 1. Introduction: A Reason to Feel Inciting emotional response has long been a strategy for politicians to frame themselves as heroes, villainize their opponents, and amplify the exigence for their policies. Moments after Donald Trump presented his bleak view of a crime-infested America at the 2016 Republican National Convention, CNN journalist Alyson Camerota sat down with former Speaker of the House and Republican strategist Newt Gingrich. Camerota began a back-and-forth by commenting on Trump?s depiction of violent crime in America in his speech: ?Some people think it was too bleak, that he painted too bleak a picture of where we are in America. Violent crime is down in America.? Gingrich interrupts, arguing that crime is not down in major cities like Chicago, Baltimore, and Washington, and that, furthermore, Americans don?t feel safe. Even when Camerota cited FBI statistics that report a decline in violent crime, Gingrich refused to concede that viewing America as a violent and crime-ridden country is not justified. ?Your view,? Gingrich says to Camerota, ?I understand your view. The current view is that liberals have a whole set of statistics which theoretically may be right, but it?s not where human beings are. People are frightened. People feel their government has abandoned them? People feel more threatened.? Camerota tries to emphasize the word feel, attempting to persuade Gingrich to recognize that while she is reflecting on reality with facts and figures, he is reflecting on sentiment. ?They feel threatened,? she says, ?but the facts don?t support it.? Gingrich replies boldly, ?Right, as a political candidate, I?ll go with what people feel, 105 and I?ll let you go with the theoreticians? (?Was Donald Trump's Speech Too Dark??). Responses to Gingrich and Camerota?s on-air argument saw Gingrich as a dangerous manipulator, fueling public emotion for political purposes. The comedian John Oliver (2016) captured this line of argument when he commented that ?[i]t?s worth taking a moment to seriously consider what Gingrich was saying?I think we can all agree that candidates can create feelings in people, and what Gingrich is saying is that feelings are as valid as facts, so then, by the transitive property, candidates can create facts.? The science communicator Ethan Siegal (2016) writes that Gingrich?s argument is ?supporting actively misleading people about the facts to appeal to their gut, their beliefs and their preconceptions, and to use those lies to gain power and enact policy.? What is striking about the general response is not necessarily its similarity to the tired topos of the evil emotionally persuasive sophist, a topos that runs through almost the entirety of Plato?s Gorgias and appears as a warning preceding Quintilian?s advice for constructing emotional appeals. What is striking is that, over two millennia later, our response to emotional appeals in political rhetoric is still to reject them as fallacious attempts to persuade without reason, to fact-check them and hold them accountable to enlightened reason, or to ignore them for lacking relevance in reasoned debate, rather than to better understand how they are constructed and how they are persuasive. In The Political Mind, George Lakoff (2008) argues that contemporary liberal political discourse too often appeals to Enlightenment ideals of rational cognition that have long since been proven to be non-intuitive systems of natural thought. Cognitive 106 biases are part of everyday rationality, and, while these biases are less than ideally rational, they are part of our default mode of cognition and work to maintain our ordinary social cognition. ?Most of us have inherited a theory of mind dating back at least to the Enlightenment,? Lakoff writes, ?namely, that reason is conscious, literal, logical, unemotional, disembodied, universal, and functions to serve our interests? (3). There are many problems with this view of rationality. To begin, work by the behavioral economists Amos Tversky & Daniel Kahneman (1973; 1981), Kahneman & Tversky (1982), and Kahneman (2011) have revealed that cognitive bias is an unavoidable part of everyday rationality. Deliberate and deductive reasoning is the exception not the norm. While pragma-dialecticists like Frans van Eemeren and Rob Grootendorst (2004) have argued that ideal models of argumentation can be used as heuristics to assess critical reasoning and to achieve rationality (123-124), assessments of rationality to argumentation theorists too often mean disembodied reasoning or reasoning without attention to affect. In this way, contemporary argumentation scholarship has reinforced the reason/emotion topos, at best asking scholars to assess the relevance of emotional appeals in a rational argument (Walton 1992, 1997, 2000) but too often casting emotion into the realm of fallacy (Walton 1992, 1997, 2000, 2015). Douglas Walton has been the guiding voice in argumentation theory in defense of serious attention to emotional appeals, yet in his enlightened defense of the emotional appeal, he still contrasts emotion with reason, claiming that ?[a]ppeals to emotion should be generally recognized as having legitimate standing as being, under certain conditions, reasonable arguments?? (2015 62, italics mine). The trouble with these views of 107 reasonableness, especially when applied to argumentative discourse, is that they are unrealistic and disembodied. Real rhetoric is, more often than not, emotive. Instead of making reason the ideal, it may be more helpful to understand how reason guides emotion and how relevant premises and rational inferences trigger predictable emotive reactions in listeners and readers. To understand emotion elicitation as something coherently and rationally constructed from language changes the way we view emotion in relation to reason. Of course enlightenment reason is not without its place in rational decision making, but it is wrong to believe that Enlightenment reason is natural reason minus all of those distracting feelings. This chapter takes a closer look at emotional appeals in political discourse, focusing particularly on the rhetoric of Donald Trump, to better understand how politicians use language to guide thought and use thought to elicit emotion. I draw on the idea of the pathetic enthymeme, which is grounded in Aristotle?s theory of emotions found in Book II of his Rhetoric, as well as supplementary contemporary workings by Jeffrey Walker (1992, 2000). The idea of the enthymeme has a range of meanings that span rhetorical histories, and I do not attempt to settle debates on, for example, whether the enthymeme is a pithy stylistic cap or a loose formed argument with a ?missing premise.? Instead, I offer an idea of the pathetic enthymeme grounded in a kind of metonymy, one where descriptions and conceptualizations of certain kinds of events and experiences bring to mind entire emotion scripts. Such enthymematic moments, I will go on to argue, require audiences to draw on personal experiences or shared common ground beliefs (what I call endoxic premises in Chapter 2) to move from conceptualizations of events to emotion scripts. Using the 108 theater of the mind model, I analyze how pathetic enthymemes construct emotional appeals that profile negative evaluations of people, shift attention to mentally simulated emotive events, and draw on personal and imagined experiences to ground conceptualization in emotive events. Examples in this chapter come from Trump speeches collected in the Clinton- Trump Corpus (Brown 2017), from Trump?s Republican National Convention speech, and from news television discourse about Trump. In analyzing pathetic enthymemes, I intend to offer a means of understanding how emotional appeals are constructed, making recourse to the script-like nature of emotion concepts, and the importance of visual imagery in conceptualizing and processing language. The goal is not to see if political arguments are wrong or right but if they are effective or ineffective in arousing emotions, and if they are effective, to offer a coherent theory why. Far from conceding that audiences are slaves to their passions, I believe that, by attending to the reason in emotions, we are in a position to sharpen an understanding of that reasoning process. 2. The Pathetic Enthymeme Donald Trump has emerged as perhaps the most excitable political orator of the twenty-first century, moving tens of thousands of supporters and protesters in his political rallies on any given day to feelings of anger, resentment, frustration, and indignation. The journalist Ed Pilkington (2018) describes the typical Trump rally as a gathering of love and hate??I love you people, it seems to say, because you hate 109 my enemy??focusing on how Trump rallies become a space for emotive connectivity and cooperative emotional suasion. By this token, Trump?s anger-inducing rhetoric brings comfort and communal purpose to his supporters. Recent rhetorical scholarship has emphasized the importance of Trump?s crafting of emotional appeals. Denise Bostdorff (2017) writes that ?President Trump?s persuasive challenge?is whether he can keep supporters in a perpetual state of anger to his benefit.? She continues, ?Even when anger is justified, it can be exhausting and, once anger dissipates among the majority who are not dedicated true believers, the thirst for revenge can dissipate, too? (698). Needless to say, Trump?s rhetorical power depends greatly on his ability to move and continue to move his audience. But how exactly does Trump craft emotional appeals? What sorts of language and argumentation strategies comprise his emotional appeals? One way to begin to answer this question is by analyzing examples of emotional appeals in political discourse and looking for patterns of language and rhetorical strategy in these appeals. Moments of name-calling in political speech? such as when Trump calls Marco Rubio ?Little Marco? or Hillary Clinton ?Crooked Hillary? (Trump.2016.7.27) 1?are clear examples of emotional appeals, motivating audiences to arouse negative feelings in relation to his political opponents. Emotional appeals are also present in utterances pertaining to positively or negatively valenced emotive events, such as when Trump tells his audience to ?[r]emember this, Hillary 1 These citations refer to Trump rally speeches in David Brown?s (2017) Clinton-Trump Corpus. The convention I am using to cite indicates the speaker first, the year second, the month third, and the day fourth. 110 Clinton essentially wants to take away your rights under the Second Amendment? (Trump.2016.8.4); this example is similarly framing Clinton negatively but doing so, via semantic composition, by depicting her as wanting to be the cause of an unfavorable event. And sometimes emotional appeals can be constituted by longer strings of discourse that include reference to negatively evaluated actors committing unfavorable acts?for example, when Trump provides narratives of innocent Americans killed by undocumented immigrants, who he goes on to characterize as murderers and rapists (e.g., Trump.2016.10.21). In all of these examples, emotional appeals focus attention to actors and situations that activate and arouse personal value systems: negatively evaluated people are understood to be enemies, and negatively evaluated actions are understood to be threatening, reprehensible, or otherwise undesirable. The contrary is also true. Political orators often frame themselves and those they are socially aligned with as heroes and protectors. Agnes Gibboney, a so-called ?angel mother,? or the mother of a child who was killed by an undocumented immigrant, was brought on stage at a Texas Trump rally to claim that ?Mrs. Trump is the only one that?s going to protect your children from being slaughtered like my son was? (Trump.2016.8.23). Gibboney frames Melania Trump as a hero by associating her with a positively evaluated action (protecting) and frames Trump?s political opponents as villains by implying that their inaction led to her son?s ?slaughter.? Finding moments of positive and negative evaluation, of downgrading and praising, or of narrative storytelling alone does not provide the full picture of an emotional appeal, however. Following Aristotle, emotional appeals require an 111 understanding of the conceptual structure of an emotion. Moments in discourse perceived as emotion-arousing plug into dynamic stories that follow emotion scripts; these scripts help experiencers understand how to feel about whom and what to do about those feelings. The strategy of constructing emotional appeals, then, is one where a rhetor intentionally makes reference to part of an emotion script?for example, the scene of a slight in a script for anger or a scene of crying over a loss in a script of sadness?in order to trigger the full script. Within contemporary discussions of Aristotelian rhetorical theory, this strategy has come to be called the pathetic enthymeme. To understand the pathetic enthymeme, it is first necessary to provide a definition of an enthymeme. ?Enthymeme? is a confusing term in rhetorical theory, since its meaning has always been in flux, so what I mean by enthymeme may align with some definitions and clash with others. Aristotle famously calls the enthymeme a ?kind of syllogism? (Rhetoric 1.1.11, 2.22.1) and, more specifically, the ?rhetorical syllogism? (Rhetoric 1.2.7), though the term syllogism may be misleading. Jeffrey Walker (2000) has proposed that by ?syllogism,? Aristotle may not have had in mind the contemporary and technical (Modern English) sense of the word?a three part deductive argument with a major premise, a minor premise, and a conclusion?but instead by the colloquial sense it had in Classical Greek, ?a putting-together of observed facts, an adding up of things? and ?nothing more than ordinary, informal reasoning and inference? (170-71). I use the term in this sense, but I take it a step further. Contemporary approaches to the enthymeme, following Bitzer (1959), have focused on how common knowledge plays an important role in cognitively 112 constructing a speaker?s meaning in arguments that are underspecified?i.e., in arguments where key premises and beliefs are left unstated and presumed to be accessible to members of the common ground. Such common ground beliefs, or endoxic premises, as I call them in Chapter 2, are a common part of everyday rationality, and noticing their intentional exploitation for argumentative purposes I believe helps to bring attention to how audiences come to the conclusions they do. That is, they help to explain and track cognition in argumentation. Given this, I define an enthymeme as an argumentative strategy where a rhetor intentionally uses language to guide audiences to reason to a conclusion from underdetermined knowledge, requiring audiences to use common knowledge and endoxic premises to ?add things up.? The idea of the pathetic enthymeme comes from a reading of the list of emotion concepts in Book II of Aristotle?s Rhetoric. William W. Fortenbaugh?s (1975) work on Aristotle?s theory of emotion sparked interest in the relationship between Aristotle?s discussion of path? and his discussion of rhetorical argumentation. According to Fortenbaugh, Aristotle?s work was part of a larger project in Plato?s Academy to ?explain the involvement of cognition in emotional response? (9). It led to questions that are still relevant today about how emotion guides reason and how reason is affected by emotion. Craig Waddell (1990) makes sense of Aristotle?s work on emotion by arguing ?the interaction between logos and pathos becomes clearer if we think of emotional appeals as enthymemes? (390). Jeffrey Walker?s (1994, 2000) work on enthymemes of emotion reemphasizes this point. Walker (1994) argues that the pathetic enthymeme works by getting audiences 113 to make premises present to mind that have the potential to shape beliefs, for example the belief that the audience has been insulted (358-59). This belief, Walker argues, will trigger a ?diffuse arousal state,? or an embodied reaction that readies the experiencer to respond to the perceived belief. While the patterns of arousal can be mapped onto specific path?, the argumentative strategies for making premises present to mind and shaping beliefs are the foundations of the pathetic enthymeme. Walker (2000) goes on to point out that examples of pathetic enthymemes often guide reasoning by highlighting ?emotively significant oppositions? like good and evil, happiness and sadness, slavery and freedom (173). These oppositions work within the argumentative strategy by spotlighting things that the rhetor and audience presuppose to be virtuous, valuable, and desirable, and then juxtaposing them with undesirable things that are presupposed to be antithetical to the moral orders of the rhetor and audience. To say that emotion activation can be an argumentative strategy requires unpacking. Joint and mass emotion activation is not by accident, especially in political discourse. It cannot be the case that the emotive reaction to an utterance is purely subjective; there must be reason behind emotive reaction and reason for emotive reaction. The reason behind the emotive reaction depends on the kinds of emotion scripts activated. In any given culture, there are shared emotive meanings and social customs associated with those meanings that are recognizable to members of that culture. Salient moments of those scripts are able to activate knowledge of the whole script. Furthermore, there is a reason for eliciting emotive reaction. The act of arousing an emotion is an act of categorization; it motivates audiences to see the 114 world differently, to recognize, evaluate, blame, identify, and otherwise reassess the social world in a specific way. In reassessing the social world, problems arise, joint intentions form, and communities are organized to address those problems. For a rhetor to be able to get audiences to shift their social alignments suggests that rhetors have the ability to influence reasoning processes. Within the theater of the mind model, the pathetic enthymeme can be thought of as an L4 strategy that a rhetor can use to guide emotion activation. It uses linguistic constructions to coordinate beliefs by realigning evaluative stances at L3, activating thoughts and feelings at L2, and ?making present? relevant actors and actions at L1. Pathetic enthymemes rely on the emotion scripts specified by audience experience and cultural customs, as well as common ground beliefs and endoxic premises that help audiences move from premises to conclusions and from conclusions to emotions. The pathetic enthymeme works because of the rich and dynamic ways our minds organize felt experiences into egocentric narratives (see, e.g., Lakoff 1996, 2000), keeping track of what things we value and what is being done to those valued entities. We are quick to sort actors into heroes?that is, people who share our values and work to protect those things we value?and villains, or people who work to harm our valued entities and interrupt our moral orders. Within the dynamic stories we create, emotions arise from conceptualized moments pertaining to valued entities and the social actors, ourselves included, that interact with them. The L4 strategy of the pathetic enthymeme, then, must be able to explain how moments of evaluation and moments where valued entities are put on stage relate to the recognition and experience of playing out complex cultural and cognitive emotion scripts. 115 3. Frame Metonymy One way to begin to analyze the linguistic and conceptual grounding for the construction of pathetic enthymemes is to look at the way that the careful placement of actors and actions on stage can trigger the activation of emotion scripts. This process of movement from a salient moment of an emotion script to the activation of the whole emotion script can be thought of as a part-for-whole frame metonymy. In Cognitive Linguistics, frame metonymy is a kind of mental shortcut that is prevalent not only in language but in thought. According to Dancygier and Sweetser (2014), frame metonymy pertains to the ability for ?characteristic parts of a frame? to evoke the whole frame (113). Crucially, activating the whole frame from part of a frame allows for inferences by allowing access to the unstated entailments of the frame, such as other frame elements and frame-based knowledge. In the famous example from Fauconnier (1985), the utterance The ham sandwich wants his check metonymically links a menu item at a restaurant to the patron who ordered that menu item by reference to the Restaurant frame. The metonymy, it should be noted, is also viewpointed: it is clear that the speaker of the utterance is a service worker within the restaurant industry and not, for example, a patron. The example may come across as odd to those who have never worked in the service industry, too, which leads to another crucial point about frame metonymies. Since frames can differ across cultures and even across individuals, frame metonymic activation, which ?depends upon shared frame structure,? is ?often culture specific? (Dancygier and Sweetser 124). 116 The activation of an emotion script from some set of described characteristics is also made available via frame metonymy. In categorizing partial and underspecified information (e.g. recognition of an enemy with a weapon), we often activate whole frames (e.g. the emotion script for fear) to try to make meaning of what we are experiencing. In a sentence like Sarah furrowed her brow, the action of brow furrowing is metonymically linked to an emotion script, such as the scripts pertaining to anger, confusion, or worry, depending on the context. In this metonymy, the description of Sarah?s physical behavior, which is conventionally understood to be part of an experience of an emotion event, brings to mind that emotion event. While some theories of emotion believe that the physical characteristics that accompany an emotion event are the emotion event itself?for example, Paul Ekman?s (2007) theory that facial contortions are constitutive of the emotion?many analysts, like Zolt?n K?vecses (2000), rightly consider them metonymies. Body parts and the emotive significance they are metonymic of often change with cultural models. For example, in modern Iranian Persian, to call someone your liver (?????)? first person, genitive?is a conventional way of expressing affection toward that person, grounded in the folk belief that warm and affectionate feelings radiate outward from the liver. To call someone your liver, then, is a kind of complex conventionalized metonymy. In this complex metonymy, the referent of the genitively inflected liver is a person whom the speaker feels affectionate about and the liver itself is metonymic of the feelings of warmth that radiate from the liver when experiencing affection. Activation of the frame of the Liver, then, can be used to activate the larger frame of Affection. While this chain of metonymies from stimulus 117 to body part to feeling may seem odd, it is worth noting that in English we do something quite similar with the word heart. Cultural models may differ but patterns of metonymic activation reveal general ways that we all think. This is to say that metonymy is not only part of how we speak but also how we think and reason about the social world. Metonymy arises in language and thought for perhaps a number of reasons, but one of them is that it helps us efficiently allocate our attention in thinking about complex situation. As the linguist Jeannette Littlemore (2015) explains, When asked to think of ?France?, people might picture a place in France they visited, or a rough map of France, or an iconic representation of France such as the Eiffel Tower. It is impossible to picture the whole of France at once as this information could not be held in one?s working memory, even if one had travelled extensively in France. (5) Being able to access complex social frames from partial information allows for reasoning about things that do not have concrete reference, including abstract concepts like freedom, justice, or any given emotion concept. Because metonymies can vary across language and cultures, they have a special function in building identity within discourse communities, quickly aligning attitudes, allowing for ?speedy communication,? and otherwise referencing shared knowledge between interlocutors (Littlemore 1). Although the literature on metonymy has distinguished a number of different kinds of metonymic processes, the notion of frame metonymy is general enough to capture different kinds of metonymy in discourse, and it provides a means of 118 articulating metonymic relationships by reference to frame elements and coordination across frames. This being said, some of the kinds of metonymies that will be analyzed in this chapter have gone by different names, such as referential metonymies (Warren 1999; Panther and Thornburg 1998), propositional metonymies (Warren 1999), and illocutionary metonymies (Panther and Thornburg 1998, 2003)?this last one relying on a ?state of affairs scenario? similar to the dynamic narratives I have argued are bound to the activation of emotion scripts in pathetic enthymemes. While a richer classificatory system may be helpful for analyzing fine-grained differences in metonymies, such detail may not make much of a difference for understanding how emotion scripts are activated for rhetorical purposes. For the sake of analysis, then, frame metonymy is understood as the chief means by which pathetic enthymemes are formed. 4. Figurative Reasoning in Pathetic Enthymemes The kinds of scenes that compose pathetic enthymemes are ones where audiences are able to quickly and effortlessly understand who they are supposed to like, what actions they are supposed to dislike, and what kinds of scenarios are significant enough to pay attention to. Many conventional referential metonymies?like the White House for the executive branch?provide shortcuts for reasoning but do not necessarily guide insight into how audiences are supposed to feel about what they are imagining. How does evaluation get encoded into pathetic enthymemes, then? Charles Fillmore (1982) identified one way that evaluations are encoded lexically when he observed that some words carry with them assessments within the 119 frames they metonymically call to mind. For example, the adjective stingy offers a negative evaluation of an actor within the Giving frame. Evaluative words like traitor, treason, and pollution, where the valence of the nominal is implicit within the word itself, are of particular interest, as these words profile an entity under a subjectively construed evaluation, serving an inherently strong role within a larger emotion script. Similarly, lexical items such as kill, poison, victory, and prince stereotypically?rather than necessarily?index entities under evaluation that serve strong roles in emotion scripts, as well. Attention to lexical choice and to the frames they bring to mind is, then, one way to understand how evaluation is coded into emotional appeals. Not all emotional appeals, however, use words that have valenced appraisals built into their semantics. In these cases, stylistic devices are often used to increase the salience of certain entities within the discourse and place them under evaluative consideration for rhetorical purposes. While, for example, being tired is not necessarily a bad thing, when Donald Trump calls Joe Biden ?Sleepy Joe,? Trump uses epitheton to reduce his political opponent to a quality that is understood to be unfavorable for someone in the high stress role of the President of the United States. The evaluations that arise from this trope still rely on metonymic reasoning, as the characteristic of sleepy for a presidential candidate brings to mind incongruities with a cultural belief of what characteristics a political figure should have, given the more general President frame. Figures of speech, then, can be thought of as strategies for reasoning, or, as Jeanne Fahnestock (1999) has called them, ?epitomes of reasoning,? as they help to organize information in certain ways and draw attention to specific discursive elements for 120 rhetorical purposes (24). This is to say that the crucial act of guiding evaluation comes in many grammatical and rhetorical forms?some more direct, such as through the use of valenced evaluative frames, and some indirect, such as through inferences made more salient through the use of stylistic figuration. Understanding the underlying metonymies of a pathetic enthymeme can provide insight into how rhetors are able to elicit emotive responses. Take for example the emotional appeal in the following infographic, taken from the September 13, 2018, broadcast of the MSNBC television program All In with Chris Hayes (2018): Figure 1. An Infographic from All In with Chris Hayes The headline reads ?Trump denies 3,000 Deaths in Puerto Rico on His Watch,? and the infographic above it shows the total deaths of Americans in Benghazi versus the total death of Americans in Puerto Rico after hurricanes Irma and Maria in 2017 and 2018. The context for the headline comes from a series of tweets on September 13, 2018, wherein Trump claims that only ?6 to 18 deaths? came about as a result of Hurricane Irma and Hurricane Maria and that the official death count, which 121 approached 3,000, was inflated by Democrats who wanted to make him ?look bad? (@realdonaldtrump). Hayes?s argument is crafted skillfully through a series of metonymies. The statement that Trump denies 3,000 deaths in Puerto Rico activates two different frames: a Denial frame, wherein Trump assesses the truth of a reported event, and the Hurricane Maria frame, which would have been a conventional news story for many Americans listening to Hayes?s program. Like all dynamic information frames, the Hurricane Maria frame develops as more is said about it, so part of the frame requires understanding that experts believe that, after Hurricane Maria, nearly 3,000 people were believed to have died. When Trump denies part of the frame, Trump can be construed as denying the whole of the frame, and because of the correspondence between the frame as a whole and knowledge about the state of affairs in Puerto Rico, Trump?s denial of the reported death toll comes across as a denial of the effects Hurricane Maria had on Puerto Rico. If the reported reality of the situation is that nearly 3,000 people died, then Trump is denying reality in his denial of the death toll. Denial of reality is not necessarily in and of itself emotion arousing. However, when the President of the United States denies reality, he violates an endoxic premise that the president is socially and politically aware of what is going on in the world. The conceptualization of an out-of-touch president provides a negatively valenced picture of Donald Trump, coupled with a potentially tense arousal at his action of denying the death toll. And it is this image of an out-of-touch president denying 122 reality that shares schematic structure with moments of emotion scripts like the feeling of outrage, the feeling of being appalled, or even a feeling of moral disgust.2 The L4 strategy in the above utterance, I hope it is clear, intends to move audiences to feel negative emotions toward Donald Trump. Words like denial and Puerto Rico allow for frames to be metonymically activated pertaining to, respectively, the evaluation of the truth of the matter and the impacts of an exigent natural disaster that affected hundreds of thousands of people. The dynamic construction of the image of Trump denying reality, furthermore, shares schematic structure with moments in emotion scripts where experiencers are unaware that someone could do something so bad, thus metonymically activating those emotions. Audiences may not themselves be moved to feel anger or contempt or outrage toward 2 For the sake of space, I do not go into detail in this chapter about what exact scenario of an emotion script is being activated and assume that the schematic similarity between the scene described and the emotion concept is transparent. Such an analysis, however, is possible, though may be tedious and trivial. For the sake of demonstration, the scene described above would factor into lines (c)-(e) of Wierzbicka?s (1999) script for outrage: Outrage (a) X felt something because X thought something (b) sometimes a person thinks: (c) ??I know now: something very bad happened to some people (d) because some other people did something very bad (e) I didn't think these people could do something like this (f) I don't want things like this to happen (g) I want to do something because of this (h) I have to do something because of this?? (i) when this person thinks this this person feels something bad (j) X felt something like this (k) because X thought something like this (94-95) 123 Trump, but they at least are able to recognize the intention of Hayes?s utterance. They do not listen to it, for example, and think that they are meant to feel joy or sorrow or even nothing at all. The utterance is understood to be an attempt to arouse emotion. The cognitive guidance of the appeal is transparent and undeniable. While the headline about Trump denying the deaths in Puerto Rico is meant to guide audiences to construe Trump negatively, the infographic above the headline has little to do with Trump, instead choosing to spotlight the inconsistent outrage of Republicans. The infographic presents two data sets: the number of Americans that died in the Benghazi terror attacks and the number of Americans that died in Puerto Rico, as well as the number of investigations into political inaction that followed the Benghazi terror attacks and that followed the aftermaths of Hurricane Maria. The choice of comparing the deaths in Puerto Rico to the deaths in Benghazi is quite excitable. A conventional Republican talking point is to blame then Secretary of State Hillary Clinton for her inaction in preventing the deaths of four Americans who died in terrorist attacks in Benghazi, Libya. The frame of Benghazi is a particularly emotive frame for Republicans, calling to mind stories of the four tragic patriots who died in the terrorist attacks, the villainous Islamists who conducted the terrorist attacks, and the neglectful and unmoved politician who did not prevent the attack. The infographic uses knowledge that Benghazi evokes an emotive frame for Republicans to argue against Republican behavior. In doing so, it is illustrative of the ways metonymic activation can guide inferences in pathetic enthymemes. For many, the frame of Benghazi metonymically links to a Republican Outrage frame, wherein Republicans become outraged because of the loss of 124 American lives due to the inaction of a careless politician. The infographic allows for the inference that the degree of outrage is measurable by the number of investigations that went into understanding the reasoning for the loss of American lives. Given the logic of the Benghazi frame, the frame activated by Puerto Rico ought to garner Republican outrage proportional to the number of lives lost due to a careless politician?that is, if four American deaths prompted ten investigations, then it should follow that 2,975 deaths would prompt 7,438 investigations. What the infographic effectively does is invite audiences to search for a logic within the Republican Outrage frame and to try to model that logic. By noting that there have been no investigations into Puerto Rico, the infographic effectively shows that the Republican Outrage frame does not conduct investigations proportionate to the number of deaths due to political inaction. From this, audiences can draw on endoxic premises to make sense of why Republicans are not outraged. Perhaps Republicans do not care about Puerto Ricans. Perhaps they are afraid of Donald Trump, who denies that Puerto Ricans died as the result of Hurricane Maria. Perhaps their outrage at the deaths in Benghazi are irrational. The open-ended quality of the pathetic enthymeme strengthens audiences? negative evaluations about Republicans, while also more directly casting negative evaluation on Donald Trump, who has been construed as someone who actively rejects reality. The MSNBC infographic organizes information through rhetorical antithesis in order to structure its argument and guide its reasoning, and, in doing so, it follows the observation about pathetic enthymemes that Walker (2000) observes, namely that the pathetic enthymeme tends to arrange information into ?emotively significant 125 oppositions? to better signal that an audience should recognize that their emotions should be aroused (173). But anthesis is but one rhetorical figure that can help guide reasoning in emotional appeals. Take, for example, the rhetorical figure of epitheton?that is, the construction of an epithet. While Quintilian argues that epithets are purely ornamental, he acknowledges that they can be useful in oratory (8.6.40- 41). In the All In with Chris Hayes example, a negative emotion-arousing image of Trump is constructed by dynamically imagining two metonymically activated frames, but such images can be constructed in much less sophisticated, yet equally effective, ways. Take for example Trump?s famous strategy of making names for his opponents. Ted Cruz was named ?lying Ted,? Marco Rubio was named ?Little Marco,? and Jeb Bush ?Low Energy Jeb.? Each of these names acts as an epithet, reducing a person to a quality that is incongruous with the Republican frame of the President. Trump?s names implicitly draw on the common ground beliefs that a president should be tall, honest, and virile. Each of these candidates is framed as lacking a quality of the President frame, and, as such, each act of framing prompts an L3 stance, negatively evaluating the potential for the named candidate to be construed as presidential. By reducing his opponents to these deficit epithets, Trump effectively frames them negatively, and the repetition of these names helps Trump to continually reinforce the idea that his political opponents lack something. these epithets, when presented in the larger frame of presidential candidates, work to cause tension between the candidates presented and what the ideal Republican candidate should be?i.e., someone without a deficit, the perfect person for the job. What Trump?s rhetoric does is sever the metonymic link between specific presidential candidates, who are construed as 126 flawed, and the frame of the President of the United States, which is construed as pristine. By being the one who names his opponents and repeats the epithets, Trump successfully reinforces negative evaluations of his opponents and is able to craft a feeling that they do not fit within the Republican frame of the president. Trump is able to make his opponents not only look like they are ill fit for the job but also encourage his audience to feel anger and irritation toward his opponents precisely because of the tension and discomfort that imagining them as president produces. Reinforcing the metonymic incongruity becomes a chief way for Trump to easily construct pathetic enthymemes and guide his audience?s reasoning about his political opponents. What should be evident so far is that pathetic enthymemes rely on activating vast amounts of rich conceptual content from a far scarcer amount of words. The idea of frame metonymy helps to explain how these words link up to that conceptual content, and various figures?from antithesis to epitheton?aid in organizing information to orient attention. Of the rhetorical figures related to the enthymeme, the figure of the rhetorical question has frequently been picked out as particularly emotive and strongly persuasive. Thomas Conley (1984) has commented that many of the examples of enthymemes that Aristotle provides in Book II of his Rhetoric are written as rhetorical questions ?raised as climatic points? (171). Manfred Kraus (2007), in his discussion of the Roman figure of contrarium?often considered a synonym for enthymeme?comments on the ?strong persuasive force? of rhetorical questions, as they invite audiences to come into agreement with their rhetors and share attitudes toward inferred conclusions (14). While the enthymemes Conley and 127 Kraus are commenting on are not pathetic enthymemes per se, his observation does highlight the role of recruiting active involvement in producing emotional suasion. This kind of recruiting of active involvement can be found in the following example, taken from a rally speech delivered in Novi, Michigan, on September 9, 2016. In addition to using the rhetorical figure of anthypophora?a rhetorical question followed by a response?Trump gets his audience to actively imagine how entire countries are actively slighting his audience: Just look how badly she [Clinton] messed up the Iran deal. And just today, we learned that on the same day as the hostages [two American naval ships arrested for sailing in Iranian water] were released, the United States signed an agreement lifting sanctions on two Iranian banks involved in financing Iran's missile program. And you know where they want those missiles to eventually land, don't you, folks? Huh? Right here. Right here. Right here in this country. That's where they want them to land. And they're taunting us with the boats. They're taunting us with the 10 sailors they captured. They're taunting us by flying right next to our airplanes. So we give them all this money, $150 billion back. We give them the deal of a lifetime. And instead of saying thank you, they feel emboldened. That won't happen with us, folks. (Trump.2016.9.30) There is more working in this example than just the anthypophora, but the rhetorical question is a good place to begin analysis. When Trump asks ?and you know where they want those missiles to eventually land, don?t you, folks,? his utterance is less asking a question and more a metonymic prompt to imagine and Iranian missile 128 strike. The imagined image of an Iranian missile strike on American soil is amplified when Trump continues: ?Right here. Right here.? While the probability of Iran firing missiles at the United States is not likely and targeting Novi, Michigan, is even less likely, Trump is able to get his audience to imagine themselves as victims of an Iranian missile strike. Through this act of imagination, Trump gets his audience to think of their immediate surroundings?the right here in the common ground?as a potential target, and, thus, to imagine themselves?the people standing right here?as in danger. The imagine scene metonymically links to potential emotion scripts like fear and panic, as members of the common ground imagine Novi, Michigan under fire from Iranian missiles. Trump?s appeal to fear works both to capture the attention of his audience and to pose Iranians as credible threats to his audience?s lives. By making Iranians a credible threat, he is able to also figure them as bad people with bad intentions. Trump goes on to amplify the villainous qualities of the Iranians: they are actively ?taunting us with the boats, taunting us with the 10 sailors they captured, taunting us by flying next to our airplanes.? The verb taunting activates a highly emotive scene of social aggression that entails an actor, who perceives themself as superior, intentionally acting to irritate an evaluee, someone who can understand the act of social aggression. The role of this actor?the they who is taunting?is filled by a homogenized identity: the Iranians. Because the role of the evaluee?the us who is being taunted?has a mapping in the common ground, the personal offense of the taunting slight is made more immediate to the audience. Furthermore, by framing the capture of the American navel fleets sailing on Iranian water as a taunt, Trump leads 129 his audience to believe that Iran is expressing a sense of superiority. This sense of superiority clashes with a potential common ground belief his audience may have that America is a superior country, even given the context. The scenes of taunting, then, can metonymically activate emotion scripts that involve agents intentionally acting inappropriately and doing so for the sake of slighting someone else. The emotion scripts of irritation, anger, and indignation are possible scripts that can be metonymically activated. And these feelings of anger and irritation are reinforced when Trump says that the Iranians ?feel emboldened? rather than thankful. The Iranians should be feeling one way but feel another way, and that other way is inappropriate and potentially dangerous. Trump activates all of these emotion scripts for a specific rhetorical purpose, and it is not simply to make his audience feel unsafe or even to make them feel angry toward Iranians. The key rhetorical intention is to direct disapproval of Hillary Clinton, his political opponent, who, according to Trump, ?messed up with the Iran deal,? thus allowing for his audience in Navi, Michigan, to feel unsafe. Trump?s rhetoric constructs this feeling of danger by getting his audience to imagine missile strikes where they are standing, gives them a way of viewing Iran?s actions toward American military encroachment as social transgressions against America, and leads them to view Iranians as proudly flaunting their potential to do bad things. And this is all because Hillary Clinton made a bad deal. When Trump concludes that ?That won?t happen with us, folks,? he frames himself as someone who is strong and will stand up to Iran. What that is and what will or will not happen is underspecified, allowing 130 audiences to make meaning by reference to the frames and the dynamic stories they are actively imagining. In all of these examples, rhetors strive to change the way that their audiences evaluate specific actors and specific actions. They do so, furthermore, indirectly: they do not tell their audience that they should be hating political opponents or fearing for their lives but instead prompt their audience to imagine complex involving undesirable people (irrationally emotive politicians, poor representations of presidents, aggressive and morally inept Middle Easterners, bad deal makers). From these acts of imagination, providing sparse information about bad actors and bad actions, audiences are led to activate larger emotion scripts. Political opponents become villains, the detainment of an encroaching military presence becomes a slight to anger, and the offer that faith in a candidate?s strength can change international policy becomes a reason for hope. 5. Afraid of Breakfast: Involved Movement Through and Across Frames In the examples above, frame metonymy acts as a way for audiences to use their imagination to build scenes, imagine how others feel, and even feel upset or discomfort when actors behaved in ways antithetical to the ways expected of them. Probing deeper into the ways that frame metonymy helps to explain how emotive reactions are constructed, this section attends to how frame activation can enrich and move imagination for argumentative purposes. If pathetic enthymemes are meant to be persuasive, the operations of the pathetic enthymeme need to be able to be explained in a way that demonstrates its usefulness. Particularly, this section addresses the idea that frame metonymy provides not only ways to construe the 131 evaluative scenes that arouse emotion scripts but involved ways to actively and dynamically feel the construal of those scenes. Specifically, the argument is that, through the mental work of frame activation and enrichment, audiences come to be involved in their mental simulations, allowing for a richer experience of an emotive event. In the following example, frame metonymy allows for an evaluative scene to be brought online as a kind of proxy for empathetic imagination, an imperative asking audiences to imagine how they would feel in such a situation. On the night of November 6, 2016, the American news channel CNN was broadcasting the 2016 presidential election. Van Jones, Anderson Cooper, Jeffrey Lord, Kayleigh McEnany, and Lou Begala were all covering the race. Like many Americans, Jones was expecting Hillary Clinton to win, but, just before 1 o?clock AM EST on election night, Jones conceded that Donald Trump had won. Upon conceding, Jones spoke freely to a general audience, presumably the television audience, about the anxieties to follow election night: People have talked about a miracle. I?m hearing about a nightmare. It?s hard to be a parent tonight for a lot of us. You tell your kids, don?t be a bully. You tell your kids, don?t be a bigot. You tell your kids, do your homework and be prepared. Then you have this outcome and you have people putting children to bed tonight. They?re afraid of breakfast. They?re afraid of ?how do I explain this to my children?. (Jones) This example is rich with evaluative frames: Miracles, Nightmares, Bullies, Fear. These frames work with other more benign frames to recast a seemingly benign 132 scene?that of sitting down with one?s children at breakfast?as a moment eliciting fear. The frame metonymy elicited by breakfast brings audiences to think and feel in a specific way, potentially eliciting feelings of empathy or maybe even shared sorrow. How this feeling is built up requires some unpacking, however. To begin analysis of Jones?s pathetic enthymeme, it is necessary to understand how Jones is able to communicate his stance that Trump?s election is a negative event with negative consequences. The passage repeatedly refers metonymically to the Parental Moral Order that parents are expected to teach children within conventional folk model of what it takes to raise good children. The moral order that parents pass on to their children includes moral maxims like teaching them not to be a bully, not to be a bigot, to do their homework, to be prepared, and so on. While reference to any one of these moral maxims may not bring to mind the entire moral order, Jones uses the rhetorical figure of anaphora to structure these moral maxims, increasing emphasis of the entire Parental Moral Order frame. Implicit in the folk model is the belief that children learn about morality by parents teaching them, and, to go further, that ethical children become successful and good people. In the background is the common ground knowledge that Trump does not follow any of the moral maxims that Jones lists. Jones here constructs an argument that frames Trump?s victory as a negative event (a nightmare) and presupposes that Trump stands for values contrary to what we teach our children. The last three sentences in Jones?s argument construct an enthymemematic moment, as Jones guides his audience to ?zoom in? to the mind of a parent and observe their thoughts and feelings. The scene rests on a number of antitheses: the 133 certain present outcome that Donald Trump will be president and the uncertain future; the political election and the personal family; the nighttime of the scene of putting children to bed and the daytime when parents will have to talk to their children; and most importantly, the moral order of the sleeping children and the threat to our children?s sense of moral order that may result from the surprising outcome of the election. Here, as with other enthymemes, we see that the enthymeme can only function with prior frames (Family, Morals, Parents, Children, Election) having already been brought online, composed, and simulated. One could think of the enthymeme, then, as the resulting conceptualization of a stylistic cap, though it does more here. It uses the argument thus far to construct an experiential and involved effect. The construal of the three sentences are represented in figure 2. As the last three sentences proceed, audiences are invited to zoom in on the construed scene. In the first sentence, the generic ?you,? who represents a moral parent that the audience can identify with, is made aware of two scenes: the negatively evaluated outcome of the election and the seemingly positively evaluated intimate bedtime scene. These scenes help to show one set of contraries: political and the personal. From here, the focus zooms in on the personal scene, a nighttime scene of parents putting their children to bed. The parent, who has knowledge of the election outcome and vivid experience talking to children about the maxims of the Parental Moral Order, is depicted as ?afraid of breakfast,? which seems like a strange thing to be afraid of, indeed. The audience uses frame metonymic inferencing to understand how the parents can be afraid of breakfast. 134 135 Figure 2. A representation of being afraid of breakfast The lexeme breakfast comes with a rich frame, one in which people (e.g. people and children) gather in a location (e.g. around a table) at a particular time (the morning) to eat food (e.g. cereal) and often to converse with one another (e.g. about the election). The Conversation frame here, and in particular the Explanation frame that may need to occur to address the election, is granted conceptual access from the Breakfast frame via frame metonymy, as Breakfast is a frame element that supplies the grounding event (a time, participants, and a location) for an explanation to occur. When Jones says that parents are afraid of breakfast and afraid of how do I this to my children, the audience is able to infer that they are afraid of explaining the outcome of the election to their children. Following this sentence, Jones zooms in even further, stating that ?they? (parents) are afraid of ?how do I explain this to my children.? The explain scene here occurs within the Breakfast frame, and thus zooms in on the communicative scene of a morning family breakfasts. It is also coherent with the previous telling children scenes, which also use the same actors in familial communicative contexts.3 Jones zooms his audience even further into this familial communicative scene with the use of the metaperformative question ?how do I explain this to my children.? This L4 embedded performative request for advice about an L3 act of explaining is referenced in the L1 event of morning breakfast, and it allows for the discomfort and affective tension felt at L2 to be understood as a reason to fear. Jones?s rhetoric cues the simulated imagination of the breakfast scene from the perspective of the shaken 3 As frames get activated, normal processing aligns frame elements in ways that allow for overlap and efficiency between frames. Paul Kay (1982) calls this the ?parsimony principle.? 136 parent. This parent is furthermore framed as a generic I, further guiding audiences to empathize. The character of the fearing parent, of course, has been an available since the third sentence: ?It?s hard to be a parent tonight for a lot of us.? Yet its rhetorical purpose is not fully felt until the end of the passage. Part of the reason why this parent?s turmoil feels so strong is because of how Jones brings his audience to go beyond observing parents interacting with children and to imagine themselves as parents having to impart moral wisdom to their children. When Jones writes ?You tell your kids [moral maxim],? he ostensibly invites his audience to imagine themselves as parents even if they have no personal knowledge or experience of parenting. As Jones?s discourse continues, the pronouns used to refer to parents shift do too: from the first person plural (It?s hard to be a parent tonight for a lot of us), to the second person (You tell your kids, don?t be a bully), to the third person plural (They?re afraid of breakfast). The figure of the parent is made available in multiple ways, moving from different kinds of invocations for identification to finally be presented as something to observe and to empathize with. The audience is not asked to identify with the parent in turmoil but to imagine the suffering of such a parent, to imagine being that parent trying to sit down to a family breakfast, and to wonder about the threat to the Parental Moral Order understanding that Donald Trump?s victory poses. To construe the discourse requires zooming in and out of identities; observing and relating to experiences; understanding the reason for a parent to feel fear, and, from that imagination the kind of sorrow such a scene invokes. The whole passage is quite a complex feat of mental coordination and metonymic activation. 137 I write that the audience feels sorrow, but the emotional response an audience may have is not so cut and dry. The audience may feel empathy for parents in difficult situations. For an empathetic emotional response to arise, the audience must have beliefs along the lines of I am a parent or I am like this person to establish the identification with the parent. The identification between the Parent?s identity or the Parent?s fear and the unstated beliefs of the audience allows for the frame metonymic activation of the empathy emotion script. Instead of empathizing, the audience may also feel the emotion script for pity or sympathy if they feel for the parent, understand the parent?s experience, evaluate that experience negatively, understand that the task that the parent has to accomplish by breakfast is difficult, and so on. A different set of common ground beliefs may be required in the case of these emotion scripts? activation. In addition, the event that the parent is feeling?one where they need to explain something but don?t know what to say or how best to explain it?could also frame metonymically stand for an event within emotion scripts like panic, and knowing that there is a short window to figure out what to say (i.e., some time before breakfast) may provide frame metonymic access to the anxiety emotion script. The audience may also feel sadness, as the truth in the Parental Moral Order is invalidated by the outcome of the election; here, the metonymy comes from the imagined conflict within the child, who has been told to do good things, upon learning that Trump has won the election, a man who has been caught saying obscene things on television and who openly bullied his opponents. If the Parental Moral Order operates under the endoxic premise that The more moral one is, the more successful one is, it may be alarming to see the role of the President of the United States, one of the highest 138 positions in the modern social world, is being held by someone who openly violates the Parental Moral Order. The fact of the matter is that there is nothing stopping the audience from construing the scene using frames of empathy, panic, anxiety, sadness, or similar emotion frames all at once or selectively. There is underspecification here over the exact way an audience may feel when processing the scene because there is uncertainty regarding how saliently an audience will construe an event as an exemplar of an event within a particular emotion script frame within the audience?s mental repertoire of frames. While language allows for the construal of semantic frames, many of these frames still are idiosyncratic and grounded in an individual?s personal experiences, categorizations, and knowledge of the world. Variation in which emotion script frames are activated may in part be due to variations in construal, in other words. Nevertheless, while there is likely idiosyncratic variation of which emotion(s) in particular Jones?s argument is meant to arouse, there are constraints on which emotions can be aroused. This is because the emotions an audience believes that the rhetor is trying to arouse have to come from the experience of mental scenes constructed, ones that themselves are built from rich frames and dynamic simulations, wherein frame elements and events simulated are evaluated in order to retain coherence through the process of cognitive coordination. Listeners who aim to understand the utterances of their interlocutors must try to evaluate the conceptual content with sensitivity to what they believe their interlocutor intends. This often requires listeners to bring online endoxic premises that help ground conceptual 139 content in presumed shared beliefs. The ideal listener of Jones?s argument wouldn?t hear Jones?s enthymematic utterance, for example, and think that parents putting their children to bed will be setting up streamers in excited anticipation of breakfast to celebrate Trump?s victory. Such a reading misses the evaluations necessary to retain interpersonal coordination of meaning between the rhetor and audience when construing the scene. It depends on endoxic premises that would be alien to the rhetor?s beliefs. Similarly, Jones?s audience probably wouldn?t hear that parents are afraid of breakfast and think that breakfast is terrifying and that they should take away from Jones?s argument that they ought to be in a state of genuine terror about the morning cereal. An audience who constructs this emotive meaning has not exercised frame metonymy or conceptual coordination between the meaning that they arrived at and the perceived and context-bound intentions of the rhetor. When processing an emotional appeal, it?s not just any feeling that the discourse is driving the audience toward, even if many feelings may arise in an audience?s mind. The emotional appeal sets up how the listener is supposed to feel such that to construe the scenes in the discourse is to come to the feelings the rhetor intends. What motivates the activation of an emotion script frame is not some intimate and private feeling but the intentional and reason-driven building of conceptual simulation, coordination, evaluation, and experience. This is why these emotion arousing strings of discourses are considered pathetic enthymemes. They are argumentative. They guide you to reason your way into an emotion. 140 What I am trying to do is address the problem of knowing when emotions are being intentionally aroused in discourse and show that attention to conceptual composition and coordination of frames provides a way of understanding the thought processes in these emotive moments. It is clearly not an accident that Jones?s pathetic enthymeme moved many of his audiences in similar ways. And it is clearly not an accident that, to feel how Jones intended his audience to feel, they need to go through the conceptual movements his language prompts. Jones would not have achieved the same emotive effect if he were simply to say, for example, ?Parents are upset because they don?t want to tell their children that a dishonest man is president.? The conceptual work and level of imaginative involvement that an audience brings to processing Jones?s rhetoric is integral for the emotional response that Jones intends. How I can explain exactly what emotion is being aroused when is certainly a thorny issue that affects not only me but all analysis of emotional experiences. Much of what it means to have an emotive experience is grounded in a personal, situated, and interoceptive understanding that is not present in a text (Barrett 2017). What emotion script is activated, how stable these emotion scripts are, and whether there is variation across people and cultures is an issue worth investigating. When a conservative television viewer laughs at Jones?s utterance, they are expressing what Alison Jagger (1989) would call an ?outlaw emotion,? and sometimes observation and theorization of outlaw emotions can provide poignant criticism of ideas about who is privileged to feel what, but even these ideas do not fully capture the problem of arguing for an understandable, normative, and stable emotion script activation. Perhaps some of these questions about what emotion script is activated exactly when 141 are best fit not for a rhetorician but maybe a neuroscientist. For now, it is sufficient to recognize that the range of intended feelings a pathetic enthymeme intends to arouse are metonymically accessible from the cognitive semantics and pragmatics implicit in the text. 6. The Constitutive Grammar of Fear One aspect of the pathetic enthymeme that has not been touched on so far is how it affects the kinds of real-world actions that audiences are guided to take. Sometimes these actions are specified in the conclusions of pathetic enthymemes themselves: vote for so and so, take up arms, sign this petition. These kinds of actions and their effects on the social world are direct enough. But sometimes pathetic enthymemes can have more profound effects on audiences and the ways that they understand their social worlds. This section looks in particular at how Trump uses fear appeals to change how people think about?and by extension, how they talk about, interact with, and argue about?migrants. This discussion feels especially exigent, given the rise of rhetorics of hate in the United States. Some political historians and legal scholars have noticed a decline in explicit race-baiting appeals and hate speech in contemporary American politics. Whereas George Wallace could run on a platform of ?segregation now, segregation tomorrow, segregation forever? in 1962, this kind of explicit racialized rhetoric is less acceptable in contemporary American political discourse. The political theorist Tali Mendelberg (2001) argues that this century?s push for inclusivity has caused racist politicians to find indirect?rather than explicit?ways to invoke racial ideologies. The legal 142 scholar Ian Haney L?pez (2015) has also commented on the current use of in-group coded language?so-called ?dog whistles??that work to secure and reinforce connections between racists in public politics. L?pez argues that racist politicians use dog whistles to at once invoke racist stereotypes while offering avenues of plausible deniability?i.e., I didn?t say that, and if you think I said that, then maybe you?re the racist. While conventional methods for studying dog whistles and implicit race- baiting appeals aim to identify racist coding in words, phrases, symbols, and hand gestures, a pathetic enthymematic analysis shows that meaning?including the meaning gleaned from racist language and sign usage?is not implicit in the language but in the metonymies that language prompts, and the inferences produced from those conceptualizations. As an example of the kind of implicit race-baiting pathetic enthymeme worth examining is this April 1, 2018 tweet from Donald Trump. Trump tweets: News commentators reported this tweet as alarmist, but if any part of it in particular is racist is hard to say. The tweet, I argue, constructs a pathetic enthymeme that implies that Latino people south of the border are dangerous outsiders, and Trump constructs 143 this appeal for political purposes. It is a very dense tweet, though, so it will need to be unpacked part by part. There are three parts of Trump?s tweet, covering three topics. The first part of Trump?s tweet is on the topic of the ?catch and release? policy. In 2001, under the George W. Bush presidency?Bush not being a liberal Democrat, it may be worth noting?the Supreme Court ruling for Zadvydas v. Davis et al. concluded that no person can be detained for longer than six months. Since the amount of detainment facilities that border patrol has is limited, the border patrol primarily seeks to detain individuals who pose a threat to national or border security. This created what has unofficially been called the ?catch and release? policy?which is not a law, it is also worth noting?where ?non-threatening? asylum seekers like women and children are ?released? into the general population under monitored supervision, while ?threatening? border crossers like suspected gang members or drug smugglers are ?caught? and detained. Within six months? time, in accordance with Zadvydas v. Davis, both populations are required to appear in court to make their case for asylum. Next, Trump?s tweet moves to the topic of what he calls the ?caravans? at the border. Trump here is referring to the thousand asylum seekers, mostly hailing from Honduras, who have fled due to drug- and gang-related violence and US-backed state violence following the controversial election of Juan Orlando Hernandez. As asylum seekers, migrants must turn themselves over to border control and make a case for asylum?which is exactly what most of those traveling did when they arrived in Tijuana on April 26, with some opting to seek asylum in Mexico. As Kirk Semple (2018) of the New York Times notes, surveys show that the migrant group that 144 Trump and his base fears so much is composed primarily of women and children fleeing violence. This population poses little if any threat. The final part of Trump?s tweet is on the topic of passing new immigration laws. In 2012, President Obama used his executive power to put in place the Deferred Action for Childhood Arrivals program?or DACA for short. This act allowed people who were brought to America when they were children to apply to register with the program, which allowed recipients to secure jobs or to go to school with guaranteed protection from deportation for a period of two years, with eligibility for renewal. Trump ended the program in September, 2017 and rejected a bipartisan proposal in December, four months before this tweet. What exactly the ?DACA deal? has to do with the catch and release policy or with the traveling caravan of women and children from Honduras is not clear? besides from the obvious fact that they both deal with brown people. Certainly, those arriving in the caravans could not seek to enter into a canceled DACA program. Yet this lack of clarity in Trump?s argument doesn?t stop his audience?us?from making meaning of the three distally related topics. Trump?s argument works in part by getting his audience to activate frames about topics like the Border, Law Enforcement, Immigration, Crossing, Liberals, Republicans, and DACA, just to name a few. These frames plug into a larger dynamic story about illegal immigration that plays out in American political discourse. His audience will undoubtedly have different levels of familiarity about all or most of the topics in these frames. Some of his audience members may be journalists who are very familiar with the topics and thus have rich frames for them, but most will 145 probably be people who aren?t following the topics as closely. Whether an expert or a novice on the subject, the activation of frames helps us to make predictions about our experiences and interactions. Frames can be activated in a number of ways, but the most common way of systematically activating frames is through the use of words. But which words are the ones motivating the implicit race-baiting appeal? A phrase like ?catch and release,? which dehumanizes migrants by borrowing from the target domain of sport fishing, or a word like ?caravan,? which is historically associated with nomadic people in North Africa and the Middle East, may be grounds for race baiting. People are not fish. But the term ?catch and release? to describe border policy has been conventionalized, and caravans are not necessarily bad. These words can?t be the sole cause of the strong reaction I and many Americans had to this tweet. Interestingly, Trump?s tweet provides only schematic details of the kind of message we strongly reacted to. Given the lack of information in this tweet, most of the meaning-making must come from his audiences? processes of inferencing, drawing on available pre-packaged narratives in the manifest common ground. Categorizing a group of migrants traveling from Honduras to the United States as a caravan makes us understand this group using the rich frame we have for caravans, prototypically a group of Middle Eastern merchants. One way you can get a sense of a word?and by extension, the frame it derives meaning from?is to figure out what other words it keeps in its company. A corpus analysis of the word caravan, using the Corpus of Contemporary American English, shows that it collocates with kinds of riding animals (camels and mules), landscapes (desserts, sand, farms, fields), words associated with traveling (routes, 146 roads, the verb move) and names of locations, primarily in the Middle East (COCA). Audiences need to blend their knowledge of caravans?prototypically masses of rural Middle Easterners traveling together on animals?with their knowledge of people who are fleeing Honduras and seeking asylum after political and gang-related violence. The categorization of these asylum seekers as a caravan structures how we see them and could call to mind other stereotypes, for example stereotypes of Middle Easterners, who have also been the target of misguided American patriotism. On top of this, because the conceptual task at hand is to try to understand the speaker?s meaning, the audience needs to think about what Trump thinks about these people. Trump?s tweet does not tell us much. It does not tell us that the caravans are coming to the border, or that the border is getting more dangerous because the caravan?s proximity, or that these people are dangerous, or that we should be afraid? only that something is getting more dangerous and caravans are coming. The audience?s predictive minds do the rest of the work, metonymically reasoning by constructing rich stories from parts of frames. By providing conceptual overlap between the migrants and Middle Eastern caravans, Trump provides some of the core frame elements in the Danger frame and invites audiences to fill in the rest. Much of the act of filling in is structured by the kinds of simulations that Trump cues grammatically. The lexeme coming activates a frame that requires mental simulation of a distal trajector (some entity that is crossing space) moving toward a landmark (an entity that the trajector relates itself to). What is profiled is a trajector which acts as an agent who intentionally narrows the gap between itself and the landmark. The decrease of space between the trajector and the landmark, furthermore, 147 allows inferences for the time it would take for the trajector to reach the landmark. The landmark of the Come frame can take the viewpoint of a proximal object, e.g. The caravan comes to the border, where the border is closer to us, as can be seen in Figure 3. Figure 3. The Come frame Or the landmark can take the viewpoint of the deictic center?the speaking subject? as is the case in utterances like The caravan is coming toward us. The progressive aspect construction, activated by the verb-form coming rather than come, alerts and arouses the audience into conceptualizing itself as the landmark, taking a local rather than a global viewpoint of the trajector?s movement. In the global viewpoint, presented in Figure 4, we watch the caravan come to the border from a distance, but in the local viewpoint, the trajector dynamically moves toward our viewpoint, since we are at the border. 148 Figure 4. A conceptualization of the progressive coming The movement of the caravan toward the border when we are projecting ourselves at the border can be construed as frightening. The word caravan does not generally profile a frightening thing, though. Trump?s appeal to fear here depends on the audience finding coherence between his claim that the border is Getting more dangerous and the claim immediately following: caravans coming. The kind of reasoning needed to establish coherence constitutes a pathetic enthymeme, in that it both guides audiences to a meaning via frame activation as well as requires audiences to provide their own endoxic premises in order to approximate Trump?s meaning. What Trump?s discourse provides are words that activate topical frames. For his audience to understand Trump?s discourse as making arguments, two inferential steps need to take place. The first step?frame metonymy?requires audiences to draw on common ground knowledge in order to fill in missing components of a frame. Trump, for example, never says what is getting dangerous. His use of a non- 149 finite verb phrase (getting dangerous) instead of a more standard clause with a predicate (SOMETHING is getting dangerous) allows him at once to activate the Danger frame, which requires some dangerous entity, as well as to circumvent stating what that entity is, allowing audiences to complete the frame themselves. We know that Trump believes border patrol agents are not able to do their jobs. We can infer that the border is not being checked, then. Because the border is not being checked, it is open to attack. Therefore, it is a dangerous space. Frame metonymy helps us to construct this implicit reasoning process. But how is it getting more dangerous? Words like more or less activate scalar reasoning. The second inferential step Trump has the audience do is to find out what the means of measuring danger are, and, as before, this is dependent on common ground knowledge, not anything explicitly stated by Trump. Since, in one utterance, Trump is talking about an increase of danger and in the next he is talking about the proximity of a distal entity? caravans?the audience must find some commonality between these two in order to make Trump?s tweet coherent. The line of reasoning is that the level of danger is contingent on the proximity of the caravan: the closer a potential threat gets, the more a potential victim is in danger. This logic becomes a grounding endoxic premise for making sense of Trump?s tweet. 150 Figure 5. A schematic of the endoxic premise the closer a potential threat gets, the more a potential victim is in danger This invited inference has the added consequence of framing the caravan as dangerous, as something to negatively evaluate and to avoid contact with. Trump effectively makes his audience turn asylum seekers into potential threats, since they are moving toward the border, which audiences have already established is a dangerous place (because of them). The parallel use of the progressive aspect in getting more dangerous and caravan coming also helps to motivate audiences to make this inference and to make it dynamically. The progressive aspect in present tense profiles an ongoing action that will continue into the future, and with actions like Come, the time of the completion of the action is inferable by observing the pace of the trajector. It is worth remembering now that there was no reason given for fearing the caravan besides its proximity to the US border and the claim that the border is getting 151 dangerous. The more one knows about the caravan?for example, that they are asylum seekers who are primarily women and children?the less dangerous it appears. Knowledge of the caravan in this case violates the invited inference that the closer the caravan is the more danger we are in, and there is little motivation to focus on the moral vulnerable people in the caravan, only to fear the caravan as a whole. What this analysis shows is that the implicit race-baiting appeal, where asylum seeking Hondurans are turned into dangerous entities, is composed by attempts to find coherence between propositions, using common ground presuppositions and activating frames that allow us to form lines of reasoning that make Trump make sense. Echoing the theory of constitutive rhetoric (Charland 1987), a frame metonymic analysis shows how pathetic enthymemes have a constitutive effect, transforming benign objects and actors into threats. Granted, this analysis may say more about how I read Trump?s tween and less about how Trump?s supporters did. Nevertheless, for them, too, affective coherence is dependent on the activation of high-level frames (like Danger) cued by low-level metonymies (such as caravan and even coming). What otherwise would appear to be a disjointed remark is understood as an intentional, vivid, completely coherent, and effective pathetic appeal. These implicit race-baiting appeals also depend on a worldview that many of his readers may not share yet have to construct by connecting disparate frames in a particular way. These activations have consequences. George Lakoff (1996) argues that the activation of a frame?in this case, several frames?cognitively and neurologically reinforces the value of that frame and the probability of that frame being activated for future acts of reasoning. When these frames are chained together, 152 they allow for coherent worldviews to be constructed, activated, and reinforced from underdetermined discourse. This is what happens when we read Trump?s tweets. Trump gets us to activate and exercise these frames in order to make him make sense. We need to understand that he thinks this caravan is threatening in order for his tweet to be coherent. Even if we do not believe that the caravan is threatening, Trump has turned a group of asylum seekers into a group whose threat-level is a topic for debate. He is inventing points of argument, effectively situating focus on a proposition from an unshared worldview. In the final part of Trump?s tweet he also calls on Republican politicians to act swiftly to pass new immigration laws and reminds his audience that there is no more DACA deal. The exigence for a so-called nuclear option for passing new immigration laws comes directly from the invented threat of these asylum seekers. To argue against the need for Republicans to pass new immigration laws here also requires the activation and negation of the framing of the asylum seekers as threatening, thus reinforcing this characterization of the asylum seekers conceptually and in political discourse. What my analysis shows is that Trump?s implicit race-baiting appeal, where asylum seeking Hondurans are turned into dangerous entities, exists as a result of his audience?s attempts to find coherence in Trump?s discourse. The bigger point doesn't just apply to race-baiting events (attempts to arouse racial animus), but to all, or most, or at the very least, very many kinds of pathetic appeals, where the affective logic of the appeal depends on the selective activation of L1 and L2 imagery in order to activate shared L3 communal evaluations. Trump gets audiences to construct the argument by having them use their own knowledge base. These appeals require a 153 great deal of activating and coordinating culturally rich semantic frames, aligning or at least entertaining the alignment of values, predicting speaker meanings, and using common ground reasoning to bring coherence to underdetermined lines of reasoning. Much of our mental reconstruction of these sorts of implicit appeals is enthymematic and frame metonymic, especially in politics and especially when we are trying to follow arguments presented in 280 characters. Cognitive enthymematic analysis?analysis of what the audience is providing to complete an argument?is helpful for unpacking emotional appeals, as well as showing how they work, often below the level of conscious awareness, to invite inferences. As president, Trump is in a privileged position where he has the ability to get some of his audience, myself included, to do a tremendous amount of cognitive work in order to make him make sense. Still, these implicit race-baiting appeals have constraints, as can be seen in attempts to translate Trump?s discourse. Japanese translator and interpreter Chikako Tsuruta (2017) has commented that translations of Trump?s words often do not capture the same meaning in English, leading to confusion in cross-cultural communication. Tsuruta attributes this to Trump?s incoherent speaking style. But, as I have shown, Trump is not incoherent, just indirect, relying on his audience to metonymically activate implicit cultural scripts that have tremendous resonance with some of his intended Right-Winged audience. Viewing Trump as an incoherent idiot, a sick-man, or an acclaim-seeker, as rhetoricians William Penman and Doug Cloud (2018) have pointed out, only detracts from the pressing issue of Trump?s role in further normalizing white supremacy. The growing scholarship on Trump, as well as the critical race scholarship surrounding 154 covert and implicit race-baiting in public politics, would do well to unpack the ways that people are invited to transform marginalized groups into stereotypes to further political agendas. Rhetorical scholarship is in a privileged place to make a difference in analyzing racist political discourse, but it still needs to find methods for analyzing implicit arguments that are systematically felt to be racist but hard to describe propositionally. Attention to the role of emotional persuasion, and, in particular, the pathetic enthymeme, may help provide a methodology for analyzing the intersections of race, politics, and affect. 7. Rising Homicides: Metonymy in Trump?s RNC Speech As I hope has become evident, to understand a pathetic enthymeme often requires a considerable amount of mental work: bringing online and arranging rich semantic frames, drawing on common ground beliefs, and coordinating underlying assumptions to arrive at a meaning that goes beyond the language a rhetor uses. These enthymemes can help to strengthen beliefs, shape categorization of events, activate emotion scripts, guide reason, and prepare for action. This chapter opened on commentary that followed one of Trump?s most notable speeches that deployed dog whistle appeals for emotive effects. Trump?s RNC acceptance speech painted a bleak picture of a crime-filled America with bloodthirsty gang members flooding through the boarders. Early in the speech, Trump presents what he calls ?the facts? in a list of declarative sentences spanning crime policy, crime statistics, violence against police officers, and undocumented immigrants. 155 Decades of progress made in bringing down crime are now being reversed by this administration's [the Obama administration?s] rollback of criminal enforcement. Homicides last year increased by 17% in America's fifty largest cities. That's the largest increase in 25 years. In our nation's capital, killings have risen by 50%. They are up nearly 60% in nearby Baltimore. In the president's hometown of Chicago, more than 2,000 have been the victims of shootings this year alone. And almost 4,000 have been killed in the Chicago area since he took office. The number of police officers killed in the line of duty has risen by almost 50% compared to this point last year. Nearly 180,000 illegal immigrants with criminal records, ordered deported from our country, are tonight roaming free to threaten peaceful citizens. The number of new illegal immigrant families who have crossed the border so far this year already exceeds the entire total of 2015. They are being released by the tens of thousands into our communities with no regard for the impact on public safety or resources. This string of discourse works to make the Obama administration look like it has been complicit and at times has aided an increase in violent crimes. It also works to make the audience believe that violent crime is on the rise and that undocumented immigrants are potentially violent criminals. The passage accomplishes the first goal by making the audience construe criminal activity increasing, though the grounding for this increase is often strategically selected or is just left up to the audience to construe. It accomplishes the second goal by strategically getting audiences to project traits of violent criminals onto undocumented immigrants. 156 The construal of violent crime as increasing is accomplished in mostly the same way. Trump densely uses numbers and other quantificational expressions, and then zooms in on the way that these figure on a scale of increase, i.e. the UP frame. Trump uses words and phrases like risen, up [on a crime scale], more than X, and almost Y [where Y is double X] to prompt the construal of a crime scale and project the increase of some negative entity (crime, murders, victims of murders) from the Crime frame onto that scale. In the first sentence of the passage, Trump cites progress?in bringing down crime, which prompts the construal of the crime scale but not the increase of crime. He achieves the effect of construing crime as increasing by stating that the Obama administration reversed this progress, thus making crime go up rather than down. The loci of crime moves, too, from DC, to Baltimore, to Chicago, before zooming in on the figure of the illegal immigrant ?roaming the country,? constructing a generalized threat across the nation. Trump?s rhetoric leads audiences to understand the frame of Crime not from any fact but from the construal of rising. The directional frame of Up, which is activated not only through lexical items like up but also from a stylistic incrementum (17%, 50%, 60%), moves audiences to understand the Crime frame in a particular way, inviting audiences to infer that the growing number of roaming criminal immigrants means that their presence proximal to the audience is more likely. Through activations of the Crime frame and the Up frame, Trump constructs a pathetic enthymeme that can guide audiences to believe that they have reason to fear and that the nation is not safe. 157 Trump?s Up scale works even when his figures are grounded in unstated points. If crime was low during the Obama administration and has risen only slightly since, crime is still low. There is no number given for what the low point or the high point is on the Up scale. When homicide rates are said to have increased by 17%, a time scale is not given. When Trump states that this statistic applies to the fifty largest cities, there is a potential implicature that this trend carries for the rest of the country as well, as fifty is a large number and the cities are large. The presented scale and magnitude drive the audience to make generalizations about the increase of homicide rates. While Trump?s rhetoric does not work through reasoned facts, that the effects of his rhetoric are explainable demonstrates that there is some reason in his argumentation strategy. 8. Conclusion Instead of trying to better understand how Trump?s rhetoric works, many political commentators were quick to point out that his facts were untrue or misleading. His argument relies on selective time scales or selective events to make elements appear more saliently and thus more likely to be trends in the increase of crime. Glenn Kessler and Michelle Ye Hee Lee (2016), writing for the Washington Post, make this point when they write that ?Trump cherry-picks data to paint an alarming picture of homicide trends, when in reality, they have been declining for decades.? Trump also cherry-picks the number of immigrant families crossing into the United States, choosing to use fiscal year data that began in October of 2015 to make it appear as if his speech in July was halfway through the year rather than 3 months from its end. 158 Trump?s rhetoric here works by getting audiences to either attach to or to guess a base estimate of what seems like a reasonable level of murders, violent crimes, or border crossings all while selectively finding data points in time scales that will reinforce the impression that crime is rising. But you can?t fact check a construal. Construing crime as on the rise and criminals as roaming free allows for the categorization of America as a potentially threatening place and citizens of America as potential victims of criminal activity. Identification with these citizens can cause audiences to fear becoming victims of violent crime. While the truth of the matter is, as Alyson Camerota expresses to Newt Gingrich in her coverage following Trump?s speech, that violent crime is on the decline according to the FBI, Trump?s rhetoric prompts audiences to construe otherwise. The mental work of pathetic enthymemes?most of which is done instantaneously and seemingly automatically?does not involve enlightened and reasoned selection of facts and figures. It involves selective attention and charitable listening with the intention of understanding. And herein may lie the danger of the emotional appeal that rhetoricians like Quintilian warned against. They do seem to limit higher order careful reasoning, sure, making gut reactions feels like common sense. But what an analysis of any pathetic enthymeme demonstrates is that what may feel like a gut reaction is, in fact, a carefully constructed way to guide attention to experience a specific feeling and to get you to do much of the heavy lifting via frame metonymy. Frame metonymy is easy, spontaneous, and very, very unconscious. It is also subject to entrenchment and alternation through repetition, following usage- 159 based principles. Put together, this means that a well-crafted enthymeme and an audience that gives the minimal effort needed to arrive at a coherent understanding will necessarily activate certain evaluative frames. These frames, furthermore, plug into the rich and complex stories that circulate, prompting us to imagine vivid mental simulations that may feel fiery, thrilling, satisfying, righteous, and even just plain right, even when the facts suggest otherwise. The groundworks for being moved are not constrained by credible sources. Rhetors can excite the imagination and guide audiences to invent their own fears when good reason to fear does not (yet) exist. And rhetors often do. With any effective rhetorical move, however, comes a potential counter. As an L4 strategy, the pathetic enthymeme is reliant on audiences to be complicit in their own emotional susasion. But audiences do not necessarily need to be complicit. By understanding the conventional strategies of pathetic enthymemes?i.e., by being able to recognize frame activations and frame metonymies?audiences can provide critical distance from the mental movements that rhetors are inciting. This critical distance diminishes the effects of an emotional appeal, but herein lies the problem that many commentators and analysts face. Explaining how an emotional appeal works, fact checking an emotional appeal, or simply labeling an emotional appeal as a manipulative or fallacious way of arguing does nothing for those other audience members who have been moved to feel differently. Perhaps the nation would do well learning to be critical of emotional persuasion, but not everyone can afford that education. 160 If we were to take Gingrich?s stance from the introduction to this chapter? that emotional reality is experiential, normal reality?what would Trump?s fear appeal look like to analysts who didn?t stop to fact check premises and criticize opinions? Rhetorical analysis in this kind of situation may illuminate much about the ways arguments systematically arouse emotions?that is, the way that a rhetor?s language works to activate rich, dynamic emotion concepts in audiences? minds. Taking emotional appeals seriously also follows from Aristotle?s call to look, in any given situation, for the available means of persuasion. If an argument can lead someone to change their feelings, then there ought to be an explanation for how that argument led to one feeling rather than another. What I am not saying is that, in our ordinary lives, we should remain impartial to emotional appeals, opting instead to find the reason that motivates them. On the contrary, while we may not be able to help ourselves from feeling, we may be able to keep ourselves from being whipped into a frenzy when hearing of those who attack and want to destroy the things we love and care for. My argument here is that analyzing emotional appeals not only shows us patterns of language and thought implicit to arguments for emotion but also represent the kinds of natural rationality we operate by day to day. With an understanding of how our minds are moved, we may be able to find ways to keep our priorities in check, so as not to be swayed by malign sophists. While the conventional rhetorical theory purports that emotional appeals are without reason or against reason, it seems to be quite the opposite. There is more charity, more of a search for a common ground from which to reason that affords effectiveness of the emotional appeal. And it?s in this charitability that rhetors who 161 use emotional appeals can get their audiences to feel for some action: to get angry to vote; to feel shared fear and empathy for our children?s future under a bigoted bully, so that we may fight the moral fight for a better tomorrow; to feel sorrow and anger toward those who are tasked with protecting us but don?t seem to care for our safety, so that, by our actions, we may elect someone who will. Categorizing mental events as instances of emotions prepares our minds and bodies to respond to the world in ways that feel immediate, common sensical, and natural, and there is power in making a vote for a political party feel natural. But there is also a lot of carefully reasoned work. One thing that we may yet learn from analyzing Trump?s rhetoric is the art of telling a story and the art of inviting audiences to imagine themselves within that story. Because commonsense knowledge is organized in a network of frames, we need to analyze persuasive rhetorical discourse as guided frame activation?that is, we need to understand how frames fit together in discourse to construct larger and more complex stories. From this, we can determine where our reason is directed to look, what to look for, and how to look at what we might find. While Trump may appear utterly incoherent, for those who are willing to follow his directions, Trump?s rhetoric prompts rich metonymies that are able to align attitudes, sharpen opinions, and reinforce feelings of community. Trump may come across as incoherent, but that it?s that perception of incoherence that masks the emotional power his rhetoric has on so many. Having a framework for systematically describing the reasoning process of those moved by his rhetoric provides a key for understanding how emotional persuasion works. 162 Chapter 4: The Function of Enargeia in Emotional Appeals 1. Introduction: Emotion in Charity Appeal Letters According to the U.S. Postal Service (2012), over half of the mail that is delivered in the United States is direct mail. A large portion of this mail comes from charities and other fundraising organizations. Direct mail fundraising letters?or appeal letters, as I will call them in this chapter?seek to bring attention to environmental, political, and social issues and claim that, with the monetary support of donors, the authoring organizations can do something about these problems. A serious business, appeal letters are responsible for a large portion of the $410.2 billion that Americans donated to charities and non-profits in 2017 (Giving USA). While different fundraising organizations have their own strategies for writing appeal letters, few organizations have tested the efficacy of appeal letters empirically. This has led a growing body of research in cognitive science, non-profit studies, corpus linguistics, and technical writing to begin to ask critical questions about the roles of persuasion, cognitive bias, genre-specific features, and rhetorical appeals in the construction and interpretation of philanthropic discourse, especially in appeal letters. Recent attention to the genre of appeal letters has pointed to the wealth of conventionalized rhetorical acts a letter is comprised of. As corpus linguists Ulla Connor and Thomas Upton (2016) describe, ?[f]undraising texts are fascinating: they persuade, inform, request, catch one's eye, wrench one's heart, and twist one's arm in a tidy attractive package? (69). One phenomenon of interest in philanthropic discourse is the discursive construction and the persuasive role of the emotional 163 appeal in appeal letters. Emotional appeals appear frequently in appeal letters, often at the beginning to draw audiences into the issues that the fundraising organization goes on to detail, as well as at the end of the letter, to motivate audiences to donate money. For example, Four Corners Equine Rescue, a volunteer non-profit located in Aztec, New Mexico, begins one of their fundraising letters like this: Do you see how this horse is crashing head over heels with a rope wrapped around its legs? That?s called horse tripping and I desperately need your help to stop it! (ALC.CA.1) Figure 4. Image of horse tripping, provided by Four Corners Equine Rescue The letter?s emotional appeal is composed of two sentences that first introduce an emotive problem and then a request of the audience to help the appeal letter writer address the problem. What is most interesting about these appeals are the ways that they direct attention to problematic scenes where vulnerable entities are put into harmful situations, often depicting these scenes succinctly but with enough detail to construct a vivid impression. 164 My aim in this chapter is to analyze how emotional appeals are constructed and how vividness emerges in the emotion-arousing discourse of fundraising appeal letters. This analysis is conducted through cognitive corpus linguistic and rhetorical analysis to provide a qualitative and quantitative methodology for analyzing rhetorical enargeia. In classical rhetorical theory, the technique of enargeia, 1 or the construction of vivid mental simulations through language, was said to ?[thrust] itself upon our notice? (Quintilian 8.3.62) and make readers and listeners into metaphoric ?spectators? (Quintilian 8.3.63, pseudo-Longinus 15.1). Specifically, enargeia involves using language to induce sensory temporal simulations in the minds of the audience, imaginatively moving them, getting them to zoom in a scene in the mind?s eye, or otherwise mentally reposition themselves to see and feel what is being described with words. The technique of crafting enargeia was thought to be useful for courtroom argumentation, where audiences could be made into fictive eyewitnesses, and it was thought to be best suited for arousing the emotions of audiences (O?Connell).2 1 Also descriptio (ps-Cicero 4.39), notatio (ps-Cicero 4.65), hypotyposis (Quintilian 9.2.40), illistratio, and evidentia (cf. Quintilian 6.2.32). Each term characterizes a different aspect of the same phenomenon: a verbal sketch of a mental image (hypotyposis), an ocular demonstration from description (description, notation), a metaphoric pictorial illustration (illustration), and, as it was often used in legal argumentation, the production of a mental image that could be used as a makeshift eyewitness evidence (evidentia) that the audience could see. 2 The belief that linguistically constructed impressions of vividness can arouse emotions is grounded in the Greek classical philosophy of emotions, which argues that mental simulations can generate emotions in certain cases. Aristotle, for example, defines fear (phobos) as a distress at a mental simulation of danger in the future (2.5.1). Assuming that different emotions prototypically have 165 Rhetoricians from classical antiquity to the Renaissance were keen to observe various linguistic features of enargeiac language?for example, vividness constructed from temporal shifting (translation temporum), the use of deictic markers and directives (e.g. look, see, here, there), the use of reported speech, and the heavy use of descriptive language when depicting things or events (see Lausberg ?812). Contemporary treatment of enargeia, such as the analyses by Ruth Webb (2009) and Heinrich Plett (2012), have tended to focus less on the linguistic features and more on enargeia?s theoretical foundations, pedagogical lineage, and various rhetorical functions. By analyzing the strategies for constructing pathetic enargeia in fundraising appeal letters, this chapter returns to the intersections of the linguistic construction of enargeia and the rhetorical functions vividness serves in the emotional persuasive philanthropic discourse of charity appeal letters. My principle argument in this chapter is that a cognitive rhetorical analysis of enargeia shows that vividness is constructed not only through descriptions of events or through embodied sensory experiences felt through mental simulation?two common beliefs about enargeiac style?but also through inserting presupposed values and beliefs within the rhetor and audience?s common ground. These presuppositions establish a fictive personal relationship between the rhetor and the different core conceptual content?fear involves some future danger, love involves some desire to be with something, and so on?a description of an event that can lead one to believe that references some core conceptual content event is better able to arouse an emotion. Enargeia thus lends itself to pathos when casting vividness on specific kinds of events: ones where actions share core conceptual content with the emotion concepts?for example, the plundering of a village arousing fear or the effects of famine on a farmer arousing pity. 166 audience, and they invite the audience to identify in a specific way. Presuppositions appear either within enargeaic mental simulations, profiling people, places, objects, and values that are being threatened and presumed to be of importance. They also work to establish relationships between fundraising organizations and their audiences through arousing enargeiac mental simulations. As evidence for my claims, I share results from quantitative corpus linguistic analyses of 128 appeal letters as well as qualitative analyses of relationship building strategies in enargeiac mental simulations. By showing how enargeia is constructed and how it serves a rhetorical purpose, the chapter opens new ways of identifying and analyzing the role of mental simulation and identification in emotional persuasion. 2. The Appeal Letter Corpus Before discussing how enargeia is constructed in fundraising appeal letters, it may be helpful to better understand the genre of the appeal letter and the role that emotional appeals more generally serve within this genre. Persuasion and the emotional appeals in fundraising letters have become an interdisciplinary topic of research. In work in rhetor and technical writing, Marshall Myers (2007) has argued that charity fundraising letters solicit donations through strong appeals to pity, going on to emphasize the rhetorical role enargeia plays (8-9). For Myers, the analysis of appeal letters reveals sophisticated approaches to persuasion that hinges on effective appeals to pathos. As Myers asserts, ?It would be difficult to imagine a social cause so great and appealing to the masses that did not in some way have an emotional element in it? (6). Corpus linguistic work conducted by researchers associated with the Indiana 167 Center for Intercultural Communication (ICIC) has also studied the role of emotional appeals and the construction of vivid images in fundraising appeal letters (Connor and Gladkov; Biber, Connor, Upton, Anthony, and Gladkov). Tagging their corpus for emotional appeals and rhetorical strategies for constructing pathos, the researchers found that, of the emotional appeals in their corpus of fundraising letters, a significant number were constructed through the technique of enargeia (Connor & Gladkov 272; Biber, Connor, Upton, Anthony, and Gladkov 135). Research in marketing has also found that experience and judgement of a product or a brand can be influenced through rich descriptions that direct sensory simulation and cause feelings of intimacy, emphasizing the links between mental simulation and emotion in persuasive marketing appeals (cf., e.g., Peck and Child; McCabe and Nowlis). While this interdisciplinary research concludes that the emotional appeal plays an important role in the persuasive discourse of fundraising appeal letters?and that mental visualization can amplify persuasive effects?it reinforces the need to conduct a closer study of how emotional appeals are typically constructed and implemented for persuasive purposes, a topic not touched on in prior research. To further analyze emotional appeals in fundraising appeal letters, I examine the linguistic ways that emotional appeals and enargeia are rhetorically instantiated in 128 fundraising letters. The letters comprise the Corpus of Appeal Letters, a 112,043- word corpus I constructed from collecting letters from 82 fundraising organizations. The letters of the corpus were sent out via direct mail to residents in Maryland and Washington D.C. between the years of 2013 and 2016. The letters in the corpus reflect direct mail letters that were sent to me; to my male, female, and nonbinary 168 friends; and to my male advisor, all of whom live in the Maryland and Washington D.C. area and have post-secondary degrees. As such, they constitute something of a convenience sample and may contain biases based on this sampling. To analyze the corpus, I used the corpus linguistic program Wordsmith Tools (2017), which was used to search concordance lines as well as generate wordlists and keyword lists, and I also used Laurence Anthony?s (2018) AntGram, a program that allows researchers to find phrase-frames (p-frames), which are partially lexically filled phrases. Keyword analyses have been used in similar corpus-driven Critical Discourse Analysis studies of political and social phenomena (McEnery, McGlashan, and Love; Mulderring), as well as analyses of the role of emotion in discourse (Bondi). In a keyword analysis, the frequency of words in a specialized corpus (i.e., a corpus constructed for a study) is compared to the frequency of the same words in a reference corpus (e.g., the Corpus of Contemporary American English or the British National Corpus). A keyword is a word that appears unusually or unexpectedly frequently in the specialized corpus given the frequency of the same word in the reference corpus. The statistical significance of the frequency is measured by finding a statistical value, the log-likelihood, which indicates whether the frequency ratio is by chance or not. If a given word has a log-likelihood value of 3.84 or above (i.e., p <0.05, or 95% certainty that the frequency distribution is not by chance), it is considered to be a statistically significant keyword of the specialized corpus and its frequency in that corpus is considered to not be by chance. The higher the log- likelihood value, the less likely the word?s frequency is by chance. For example, if a 169 corpus of Gilligan?s Island episodes were to be measured against a reference corpus of general American English, words like Gilligan, Skipper, island, ship, and professor would appear as key words with high log-likelihood values, meaning that the distribution of those words appear more frequently in the Gilligan?s Island episodes than in general American English. According to Mike Scott (2001), keywords can be thought to illustrate the ?aboutness? of a corpus?that is, each keyword indicates the topics, elements, situations, and concepts that comprise what the corpus is about. Gilligan?s Island is thus ?about? Gilligan, the Skipper, an Island, and so on. To find the words that are unique to appeal letters, I used the written sub- corpora of the American National Corpus (2015) as a reference corpus. The American National Corpus is a large 22-million-word corpus of American English, and the written sub-corpora consist of 11,782,177 words, taken from journal articles, websites, technical science and medical writing, the 9/11 Report, non-fiction books, personal letters and email correspondences, and travel guides. Comparing word frequencies in the Corpus of Appeal Letters to the entire sub-corpus of the American National Corpus helps to pick out features of appeal letters that are unique to the genre and, thus, are not features of academic journal articles, email correspondences, technical reports, or general non-fiction writing. 3. Crafting Intimacy Keyword analysis is helpful for getting a sense of what sorts of discursive and rhetorical moves are central to the appeal letter genre. The words in Table 1 below show the top 25 keywords when comparing the Corpus of Appeal Letters to the 170 written sub-corpus of the American National Corpus. By themselves, these words are not inherently revealing, but when analyzed through the theater of the mind model, the words sketch a kind of basic scene frequently shared across appeal letters. This scene involves a focused attention to crafting a direct appeal (help, please, support, need, protect) to an audience (your, you) by a rhetor who stands in for an organization (we, us) that advocates or fundraises for some valued entity (NRA, gun, food, wildlife, wilderness, people, hunger). Rhetors often request money as donations that should be sent with the prepaid envelopes they provide (enclosed, gift) and presuppose an audience?s cooperation in donating or otherwise supporting to the organization (generous, thank, join). Key word ALC Freq. % Texts ANC-W Freq. ANC-W. % Log-Likelihood 1 OUR 1245 1.11 124 12779 0.11 3458.94 2 YOUR 1053 0.94 123 8824 0.07 3293.72 3 YOU 1547 1.39 124 24023 0.20 3219.14 4 HELP 538 0.48 116 3457 0.03 1930.81 5 PLEASE 303 0.27 102 1056 0.00? 1406.60 6 GIFT 276 0.25 83 829 0.000 1350.26 7 WE 1051 0.94 122 29540 0.25 1218.61 8 SUPPORT 383 0.34 107 3557 0.03 1129.96 9 TODAY 344 0.31 111 3581 0.03 947.42 10 ENCLOSED 132 0.12 55 287 0.00? 715.68 11 GENEROUS 124 0.11 68 229 0.00? 704.38 12 NEED 295 0.26 101 3820 0.03 703.07 13 NRA 92 0.08 16 39 0.00? 700.15 14 GUN 152 0.14 16 556 0.00? 693.15 15 US 341 0.31 107 5536 0.05 684.73 16 THANK 122 0.11 79 424 0.00? 566.93 17 WILL 603 0.54 115 19913 0.17 563.81 18 FOOD 182 0.16 36 1651 0.01 544.33 19 WILDLIFE 88 0.08 21 124 0.00? 536.19 20 WILDERNESS 82 0.07 12 88 0.00? 531.78 21 PROTECT 134 0.12 45 784 0.00? 502.72 22 EVERY 239 0.21 91 3931 0.03 474.82 23 HOPE 146 0.13 71 1160 0.00? 470.15 24 PEOPLE 370 0.33 88 9679 0.08 467.58 25 JOIN 110 0.10 53 496 0.00? 462.25 171 Table 1. Keywords of the Appeal Letters Corpus As the words with the highest log-likelihood values illustrate, the most keyed words in the corpus are personal pronouns, drawing attention to the relationship between the speaker and hearer in the common ground. In the theater of the mind model, the profiling of this relationship is at the third level (L3). This level is also where common knowledge is negotiated, presuppositions are projected, and values between writers and readers are aligned. Personal pronouns like our and your are also interesting because they bring the participants of the common ground on stage, and they do this to profile a possessive relationship to some nominal thing. In the prototypical speech situation, the speaker and hearer are in the same proximal space and thus are able to ascertain what sorts of things are possessed by either of the speech participants. In letters and other forms of asynchronous communication, however, writers and the addressees do not share a physical space, so the onstage profiling of possession marked by [our NP] or [your NP]3 project presuppositions into the common ground about who knows who has what. Given the asynchronous communication, the letter writer holds the floor and thus is able to manipulate the common ground for persuasive purposes. Determinatives like our and your are often thought of as simply referential deictic markers that profile speakers? and hearers? relationships to the world, in particular to some nominal. Cognitive Linguistics, however, has shown that the meanings of determinatives are far more complicated. Langacker (1991b) has 3 Within conventional linguistic shorthand, NP means noun phrase and Adj means adjective, V means verb, and Ving is the progressive particle form of a verb. 172 demonstrated that determinatives say a lot about how language users relate elements of linguistic expressions to the here and now of the speaking situation (i.e., the ground). As such, determinatives are markers of intersubjectivity. Grounding predications, like our cause and your immediate contribution, do more than just reference some possession; they presuppose a refence in the real world? correspondents to the our?and focus readers? attention to that presupposed reference and their presupposed possession. The speech act of making this predication has a further cognitive function of coordinating attention and communicating assessments about the world (Langacker 1991b, 322). Even something as simple as a personal pronoun, which may seem like it merely identifies participants in the ground, expresses complex intersubjective meaning. Langacker (2007) writes that ?[i]n the case of I and you, the mutual apprehension of other minds is not just crucial but pivotal,? since ?[a] central if not a defining feature of intersubjectivity is the apprehension of other minds and what they apprehend? (182). Given that one speaker?s I is another hearer?s you, discourse participants are keen to attending to the grounded relationship, imagining what interlocutors believe about themselves, believe about their interlocutors, and believe about their shared discourse space. Grounding predicates like [our NP] and [your NP] go further, in that they put on stage an image of the reader objectively construed in the background of a bad situation, inviting the actual reader to subjectively identify with that image. A phrase frame (p-frame) analysis of the Appeal Letter Corpus shows various kinds of grounding constructions that help the letter writer to manipulate the common 173 ground. In addition to [our NP], which appears 842 times over 123 of the 128 letters, other grounding constructions include the more specific [our [Adj NP]], [the [Adj NP]], [you and I [VP]], [your [Adj NP]], and [Ving [(to) you]. The words that fit into empty slots in these p-frames can be found in Table 2.4 p-frame Freq. Words that fill slots [our NP] 842 cause, abilities, air, loved ones, guns biggest, collective, historic, most fundamental, [our [Adj NP]] 233 amazing, important, precious [the [Adj NP]] 794 accepted facts, actual cost, lying Democrats [you and I [VP]] 18 celebrate, hear, enjoy, know, see, understand immediate contribution, ongoing support, sincere [your [Adj NP]] 150 kindness [Ving [(to) you]] 104 writing, turning, speaking, sending, asking, urging Table 2. Frequent p-frames that ground Far from objectively stating what is presumed to be common knowledge to the reader and letter writer, moments of L3 grounding configure, (re)align, and propose a common ground from which a letter writer?s argument may begin. When the global women?s rights organization MADRE invites their audience to ?join our amazing 4 p-frame analysis normally finds frequent n-grams (bigrams, trigrans, etc.) where one word is replaced with a wildcard, represented by an asterisk (*), but p-frame analysis doesn?t always pay attention to syntactic phrase structure. To analyze more complex noun phrase constructions (e.g. [our [Adj NP]] as well as [our [NP]]), I used a version of the Appeal Letter Corpus that has tagged each word and character for its part of speech. The tagger used was Helmut Schmid?s TreeTagger program. 174 organization? they amplify the value of the organization while introducing it into the common ground (ALC.H.26). When the Wounded Warrior Project thanks the audience for ?your sincere kindness? (ALC.H.46), the audience is not only supposed to understand that they have been being kind but also that this kindness is sincere, persuading them to align their intentions with the letter writer and the organization?s in a more vulnerable and transparent way. Being able to control the ground can rhetorically affect how letter readers understand themselves as intended addressees. When the Parents and Friends of Lesbians and Gays (PFLAG) write ?Our strength is based on the firsthand experience of seeing our loved ones discriminated against and hurt? (ALC.H.33), the letter writer projects some presuppositions that affectively shape the way that the addressee is meant to identify. The use of the [our NP] phrase here profiles a collective strength that the letter writer and addressee both share due to experiences of witnessing discrimination of gays and lesbians. Controlling the ground here means being able to hail the addressee to identify with a friend or family member of gays and lesbians who experience hardships due to their identities. It establishes a proximal relationship to gay and lesbian friends and family, and, by doing so, motivates a reason for caring about gay and lesbian issues. For PFLAG, LGBT allies and family members are the target audience, so being able to discursively establish a relationship with the target audience?and to be able to do so early on?is an integral strategy for persuasion in the appeal letter. It is, in a sense, a prerequisite for argumentation on issues of gay and lesbian protection to begin. 175 As should be evident, the L3 grounding constructions are part of a larger L4 rhetorical strategy, wherein readers are guided to identify with a constructed or ?invoked? addressee (Ede and Lunsford). I call these L4 strategies engagement strategies, as they make explicit the intersubjective relationship of members of the common ground and by doing so strengthen it. At the most basic level, engagement strategies linguistically signal aspects of discourse participants? ownerships (our, your, my, etc.), construing representations of members of th ground on stage (I, me, you), or otherwise bring attention to the interaction of discourse participants. This can be through speech acts (e.g., requests, directives, questions, greeting, etc.), through strategic interpersonal framing, as well as through the use of more complex rhetorical strategies like enargeia. In the later half of this essay, the concept of enargeiac engagement is introduced to demonstrate how appeal letters use mental visualization and the crafting of vividness to craft intimacy between letter writers and addressees, as well as to strategically construct identification for the purpose of larger argumentation. While the audience that is constructed differs across letters, the process of constructing audiences through engagement strategies remains constant: what the audience is expected to know and what values the audience is expected to have are built into the common ground through constructions like [your NP] that presuppose something about what the reader has, what the reader believes, and what the reader already knows. Due to the conventions of the appeal letter genre, which formally borrows from the genre of the personal letter, writers of appeal letters take advantage of feigned intimacy with their audience, allowing them to make these presuppositions 176 about shared values and knowledge without question or a feeling of intrusion. Letters will sometimes refer to a history of the reader with the organization, and, since the letter writer metonymically stands in for the organization, this history can metonymically also be transferred to the feigned intimate relationship between the letter writer and the addressee. Addressees are not just readers but friends, the presupposition that the reader is a ?friend? appearing 67 times across 59 letters in the corpus. When they are not friends, they are neighbors, members, supporters, and fellow lovers. These words help to define the kind of relationship the writer and addressee are presupposed to have, thus allowing the letter writers to strike intimate tones with the addressee. To call someone a friend presupposes a certain kind of relationship. Friends listen to one another, at the very least. They are there for each other in times of stress and hardship, and they do what they can to help. Friends are keenly attuned to each other?s feelings. Casting the addressee of an appeal letter as a friend encourages readers to take up a sense of affection and empathy with what the concerns of the letter writer. This affection potentially motivates readers to allow themselves to be guided to attend take serious the issues of concern letter writers voice. And, since the concerns of the letter writer are the concerns of the fundraising organization, the reader is encouraged to metonymically become entuned with the large non-profits and campaign organizations that are soliciting money. The crafting of a friendship in the common ground is in a way the chief emotive strategy of the fundraising appeal letter, on which all other emotional appeals play out. 177 Within the common ground of the appeal letter, the letter writer and the addressee engage in what Esther Pascual calls a ?fictive interaction.? For Pascual, fictive interaction involves discourse that simulates an imagined conversational relationship that is maintained through quasi-dialogic discursive moves that are ?non- genuine? in that they are never uttered with discursive intent. Pascual argues that fictive interactions are an often-overlooked aspect of discourse that shape cognition and serves specific rhetorical functions?from large-scale fictive interactions like the pedagogical dialogues of antiquity; to sentence-level constructions like the rhetorical question, which use the interactive forms associated with information seeking fictively to make rhetorical points; to even smaller scale fictive exchanges, like the quasi-quotational (be) like construction in ?I was like ?come here? and he?s like ?no way!?? (Pascual 6-7; 41-48). In each case, non-genuine discourse makes clear that the discursive interaction being signaled is not to be taken as a real interaction. In the case of appeal letters, the non-genuine discursive moves that signal the fictive interaction between the letter writer and reader include the repeated references to the addressee as a friend; the usually frequent reference to members of the conversational ground through personal pronouns, such as you or we; the use of speech acts and discursive moves that rely on already established relationships, for example the utterance Can I count on you (to do X) (ALC.H.34, ALC.H.2, ALC.CA.E.1, ALC.P.6); and the frequent grounded presuppositions of familiarity and joint alignment with controversial matters social, political, and moral. In reality, the work that goes into compressing the appeal letter reveal a very different picture. The letter writer for an organization may be a team of writers, who are compressed into a 178 single, named author?sometimes even a famous person, like Jane Goodall (ALC.CA.3), Sigourney Weaver (ALC.CA.8), and even The Dalai Lama (ALC.H.22). Given that the reader can be anyone who the letter is sent to, the actual readers of an appeal letter, like myself, may also not agree with the views and values that the fictive reader is presupposed to hold. Nevertheless, the L3 grounding strategies that entertain a fictive interaction invite an identification with a specific kind of reader, one who will not question the propositions and logic of valuation to be accepted into the common ground. The read reader may disagree, but to be able to understand the thrust of the argument being made, the reader must understand what the letter writer would like the addressee to be thinking and feeling. With this ground work established, the reader can jointly attend to the various L1 events being staged in the theater of the mind in a way that will maximize rhetorical effect (Figure 2). 179 Figure 2. The Fictive Common Ground of the Appeal Letter The diagram in Figure 2 depicts the Appeal Letter frame, which involves a fictive interaction between a ?you,? the constructed addressee, who, of course, shares the values of the organization and has money to donate to the organization, and an ?I,? the constructed letter writer who is often depicted as a good (famous) person, fighting for a good cause. The addressee and the letter writer jointly attend to L1 events and experience L2 effects from perceiving and evaluating those L1 events. The letter writer puts these events on stage for specific sorts of rhetorical purposes, i.e. L4 argumentative strategies. The rhetorical construction of a fictive common ground is one of the most vital rhetorical acts of the appeal letter, as it positions the letter writer as a de facto ally, one who shares thoughts, feelings, and moral value systems with the audience. Crucially, the shared value system (L3) is what causes the writer and addressee to experience similar thoughts and feelings in similar circumstances. The value system that is presupposed within the fictive common ground is necessary for evaluating L1 events in ways that align with, deviate from, and otherwise draw attention to the presupposed value system. Deviations from the presupposed moral order can lead readers to feel upset and confused. They may desire a different course of action and pay special attention to actions and actors in the events depicted that lead them to construct emotion events. Through the deliberate evocation of emotion events, letter writers can bring attention to their cause, propose future actions to address causes, or highlight efforts that are going underfunded. Letter writers can then implore 180 addressees to join the cause, support the fundraising organization, and protect the values that the writer and reader share. It is no wonder why words like please, support, join, and protect?all words that profile L3 relationships between the addressee and the fundraising organization?appear in the list of top keywords in the Appeal Letter Corpus. L3 relationship building is part of the generic framework of the appeal letter genre; while political orientations and cause types may change, the L3 appeal to showing support for a cause and strengthening the fictive relationship between the addressee and the letter writer does not. The way that addressees are instructed to show support is by making a gift using the envelopes or forms enclosed in the mailed appeal letter package. While the cultivation of fictive relationships appears to be a key component of fundraising appeal letters, emotion arousal is often a product of larger discursive strategies. Some of the keywords that profile L3 relationships, like help and please, are metonymically linked to emotion scripts associated with making pleas. Yet the most salient sorts of emotion?anger, sorrow, fear, pity, and care?are no profiled in social relationships between the ideal reader ?you? and the letter writer ?I? in the L3 common ground. Keywords that profile thoughts or feelings in L2?like need, will, generous, and hope?do not touch on appeals to anger or sorrow or fear, either. Through analyzing how L1 keywords that profile objects or agents appear on stage? words like NRA, gun, food, wildlife, wilderness, and people?we can get a better sense of how emotional appeals are constructed. Some keywords operate at multiple levels of the theater of the mind, it is worth noting. For example, generous has 181 subjective emotive weight but involves the activation of the Give frame, which is intersubjective, and thus L3. Likewise, people and animate agents in general typically involves activation of theory of mind, and thus potentially operates at L2 and L3, as well. In fundraising texts, these words are not presented as neutral but as value-laden terms that help to align attitudes and moral orders between the addressee and letter writer. Their emotive significance, furthermore, occurs not because of the word but because within inferential chains that cued by the frames they appear in and the global narratives those frames support when fit together. While keywords like gun and wildlife locate L1 actors and objects, within their respected letters, they single out specific objects presupposed to be of value to the reader, like the horses in the horse tripping letter at the beginning of this chapter. These valued objects are, furthermore, the main beneficiaries of the causes that charity organizations are soliciting money for. Early references to valued objects are often found within vivid narrations and emotionally charged depictions of bad situations. For example, this fear appeal introduces guns as the good things that the reader and letter writer care for and that politicians in Washington are trying to take away: President Barack Obama and the media are conspiring to DESTROY YOUR FREEDOM. And unless you take action now, your right to own a gun will soon be a thing of the past. (ALC.P.37) Here, the valued object of the gun is associated with the valued object of freedom. The rhetor and the audience are both presumed to be in alignment with recognizing the value of both guns and freedom, which are both put in danger by nefarious forces. 182 Furthermore, having access to guns is presupposed to be a condition of maintaining freedom, such that to be against gun access is to be against maintaining freedom. The good actors protect and cherish valued objects and thus are for maintaining freedom. Bad actors want to destroy them, threaten them, or take them away. They are in opposition to freedom. L1 valued object keywords often show up in narrative that reinforce demagogic rhetorical techniques (cf. Roberts-Miller), reducing complex social issues to narratives about good guys vs. bad guys. The placement of valued objects in good and bad situations is the primary way letter writers construct emotional appeals in appeal letters. By focusing attention on keywords like horses or guns or children, letter writers can align themselves with those good forces who protect valued objects, and they can construct depictions of evil villains who want to destroy the same valued objects. Letter writers also use valued objects to construct emotional appeals. Appeals to anger, fear, and pity involve valued objects construed in bad situations, whereas appeals to hope, beauty, and care typically involve valued objects construed in good situations. Figure 3 represents the basic appeals letter frame with the valued object on stage. A dotted line connects the valued object to either a ?good situation? where it is safe or a ?bad situation? where it is in danger. These two situations are also metonymic of the moral orders of good and evil.5 5 It should be noted that these scenes are not intended to represent static moments but dynamic and immersive stories that are unfolding in discourse. In this sense, the resemble the kinds of hero and villain stories in the political discourse analyzed in Chapter 3. 183 Figure 3. Depictions of Good Situations and Bad Situations in L1 In actual usage, appeals present complex situations that involve fluctuations of valence, subtly layering hopeful and fearful attitudes towards cherished agents and values objects. Amnesty International, for example, strings together a series of complex scenes to construct this emotional appeal: Children in every corner of the world deserve an education and the opportunities that come with it. We need to stay powerful advocates for the world?s children?they are our most vulnerable citizens and our greatest hope for the future. Your continued generosity will help. (ALC.H.3) The appeal first implies that not all children have education and opportunities, placing the valued object of children in a bad situation. The next sentence focuses not on children but on the commitment of the ?we? of the common ground, aligning the addressee and the letter writer with the goals of the organization and constructing a feeling of solidarity with the cause of educating children. The letter writer comments 184 that children are ?our most vulnerable citizens and our greatest hope for the future,? two noun phrases that utilize the [our [Adj NP]] grounding construction (see Table 2); this construction highlights aspects of the keyword children that are meant to already be in agreement: we all believe that children need protection and education, and we all want good things for children in the future. The appeal ends by implying that giving money to Amnesty International?which is an act that has already been decided will happen?will lead to the future good situation that ?we? want. The appeal shifts from a bad situation, to situations that focus on the ground, to a request for donations with a strong implication that fulfilling the request will result in a good situation. The reader is also pulled through the bad situation into the good situation through the grounding, providing an experience for the reader (see Figure 4). Figure 4. Moving from Bad Situations in the Present to Good Situations in the Future 185 This keyword analysis above provides is not simply a list of unusually frequent words in the Appeal Letter Corpus but quantitative evidence of the general rhetorical structure of the fundraising appeal letter. Fundraising letters take place in a fictive interaction between a letter writer, who speaks on behalf of an organization, and an addressee, who the actual reader is guided to identify with. The values and beliefs of the addressee have already been presupposed by the letter writer, and these values and beliefs are central to the arguments for supporting an organization?s initiatives to protect some valued object at risk. What happens to valued objects, how the letter writer and addressee position themselves in relation to what is happening to valued objects, and the kinds of proposed future actions letter writers put forth for addressing what is happening to valued objects compose the main argument of any appeal letter. These arguments are, furthermore, often constructed as emotional appeals, guiding audiences to construe the things they are supposed to think favorably about in bad situations, acted on by bad people or bad forces. Without this common ground, emotional appeals may not be as effective, the success of the letter at getting a (real) addressee to identify with the letter?s implied ideal reader depends on the addressee sharing the value system presupposed in the way that the letter is written. As the next section will show, grounding and simulated experience also play a crucial role in constructing engaging emotional appeals as well. The kinds of linguistic techniques that rhetors use to bring valued objects on stage, put them in focus, and detail bad situations often rely on the construction of imagined sensory experiences. 186 4. Crafting Enargeia through Focus and Positioning When valued objects are put on stage, things happen to them. They are attacked, starved, cut down, neglected, or harmed in some way. The appeal letter narratives that depict bad situations will sometimes go into vivid detail, intentionally putting addressees in the uncomfortable position of a helpless witness to appalling scenes in the theaters of their mind. Discussions of bad situations have a way of turning audiences into experiencers. Addressees do not just read about the horrible things that are happening to the things they are presumed to value; they imagine the actions and feel revolted and despondent as they play out in their minds. Through this process, addressees undergo a kind of sensory experience: seeing, feeling, and moving with the discourse. This is because linguistic cues literally activate brain regions involved in sensory-motor processing (cf. Bergen). Recent scholarship in rhetoric and writing studies has sought to understand ?how?sensation figure[s] into rhetorical processes? (Hawhee 2015, 12). Much of the attention so far has been on how sensation gets encoded into language?how writers ??translate? from sense to print,? to use Carole Blair?s phrasing (274). But there is a corollary, too, that deserves attention: how sensation arises from language. Some scholars, like Monica Westin (2017), have argued that attention to classical theories of language?especially Aristotle?s theory of phantasia, where language and sensation co-occur?may reveal ways of understanding how rhetors can make us feel argumentation (Westin 260). Deborah Hawhee (2015) asks us to think more about how rhetoric feels and questions whether rhetorical scholarship has developed methods for understanding the complexities of individual and collective experience 187 that comes from rhetorical processes (12). These questions are especially important for understanding the emotive quality of enargeia, where sensory simulations can make it feel as if interaction with linguistically depicted conceptual content is being experienced. Cognitive linguistics and the cognitive sciences have much it can contribute to the study of rhetoric and sensation. For one, how language constructs mental simulations is a major topic in contemporary cognitive science (see Bergan). The theater of the mind model here is helpful for understanding how language cues sensory responses and what kinds of rhetorical processes those effects work in service of. What is most pertinent to this project is the question of how language can build up sensory simulations for the purpose of crafting rhetorically motivated mental visualizations. In other words, what can the cognitive rhetorical approach to pathos tell us about enargeia? Before analyzing enargeia in appeal letters, it may be worthwhile to review enargeia as it pertains to the linguistic construction of vividness. In his discussion of bringing-before-the-eyes (pro ommat?n poiein), the theoretical precursor to enargeia, Aristotle argued that certain senses of time presented conceptual content in ways more immediate. Aristotle observed that describing things in the process of undergoing rather than having already undergone made could craft vivid mental visualizations (3.10.6)?in essence drawing attention to the ways that tense and aspect affect mental visualization. This ?rhetorical vision,? as Deborah Hawhee (2011) calls it, was thought to be a persuasive strategy insofar as audiences already had rich and intimate connections with the content being conceptualized. Focusing 188 more on lexical and conceptual content, Hawhee notes that the vivid mental visualization of moments of bringing-before-the-eyes ?[occur] on the part of the audience as the result of the ?bundle? of ideas contained in the images [conceptualized]? (155). Thus two important observations about linguistically cued mental visualization can be drawn from Aristotle: firstly, given that what is undergoing is easier to vividly construe than what has undergone, construing a distal imperfective event closer to the time of the common ground can be thought to be more enargeiac; and secondly, given that mental visualizations are crafted from bundles of ideas contained in images, the meanings of mental simulations are generated by audiences, their experiences, and their associations with the content being visualized. As the rhetorical techniques for crafting mental visualizations developed in rhetorical theory, however, the pedagogical advice for constructing mental visualizations also changed to focus less on argument and language and more on the mind of the rhetor. While Aristotle theorized mental visualization as an effect that language has on those who hear it, by the time Quintilian writes on visualization, it is less a technique of language and more of a strategy of invention from emotive imagination (6.2.30-31). This emotive imagination is translated into language by the rhetor; as pseudo-Longinus says, ?you see what you describe and bring it vividly before your audience? (15.1). If the language describes something credible and real enough, it was thought to produce mental visualizations in the mind of the audience as well. It?s not so much that language disappeared as a focus of enargeia; it?s that mental tricks, like emotive imagination, were thought to be more practical 189 pedagogically than studying linguistic techniques. The model for understanding enargeia thus conceived of the technique as working via a transference of thoughts and feelings from the rhetor?s mind to the audience?s mind via the conduit of language, focusing less on the intersubjective aspect that language has in producing simulations in both rhetor and audience (cf. O?Connell 231). While the model for understanding enargeia changed, it was a popular enough technique at the time of the Roman Republic for Quintilian to discuss it in multiple volumes of his Institutio Oratoria, and it had been studied long enough for stylistic and grammatical observations to make it into Quintilian?s review of literature. The technique was, after all, thought to be quite useful, as it could make second-hand information appear as if it was first-hand and thus make arguments appear more credible. The technique was used when a rhetor wanted to make something appear more real and more probable, especially in the cases where a simple statement of facts would not suffice. In doing so, the rhetor makes the imagined case plausible, since audiences are able to ?see? the situation unfolding in their mind?s eye (cf. Quintilian 4.2.31). As such, the technique was thought to be best suited for narrations (narratio) of circumstances and events within arguments (Lausberg ?294). By late antiquity, it is considered a way of making the narration ?delightful? through creating epistemic distance, using hedges of varying epistemic commitment, such as ?perhaps,? ?I think,? and ?probably? (Anonymous Seguerianus 96-98). The inclusion of these kinds of hedging devices, which highlight how language affects interpersonal relationships, in the classical literature on vividness stresses the relevance of thinking 190 of vividness as something produced between conceptualizers?that is, cooperative imaginers?rather than something intrinsic within a static set of words themselves. Important here is that what is described in a passage marked by enargeia is something that is not readily observable?something not present. To bring something before the eyes, then, requires rhetors to direct attention to some kind of fictional mental space, where narrations of events can play out vividly in the simulations of the mind?s eye. The classical techniques for constructing enargeia present a theory that links mental visualization to language in interesting ways. Techniques clustered into two main types of strategies: ones that involve breaking the representation of summative events into sequential events that can each be attended to with differing degrees of focused granularity, and ones that involve moving the subjective viewpoint closer or farther away from some perceivable object or event within a mental simulation. I call the first strategy the granularity strategy and the second strategy the repositioning strategy. Both of these strategies construct vividness by directing attention but do so in different ways. When constructing enargeia through granularity strategies, attention is directed from a summative, zoomed out construal of an event, to the ongoing individual parts of the whole?for example, describing the storming of a city in terms of parts of that storming, like ?flames pouring from the house and temple, ? the crash of falling roofs, ? the wailing of women and children? (Quintilian 8.3.68). Worth noting is the use of the progressive aspect in the representation of the individual parts, which are in agreement with Aristotle?s advice that representing 191 events as undergoing rather than undergone makes what is being represented more enticing to the mind. In Cognitive Grammar (Langacker, 2001), and, by extension, in the theater of the mind model, the English present progressive aspect leads to the construal of an imperfective process. In imperfective processes, the construed event remains stable within the immediate scope of the scene without being completed (Langacker 2013, 147-8). The utterance the dog is jumping over the fence zooms into one particular aspect of the event and focuses in on the action, which is construed as an ongoing and unbounded process. The construal does not indicate whether the action was successful or not, whether the dog got to the other side of the fence. It simply situates the observer within a freeze frame of the action as it is happening. The imperfective construal of processes thus can be thought of as immersive, bringing focused attention on unbounded activities. As will be seen, certain verbs of thought and mental perception, like imagine, are imperfective constructions used in enargeiac moments to situate viewpoints. From this act of situating, more granular mental simulations can be experienced. The strategy of repositioning the subjective viewing scene so that it is proximal to the common ground appear in several techniques for constructing enargeia. For example, Greek and Roman rhetoricians observed that shifting grammatical tense of past events into the present tense?a technique called translatio temporum?brought what was being described vividly before the eyes (Quintilian 192 9.2.41, ps-Longinus 1.25). 6 This technique, as Wallace Chafe (1994) notes, takes imagined perceptions of events that are distal from the speech environment and represents them as if they were proximal in the act of speaking (209). The result is a kind of active representation stemming from the introverted consciousness of the speaker, which, in turn, relies on an active imagination of the described event from the hearer. To provide a quick list, other forms of repositioning that cause enargeia can be found in viewpoint shifting (ps-Longinus 1.26), the use of reported speech (see Lunde, Lausberg ?816-18), the use of deictic adverbs to reference toward or away from the common ground (Lausberg ?815), and the use of specific figures, like nature metaphors (Quintilian 8.3.71, cf. ps-Longinus 1.35) and novel similes (Rhetorica ad Herennium 4.47.60-48.61), both of which displace certain qualities to amplify others. Quintilian also discusses the use of phrases that instruct audiences to retreat into their mind?s eye (9.2.41), a kind of repositioning strategy that will be discussed in the next section. In practice, enargeiac discourse often uses both granularity strategies and repositioning strategies. The two strategies are not mutually exclusive, in other words. A rhetor can reposition the audience to observe a past, future, or fictive event by 6 In particular, rhetoricians believed, like Aristotle, that the present tense was most vivid, so depicting past events in present tense was the prototypical example of translatio temporum. Demetrius seems to be the lone classical theorist who believes that the past tense is more vivid than the present (see On Style 214). It should be noted, however that his example is strange, since he juxtaposes two events in a process, ?dying? vs. ?dead?, and concludes that saying ?I am dead? is more vivid because it is in the past tense. What he calls the past tense is actually the present?the utterance involving a blend of a past event being uttered in the present tense?and what he calls the present is actually the present progressive. 193 making it more proximal to the time of the common ground. Both strategies are grounded in embodied human experience, furthermore. When we want to get a better view of something unclear, we can try to intensely focus on it (the cognitive grounds for granularity strategies), and we can move our physical bodies to get a different viewpoint (the cognitive grounds for repositioning strategies). Often, we do both, moving our bodies and increasing focus and attention to the scene observed. What the enargeiac strategies do is represent how these kinds of embodied activities are guided conceptually by language, particularly when conceptualizing simulations in the mind?s eye. These strategies only work produce enargeia if what is being focused on is of interest to the conceptualizer. As discussed in Chapter 2, there is little advice in the classical literature about what makes for an interesting imaginative scene. The discussion of evaluative frame activation and frame metonymies in the pathetic enthymeme, however, provides an explanation for why certain images move us and why others do not. We are drawn to focus on scenes that strongly align or break with our moral orders, thus making the description of these scenes more likely to cue vivid imagination. As we frames come online in discourse and our minds move us with metonymy, we come to experience the language we are processing. The kind of feeling of immersion that comes with enargeiac movements can furthermore be a useful experience to exploit from the perspective of the rhetorician. A rhetor, who uses language to guide the repositioning and focused attention of the audience, is granted a way to control the mental experiences of audiences. This guiding can be used for larger argumentative purposes. If experiences in enargeiac 194 moments become part of the common ground?for example, if a rhetor references shared feelings, perspectives, or experiences?the rhetor can use enargeia as a kind of engagement strategy that motivates or otherwise influences the relationship between the discourse participants in the common ground. I term this rhetorical usage of simulated experience enargeiac engagement. As will become evident, enargeiac engagement can have further rhetorical functions, not only establishing a relationship between discourse participants and a mental simulation, but also using experiences from a mental simulation as evidence for arguments in the common ground discourse. 5. Imperative Imagine, Irreality, and Enargeaic Engagement One classical technique for constructing enargeia that does not get as much attention in contemporary discussion is the use of words like ?imagine,? which, according to Quintilian, can guide audiences into their mind?s eye to experience what they can?t see with their bodies (9.2.41).7 This technique can be thought as primarily a repositioning strategy, as it draws audiences into an introverted consciousness to simulate the irreal events imagined, but it is also often used in tandem with granularity strategies, too, as it construes an imperfective process and thus situates attention on the imagined scene. Telling someone to imagine some events allows a rhetor to begin a narration regarding things not real, with the intent to zoom in to more specific aspects of that narration, such as actors or events. Asking someone to 7 Technically, Quintilian attributes this technique to Cicero, who he quotes as saying ?Though you cannot see this with your bodily eyes, you can see it with the mind's eye? (9.2.41). This quote does not appear in the extant corpus of Cicero?s writing, however. 195 imagine something?or using another verb of thought or feeling that would be spotlighting in L2 of the theater of the mind?has a way of quickly getting audiences to comply in constructing mental visualizations of scenarios that audiences would be less inclined to simulate on their own. Words like imagine, especially when used in imperative clauses, are called mental space builders in cognitive semantics, and they, as well as the other linguistic components of enargeia, frequently co-occur with moments of vividness within the Appeal Letter Corpus. Within Mental Space Theory (Fauconnier 1985), Conceptual Blending Theory (Fauconnier & Turner 2002), and Cognitive Grammar (Langacker 2013), a mental space builder is defined as a word or phrase that sets up a new mental space in the mind in order to categorize and simulate cognitive phenomena outside of baseline reality. Assuming that coherent consciousness involves shifting attention between the real and the unreal, the probable, or the impossible, language users need some way to signal these kinds of thought processes to their interlocutors. Mental space builders do much of this work by shifting and focusing attention on irreality, which is profiled against a background of a shared sense of reality. For example, in the utterance In the picture, the girl with the blue eyes has green eyes, the prepositional phrase in the picture is a mental space builder that instructs readers to build a mental space to categorize a girl with green eyes, who does not exist in baseline reality. Referents in mental spaces, like the girl with green eyes, often have explicit or inferred correspondences with elements in reality, e.g. the girl with blue eyes. Mental spaces get built up in discourse in ways that help us sort out what is real, what is not real, what is potential, what is probable, and whatever else we may want 196 to imagine. This imaginative process is not trivial, moreover. Mental space congifgurations consist of both realis?our real, factual experiences?and irrealis? what could potentially happen, what is not actually the case, what we can imagine. We organize and keep track of mental spaces in order to help us not only sort out fact from non-fact but more importantly to provide a richness and attention to ongoing evaluations and searches for meaning being done at the level perceptions of our baseline realities. Thinking through the consequences of actions in these irrealis mental spaces can, furthermore, help to reason about elements and events in baseline reality, too, as has been shown in analyses of mental spaces in conditional constructions (see Dancygier and Sweetser 2005). By playing out events in the mind, experiencers can learn from simulated experiences and imagine actions, their consequences, and emergent feelings. The mental space that comes from an imperative imagine mental space builder serves two related purposes. The first is to situate attention on an empty canvas in the mind, on which rhetors can place valued objects on stage in kinds of thought experiments that are designed to result in the emergent experience of feeling; that is, addressees use the elements they are directed to imagine by letter writers in order to simulate viewpoints and experiences which, given the context, lead to thoughts and feelings about those simulations. This aspect of the imperative imagine is obviously enargeiac and is in line with the discussion of the relationship between mental simulation and enargeia (see Webb, Plett). But there is another rhetorical purpose of imperative imagine and other kinds of enargeiac constructions that receives less attention: given that the imperative form is an L3 construction?i.e., a directive from 197 a writer to a reader to perform an L2 action?the imagine space serves an intersubjective function, too. Paradoxically, it requests audiences to imagine scenes that are not real in order to take a particular stance on some aspect of reality. Shifting from mental space back to the common ground allows for enargaeiac engagement, a way for imagined experiences to be meaningful for tending to a relationship in the common ground. Moments of enargeiac engagement transport knowledge and experience from mental spaces into the shared common ground, allowing the rhetor to draw on these imagined phenomena and use them to align beliefs, guide future actions, and further lines of argument. In the case of appeal letters, mental space builders guide audiences to construct mental visualizations, often of bad situations, that motivate moral ways of thinking. The use of the imperative imagine, for example, is used as a way of introducing valued objects on stage in worlds that are presupposed to be unfavorable and undesirable, e.g. Just imagine what four more years of Barack Obama will bring (ALC.P.27). The imperative imagine mental space builder signals an intersubjective L3 move, where a rhetor instructs an audience member to perform the L2 action of generating thoughts. Depending on the context, repositioning a subjective viewpoint on Obama?s future actions could lead to conceptualizations of good situations or bad situations. In the above example, however, the NRA uses Obama is a metonym for the government?s efforts to control firearms. Imagining four more years of Obama, thus, is meant to reposition viewpoints to observe the audience-imagined outcomes of an unfavorable future. The moral order activated is one where gun restrictions are more important than gun rights, and following this moral order is meant to lead 198 addressees to experience negative feelings because they do not want this negative future to occur. Through enargeia, the rhetor thus strengthens addressees? presupposed moral orderings that gun rights are more important than gun control. The thoughts licensed by imperative imagine are often admittedly not grounded in reality or even probable reality. They are meant to motivate feelings and use those feelings for argumentative purposes. Take for example this passage from the Gentle Giants Horse Rescue: Imagine the rightful outrage Americans would express if they knew that a retired Clydesdale who had worked in commercials for Budweiser was going to be sold for slaughter! (Thankfully, this hasn't happened.) (ALC.AC.2) The rhetor in this example asks the audience to imagine the ?rightful? emotive reactions of people that they are meant to identify with when they come to learn about something bad happening to a presupposed valued object. That the letter writer parenthetically notes that what the audience has just simulated has not happened goes to further support the argument that these mental space builders do not merely guide thoughts about the conceptual content within the imagine mental space but also signal an argumentative purpose of constructing the mental space. They are rhetorical moves, argumentative moments of enargeia that engage with the audience to shift and align modes of interpretation by getting audiences to activate frames they may already accept but rarely if ever experience so vividly and compellingly. In the Gentle Giants example, the feelings of outrage toward those who would kill retired horses acts to put the letter writer and addressee in the same emotive state, which helps to establish a rapport with the fundraising organization and align moral views of what 199 should or should not be done to the valued objects in question. In other words, the emotive reactions here serve to establish enargeiac engagement. Mental space builders like imagine frequently occur in constructions with other kinds of mental space builders, such as the negative particles not and no, which guide audiences to understand what aspects of baseline reality transfer into the imagine space and what aspects don?t. In Mental Space Theory, negative particles like not and no are said to construct two mental spaces: one being a negative scene where something is not there, and the other being a positive alternative where something is (see Fauconnier 96-98; Dancygier 2012). The necessity of the positive alternative helps to explain the kinds of pragmatic implications that negating a proposition makes available. When a rhetor negates something, such as the negation of a noun phrase with a nonfinite clausal complement in [T]here is no program to release cougars back into the wild (ALC.AC.13), the utterance is thought to presuppose that there was an expectation in the common ground for there to be such a program. Through negation, the thing that is negated becomes salient, and audiences can infer, given the context, that a speaker is requesting, complaining, or suggesting there should be a program for releasing cougars back into the wild (see Dancygier 2012, 69-70). Understanding the meaning of the negation involves understanding what the absence of that thing means for the speaker, or, to use Hawhee?s words, understanding the ?bundle of ideas? that a speaker may associate with its absence (Hawhee 2011, 155). Notice the effect of the negation in this example from the nonprofit Feeding America: 200 Right here in America, mothers and fathers have to make impossible choices every day? things most of us can't even imagine. Do you pay your rent or do you buy groceries? Do you pay your utilities or make your car payment? Which child doesn't get to eat today? (ALC.H.12, underline mine) Compare the clausal negation Which child doesn?t get to eat today with its logical counterpart Which child gets to eat today. Both utterances describe the same event, yet, by focusing on the withholding of food from children, the construal of the negative utterance clearly provides a more emotive meaning by focusing attention on the absence of something expected to be present. The use of the interrogative form, furthermore, invites audiences to imagine being in the position of making the choice of which child doesn?t get to eat, a kind of repositioning strategy. In particular, audiences are invited to imagine themselves as ?mothers and fathers? making ?impossible choices.? The desire in the baseline mental space of wanting all children to be able to eat is limited by the circumstances in the irrealis mental space of the parents who have to choose which child eats. Feeding America never states why these parents have to make this decision; the audience is made to infer this through their knowledge of why some parents can?t feed all of their children. This elaborate meaning making works through constructing mental spaces that lead to potentially moving thoughts about what it means for some thing to exist, what it means for it not to exist, and how the experience of having that thing taken away may feel for someone. 201 These mental space exercises can serve as the enargeaic bases for larger and more elaborate acts of emotional persuasion that work in service of petitioning addressees to join a cause. Greenpeace instructs their reader, for example, to Imagine our world without animals like whales, tigers, and seals. Unfortunately, in the last four decades we?ve come dangerously close to losing them forever. And, without concentrated ACTION from the people of the world, we might. Greenpeace takes action for the earth?and the people and animals that call it home?every day. But we?re only as strong as the people who stand with us. Will you stand with Greenpeace? (ALC.CA.6, underlines and italics mine) In the example, mental spaces work together in order to guide imagination of a world where certain animals have died out and to provide exigence to act now in order to stop this imagine world from becoming reality. In the first sentence, two mental spaces are activated: an irrealis space, built from the imperative imagine, that situates attention on an imagined world; and a negative space, built from the space builder without, that puts constraints on what exists in the imagined world. In the second sentence, the prepositional phrase in the last four decades, activates a past realis mental space, which prompts simulations of animals dying in baseline reality to match the lack of animals in the irrealis space of the first sentence. In the third sentence, the mental space builder without behaves differently from in the first sentence; in the first sentence, it restricted what should be imagined for the irrealis mental space, but in the third sentence, it negates the set of actions needed to prevent 202 the unfavorable immediate future, which is a potential future that is cued by the space builder might. In three sentences, the rhetor guides the addressee to imagine an unfavorable future where certain animals do not exist, contrast it with present reality where those animals are nearly died off, a past reality where those animals slowly died off, and an immediate future where actions are necessary to protect animals. The flurry of mental space shifting leads audiences to imagine different viewpoints on the same topic: living in a world without certain animals, watching the animals die out, knowing the animals are dying, and acting to prevent them from dying. Imagining a world without whales, tigers, and seals, then, acts as a vivid contrast to baseline reality and helps to reason about past observations and motivate potential future actions. This intentional act of invoking loss through enargeia is not merely meant to invoke feelings of sadness but to set up a larger argument for joining in Greenpeace?s cause: if you do not want this imagined world that causes you sadness to become reality?and mind you that it has almost become reality?then join Greenpeace and take action. The mental space builder here provides a way to set up an emotional appeal for the larger rhetorical purpose of petitioning audiences to join a cause. First and second person grounding constructions (our, us, you) bring attention back to the common ground to align stances over what was just simulated. Greenpeace?s argument, thus, uses the experiences of mental simulation to engage with the audience, crafting enargeiac engagement. This relationship between enargeiac mental visualization and engagement in the fictive common ground helps to explain one way that enargeia is used 203 argumentatively. Effectively, readers of appeal letters can imagine experiences in a mental space that rhetors can draw on as evidence for arguments, once that experience has been grounded in the common ground. This strategy is used quite effectively by the Sierra Club, for example, when it begins its appeal letter through a series of anaphoric imperative imagines. Through the course of construing the events and experiences that take place in the imagined space, addressees come to shift their identity and empathize with the experience of nonhumans: Imagine that, little by little, your home was taken away from you. The forests and mountains where you once roamed freely disappeared, replaced by roads and concrete buildings. What if politicians in suits, someplace far away, decided your fate... decided that you, your family, your friends, and neighbors had become a nuisance a menace?to those who had invaded your home? And so now, you must die. Imagine these politicians rallying for your slaughter... ignoring what science has told them, encouraging citizens to hunt you down and kill you. Imagine your family under attack. Defenseless, with nowhere left to hide, you must dodge bullets from the ground and sky, just to find food for yourself and your young children. Imagine that, in one of these public hunts, you watched your offspring die. Then you will know the terror that wolves face every day... and why we so desperately need your help. After all, you and your fellow humans are the only ones who can save us. Our fate is in your hands. So I hope you will answer this cry for help. You are our only hope. And time is running out... (ALC.CA.11) The argument that the Sierra Club sets up is one that is dependent on simulated experiences. It asks the addressee to imagine X, and by imagining it, then you will 204 know Y. Presumably, the knowledge that is gained in the then you will know clause stems from what is imagined. There are some interesting mental movements that must work in tandem for the simulated experience to provide evidence for believing what is presumed to be learned, though. For instance, the ?you? of the imagine mental space is not the addressee, and very little of the addressee?s identity is meant to be projected into the mental space. The identity of the ?you? is ascertained by the simulations within the imagine space: in the mental space, you live outside, you have a family, you are hated by politicians, you are being hunted, and so on. The experience of the ?you? is also made immersive through granularity strategies that stem from the construal of imperfective processes through simple present tense events (you must die, your family under attack) and present progressive events (politicians rallying for your slaughter, ignoring science, encouraging citizens to hunt you), all highly emotive scenes. When the discourse moves the addressee out of the mental space simulation back into the common ground?i.e., once the discourse arrives at the then you will know clause? the mental experiences and the identity that was crafted in the mental space transfers into the common ground as knowledge available for making inferences. When ancient rhetorical theorists discuss enargeia as providing evidence, this process, which is dependent on enargeaic engagement, is possibly what they are referring to. The process of enargeiac engagement is complicated further in the Sierra Club?s argument once the reader comes to the then you will know clause. What the mental simulations from the imagine space are supposed to provide evidence of is the ?terror that wolves face every day...? Construing the events of the imagine mental 205 space can certainly lead to the categorization of those events as terror inducing; it construes the scene as bad, a something that an experiencer very much so does not want to experience, and as a threat to the experiencer?s life or the lives of those that the experiencer cares for. These are all part of the emotion script for the English emotion word terror. However, the then you will know clause continues: ?and that is why we so desperately need your help.? The ?we? here is unusual. At first, it seems to reference the Sierra Club, who one could infer is working to protect wolves in the wild. But as the letter continues, reference to the addressee as a human (?you and your fellow humans?) leads to a backwards projected interpretation of the ?we? as wolves. The common ground being built here is between a ?you? addressee and a wolf letter writer. Again, as with earlier examples, first person pronominal grounding constructions work to construct a fictive common ground necessary for making inferences. In the imagine X, then you will know Y argument, X is imagined fearful experiences, and Y is the first-hand experience of being a wolf. Given that there are no wolves that can dictate their first-hand experiences, the registering of the wolf letter writer in the fictive common ground acts as a source of verification that, indeed, X can lead to knowledge of Y: imagining the terror of a wolf can lead to knowledge a wolf?s experiences.8 The result of moving the experience of being hunted from the imagine space back into the common ground?the enargeiac engagement?constructs a kind of shared grounding of experience between the addressee and the letter writer. Of 8 I really like the idea that this wolf works for Greenpeace. Imagine the interview. 206 course, the letter writer needs to set up this engagement, and the set up itself requires quite a lot of work for its rhetorical payoff. Given that wolves can?t write letters, the letter sets up a fictive common ground where the letter writer is a wolf, a further removal from the experiences of reality. Zooming out from the irrealis L1 space, where ?you? are experiencing a series of bad situations, to an interpersonal L3 space (that is not explicitly marked as irrealis) moves the audience from the body of a wolf in their imagination to the body of a human talking to a wolf in a fictive common ground. The L2 emotions of terror that the human-identifying-as-wolf experience in imagine space become the feelings that the wolf-letter-writer have experienced in the fictive common ground, bonding addressee and letter writer through the shared experience of being a hunted wolf. Though the conversation is obviously pretend, it is meant to be taken seriously. The ethopoeia of the wolf narrator here, which is a fiction that is not explicitly embedded in a mental space besides the baseline reality shared by discourse participants, has a further enargeaic function: it makes the audience imagine that they are being addressed by a real wolf, a striking image if there ever was one. The effect of all of this mental space building and movement is not only to get audiences to consider the experiences of wolves, not only to imagine the sensations and construed emotions that come from being hunted, but to empathize with wolves as members of the common ground, as people who can guide joint attention and who can make arguments. The real reader may (and should) very well know that wolves are not people and wolves cannot make arguments, but the argument gains persuasive 207 force from the letter-writing wolf ethopoeia, a blend of the conventional letter writer and a wolf who experiences the subject matter first-hand.9 The enargeaic engagement allows for the letter writer to align a moral order through their shared fictive wolf experiences, and this moral order provides the argumentative reason for the plea for help that the wolf makes. From this engagement, the addressee can trust that their simulated experience is a faithful representation of the experience of wolves, strengthening the final act of solicitation. If the addressee knows what it?s like to be hunted, then they know why they need to give money to help the wolf letter writer. The argument is thus a moral one that presupposes that wolves are valued objects and that it is the job of humans to protect wolves. Given that the letter writer is also a wolf, the final plea for help is made all the more urgent. If kind humans (who know what it is like to be hunted, etc.) have a responsibility to help wolves, and if the human addressee is in a conversation with a wolf letter writer who is asking for help, then it is the moral duty of the human to help the wolf. As is evident by the Sierra Club example, embedding mental spaces and shifting identity between mental spaces can be an effective rhetorical strategy for building enargeia. This strategy involves linguistically setting up a mental space in 9 Note that ethopoeia has a long history of being thought of as an emotive and imaginative genre as well. Pseudo-Hermogenes specifically notes the ?image-making? (eidolopoiia) effect of attributing words to those who cannot speak (20), and he notes the role of pathos in the genre, too (21). As for ethopoeia using animals, Deborah Hawhee?s (2016) Rhetoric in Tooth and Claw: Animals, Language, Sensation also presents a long history of rhetoricians using animals as ethical mouthpieces, from Aesop to Erasmus. 208 which non-real experiences can be simulated and then returned to the common ground with a shared sense of experience that can be made available to be acknowledged by the discourse participants in the common ground. Figure 5 illustrates how these mental spaces are set up in the Sierra Club example. Figure 5. Sierra Club Wolf Experience Example What is of importance here is the ways that experiences transfer across mental spaces, from the imagined space to the fictive common ground and then finally as the grounds for an argument. The audience is made to understand the fear that wolves experience because they have themselves simulated the fearful events. These grounding experiences, motivated by an imagined ?you? in the simulated experiences within the construed mental space, provide a more embodied way of understanding 209 the valued object, which can further allow for a stronger relationship to become established between the interlocutors in the fictive common ground. The letter writer then uses the shared experience to make an argument: if you know this experience, then you must help. The result is an engagement in the common ground, leading to greater appreciation of the relationship addressees have with their letter writers and, by extension, with soliciting fundraising organizations. 6. The Identification Functions of Vividness in the Historical Present Enargeia serves an intersubjective function that has often been underemphasized in rhetorical literature, as it is this intersubjective function that allows for vivid mental visualizations to be repackaged as evidence for a rhetor?s argument. They can bring addressees to imagine terrible futures, or they can make them simulate impossible fictional experiences, and these imagined sensations can further function as grounds for other arguments. Repositioning and granularity strategies work together to help construct enargeia, but enargeia?s rhetorical force comes from the frames it metonymically activates. The audience constructs the evidence and feels the feelings, but the rhetor directs the mental visualization and does so for a purpose. What makes enargeia work is, more importantly, not a rhetor?s intention but a rhetor?s strategy; selective sensorial and viewpointed frame activations engage the attention on some evaluated scene. Framing, in a sense, predetermines the evaluative inferences. 210 Seeing the rhetor?s purpose can be difficult, since attention flows with the discourse, moving in and out of mental spaces, focusing and relaxing attention on events and objects, inviting witnesses to feel the effects of their directing. The question is how audiences are able to make so much sense from noisy data and cohesionless discourse, and the answer to this is through strategic frame activation and reliance on common ground beliefs. Attending to the work that enargeia does? presupposing values, aligning moral orders, and engaging relationships?reveals ways imagination, guided by frame activation, works with reason and emotion to produce persuasive and immersive effects. Rhetors can direct frame activation, but audiences must persuade themselves, and seeing is believing. The imperative imagine construction is one but of many linguistic strategies for constructing enargeia. Other strategies provide different ways of bringing focus to events in the mind?s eye. Many of the repositioning strategies discussed in classical rhetorical theory work not by transferring the mind into an irrealis mental space but by displacing space, time, or the body away from or toward the common ground. In other words, these strategies use L1 descriptions of events for L3 effects, shifting and repositioning viewpoint in the common ground for strategic purposes. While the literature on classical repositioning strategies (correctly) emphasizes the effect of bringing things before the eyes, rhetorical strategies like temporal shifting (translatio temporum) also tend to have a rhetorical function that is relative to the argument for which they are recruited. Within the appeal letters, repositioning strategies often prompt audiences to entertain viewpoints that are meant to be difficult to watch. By repositioning a viewpoint to move subjective 211 conceptualization into another physical, imaginative, or temporal location, rhetors can make distal scenes present in the here and now. One well-studied repositioning strategy for constructing enargeia along these lines is the use of the historical present. It has become commonplace to understand the function of the historical present as a construction that ?puts before the eyes,? even in contemporary linguistics. The famous grammarian of the English language Otto Jespersen (1929) echoes the common ?as if before the eyes? wording of enargeia when he writes that, ?[i]n the historic present, the speaker, as it were, forgets all about time and imagines, or remembers, what is recounting, as vividly as if it were now present before his eyes? (19). Quirk, Greenbaum, Leech, and Svartvik, in their Comprehensive Grammar of the English Language also echo the wording associated with enargeia when they write ?The historic present describes the past as if it is happening now: it conveys something of the dramatic immediacy of an eye-witness account? (181). As mentioned before, Wallace Chafe (1994) also discusses the historic present as a mode of representing memories or acts of imagination through the introverted consciousness of a speaker. Construing the historical present represents the distal memory proximal to the here and now of the speech event, often drawing on proximal deictic markers like now (Chafe 209). This representation of ordinary consciousness is used for kinds of ?special effects? in moments that are considered to be of importance, making the scene stand out (209-210).10 10 Heinrich Lausberg, surveying the classical techniques for constructing enargeia, also notes the use of deictic adverbs in moments of enargeia (?815), though he does not recognize this displacement strategy as part of the historical present. 212 Rhetorically, the historical present can do more than just make a scene stand out. In the Appeals Letter Corpus, moments of discourse in the historical present serve an indirect argumentative function. Consider the function of this passage, written in the historical present, from Four Corners Equine Rescue: Now the hungry and dehydrated horse stands waiting in a cramped chute next to the circular track. Suddenly a loud horn goes off, the chute door flies open, and a cowboy shocks the horse with a "Hot Shot" electric prod sending 5,000 volts screaming through its body. Seconds later the cowboy lassos the back legs causing the terrified horse to crash head over heels into the ground. Then the horse gets up and gallops in the opposite direction only to be lassoed and tripped again...and again?and again. By now rope burns have carved away inches of her skin and her blood stains the rope and dirt. (ALC.CA.1) The passage above is inserted in the appeal letter between general discussions of what cowboys in illegal New Mexican rodeos habitually do, both of which are also in present tense but are not in the historical present. When the discourse shifts into the historical present, the tense does not shift, but the use of the deictic adverb now signals the translatio temporum. The reader construes the distal scene as if it were happening now, and adverbials like suddenly, seconds later, then, and by now help the reader to organize the time signatures of the global simulation, as if the bad situation being visualized were dynamically playing out event by event with close 213 attention to the timing of the staging. The shifting of perfective and imperfective construals of scenes also aids in the immersive quality of the mental simulation. The situation begins with the imperfective process (the?horse stands) and, with the suddenly, a series of perfective processes occur. That the scene is situated imperfectively may draw attention to the horse, who undergoes the cowboy?s actions. And given that the horse is the valued object, this imperfective/perfective shift may amplify the appraisal of the bad situation being construed, providing an establishing shot from which the scene can play out. Like most instances of enargeia, the bad situation represented in the historical present makes clear to the addressee what attitude they should have about the scene they are watching in their mind?s eye. Granularity strategies of L1 phenomena, in particular the use of (perfective) event-evoking past participle attributive adjective [Ved [NP]] constructions (dehydrated horse, cramped chute, terrified horse) and the use of (imperfective) progressive participle [Ving] constructions (waiting, sending, screaming, causing), construe the valued object with reference to two different temporal relations. The instances of [Ved[NP]] construe horses as already in a state of physical distress, the actions that put them in these states having already happened. The actions depicted by the instances of [Ving], on the other hand, depict the horses as undergoing physical distress. These construals are meant to bring attention to the discomfort of the situation. They signal negatively valenced core affective experiences, ranging in levels of activation, from the stiffness of waiting in a chute to the excited reaction to the 5,000 volts of screaming current jolting through the horse?s body. The horse in this scene may be terrified of the cowboy, but the audience is not 214 meant to be; looking at the scene in the mind?s eye, the audience is disgusted, angered, aghast, indignant, empathetic, but not terrified. The horses feel one way, but the audience is meant to feel another way, focusing on the bad situation unfolding without identifying with the horse. The rhetorical effect of this passage is the repositioning of the addressee from the common ground into the stands of a wholly imagined illegal New Mexican rodeo. The historical present situation orients the reader in a way so that their viewpoint affords focused attention on the horse chute and circular track, allowing for a more immersive construal of the interaction between the horses and cowboys on stage. And watching the rodeo in the theater of the mind has its advantages. For one, observers in the theater of the mind can zoom in close, and with full attention, to the suffering of the horses, in a way that would be difficult when physically observing them. In the theater of the mind, readers can zoom in on the trembling horses, can imagine the faces of the horses when they are shocked, and can otherwise complete the depicted events through imaginatively simulating the scene. To use Chafe?s term, this is part of the ?special effect? of the historical present: it allows for attention to more detail through the use of imagination. While the rhetorical effect of enargeia in this passage is to displace audiences, the rhetorical function I believe is to motivate disidentification with other spectators. To get a perspective like the one imagined, a reader would have to be a real spectator, and real spectators go to illegal New Mexican rodeos for entertainment. Given that horses are valued objects, part of the presupposed moral order, both the addressee and letter writer are resolute in their disgust toward the abusive actions they jointly attend 215 to on stage, and thus they are indignant toward the spectators who cheer on the cowboys. The addressee dissociates themself from identifying as the kind of person who would take pleasure in watching an illegal New Mexican rodeo, and the letter writer indirectly guides the addressee to construct this disidentification via the enargeia. Repositioning strategies like the use of the historical present, then, can serve rhetorical purposes beyond simply making events stand out or making events feel more immersive. They can shape identity, and, by doing so, they can aid in larger argumentative goals, for example, to shape emotive reactions to people, events, or organizations and to guide future actions when it comes to these people, events, or organizations. As is evident, the function of enargeia is not simply to make objects and events present in the mind?s eye but to shape how people continue to view these things. They have a more formative role in the rhetorical shaping of alignments in the common ground and evaluations of experiences in a shared social world. 7. How Vivid is Vivid? So far, I have presented a baseline argumentative frame for how emotional appeals are constructed in appeal letters, paying special attention to the ways that objects with presupposed value are introduced into bad and good situations on stage and how presuppositions about the relationship between the fictive reader and letter writer encourage specific kinds of identification between the real reader and the ideal addressee. Emotional appeals in fundraising appeal letters, furthermore, depends on the strategic use of words and images to craft enargeia that makes imagining bad 216 situations immersive and moving. These enargeiac strategies have a function in the common ground, too; they help to align moral orders and build solidarity between the letter writer and addressee. While all strategies work by setting up a viewpoint in a mental space in the theater of the mind, some viewpoints make an evaluative scene feel more immediate, while other viewpoints position the viewer as more removed and distant. There is a question worth addressing, however. How vivid is vivid? This question is pertinent because it?s easy to observe different degrees of vividness in linguistically prompted mental visualization. Take, for example, the imperative imagine construction, in which an L3 imperative cues the construction of an L2 fictive mental space, where L1 situations are simulated. As we have seen, these moves in the theater of the mind serve L4 argumentative purposes in the common ground, outside of the mental space constructed. When simulating events in the irrealis mental space, presuppositions about the value of the specific conceptual elements invite a moral evaluation. The simulated experiences themselves can be brought back into the (fictive) common ground for different rhetorical functions, such as emphasizing shared stances and feelings toward the experiences?constructing enargeaic engagement?as well as using simulations as a kind of ?evidence? (evidentia) for crafting arguments. Sometimes, what is imagined in an irrealis mental space and in the fictive common ground can be striking, like imagining yourself being hunted or imagining being talked to by a wolf. Other times, what is imagined is less vivid, like imagining a world without certain animals; unless the reader really loves 217 whales, tigers, and seals, imagining a world without them may not be so shocking.11 Without compelling reason to focus attention, to do the mental work of vividly simulating and imaginatively investigating that simulation, certain enargeiac strategies may yield less useful mental visualizations for emotional persuasion. The imperative imagine strategy is only but one of a myriad of strategies that letter writers use to craft vividness in appeal letters, and many strategies can be used together, too. The technique of crafting translatio temporum in appeal letters, for example, includes a range of linguistic strategies, from the use of the historical present above, which stages temporally distal events using the present tense and proximally deictic adverbials (e.g., Imagine, I am standing there watching the horses, and now they begin to shake), to the use of counterfactual conditionals (e.g., Imagine if you had been there when it happened), which prompt simulation of how events would play out if some past action occurred differently. Both of these strategies can work in tandem with the imperative imagine strategy to achieve the desired effects of the rhetorical situation. Recasting the time of events in the present can cause an immediacy that could intensify the shock of witnessing a bad situation, as Four 11 A related topic, many current strategies for combatting climate change often make emotional appeals by asking people to imagine a world without polar bears. This argument tends to fall flat on people who don?t have a special connection to polar bears, have never seen one in person, and don?t understand the relevance of keeping them alive. Anna Westerstahl Stenport and Richard S. Vachula (2017) have noted that appeals to emotion through depictions of polar bears has led to desensitization, as audiences become aware of the link between polar bear suffering and political and ideological goals to combat climate change. It?s worth noting that espousing a political ideology does not preclude a genuine desire to eliminate suffering, thus meriting the use of emotional appeals, yet skeptics have difficulty being persuaded, nevertheless. 218 Corners Equine Rescue does when they use the historical present repositioning strategy to bring enargeia to an event where cowboys prod, trip, and abuse dehydrated horses for entertainment purposes in an illegal New Mexican rodeo (ALC.CA.1). Likewise, counterfactual constructions can allow for events to imaginatively play out differently. In an appeal letter by the Arizona-based animal sanctuary Keepers of the Wild, a counterfactual space is set up where the addressee is cast in the role of the letter writer when she came across a mangy and abused lion named Sabu (ALC.CA.9). In each case, the transport of time and viewpoint aids in constructing an emotive effect by getting readers to construct and experience complex scenes outside of what is available in the here and now. These repositioning strategies provide full access to simulated events, gifting the reader with the ability to imagine the facial contortions of a shaking horse as it is being prodded by a smirking cowboy or the sound of a whimpering lion, struggling to meet your gaze. These larger and more complex strategies tend to get much of the attention when it comes to enargeia, but they are usually the result of smaller language patterns and cognitive phenomena. For example, the historical present, as has been shown, makes frequent use of adverbials that profile a proximal deictic relation to the common ground and thus motivate a specific viewpoint. The progressive participle [Ving] and the generic present tense are two other constructions that are generally ignored when it comes to linguistically describing the ways that enargeia is composed, even though they construe a scene imperfectively. Construing imperfective processes focuses attention on a narrow aspect of an event, one where the action is undergoing. This kind of construal has an immersive quality to it, but it 219 is perhaps so frequent that its ability to construct vividness is missed. Yet, as has been shown in examples of both the imperative imagine and the historical present, these kinds of construals play important roles in the larger and more complex enargeiac phenomena. Beginning to describe the linguistic grounding for vividness is a step forward in understanding not only the linguistic construction of enargeia but the ways different enargeiac strategies are used argumentatively. The existing research on the construction of vividness in fundraising letters has yet to provide a theoretical account for how language cues mental simulation. The Indiana Center for Intercultural Communication (ICIC), as well as other scholars, like Douglas Biber, have analyzed enargeia in a corpus of fundraising letters that looks much like the Appeals Letter Corpus. In two separate studies, researchers coded the ICIC corpus for appeals to logos, ethos, and pathos, providing definitions for each code applied. Appeals to pathos, for example, were broken into four different kinds of rhetorical moves: assertions of empathy with audience, discussion of what the audience values, the use of vivid pictures (which equates to enargeia), and the use of charged language (Connor & Gladkov, 259; Biber, Connor, Upton, Anthony, & Gladkov, 125). The results found that over a quarter (27.67%) of appeal letters were made up of appeals to pathos, at least in accordance to how the researchers were defining pathos.12 When analyzing the specific rhetorical moves used to arouse pathos, however, the ICIC 12 My own approach to defining pathos is detailed in Chapter 3 of this dissertation. As an argumentative strategy, pathos is the metonymic activation of the whole of an emotion script via the categorization of some construed event in terms of part of an emotion script. 220 researchers were unable to characterize the linguistic strategies involved in crafting vivid pictures, because, by their analysis, enargeia appeared only in about 5.5% of the corpus, making a keyword analysis ineffective (Biber, Connor, Upton, Anthony, & Gladkov, 135). Interestingly, some rhetorical moves that are coded for in the analysis of logos?moves like the use of narrative examples, comparison, contrast, (intensification by) degree, and authority, five of the twelve logos codes?are also conducive to constructing enargeia. Because enargeia appears in the narration, narrative examples can be a source of enargeia, and because comparison and contrast focus attention to what is present or absent, alike or dissimilar, these, too, can have enargeiac effects. Intensification by degree imbues an assertion with emphasis and can invite audiences to pay closer attention to what is being asserted, potentially also leading to an enargeiac construal. And if these are all aspects of logos, I am led to wonder if the ICIC researchers were mischaracterizing how pathos and especially enargeia are constructed. Just like how pathos doesn?t announce its own effect, enargeia doesn?t either. This can make enargeia and similar pathos strategies hard to search for using corpus methods. Guided by simulated experiences, emotional appeals depend on audiences to figure out how they ought to feel in a given situation. This last point is further corroborated by the frequency data of emotion words in the Appeal Letter Corpus. An analysis of emotion words shows that direct expression of emotion labels like anger and sadness as well as associated words (e.g., angry, sad) are not very frequent. Figure 6 shows a result of emotion word cluster frequencies. As the figure shows, aside from a few words?i.e., care, love, suffering, 221 joy, and pride?emotion words are scarce. Care is the only word among these that is key (see Table 1). 300 239 225 150 107 75 48 45 37 29 28 24 24 14 12 8 8 7 7 5 3 3 2 3 2 0 Care Love Suffering Joy Pride Loneliness Hatred Gratitude Fear Happiness Sorrow Shame Terror Depression Surprise Grace Regret Frustration Disgust Anger Disgust Figure 6. Frequency of Emotion Words in the Appeal Letter Corpus Many of the letters in the Appeal Letter Corpus are intended to make audiences angry or sad so that they will want to act on fixing the kinds of perceived injustices depicted in the bad scenarios that befall valued objects, yet words like angry and sad are rarely used in appeal letters. Of these usages of anger, only one usage correlated to an intentional expression of anger: ?Quite simply, I was angry. I was depressed,? says Sandra. ?But when I was ready to give up, the people at the Lighthouse would not? (ALC.H.M.9). The usage appears next to an expression of another emotion concept, depression, and neither anger nor depression are the intended emotional response from the utterance. Instead, references to anger and depression are used to depict a bad emotion state 222 Sandra was in before Lighthouse, an organization that helps children and adults struggling with vision loss, began helping her. The appeal, then, asks readers to imagine someone who moves from a bad emotion state to a good one, with the effortful help of the fundraising organization. This is far from an appeal to anger. It is arguably a construction of enargeia. The passage uses direct speech to let audiences hear from Sandra herself about how an organization lifted her spirits, thus providing a kind of first-hand testimonial that can be used as evidence for reasoning about the ethics and the efficacy of the organization. Contemporary investigations into the pragmatic functions of enargeia have paid special attention to reported speech for this reason (see Lunde). Often to get a sense of the rhetorical effect of enargeia requires audiences to attend to the range of experiences that mental simulations present. At its core, enargeia is a strategy that is closely tied to the pathetic enthymeme. Like the pathetic enthymeme, which operates via frame metonymy, enargeia depends on sparse moments in a discourse cuing vivid mental experiences, which are enriched through their relations shared value systems and alignment in the common ground. In a sense, enargeia can be thought of as a special case of frame metonymy, one where granularity strategies and reposition strategies position and focus viewpoints on evaluative scenes to activate emotion scripts. They constitute a more situated approach to activating emotion scripts than most cases of the pathetic enthymeme, but the two can work in tandem: the frames that are activated to construe valuative events in the pathetic enthymeme are experienced vividly through the situated viewpoints and sense of audience-rhetor engagement that enargeia provides. 223 If scholarship continues to treat enargeia only as a visual phenomenon?as, for example, the ICIC researchers suggest when they title it the strategy of making ?vivid pictures??many of the other rhetorically persuasive aspects of vividness get lost, such as experiential vividness and sensational vividness that do not rely on fine- grained details and granularity strategies. Moreover, the purpose of crafting vividness is less evident if all that is being sought out is the construction of a detailed mental image. As has been shown, the experiential component of enargeia can work to strengthen the bonds between rhetors and their audiences, motivate specific kinds of identification, and even be used as evidence in an argument. Rather than narrowing the scope of what counts as enargeiac by focusing only on vivid images, more work needs to be done to capture the ways that rhetors invite audiences to simulate experiences through regular linguistic constructions. The question shouldn?t be ?how vivid is this,? but ?is this vivid,? followed by, ?what does this vividness do?? By focusing on the enargeiac quality imbued in language, we will come to a better understanding of the relationship between language, experience, visualization, and persuasion. 224 Chapter 5: Defensive Stance in the Self-Assessments of Student Reflective Writing 1. Introduction: Rhetoric in Reflective Writing Reflective writing is a common and well-regarded educational practice used to increase students? self-critical awareness?and metacognitive awareness more generally?by getting them to think about their actions, consider their consequences, and assess their abilities. While reflective writing has been a topic of study since John Dewey?s (1910) How We Think, reflection research is currently undergoing a renaissance. Kathleen Blake Yancey (2016a) hails what she calls the ?third generation in reflection? as a pedagogical endeavor that moves beyond seeing reflection as a moment of the writing process or as a means of externalizing reflection for assessment purposes, the topics of the first two generations (9). While the first generation of reflection focused closely on personal reflective practices, and the second on classroom practices for assessment purposes, the third generation of reflection focuses on the metacognitive, rhetorical, and epistemological functions of the genre of reflective writing (Yancye 2016b, 303). Current scholarship, like that by Elizabeth G. Allan and Dana Driscoll (2013), Naomi Silver (2013), Anne Beaufort (2017), and Kathleen Blake Yancey (1998), blurs the boundaries between reflection and metacognition, understanding reflective writing as a kind of intentional self- monitoring that helps to actively connect ideas and develop understanding through ?epistemological practice? (Yancey 2016b, 302) This focus on the metacognitive and epistemological quality of reflection draws attention to the streams of thoughts and feelings that student writers experience 225 when writing; however, it also tends to forget that reflective writing is, first and foremost, writing. As Elizabeth Allan and Dana Driscoll observe, reflective writing is ?too often treated as a transparent and universal skill? (48) or a means of directly accessing unmediated thoughts and feelings. As a genre, reflective writing necessitates the use of certain recurrent rhetorical moves, typified by linguistic and stylistic choices, that often go overlooked and underexamined. Recent criticisms of the genre by Jeff Sommers (2017), Rachel Ihara (2014), and Michael Neal (2017) have sought to bring attention to the ways that student writers use reflective writing for their own goals and how those may differ from the way that instructors read and understand that writing. Sommers (2017), for example, points to the potential conflicts between students? and instructors? objectives: while instructors want to see what students are understanding, students? motivations are not altogether transparent. Rachel Ihara (2014) has drawn attention to students? compulsions to use reflective writing to ?construct a narrative of growth? (266), for example. This ?progress narrative,? as Lindenman, Camper, Dunne-Jacoby, and Enoch (2018) call it, ?talks the talk? but is rarely backed by evidence (589). While the genre invites reflective writers to assess themselves, the rhetorical component of reflection can compel them to perform to their readers, and this move has arguably more of a persuasive than metacognitive thrust. This chapter focuses on the ways that students use the space of reflective writing to shape how their instructors feel?feel about them, about their writing, and about their assessments of their writing. In particular, it focuses on how emotional appeals to instructors are formed from a narrow set of rhetorical moves that occur in 226 acts of self-assessment in reflective writing. While teacher/student emotional appeals have been studied as acts of fallacious reasoning through appealing to pity (e.g., Castell; Walton 1997, 15-20), the kinds of appeals that are analyzed in this study of student reflective self-assessments are less direct, more nuanced, and, as a result, more in need of careful analysis. One move that will be analyzed in detail is the epistemic stance marker I think, which mitigates the expression of certainty made by the student in acts of self-assessment. While mitigation is not often thought of as an emotional appeal, the way that this conventionalized mitigative move functions rhetorically, I will argue, is intentionally intended to have an effect on readers? feelings. Through this mitigative move, students form a kind of stance that can influence the ways that their audiences evaluate their actions. In doing so, they their audience become more prone ?to differ in their judgments,? to quote what Aristotle writes about the emotional appeal (Aristotle Rhetoric, 2.2.8). What this chapter contributes to the third generation of reflection is an attention to the rhetorical and emotive aspects of reflective writing. Through corpus and cognitive rhetorical analysis of ten students in a first-year writing classroom, I show how mitigated expressions of self-assessment function argumentatively and emotively. While this analysis may not reveal much with regards to the purely metacognitive qualities of reflection, it draws attention to the rhetorical aspect of reflective writing, focusing on subtle linguistic moves and their potential emotionally persuasive effects. In doing so, my aim is to better understand the rhetorical function of mitigation in reflective writing, the rhetorical contexts mitigation indexes, and the pedagogical implications that analysis of mitigation suggests. 227 2. Corpus Construction and Methods of Analysis In order to analyze recurring mitigative moves in reflective writing, this chapter uses a top-town mixed methods approach, combining corpus methods with qualitative analysis. The Corpus of Reflective Writing gathers the reflective writing of ten students across one semester. 1 The first-year writing curriculum at the University of Maryland has students write six assignments throughout the semester: a summary assignment, an inquiry assignment, a rhetorical analysis, a digital writing assignment, and an argumentative position paper, ultimately culminating in a final revision and reflection assignment that asks students to reflect on what they have learned and to use this reflection to substantively revise one prior assignment. To help students develop their reflective skills, short ungraded reflections are assigned as a part of each of the five assignments leading up to the revision and reflection assignment. The idea of using reflection to guide revision is common in ?reflection-in-action? exercises, or reflective practices that focus attention to the choices, outcomes, and possibilities that writers encounter in writing (Sch?n 1983, Yancey 1998). In addition to reflective writing accompanying assignments, students were asked to compose writing autobiographies, preliminary reflective writing at the start 1The data was taken from the class of Justin Lohr, an instructor of first year composition at the University of Maryland. This data was made available as the result of an IRB study conducted by myself, Jessica Enoch, Elizabeth Miller, and Justin Lohr, at the University of Maryland. The study is currently being written into an article, titled ?Thinking About Feeling: The Role of Emotion in Reflective Writing.? 228 of the semester that asks students to reflect generally on their relationship to writing. To avoid students constructing ?progress narratives? (Lindenman, Camper, Dunne Jacoby, and Enoch 2018; Ihara 2014) for their final reflection assignments, the curriculum also requires students to produce an ungraded progress narrative as a separate assignment from the final reflection. The curricular change follows the observation, made by Heather Lindenman, Martin Camper, Lindsey Dunne Jacoby, and Jessica Enoch, that the progress narrative feature of reflective writing rarely reflects the actual learning and skill development that students have undergone (591). For the purposes of this study, the progress narrative was not included in the Corpus of Reflective Writing, as it can be difficult to determine if the writing produced is reflective in nature or a mutually understood exercise in ?schmoozing? (see Yancey and Weiser 1997). Each of the ten students in this study reflected on what actions they took when writing and revising, what they chose to focus on, how they thought they did, and what they believe they still need to work on. The prompts for each reflective writing exercise are included in Appendix A. As the prompts show, students were instructed to assess their successes and challenges, as well as point to areas that they would like their instructor to respond to assignment-specific questions are also asked in reflective writing prompts, including questions about how students integrated stylistic components that they were learning in class. In total, six pieces of reflective writing were requested from all ten students. Because one student did not write a reflection on her final position paper, this resulted 229 in 59 pieces of reflective writing collected across all ten students. All documents produced in the Corpus of Reflective Writing total 46,164 words (2,640 lemmas). Following its construction, the corpus was loaded into the qualitative software Dedoose (2018), where a team of writing program administrators at the University of Maryland consisting of myself, Elizabeth Miller, Justin Lohr, and Jessica Enoch read through each document and coded them according to recurrent patterns we were observing in the reflective writing. These patterns consisted of expressions of writerly identity (including discussions of goals and aspirations, declaration of writing knowledge, discussions of writing processes, and statements of self-definition), engagement with feedback (including appeals to the instructor, assessment of instructor feedback, directing instructors to look at specific content, acknowledgment of integration of instructor feedback, and acknowledgement of peer feedback), moments of reflection (including reflections of actions taken, assessments of those actions, rationales for taking those actions, discussions of possible directions their writing may take, and realizations that came about from writing), as well as explicit expressions of emotion. The aggregate collection of tags from these codes was used to create subcorpora of rhetorical moves. Because the tags represent discursive moves that students are making in their reflective writing?e.g., engaging with feedback or assessing their actions?each subcorpus is assumed to be able to capture the kind of language typical of that move. For the purpose of this study, only the Reflection subcorpus is analyzed, which is further subdivided into five subcorpra: reflection on 230 actions, assessment of actions, rationale for actions, possibility of different actions, and realization through reflection. Using the corpus linguistic program Sketch Engine (2018), I analyzed each subcorpus and compared it to the total number of words in the full corpus to produce a ratio that represents the amount of time students spent on each of these moves in their reflective writing (see Appendix B). The Representation value was calculated by dividing the number of words in the subcorpus by the size of the total corpus. To analyze the kinds of language that characterizes each reflective move, I conducted keyword analyses for each subcorpus, providing sketches of what each reflective move was ?about? (Scott 2001). This ?top-down? methodology of constructing subcorpora of rhetorical moves and conducting keyword analyses of each subcorpus was inspired by Biber, Connor, Upton, Anthony, and Gladkov (2007), who conducted keyword analyses on rhetorical appeals within fundraising discourse. Given that the focus of this study is specifically on the ways that emotional appeals are constructed in mitigated moments of self-assessment in reflective writing, the main subcorpus of interest is the Reflection subcorpus, with special attention paid to the Assessment of Action subcorpus. These corpora are analyzed in more detail in the next section. In the following section, the reflective writing and self-assessment corpora are analyzed using the theater of the mind model. Using the model, I show how linguistic moves that focus attention to L2 thoughts in feelings about actions simultaneously construct stances registered as L3 social interactions. The attenuative quality of these social interactions, furthermore, index a conscious attention to monitoring the 231 emotive force of students? evaluations and thus provide a window into the ways students use language to monitor and shape their instructors? feelings. The model is thus not only helpful for organizing the layered complexities of argumentative moves but also helpful for describing where and how different kinds of emotive meaning arise in argumentation. 3. Mitigation, Hedging, and First Year Writing Mitigation?the reverse of amplification?has a long history as a rhetorical device that serves a number of functions. If the goal of amplification is to ?[endow] an element with conceptual importance by making it salient in a text and prominent in perception? (Fahnestock 2011, 16), one might figure that the goal of mitigation is the opposite: to decrease attention to a concept and to avoid salience. This is, however, not the definition that is found in the Rhetorica ad Herennium. There, mitigation (deminutio) is defined as the figure that ?occurs when we say that by nature, fortune, or diligence, we or our clients possesses some exceptional advantage, and, in order to avoid the impression of arrogant display, we moderate and soften the statement of it?Therefore just as by circumspection we escape jealousy in life, so by prudence we avoid antipathy in speaking? (4.38.50). While the ad Herennium does mention the ?moderation? and ?softening? of a statement, it also foregrounds the interactional and, indeed, emotive contexts that invite this softening. Similar links between mitigation and emotion can be found in the classical handbooks of the Anonymous Seguerianus?s writing on argumentation (160) and Cicero writing on the constructing a peroration (De Partitione Oratoria 53). As Claudia Caffi (2007) notes, mitigative 232 strategies (e.g., attenuation, downgrading, indirectness) are often ascribed an emotive component to their usage, managing the reactions of the audiences to save face or come across in a favorable way (40-44). Yet this emotive aspect is not touched upon much in the contemporary literature. In contemporary linguistics, most research on mitigation focuses on the use of hedging, which originated not from attention to its emotive effects but its logic. It is surprising that the metaphor of a ?hedge? ever became a technical term in linguistics in the first place. Given its history, it may have been inevitable that the term varied slightly across scholars and disciplines since its introduction to linguistic pragmatics, sociolinguistics, and pedagogical linguistics. The term was introduced by George Lakoff (1972) who defined a hedge as ?words whose job is to make things fuzzier or less fuzzy? (195).2 Drawing on insights from Lotfi Zadeh?s mathematical work on ?fuzzy sets? and ?fuzzy logic,? Lakoff noticed that conventionally used words and phrases, such as a little, sort of, in many cases, kind of, generally, and maybe, insert fuzziness and uncertainty into the membership of a nominal into a conceptual category (e.g., A bat is sort of a bird).3 This sort of vagueness about propositions was found to be only one of the many ways speakers and writers mitigate their assertions. The linguist Ellen F. Prince, together with Joel Frader, a professor of Pediatrics, and 2 It is certainly worth noting that Lakoff?s definition included both mitigating and intensifying constructions, whereas current scholarship separates the two, calling only the mitigating constructions hedges. Intensifying constructions are aptly named intensifiers. 3 Lakoff was also interested in other constructions that affect impressions of what does or does not count as a member of a category (e.g., essentially, technically, strictly speaking, more or less, etc.). 233 Charles Bosk, a Sociologist, constructed a corpus (1982) of physicians talking to other physicians about patients during their morning rounds to better understand the role of hedging in physician-physician discourse. They keenly note that not all hedges are the same: some make the propositional content fuzzy, while others correlate ?with fuzziness in the relationship between the propositional content and the speaker, that is, in the speaker?s commitment to the truth of the proposition conveyed? (85). This second class of hedges are captured in constructional schemas that hedge speaker commitment such as It may be the case that X, I think that X, It seems that X, and so on. For those who categorize hedges, the kind of hedge that fuzzies propositional content has come to be called a ?propositional hedge? (Fraser 2010) or a ?bush? (Caffi 2007), while the kinds of hedge that correlate with the expression of commitment to the truth are simply called ?hedges? (Caffi 2007).4 For Caffi, bushes and hedges serve similar functions?i.e., to make things fuzzy?but they do so in different ways and thus should be treated as separate mitigative devices. She adds a third device, too: shielding.5 A shield makes fuzzy the deictic origin of an assertion, 4 Or they are called ?attributional shields? (Prince et al. 1982, 85). To avoid confusion, I will not use the term ?shield? to refer to this kind of mitigative device. See Note 2 below for a review on terms. 5 It may be helpful to know that these terms (bushes, hedges, and shields) are often confused or redefined. What Prince et al. (1982) call approximators are what Caffi (2007) calls a bushe and Fraser (2010) a propositional hedge. What Prince et. al call a plausibility shield, Caffi (2007) calls a hedge. Caffi?s term of shield corresponds to what Prince et. al call an attribution shield. As Caffi notes, she came to these terms without knowing of Prince?s work on the subject (69-70). She is also critical of plausibility shields and attribution shields both being called ?shields? as they work fundamentally differently, systematically mitigating different sorts of things (70). I find Caffi?s system of bushes, hedges, and shields to be more intuitive, and I believe Caffi in saying that she was unaware of Prince et al.?s work on this subject. 234 attributing an assertion of a proposition away from the speaker, for example The book says, One may think, As we know, and so on (Caffi 2007, 66). Contemporary studies of hedging have investigated the ways that different social groups hedge for different purposes. The sociocultural strand of research on hedging began with Robin Lakoff (1973, 1975), who sought to understand the relationship between hedging and accommodating to interlocutors? expectations within a speech event. For Lakoff, hedges were part of the gendered expression of women?s language, and an analysis of hedging worked to show the ways that women sought to be polite and to avoid conflict. Lakoff?s work and the work of others who investigate the role of hedging in social interactions illustrates the ways that speakers? and writers? language change across contexts, genres, and styles of communication. For example, Janet Holmes (1995, 2006) has followed Robin Lakoff?s work to observe the role of hedges in the communicative strategies of women and men in New Zealand, as well as the role of mitigation in indirect workplace interactions. Specifically, Holmes focuses on two different kinds of motivations for hedges: hedges that mitigate the certainty of a statement (e.g., That might be right, in the context where the speaker does not know the answer), which are called epistemic hedges, and hedges that introduce indirectness to soften the force of an expression (That might be right, in the context where the speaker know the right answer but does not want to come across as forceful), which are called affective hedges. Holmes?s work shows how hedges can index power relations in gendered interactions. Affective hedging as a strategy falls under what Penelope Brown and Stephen Levinson (1987) call ?positive politeness,? whereby speakers linguistically appeal to 235 hearers? ?positive face,? avoiding face-threatening moves and allowing hearers to continue to feel good (101-103). Karin Aijmer (2002) concurs that this kind of hedging practice is ?an important resource for the realization of politeness strategies? (8) and notes that providing such hedges indicates the speaker?s desire not to be responsible for the truth of their utterance. While epistemic and affective hedges may seem distinctly different, phrases like I think can serve both an epistemic function and an affective function. For example, in the sentence I think you?ve done a great job here, the hedge of the evaluation serves an affective function, as it mitigates the force of the assertion in the interpersonal dimension between the speaker and hearer, while, conversely, I think I?ve done a good job, when uttered by someone who is not in the position to evaluate their performance to someone who is, acts more as an epistemic hedge. It is precisely this kind of hedge that will be of interest in analyses of students? self-assessment in their reflective writing. The effect of hedging is something that has been known in writing studies for a very long time, but recent quantitative methods borrowed from corpus linguistics has allowed researchers to better understand how patterns of hedging vary systematically across genres and social contexts. Scholars like Ken Hyland (2000), Douglas Biber (2006), and Laura Aull & Zak Lancaster (2014) have analyzed how writers at various levels of academic proficiency use hedges to create nuanced stances, tracking out the development of writers from early in their academic careers to seasoned academics. Their work shows that students learn to hedge with regards to the demands of their discipline. For example, the students in the sciences tend to hedge more, and students in the humanities tend to hedge less. First year academic 236 writers, as Aull and Lancaster show (2014), seem to be less adept at using hedges to craft an epistemically well-tempered ethos. They show that first-year writing students tend to hedge less than their upper-level student counterparts and far less than published academics in all fields. As they argue, hedging is a valuable resource in academic writing for ?decreasing epistemic commitment and expanding discursive space? (10). Hedging is one of the ways that skillful writers nuance arguments and establish stance. As novice writers develop a sense of academic voice, they come to understand the value of nuance in crafting strong arguments, using hedging and other mitigative devices to develop more carefully constructed stances. What writing studies research?as well as the sociolinguistic and discourse analytic research? demonstrates is the ways that writers and speakers rely on mitigative devices as tools both for the construction of personal identity and styles, as well as for the negotiation of interpersonal sociality. Interestingly, the kinds of genres of academic writing that are frequently analyzed by applied linguists and writing studies specialists interested in stance tend to be argumentative in nature. As Aull and Lancaster (2014) show, in argumentative first-year writing, students tend to express more certainty than nuance and thus tend to boost more than they hedge. This is not necessarily the case for reflective writing in first-year classrooms. Table 1 shows the normalized frequencies6 of the top eight markers of epistemic and affective hedging per 10,000 words in the first-year student 6 Normalized frequencies help to compare frequencies of words across corpora of different sized. To obtain normalized frequencies of a hedging construction per 10,000 words, the frequency of the construction is multiplied by 10,000 and then divided by the size of the corpus it is derived from. 237 Corpus of Reflective Writing and compares them to the same markers in the Michigan Corpus of Upper-level Student Papers (MICUSP). MICUSP is a 2,121,001- word corpus of upper-level undergraduate and early graduate student papers in various disciplines that span different kinds of academic writing tasks?from lab reports to reviews to argumentative papers?that is often used to analyze college student writing. As the table shows, the reflective writing of the Maryland first-year writing students indicates a significantly high frequency of hedges. Corpus of Reflective Writing MICUSP marker Freq./ 10,000 words Freq./ 10,000 words I think 26.21 1.32 can 19.93 28.02 could 18.63 12.57 I feel 18.41 0.33 I thought 13.65 0.22 I felt 7.80 0.18 may 7.80 16.42 seem 4.80 2.44 Table 1. Top 8 markers of hedging These hedges consist of verbs of thoughts and feeling (think, feel, thought, felt), verbs of appearance (seem), and modal auxiliaries (can, could, may). While not all hedging markers are more frequent in the reflective writing corpus?indeed, the modal auxiliaries can and may are considerably higher in MICUSP?verbal hedges are notably higher in reflective writing than other genres of academic writing. I think, for example, appears twenty times more frequently in reflective writing, and I thought is 62 times more frequent in reflective writing. That the corpus would highlight first person verbs of cognition is, to some extent, to be expected, given that reflective writing requires thinking. Markers like I feel, I thought, and I felt almost never show 238 up in academic writing. The frequency of markers in reflective writing thus reveals a characteristic rhetorical component of the genre?namely, to focus on personal thoughts and feelings. It is worth noting that while the words and phrases in Table 1 may indicate hedging, they have other conventionalized meanings as well. The epistemic modal auxiliary verbs listed, for example, are frequently used to hedge, but they are also frequently used for purposes other than hedging. Phrases like I think, I feel, and I thought, too, can be used entirely referentially, entirely meta-performatively (i.e., as a hedge), or (usually) a bit of both. The question that remains, however, is what is being hedged and why. What are the hedges that accompany I think and I feel doing for student writers in their reflective processes? What is compelling students to mitigate their assertions in their own personal reflections? And what light can these mitigative moments shine on how students are using the space of reflective writing? Given that epistemic and affective hedging occur in social contexts where speakers and writers are held accountable for the validity of their claims or want to tend to the interpersonal relationships between themselves and their interlocutors, to presume that reflective writing is an asocial, wholly metacognitive activity does not seem to follow from these interactive and socially motivated markers. In the following section, I look at the place where mitigation tends to cluster: reflective statements of self-assessment. There I ask if these mitigative moves are serving a broader and perhaps more argumentative rhetorical function. 239 4. Analyzing Reflection The Corpus of Reflective Writing allows questions to be answered with regards to how students understand themselves as writers, their revision practices, and their feelings about their writing. It also allows us to analyze how much time students are spending performing reflective acts. Table 2 shows how frequently, based on wordcount, students performed each of these reflective writing tasks. In the first column is the name of the subcorpus, comprised of tagged moments of a particular reflective writing move (e.g., reflection, description of action, rationale for action). Since providing a rationale for action often requires a description of action within a moment of reflection, the same sentence may have multiple tags. In the second column are the total wordcounts of each subcorpus. The third column provides a value representing the percentage of the words in the full corpus that the reflective move takes up, and the fourth column shows the percentage relative to only the tagged reflective writing subcorpus. 240 % of words % of Dataset subcorpus wordcount in full words in corpus Reflection subcorpus Reflection 26044 56.41 % 100 %7 Description of Action 21458 46.48 % 82.39 % Assessment of Action 7420 16.07 % 28.49 % Rationale for Action 5256 11.39 % 20.18 % Possibility of Difference 684 1.48 % 2.62 % Realization 1754 3.80 % 6.37 % Table 2. Frequency of Reflective Moves Of the writing collected that comprises the Corpus of Reflective Writing, a majority (56.41%) was tagged as writing where students reflected on ongoing writing projects. While, given the name of the corpus, one might expect that the entire corpus consists of reflective writing, students often do more in their reflective writing than just reflect?for example, plan future activities, express their writerly identity, and convey emotion to their teachers. 82.39% of this writing involved students describing actions they made when writing or revising their papers. For example, when one of the students, Dwight, writes 7 This value may be confusing to some, since it is entirely trivial. What it means is that 100% of the words in the Reflection subcorpus consist of moments that were tagged as Reflection. 241 I used the website to discuss 3 different methods of second language acquisition. I used popular sources to show why people find certain methods to be optimal. (10.DF.Ref) 8 this is an example of the Description of Action tag. This kind of reflection is what Kathleen Blake Yancey (2017) has called ?routine descriptive reflective writing? (308). It provides a point-by-point narrative of what a student did with little rationale for why one course of action was taken over another and little self-assessment of the actions taken. The observation that students are good at describing what they do but not why they do it?a finding from Lindenman, Camper, Dunne Jacoby, and Enoch (2018) from the same university and course population?led to the breaking down of reflective writing into moments where students assess their actions as well as moments where they provide rationale for the course of action they took. Moments where students provide rationale for actions demonstrate that students understand that their teachers want to see writers provide reasons for their actions. For example, Alana provides the reasoning for why she included a counterargument to her position paper by stating: I thought it was important to address both sides to this issue and to go in further depth why people wish to lie and what their motives are within the online dating world. (5.PP.Ref) 8 The organizing scheme for this corpus is to write the number of the student, followed by the assignment, and finally the task. In this case, Dwight was student number 10, he was writing about his Digital Forum assignment, and his writing was a reflective memo. 242 The Rationale subcorpus illuminates how frequently students provided reasons for their actions. Notice that Alana uses I thought in the above example, but this kind of move is not considered a hedge precisely because it involves genuine reference to the writer?s mental state and act of recollection. The rationale moves differ in purpose and meaning from the assessment moves. The assessment subcorpus shows moments where students pause to evaluate the efficacy of their actions. For example, when Dawn writes In my analysis, I thought I did a good job at analyzing how the author?s language created meaning and displaying how the language was a choice in strengthening the argument. (1.Sum.Ref) this was tagged as an assessment. In this example, Dawn?s use of I thought is considered a hedge because what follows the construction is not a recollection but an evaluation. Dawn is not providing a reasoning but an assessment of her work, and the marker here thus hedges that asserted evaluation. Interestingly, by adding the frequency of self-assessment and combining the frequency of rationale, it becomes evident that 48.67% of the reflective moves are going beyond simple descriptions of actions taken.9 This illustrates that the students in the course are engaged in more than ?routine descriptive reflective writing.? They are spending a considerable amount of time evaluating their actions and explaining the rationales for the course of actions that they chose. 9 This value was arrived at by adding the Assessment of Action value to the Rationale of Action value. Assessment of Action (28.49%) + Rationale for Action (20.18) = 48.67% 243 The ratio of description, assessment, and reasoning in students? reflective writing provides insight into students? metacognition regarding their learning processes. In their now landmark How Learning Works, Susan A. Ambrose and her colleagues Michael W. Bridges, Michele DiPietro, Marsha C. Lovett and Marie K. Norman (2010) emphasize the roles of self-monitoring, self-evaluation, and active revision in the learning process (91-92). Self-assessment is a crucial part of learning, and it also comprises a sizable portion of the reflective acts that the students are performing in their reflective writing. Given that self-assessment is such a crucial part of learning through reflection, it is necessary to closely examine how students talk when they assess. To better understand what patterns of language students use to assess their actions and work, a keyword list of the Assessment data subcorpus was created against the full corpus. These words highlight what is unique about the Assessment subcorpus relative to the full corpus. Because not all words are evenly distributed throughout the corpus?e.g., one student writes about Disney princesses frequently, whereas the rest of the students do not?word frequencies were derived using an averaged reduced frequency (ARF) measure. An ARF measure further normalizes the distribution of words in the corpus, deemphasizing effects of rare words that are idiosyncratically used by one student and not the rest. The resulting list of keywords in Table 3 represents words that stand out as more ?about? student self-assessment and thus are presumed to be more useful to students who are making self-assessments in their 244 reflective memos than writing any other part of a reflective memo, writing an autobiography, or writing a revision plan. 10 Assessment sub-corpus Full corpus word ARF ARF/10,000 ARF ARF/10,000 Keyness words words Score strengths 6.20 7.764 0.780 1.564 5.0 think 37.00 46.555 53.20 10.613 4.4 successful 5.30 6.684 0.760 1.526 4.4 struggled 9.20 11.598 1.430 2.854 4.1 hard 5.90 7.421 1.050 2.104 3.5 felt 10.80 1.3503 2.200 4.399 3.1 strong 5.60 7.053 1.160 2.308 3.0 allowed 5.70 7.163 1.210 2.405 3.0 feel 16.70 21.079 36.00 7.188 2.9 really 9.70 12.187 21.50 4.282 2.8 Table 3. Keywords of Assessment Sub-Corpus Measured Against Full Corpus What is interesting about the keywords in Table 3 is what they tell us about the kinds of recurrent patterns of language and their accompanying patterns of thought that come with assessment. As the keyness score indicates, strengths is on average 5 times more likely to appear in a moment of assessment than anywhere else in the student writings collected. I think, too, is nearly 4.5 times more likely to show up in self-assessment than anywhere else. This is impressive, given that, as Table 1 10 Unlike other corpus linguistic software, Sketch Engine, created by Adam Killgarriff, calculates keyness not by log-likelihood or statistical significance but by a raw ratio of frequency in a specialized corpus versus frequency in a reference corpus. Killgarriff has written on why a raw ratio is a more honest measure for determining keyness (see Killgarriff 2009). Sketch Engine was chosen for this analysis, however, because of its ability to create Average Reduced Frequencies (ARFs), which help to filter out some of the project-specific words that perhaps one student uses but no other students use. 245 illustrated earlier, I think is twenty times more likely to show up in reflective writing relative to the academic writing collected in MICUSP. Words like strengths, successful, and struggled are likely primed by the questions asked in the reflection prompts students were given, which as, for example, ?What strengths do you see?? and ?What did you struggle with?? (see Appendix A). However, words like think, felt, and feel?all of which can be markers of epistemic and affective hedging?are not prompted by any question. Nearly every instance of think, felt, and feel is also accompanied by the first-person pronoun I, indicating that the students are taking up personal stances. The things that students think and feel about are, perhaps not surprisingly, also things that they are evaluating. Think tends to co-occur with positive evaluations?e.g., ?I think I achieved what I set out to? (9.Inq.Ref)?and feel with neutral or negative ones, e.g. ?I feel I spent too long in my introduction using anecdotal examples instead of defining key terms? (9.PP.Ref). I feel, however, also tends to appear in moments of positive self-assessment, too, as in ?With these changes I feel confident that my paper will transform from hot garbage to more of a mild to moderate garbage fire? (8.Inq.Ref). Because I think tends to index moments of positive self-assessment, it may be helpful to understand how hedging works in these moments and what rhetorical purpose they serve. The following example illustrates the way one student, Gale, uses I think (twice in one sentence) to hedge strongly positive assertions. I think that I most successfully expressed how the relationship/issue between academics/athletics affects more than just athletes, and that other non-athletes can/should care about this topic, in conclusion I think I established a topic?s 246 exigence very well even though not very many people would care about this topic at first glance. (6.Inq.Ref) The use of I think in this example illustrates Gale taking a weakly assertive epistemic stance toward her evaluation of the quality of content in her essay. Intensifiers like most in ?most successfully,? positive evaluative stance markers like successfully, and the deontic modal should in the following clause ?other non-athletes can/should care about this topic? position Gale as someone who believes her work is good and important. By leading this sentence with I think, Gale makes her assertion weaker, grounding it in her epistemic stance, her thoughts and beliefs. In doing so, Gale also makes her statement less vulnerable to contradiction, so long as her belief is thought to be sincere. By not baldly asserting these very glowing evaluations, Gale comes across as more modest, maybe even humble, certainly not arrogant. In the final clause of this excerpt, Gale assesses the exigence of her writing but again uses I think to put herself on stage as someone performing a thought. This may lead one to wonder what is motivating students to hedge their self-assessments. 5. I Think As An Emotional Appeal It?s easy to forget that when students write reflective memos, they are not only reflecting but also communicating to an audience. Specifically, they are responding to prompt questions?such as ?What did you struggle with?? and ?What would you like me to notice???that are clearly coming from a place of authority (Appendix A). In these reflective assignments, then, students take stances on their writing to assert their views, to create opportunities for conversation, to signal shared values, and to update 247 their instructors. In other words, student writers use the space of reflective writing for their own rhetorical purposes. An analysis of the kinds of stances that the writers took in this study reveals recurrent patterns that speak to the ways students evaluate their work, negotiate power with their teachers, and often express discomfort when asked to assess the quality of their writing. The dialogic component of reflective writing is important because it helps to understand the mitigative role that hedging plays for students who write knowing that their teachers will read their reflections. While in principle, what is expressed in reflective writing need not be hedged?that is, if students want to evaluate their actions, it is not necessary for them to make their evaluations more polite?the dialogic aspect of hedging acts as a reminder that reflective writing is interactional and intersubjective. Writers write to produce effects in their audiences, and in reflective writing, this is no different. Students are keenly aware of how their evaluations will be perceived and are thus paying special attention to the potential effects of their speech acts. Consider this hedge in the reflective writing of a student named Alana. Alana is writing a reflection on her summary assignment, a short summary of a chapter from Joseph Harris?s Revising. Worth noting is that the goal of the summary assignment that Alana is reflecting on is to construct an informative 300-word summary that reflects the main ideas and arguments of Harris?s chapter. After providing a one- sentence summary of what she believes Harris?s argument is, she writes ?I think I was successful in creating a summary that was able to convince the audience to revise their work and to know how to do so based on what Harris writes? (5.Sum.Ref). 248 Given that the goal of the assignment is not to convince audience of anything or to produce a persuasive document, Alana?s moment of evaluation draws attention to her verbalization of her gut feelings. I think here serves a number of rhetorical functions. At its most basic level, it hedges a strong claim: that her summary assignment will persuade audiences to revise their work because of what she says about Harris. Alana not only puts this bold assertion on stage but puts herself thinking it on stage, too. She effectively creates an L3 move?focusing the reader toward the writer?s evaluation? but does so through an L2 move?the expression of a personal thought. What spotlighting herself thinking does for Alana is not only to allow for epistemic focus on the bold assertion that she is making, but also allow Alana to create a distance between the object being evaluated and the reader. If Alana were simply to state her evaluation as ?I was successful in creating a summary that was able to convince the audience to revise their work?? she would allow for her stance to be countered or put into question. The adjective successful marks a positive evaluative stance, inviting her readers to either accept or take their own stance on the object being evaluated. Her instructor could align with Alana?s stance and concede the positive evaluation. Her instructor could alternatively go and read her summary and think that, in fact, no, Alana has not successfully created a persuasive summary, marking a disalignment of stances. What is important is that Alana allows an avenue for disagreement and alternate perspectives, broadening the potential for further lines of discourse. What using the hedge I think does is protect Alana from being contradicted. Chiefly, for Alana?s utterance to be false, her instructor would need to be able to 249 access Alana?s thoughts. If her instructor disagrees with her assessment, that?s fine. That?s their perspective. But by putting herself on stage epistemically evaluating, Alana makes her own personal and internalized thought process the object of evaluation. This complicates how instructors can respond. In Figure 1(a), Alana is represented evaluating her summary assignment without the I think hedge. Notice that Alana places her evaluation on stage and invites her audience to join in alignment with her. This is different from the kind of move that is put on stage in 1(b), which uses the I think epistemic hedge. In 1(b), Alana objectively places herself on stage and obstructs direct access to the object of evaluation. Figure 1(a). ?I was successful in creating a summary that was able to convince the audience to revise their work and to know how to do so based on what Harris writes? 250 As the figure shows, instead of evaluating how Alana did on her assignment, the hedges assertion rearranges the stance triangle to assess Alana?s assessment. While in 1(a) the frame that Alana and her audience are positioning themselves in accordance with is a positively evaluated one (as indicated by the use of green), in 1(b), Alana and her audience position themselves in relation to Alana having thoughts. The positive evaluation is in a further embedded space, mediated by Alana. The primary focus also shifts from a past event (I created) to a present event (I think), further emphasizing the importance of aligning not with what Alana is evaluating but with Alana?s personal epistemic stance. Figure 1(b). ?I think I was successful in creating a summary that was able?? Alana is not alone in hedging evaluations in this way. When students hedge their evaluations, and, in effect, make their evaluations hard to disagree with, they hedge themselves, their thoughts, and their feelings between what is being evaluated. In this 251 way, they are crafting what I call a defensive stance. This stance is not defensive in the sense that Alana and other students who hedge their positive evaluations are feeling under attack but because it preemptively defends positive assessments from an instructor?s potential counter assessments. It uses an L2 move (e.g., I think, I feel) to keep the conditions for alignment centered on the speaker?s personal attitudes. In Brown & Levinson?s landmark work on politeness, the sort of hedge that Alana is producing falls under the heading of ?hedging to avoid disagreement? and is characterized as a strategy that speakers use to preemptively save face when making assertions that they may not have the authority to make (151). Within moments of self-assessment, the kinds of utterances that are hedged are often utterances discussing the completion of a paper. Students hedge that their actions were ?successful? (1.Sum.Ref, 1.Inq.Ref, 5.Sum.Ref, 6.Sum.Ref, 10.RA.Ref), that they did a ?great job? (5.RA.Ref, 5.DF.Ref) or a ?good job? (1.RA.Ref, 10.DF.Ref), that they know how their audience will respond to their writing (5.DF.Ref), and that they are improving (4.Inq.Ref, 4.RR.Ref). Students do not have the final say in making these assessments, yet making these assessments work favorably in indexing the idea that students are becoming stronger writers as a result of their writing assignments. Moments of hedged self-assessment, then, frequently show up in statements of progress and moments that feed into general progress narratives. By hedging to avoid disagreement in these moments, students mitigate potential for their instructors to take issue with their progress. While it may only have a minor effect, hedges can potentially help to reframe how instructors see their students and to understand them and their works positively and without critical scrutiny. Students frame themselves as 252 good students, presenting themselves as the favorable product of their instructor?s pedagogical efforts. It is worth noting that not all hedged self-assessments appear in moments where students frame themselves and their work positively, however. For example, Hera, who has just received the grade of her Inquiry Assignment, reflects on how she wrote that paper. Hera writes ?I think I achieved what I set out to do in my revision plan but I had a flawed revision plan because the corrections I made were mostly unimportant or meaningless to the actual inquiry? (9.Inq.Ref). Hera begins with a hedged positive self-assessment?she did what she planned on doing?but continues on to discuss her shortcomings, calling her corrections ?unimportant? and ?meaningless? and providing some explanation for the problems she believes her instructor might see in her final draft. I think here not only works to mitigate the positive evaluation in the first clause; it allows the negative evaluation in the second clause to feel balanced, salient, and also mitigated. Of course, it may very well be the case that Hera could have improved on her revision, as what she set out to do was probably not to produce something she was not satisfied with, but this moment of self-assessment is made ?fuzzy,? to use Lakoff?s term, so that it can recede into the background. Similarly, pairing the positive self-assessment and the negative self- assessment within the scope of the hedge I think softens the force of the negative evaluation and shortens the distance between the positive and negative, making the positive feel less positive and the negative feel less negative. Within her assessment, Hera uses this negative assessment to express frustration and disappointment with her 253 performance, while still constructing an ethos that is meant to appear rational and in evaluative alignment with her instructor. Hedged self-assessment, it seems, is often serving this mitigative role of either reconfiguring self-evaluative stance so that instructors are observing students performing stancetaking?as we see with Alana?thus inviting instructors to feel good about their students feeling good, or backgrounding, redirecting, or obscuring certain emotive incongruities in order to salvage student-instructor evaluative alignment, as is the case with Hera?s hedged self-assessment. In both cases, students are not necessarily trying to move their audience to arouse strong feelings. On the contrary, they are mitigating potential disagreements with summative positive evaluations. When Artemis writes ?I think overall there?s not much more to be done, all things considered? when assessing her assignment (8.DF.Ref), she hedges an assertion that her assignment is almost finished. Her hedge does not invite her audience to engage with what more can be done, and, in that, it helps her to avoid potential negative feedback. Commenting not on reflection but on academic writing, Marc Silver has also notices the role of this sort of hedge, calling it a meta-pragmatic ?attempt to anticipate and redirect the reader?s potential rejection of his/her claim? (365). I think here reconfigures the stance triangle so that interlocutors are blocked from direct access to evaluations. In the context of reflective writing, the impetus for using this meta-pragmatic strategy may come from discomfort with making assertions with certainty in a context where, in the student?s mind, she doesn?t get the final say on the quality of her work. 254 This kind of uncertainty and hesitancy to take assertive stances toward the assessed content of reflection appears time and again in the Assessment dataset subcorpus. While reflective writing encourages students to express their subjective thoughts and feelings, the use of key epistemic and affective stance markers like I think and I feel show that students are often adding an extra layer of metacognitive depiction in their expressions of thoughts and feelings in order to perform a preemptively defensive stance. In other words, student writers are signaling to their teacher readers that they are having thoughts and feelings and that these thoughts and feelings are their own; if teachers disagree with the things the students are thinking and feeling, then that?s fine. That?s their opinion. The motive for this defensive stance may very well be to intentionally or unintentionally imbue evaluative moves with an expression of frustration for being asked to perform self-assessment while not having the final say on the quality of work being assessed. Hedges do not appear apropos of nothing; their use has meaning. Students are led to hedge their self-assessments due to the powerlessness they may feel when positioning themselves with their instructors during evaluative acts. It is in this sense that I think may work as a mitigative device that indexes emotive attitudes and further works to influence the emotive attitudes of interlocutors. When students use hedges like I think in self-assessment, it is often to avoid participation at the highest level of rhetorical engagement. Students mitigate presenting their evaluations as objects that can be argued over, and they are able to do so by politely constructing defensive stances, allowing them to avoid disagreement and potential face-threatening tension. 255 Instructors can align with students? stances towards their evaluations or not, but if they decide to insert their own stances?to counter, to question, to disagree, to disalign?there is a potential for pushback. To disagree with a student?s thoughts on their positive evaluations means to interrupt a student?s positive feelings. To construct a stance of disagreement with someone who states that they privately believe they have done something well is not affectively the preferred position to take. To avoid taking up this position, an instructor may feel that it is easier to, in that instance, align with the student and allow for the positive evaluation to become embedded in the aligned stance. As a result, the instructor?s thoughts and feelings about the object under assessment are influenced, allowing for the positive framing of the object. Many instructors who are conflict avoidant may be persuaded through this strategy. Indeed, if reflective writing is presented as an ungraded activity that promotes thinking about writing, then there is little motivation for instructors to come to reflective writing with a critical attitude. Alternatively, instructors may ignore responding to students? assessments or they may not take students? thoughts and feelings seriously, aware that students are ?schmoozing? with these hedged positive evaluations (see Weisser 1997). Such an approach is also not preferable, this time not from an affective sense but from a pedagogical sense, as it defeats the metacognitive purpose of reflective writing. If reflective writing is meant to show conscious and monitored reflection on actions and assessments, and if instructors cannot trust students? expressions of self-assessment, then the expression of hedged self-assessments may indicate flaws within the practice. 256 I assume few teachers think about the ways that hedged self-assertions can influence their thoughts and feelings, or how mitigative moves in students? reflective writing can shape the way they take up their students? perspectives. I wonder, however, if this is not because we too often neglect the emotional dimension of reflective writing. In her 2003 article ?A Feeling for Aristotle: Emotion in the Sphere of Ethics,? Ellen Quandahl criticizes the shortcomings of contemporary rhetoric and composition scholarship when issues of emotion and ethics are involved. ?[T]he field is still silent on how emotion might figure in rhetorical pedagogy,? Quandahl remarks, ?because the current work on ethics in Composition has not yet deeply considered the emotional dimension? (21). How might understanding hedged self- assertions as emotional appeals influence the ways that we teach and read reflective writing? Suggesting possible avenues for integrating emotion into the work of writing pedagogy, Quandahl recommends developing ?opportunities to observe repertoires of feeling represented in language? (21). To do so, however, requires instructors to understand more than just the words that evoke emotion but also the strategies that realign and reposition, influence, mitigate, and otherwise indirectly manage emotion. 6. I Think Across the Semester While this study was set up to look at how students? reflective practices change across the semester, the evidence suggests that, for the most part, the tendency to imply uncertainty and perform a defensive stance in self-assessment writing does not seem to change. Figure 2 shows the frequency of I think across all students in various reflection events over the semester. 257 300 250 200 150 100 50 0 Summary Inquiry Rhetorical Digital Forum Position Paper Final Reflection Reflection Analysis Reflection Reflection Reflective Reflection Memo chronological reflection event Dawn Jefferson Mirage Ariel Alana Gale Zeus Artemis Hera Dwight Average Figure 2. I think across the semester Looking at the average student?s usage of I think, it is clear that students are consistently using I think throughout the semester, though there is a pretty sharp decline on the final reflective memo. In fact, there is a drop in many of the hedges that express a defensive stance in that final reflective memo. The memo is turned in alongside students? final revision assignment, and it is considerably longer than any other reflective memo students are asked to write throughout the semester. While students may be unusually more confident in the assertions they are making in their final reflective memo, it may be something else. Because this piece of reflective writing is graded as an assignment, students may be treating this piece of reflective writing less as a moment to emote or express discomfort or uncertainty and more as a moment to fall into detached, routine descriptive writing. Students also workshop these memos, adding readers to the memo who are not in a position to affect the grade of the assignment and who may find it strange to see classmates emoting and expressing discomfort in acts of self-assessment. A shift in the context and discourse 258 frequency per 10,000 community of the reflective memo may be affecting the language students use when performing rhetorical moves in this memo. The kind of metacognitive bleaching that can come with treating reflection as a workshopped paper has been observed by Christian Russell McDonald, as well, who writes that ?the reflective essay as a culminating assignment, through sheer repetition, can be reduced unintentionally to a standardized form? (220). It can be treated more like a polished piece of academic writing than a genre intended to help students learn more about themselves and their writing. This leads me to ask: what is the role of I think in reflective writing? This question may sound absurd, given that students obviously ought to be thinking when they are reflecting. But I think puts the writer on stage in a way that frequently indirectly implies uncertainty, hesitation, discomfort, and perhaps a need to explicitly signal personal ownership of her assessments. Embedding a sentence under I think adds nothing to the utterance that is not implicit within the conditions of its assertion; it is understood that the writer is thinking and it should be expected that the writer believes that what is being said is true. I think is often redundant if the goal is simply to assess, but this redundancy places a spotlight on the student reflector. At least to me?if I may hedge my own claim?it signals to the teacher reader that students, in their moment of reflection, are uncertain, perhaps uncomfortable assessing their pre- graded work, yet know they are expected to make an assessment regardless. Still, we may want to imagine a time when students are more willing to honestly express their feelings about their work without fears of teachers? finger wagging and judgments when student assessments and teacher assessments do not align. 259 7. Conclusion This chapter adds an important dimension to the analysis of emotion that is not frequently considered: the way that rhetors negotiate stance, mitigate emotive reactions, and construct nuance in accordance to their persuasive purposes. Within the classical rhetorical tradition, understanding how construct emotional appeals also meant learning how to calm and cancel emotive reactions, too. Student reflective writing, while often analyzed for its metacognitive value, also provides a space for observing genuine emotional appeals in the wild. Observing how mitigation works in moments of self-assessment has, I hope, clarified an under-appreciated but potentially very rich resource for thinking about the roles that emotion has to play in discourse and reason. If, as Laura Micciche says (2007), ?emotion takes form between bodies rather than residing in them? (13), then it is imperative to understand the role of conventional linguistic moves that reposition, shift stances, and index our complex social relationships in any given situation. While this study is limited in terms of the definiteness of its conclusions, it provides an outline for future research. Ideally, more data from more students would be helpful. A larger corpus would allow for a better understanding of how frequent and how unique hedged self-assessments are to reflective writing and what further functions they may serve. Introducing further qualitative methods, like interviews with students on how they understand self-assessment and the power dynamics of their classrooms may also further our understanding of the emotive role of mitigation in reflective writing. Analysis of other markers of epistemic and affective hedging, 260 such as I feel, as well as affective stance markers like emotive adverbs and ?private verbs? (Biber 1988, 105) that collocate with hedges may also provide a richer picture of where and why students mitigate in reflective writing. Moving forward, we as instructors may want to spend more time doing exactly what Allan and Driscoll suggest we do: talk about reflective writing in our classrooms. Teach it, but also illustrate pitfalls and open up conversations about the role and value of self-assessment in reflective writing. This means that instructors need to first establish what the goal of reading students? self-assessments is for themselves, a point that reflective writing pioneer Jeff Sommers (2016) states he only recently came to recognize as significant, after three decades of assigning reflective memo writing. To be sure, it?s not bad that students are indirectly expressing uncertainty or discomfort when performing self-reflection, but it?s telling that they feel the need to do so indirectly, taking on preemptively defensive stance and significantly reducing the scope of certainty of their assessments. Given that some students are afraid of overly schoolmarmy teachers telling them that they are wrong to believe that they have produced good work, it may not be enough to just have students write more reflective memos with better reflective prompts. It may be that we need to engage in conversations about our roles as teachers, their fears as students, and our combined honest expectations of what reflective writing can do for developing awareness?not just performing awareness?of writing knowledge. Moreover, understanding how small linguistic moves can index larger rhetorical and emotive situations needs to be a priority for reflective writing scholarship. If we are to understand reflective writing as a window into students? 261 developing minds, reflecting how they write and how they think about writing, we need to also be able to understand when what they say in their reflections is meant to provide access into metacognition and when it is meant to influence our thoughts and feelings for rhetorical purposes. Naively believing that reflection is a genre that produces metacognition may not be helpful if students approach reflective writing with a different set of motives. The objective of this chapter was to understand how linguistic moves can influence emotive reactions, but it took a different approach. While earlier chapters focus on how language can arouse emotions and strong feelings, this chapter instead focuses on how words can mitigate feelings, persuade instructors not to find problems, not to be moved to respond, and not to engage critically. To put it another way, this chapter explores a context in which emotions influence the ways people write and writing influences the ways people feel. While some students are in fact styling their writing to ?schmooze? to their teachers, much of the rhetorical strategy of performing a defensive stance may very well be unintended by student writers. Nevertheless, as Robin Lakoff (1975) famously writes, ?language uses us as much as we use language? (3). The same can be said about emotions and rhetoric. 262 Concluding Remarks There is a maxim that often gets attributed to Marcel Proust: ?It is the lack of imagination that keeps a man from suffering too much.? This maxim shares characteristics with other more well-known expressions in English, like ?curiosity killed the cat,? and ?ignorance is bliss,? all of which draw attention to the relationship between imaginative activity and emotion arousal. In a sense, then, the argument crafted throughout this dissertation?that intentional and rhetorical strategies for emotion arousal are dependent on guiding imagination?is not original. The argument was made by Aristotle in antiquity, returned to by Roman rhetoricians like Quintilian, and its appearance in modern day idioms suggests that it was not lost over the years. Indeed, George Campbell made a similar argument in the 18th century when he claimed we need to please the imagination before we can move the passions (1). So what exactly does this dissertation contribute to the analysis of emotion in rhetoric? I suggest the contribution is less so in the observation that emotions play out in the theater of the mind and more so in the ways that language positions us within that theater. It is one thing to know that a rhetor can achieve emotive affect by drawing on an audience?s imagination, and it is another entirely to be able to show the linguistic constructions in a language that prompt imagination and signal emotive suasion. Moreover, as this dissertation has demonstrated, the broad category of imagination may refer to several senses, such as mental simulation and inductive reasoning, that need further unpacking. By introducing the theater of the mind model, this dissertation has provided a way of understanding lexicalized human experience? the extent of our semantic space?in terms of physical, mental (meta-physical), to 263 communicative (meta-mental), and rhetorical (meta-communicative) representations. The interconnection of these kinds of representations in discourse is part of what has been so elusive to rhetorical theory and rhetorical analysis, especially in the context of emotional appeals, which lose efficacy when they announce themselves in persuasive events. From analyzing emotional appeals using the theater of the mind model, I have shown how rich emotive events can be cued from lexical and grammatical constructions that are not themselves emotive at all. The pathetic enthymeme draws readers to activate emotion concepts from lexical reference to salient parts of an emotion event, revealing emotion arousal?s reliance on metonymy and reasoning from cultural and discursive common ground. In analyzing the grammatical features of enargeia, I show how grammatical construals at one level of the theater of the mind and mental space configurations of irrealis events at another level are used in persuasive discourse to focus in on and reposition audiences within emotional scenarios. At the level of communicative interactions, writers and speakers can mitigate their commitment to expressions and assertions, a meta-mental move, in order to soften, redirect, or otherwise circumvent the arousal of an emotion event. Language, then, plays an important role in tending to the interpersonal dynamic of shaping emotion through the alignment of stance. More than just simply stating that emotional appeals rely on pleasing the imagination, I have demonstrated how tending to the physical, mental, and communicative representations that discourse entails can help to show how rhetorical effects are produced systematically and strategically for emotive effect. 264 Having spent so much time on this project, I am starting to see the strategies outlined in the chapters of this dissertation everywhere. A few weeks before writing this concluding chapter, I read that the Trump Administration was planning on revoking healthcare anti-discrimination protections for trans and non-binary people in the United States, and, outraged by this, I set off to Twitter to read what people were saying about this news. Almost immediately, I came across a viral Twitter thread,1 written by Sam Dylan Finch, that composed a compelling emotional appeal. This emotional appeal, interestingly, used every strategy that I have presented. The argument that Finch makes in his Twitter thread is that transgender people are often persecuted by healthcare workers, and that if some thinks this is fine or unimportant, then that person is immorally deciding whose experiences matter. Finch?s argument begins by invoking an oppositional you in the common ground who ?really [doesn?t] understand? what the exclusive we is arguing: I want to explain what we're talking about when we say that transgender people need anti-discrimination protections in health care. Because I think some of you really don't understand what this means, or might perceive our reaction as dramatic. But it's not. (@samdylanfinch) A fictive common ground is established in these two sentences, where I am being guided to imagine myself as someone who is unfamiliar with trans discrimination. Perhaps because the audience of this online asynchronous exchange is not fully predictable, Finch uses a hedged I think to mitigate the force of his assertion that ?you 1 For the full text of the Twitter thread, see the Works Cited, under Finch. 265 don?t understand what this means, or might perceive our reaction as dramatic.? Because, as one might expect, calling someone out for being ignorant on an issue is a potentially emotion-arousing move in discourse, the hedge here helps Finch to establish the proper kind of interaction with his audience. After this opening tweet, Finch moves into a series of stories involving discrimination against trans women and men from EMTs and doctors. Finch begins with a story of someone he knows and then moves into two other stories of discrimination by healthcare workers. I remember when I lived in Michigan and a trans woman that I knew had to call 911, because a serious wound she had (from an unrelated medical condition) started hemorrhaging overnight. One of the EMTs went upstairs to her room, and when that EMT realized she was transgender, was visibly disgusted and left the room. Her mom overheard the EMT mocking her as he spoke to the other EMT, referring to her as an ?it.? But the worst part of it was that, when it was decided that she needed to go to the hospital, they made her walk down the stairs herself without helping her. She was hemorrhaging blood from a leg wound. They stood impatiently and just watched her struggle. (@samdylanfinch) This is the first of three examples of discrimination that Finch provides. The second and third examples are more well-known: the second being the story of Tyra Hunter, a trans black woman from Washington DC who died of injuries caused by a car accident in 1995 after EMTs refused to treat her because she was transgender, and the 266 third of Robert Eads, a trans man whose diagnosed ovarian cancer went untreated by twenty separate doctors, eventually leading to his death. Finch does not tell his audience that they should be feeling angry or that they should be feeling indignation. Instead, the descriptions of events in the stories lead audiences to construct feelings of humiliation, neglect, and indignation. For instance, in the example of the paramedics who forced the hemorrhaging trans woman to walk herself down the stairs while they looked on, Finch only needs to tell us that she is injured, that she needs to go to the hospital, and that the paramedics were not helping her down the stairs. Through describing aspects of the physical event (the motion and viewpoint of go down the stairs, the event of hemorrhaging blood, the tense force dynamics of struggle, etc.) and mental events (the evaluation of worst part, the reference to a reasoning process in decided, and the affect indexed by impatiently) the audience is able to quickly and effortlessly simulate the scene and categorize their emotion state. They do this by reasoning their ways through the rest of the script, drawing on endoxic premises when necessary: being upstairs, the woman needs to go downstairs, but this is difficult for her because she is injured; the paramedics are there to help her go downstairs; the paramedics are not helping her, and, in fact, they are impatiently watching her; being watched impatiently while struggling to go down the stairs is humiliating; the paramedics? impatience is unjustified because they are not doing their job; the trans woman struggling to go down the flight of stairs should not have to be struggling because the paramedics should be helping her; this is a bad action, and I don?t want the paramedics to do this; I want to say something against this action; etc. Simulating this complex script also helps audiences not only to 267 register the intended emotion a rhetor wants to arouse but also to align shared attitudes and feelings and to react to perceived injustices. Later in the Twitter thread, Finch refers back to the examples of trans men and women who were discriminated against by healthcare workers, and he does so through enargeiac attempts to get audiences to mentally simulate themselves in the situations already described: This is not about politics. This is about fundamental human rights. I want you to imagine getting into a serious car accident, and as you are literally dying before someone?s eyes, they are MOCKING you when you thought they had come to help you. I want you to imagine the humiliation of hemorrhaging blood, and being made to crawl down a staircase, while two ambulance workers that you called for help refer to you as ?it? and look at you with disgust. I want you to imagine getting a cancer diagnosis, and going to doctor after doctor, TWENTY TIMES, to no avail. Imagine one callously remarking that maybe the cancer would teach you a lesson. Imagine the time is ticking, and no matter how much you plead, no one will help. (@samdylanfinch) The irrealis mental spaces constructed in imagine constructions work to reposition audiences as experiencers in the emotive events, and these repositionings work alongside the metonymies that guide the pathetic enthymemes (e.g. serious car accident, literally dying, MOCKING, come to help you). The imperfective construals prompted by progressive aspect constructions (getting into a serious car accident, dying, MOCKING, hemorrhaging, being made to crawl) and simple present 268 constructions (refer, look, plead) further work to situate conceptualizers in imperfective events, prompting them to simulate the completion of the events as they are unfolding in the theaters of their minds. Mental space shifting between tenses, aspects, mood, and negation quite literally moves audiences in the mind?s eye between mental spaces, directing them to shift how they are looking. As I have argued in this dissertation, any analysis of pathos that does not account for the effects of frame activation, as well as the grammars of tense, aspect, mood, and negation, cannot account for how language moves because it does not understand the role that cognition plays in making audiences simulate their own movement. Providing a way to analyze this kind of movement, then, is paramount to the project of understanding the emotional appeal. While the chapters of this dissertation have focused on individual rhetorical strategies and individual cognitive, linguistic, and emotive elements, Finch?s argument demonstrates that these strategies often run together: the pathetic enthymeme is heightened by enargeiac effects, and intersubjective discourse moves like hedging help to mitigate and fine-tune the kinds of emotional distance and engagement rhetors want to establish with their audiences, given the sets of assertions their arguments entail. These moves, moreover, are available to any speaker, writer, or rhetor. They are part of our language, and they are one of the many ways we use language to move one another. That is, the strategies presented in the chapters of this dissertation do not reflect language usage of a particular rhetor or ideology?they are not simply strategies that the NRA uses or that reflective writers use. They reflect shared, 269 mutually intelligible rhetorical strategies. Because a rhetor cannot get their audience to change their emotion state by commanding it?demanding someone to ?feel sad? will likely lead to their confusion, if anything else?speakers and writers rely on linguistic strategies to move their audiences. The same rhetorical strategies used by the NRA and Donald Trump are used by a trans rights activist to achieve similar emotive effects for very different political purposes. It is my hope that, in the course of this study, some of these strategies have become apparent. And it is my wish that, by understanding these strategies, we finally begin to address the feeling of persuasion not as something nebulous or nefarious, nor as something mercurial, always just slipping from our grasps, but as something describable, analyzable, even producible, but most importantly reasonable. 270 Appendices Appendix A The following are the reflective writing prompts that students responded to. Writing Autobiography Reflect on your experiences with writing and academic writing throughout your life. How would you evaluate and define these experiences? How do you see yourself as an academic writer? What excites and interests you about a class like this? What concerns do you have? What do you want me to know about who you are as a writer or what your experiences with writing have been? **If you?d prefer not to share your experiences with the class via Discussion Board, you?re welcome to email me your response. Reflective Writing Assignment #1 for Academic Summary Please respond to the following prompts: 1. Describe your writing process for this assignment. a. How did you move from your reading and analysis of the Harris essay to the completed and final draft? b. What did your first draft look like compared to your final draft? c. How did the revision workshop shape your drafting process? 2. What did you think was successful about your writing (what did you do well)? 3. What was challenging about the assignment? 4. Focus on Style: How did you work on the strategy of cohesion in your summary: 5. What would you like me to notice and comment on? 271 Reflective Writing Assignment #2 for Inquiry Essay Please respond to the following prompts: 1. Review the responses I?ve made to your Summary and Inquiry essay. What am I asking you to focus on? What patterns do you see? What differences do you see? 2. Review the comments from your peers in both draft workshops, what are you noticing about your writing given the feedback you received from your peers? 3. Re-read your revision plan from your draft workshop. What did you want to focus on then? How did you carry out that plan? What might you have done instead? 4. What did you think was successful about your writing (what did you do well)? 5. What did you struggle with? Why? 6. Style in Focus: How and where did you work on hedging and boosting? How effective were these attempts? Why? 7. How is class going for you thus far? What questions or concerns do you have? Reflective Writing Assignment #3 for Rhetorical Analysis Please respond to the following prompts: 1. Think back over the class activities for this unit, what learning experiences helped you as you drafted and revised this essay? Be specific. 272 2. What strengths do you see in your Rhetorical Analysis? 3. What did you struggle with? 4. What would you like me to notice and respond to? 5. Style in Focus: How did you incorporate what you learned about paragraphing into your draft? Point to specific places in your essay where you leveraged this knowledge and skill. 6. Given what you?ve learned from the Rhetorical Analysis, what are your goals for the next writing assignment? Reflective Writing Assignment #4 for Digital Forum Please respond to the following prompts: 1. What new knowledge or skills did you gain by working in a digital context? What was exciting, interesting, or challenging about working within this environment? 2. How did you manipulate the affordances of the website? Why did you make these choices? 3. What lines of argument do you employ in developing your stakeholder positions? How did using these lines of argument broaden your sense of the argumentative choices available to you? 4. What strengths do you see in your website/forum? 273 5. Style in Focus: How did you leverage your knowledge about language varieties as you wrote and revised your Digital Forum? Be specific. 6. What did you struggle with? 7. What would you like me to notice and respond to? Reflective Writing Assignment #5 for Position Paper Please respond to the following prompts: 1. What are the strengths of this essay? What are the weaknesses? 2. How did you take into account my feedback on your earlier assignments? 3. How did the Inquiry essay, Rhetorical Analysis, and Digital Forum affect the way you wrote the Position Paper? 4. Think back over the class activities for this unit, what learning experiences helped you as you drafted and revised this essay? Be specific. 5. Style in Focus: How do you maintain global coherence through the argument? What did you do to make sure that your supporting arguments clearly connect back to and extend the thesis? Revision and Reflection Assignment Sheet The Assignment: You will substantially revise one of your major assignments: your Inquiry essay, Rhetorical Analysis, or Digital Forum?the choice is yours! You will also compose a 2-page single-spaced reflective memo detailing the revisions you made and explaining the rationale behind the changes. 274 Purpose: As we?ve discussed throughout the semester, revision and reflection are key components to growing as a writer and thinker. This final assignment for the course is an opportunity to enact the revision strategies you?ve been learning all semester in a sustained way by revising an essay of your choosing. In addition to this revision, you will also compose a reflective memo that details the changes you?ve made in the revision and the reasoning behind these changes. In the memo, you want to explain not just what you revised but why you revised in the way you did. Audience: I am the audience for the reflective memo, so you should direct your comments to me. Writing the Revision and Reflection 1. To guide your revision, you will consider the rhetorical concepts and strategies you?ve learned throughout the semester, review the comments I have made on your documents, reconsider the suggestions students have made on your drafts, and reflect on your own ideas and concerns that you articulated in your reflective writing assignments. 2. You will devise a revision plan based on these observations you?ve made about your writing and carry out your revision. Your revision will demonstrate your assessments of your writing and your mastery over the skills you?ve identified as critical to the revision. The idea here is that you show how you?ve developed and fine-tuned your rhetorical skills over the course of the semester and that you can now use these new understandings to revise your work. Your revised essay should reflect your end-of-the-semester knowledge and rhetorical expertise. 3. The revisions you make to the project of your choice must be substantive and effective. While you will be expected to bring our style and grammar work to bear on this revision, these are not the only changes you should be making. This means you are not only editing your essay. Rather, your work in this revision is to rethink major parts or aspects of the essay such as appeals, arrangement, introductions and conclusions, integration of research and so on. The goal is to create a stronger essay by tackling a major rhetorical concern in your revision. 4. After you have completed your revision, you will compose the reflective memo. The expectation here is that you explain to me your goals for the revision and discuss the specific changes you?ve made to meet these goals and the reasoning behind these changes. For example, you may find that your arrangement in your Rhetorical Analysis lacked cohesiveness and purpose, and after working on arrangement during the Position Paper unit, you can now devise a better arrangement strategy. In your reflective memo, you would discuss how you thought through issues related to cohesiveness and purpose, how and why you identified arrangement as a possible solution to these problems, and how you created a new organizational strategy that allowed for greater sense of purpose and cohesion. 5. Detailed discussions in your memo are vital because they highlight for me the thinking and decision-making that directed your revisions. Here is where I 275 gain a sense that you are making conscious, careful, and rhetorically effective decisions. Furthermore, the idea behind the reflective memo is that you offer me a guide to reading your revision. In essence you?re saying, ?Here?s what I did and here?s why I did it.? There are a variety of ways you can incorporate detail into your memo: You can offer evidence of the changes you?ve made by quoting your earlier and revised drafts. You can bring in the specific advice I have offered or suggestions you received from peers or tutors at the Writing Center. Or you may include identification of specific rhetorical strategies discussed in class or through readings. The idea is that you want to show me the specific reasons why you made the revisions you did. 276 Appendix B Reflective moves in the Corpus of Reflective Writing Words % of Reflective move words in full corpus Writerly Identity 9026 19.55 % Goals and aspirations 1590 3.44 % Declaration of writing knowledge 2601 5.63 % Processes 3138 6.80 % Definition 2863 6.20 % Engagement with Feedback 7013 15.19 % Appealing to the teacher 423 0.91 % Assessing teacher feedback 1154 2.50 % Directing teacher feedback 1129 2.44 % Integrating teacher feedback 1129 2.44 % Peer feedback source 3239 7.01 % Reflection 26044 56.41 % Action 21458 46.48 % Assessment of action 7420 16.07 % Rationale for action 5256 11.39 % Possibility of different action 684 1.48 % Realization through reflection 1754 3.80 % Expression of emotion 11081 24.0 % 277 Bibliography Aijmer, Karin. English Discourse Particles: Evidence from a Corpus. John Benjamins, 2002. Allan, Elizabeth G., and Dana Lynn Driscoll. ?The Three-Fold Benefit of Reflective Writing: Improving Program Assessment, Student Learning, and Faculty Professional Development.? Assessing Writing, vol. 21, 2014, pp. 37?55. Ambrose, Susan, Michael W. Bridges, Michele DiPietro, Marsha C. Lovett, and Marie K. Norman. How Learning Works. John Wiley & Sons, 2010. American National Corpus Project. Open American National Corpus. 2015, http://www.anc.org/. Anonymous Seguerianus. ?Art of Political Speech.? Two Greek Rhetorical Treatises from the Roman Empire: Introduction, Text, and Translation of the Arts of Rhetoric Attributed to Anonymous Seguerianus and to Apsine of Gadara, edited by Mervin R. Dilts and George A. Kennedy, Brill, 1997. Anscombre, Jean-Claude, and Oswald Ducrot. L?argumentation Dans La Langue. Mardaga, 1983. Anthony, Laurence. AntGram. 1.0, Waseda University, 2018. Aristotle. On Rhetoric: A Theory of Civic Discourse. Translated by George A. Kennedy, 2nd ed., Oxford UP, 2007. Aull, Laura, and Zak Lancaster. ?Linguistic Markers of Stance in Early and Advanced Academic Writing: A Corpus-Based Comparison.? Written Communication, vol. 31, no. 2, 2014, pp. 151?83. 278 Barcelona, Anthony. ?The Case for a Metonymic Basis of Pragmatic Inferencing: Evidence from Jokes and Funny Anecdotes.? Metonymy and Pragmatic Inferencing, edited by Klaus-Uwe Panther and Linda L. Thornburg, John Benjamins, 2003. Barrett, Lisa Feldman. ?Solving the Emotion Paradox: Categorization and the Experience of Emotion.? Personality and Social Psychology Review, vol. 10, no. 1, 2006, pp. 20? 46. ---. ?The Future of Psychology: Connecting Mind to Brain.? Perspectives in Psychological Science, vol. 4, no. 4, 2009, pp. 326?39. ---. How Emotions Are Made: The Secret Life of the Brain. Houghton Mifflin Harcourt, 2017. Beaufort, Anne. ?Reflection: The Metacognitive Move toward Transfer of Learning.? A Rhetoric of Reflection, edited by Kathleen Blake Yancey, UP of Colorado, 2016, pp. 23?41. Bergen, Ben. Louder Than Words: The New Science of How the Mind Makes Meaning. Basic Books, 2012. Bergen, Ben, and Kathryn Wheeler. ?Grammatical Aspect and Mental Simulation.? Brain & Language, vol. 112, 2010, pp. 150?58. Biber, Douglas. ?Stance in Spoken and Written University Registers.? Journal of English for Academic Purposes, vol. 5, no. 2, 2006, pp. 97?116. ---. Variation Across Speech and Writing. Cambridge UP, 1988. Biber, Douglas, and Edward Finegan. ?Styles of Stance in English: Lexical and Grammatical Marking of Evidentiality and Affect.? Text, vol. 9, 1989, pp. 93?124. 279 Biber, Douglas, Molly Anthony, Ulla Connor, Thomas Upton, and Kostyantyn Gladkov. ?Rhetorical Appeals in Fundraising.? Discourse on the Move: Using Corpus Analysis to Describe Discourse Structure, John Benjamins, 2007, pp. 120?51. Bitzer, Lloyd. ?Aristotle?s Enthymeme Revisted.? Quarterly Journal of Speech, vol. 45, 1959, pp. 409?14. Black, Richard. ?An Engaging Dissection of Disaster.? BBC News, 15 Sept. 2006, http://news.bbc.co.uk/2/hi/science/nature/5348692.stm. Blair, Carole. ?Reflections on Criticism and Bodies: Parables from Public Places.? Western Journal of Communication, vol. 65, no. 3, 2001, pp. 271?94. Bloom, Paul. Against Empathy: The Case for Rational Compassion. Ecco, 2016. Bondi, Marina. ?Keywords and Emotions: A Case Study of the Bloody Sunday Inquiry.? Discourse and Contemporary Social Change, edited by Norman Fairclough, Peter Lang, 2007, pp. 407?32. Bostdorff, Denise M. ?Obama, Trump, And Reflections n The Rhetoric Of Political Change.? Rhetoric & Public Affairs, vol. 20, no. 4, 2017, pp. 695?706. Brown, David W. Clinton-Trump Corpus. 2017, http://www.thegrammarlab.com. Brown, Penelope, and Stephen C. Levinson. Politeness: Some Universals in Language Usage. Cambridge UP, 1987. Burke, Kenneth. A Grammar of Motives. U California P, 1969. ---. Counter-Statement. U of California P, 1968. Caffi, Claudia. Mitigation. Elsevier, 2007. Campbell, George. The Philosophy of Rhetoric. Edited by Lloyd Bitzer, Southern Illinois UP, 1963. 280 Castell, Alburey. A College Logic. Macmillan, 1935. Chafe, Wallace. Discourse, Consciousness, and Time. U of Chicago P, 1994. Charland, Maurice. ?Constitutive Rhetoric: The Case Of The Peuple Qu?b?cois.? Quarterly Journal of Speech, vol. 73, no. 2, 1987, pp. 133?50. [Cicero]. Rhetorica Ad Herennium. Translated by Harry Caplan, Loeb Classical Library, Harvard UP, 1954. Cicero. The Verrine Orations. Translated by L. H. G. Greenwood, Vol. 2, Harvard UP, 1935. Cicero, Marcus Tullius. ?De Partitione Oratoria.? De Oratore Book 3?; De Fato?; Paradoxa Stoicorum?; De Partitione Oratoria, translated by Harris Rackham and EW Sutton, Harvard UP, 2014. Clark, Herbert. Using Language. Cambridge UP, 1996. Conley, Thomas M. ?The Enthymeme in Perspective.? Quarterly Journal of Speech, vol. 70, 1984, pp. 168?87. Connor, Ulla, and Kostyantyn Gladkov. ?Rhetorical Appeals in Fundraising Direct Mail Letters.? Discourse in the Professions: Perspectives from Corpus Linguistics, edited by Ulla Connor and Thomas Upton, John Benjamins, 2004, pp. 257?86. Connor, Ulla, and Thomas Upton. ?Linguistic Dimensions of Direct Mail Letters.? Corpus Analysis: Language Structure and Language Use, edited by Pepi Leistyna and Charles F. Meyer, Brill, 2016, pp. 71?86. Croft, William, and D. Alan Cruise. Cognitive Linguistics. Cambridge UP, 2004. Crowley, Sharon. Towards A Civil Discourse: Rhetoric and Fundamentalism. U of Pittsburgh P, 2006. 281 Dancygier, Barbara. The Language of Stories. Cambridge UP, 2012. Dancygier, Barbara, and Eve Sweetser. Figurative Language. Cambridge UP, 2014. ---. Mental Spaces in Grammar: Conditional Constructions. Cambridge UP, 2005. Darwin, Charles. The Expression of the Emotions in Man and Animals. John Murray, 1872. Davies, Mark. Corpus of Contemporary American English. 2019, https://corpus.byu.edu/coca/. Dedoose. 8.0.35, SocioCultural Research Consultants, 2018. Dewey, John. How We Think. Heath, 1910. ---. How We Think. D.C. Heath & Co., 1910. Dimmock v Secretary of State for Education & Skills. CO/3615/2007, 10 Dec. 2007. Dixon, Thomas. ??Emotion?: One Word, Many Concepts.? Emotion Review, vol. 4, no. 4, 2012, pp. 387?88. ---. ??Emotion?: The History of a Keyword in Crisis.? Emotion Review, vol. 4, no. 4, 2012, pp. 338?44. Dobbs, Michael. ?An Inconvenient Truth for Al Gore.? The Washington Post, 12 Oct. 2007, http://voices.washingtonpost.com/fact- checker/2007/10/an_inconvenient_truth_for_al_g_1.html. Du Bois, John. ?The Stance Triangle.? Stancetaking in Discourse: Subjectivity, Evaluation, Interaction, edited by Robert Englebretson, John Benjamins, 2007, pp. 139?82. Dur?n, Juan I., Rainer Reisenzein, and Jos?-Miguel Fern?ndez-Dols. ?Coherence Between Emotions and Facial Expressions: A Research Synthesis.? The Science of Facial 282 Expressions, edited by Jos?-Miguel Fern?ndez-Dols and James A. Russell, Oxford UP, 2017, pp. 107?32. Ekman, Paul. Emotions Revealed: Understanding Faces and Feelings. Owl Books, 2007. Fahnestock, Jeanne. ?Rhetoric in the Age of Cognitive Science.? The Viability of Rhetoric, edited by Richard Graff, SUNY P, 2005, pp. 159?79. ---. Rhetorical Style: The Use of Language in Persuasion. Oxford UP, 2011. Fauconnier, Gilles. Mental Spaces: Aspects of Meaning Construction in Natural Language. MIT Press, 1985. Fauconnier, Gilles, and Mark Turner. The Way We Think: Conceptual Blending And The Mind?s Hidden Complexities. Basic Books, 2002. Fillmore, Charles J. ?Frame Semantics.? Linguistics in the Morning Calm, edited by The Linguistics Society of Korea, Hanshin, 1982, pp. 111?37. ---. ?Frames and the Semantics of Understanding.? Quaderni Di Semantica, vol. 6, 1985, pp. 222?54. Finch, Sam Dylan (@samdylanfinch). ?I want to explain what we're talking about when we say that transgender people need anti-discrimination protections in health care. Because I think some of you really don't understand what this means, or might perceive our reaction as dramatic. But it's not.? 24 May 2019, 12:58 p.m. Tweet. ---. ?I remember when I lived in Michigan and a trans woman that I knew had to call 911, because a serious wound she had (from an unrelated medical condition) started hemorrhaging overnight.? 24 May 2019, 1:01 p.m. Tweet ---. ?One of the EMTs went upstairs to her room, and when that EMT realized she was transgender, was visibly disgusted and left the room. Her mom overheard the EMT 283 mocking her as he spoke to the other EMT, referring to her as an ?it.?? 24 May 2019, 1:03 p.m. Tweet. ---. ?But the worst part of it was that, when it was decided that she needed to go to the hospital, they made her walk down the stairs herself without helping her. She was hemorrhaging blood from a leg wound. They stood impatiently and just watched her struggle.? 24 May 2019, 1:05 p.m. Tweet. ---. ?That's not even the worst story that I know of. A transgender woman in DC named Tyra Hunter was seriously injured in a car accident, and ambulance workers, upon realizing she was trans, outright refused to treat her. They mocked her. She later DIED from those wounds.? 24 May 2019, 1:06 p.m. Tweet. ---. ?A trans man named Robert Eads died of ovarian cancer after TWENTY SEPARATE DOCTORS refused to treat him. Lambda Legal reported that one of the doctors said the cancer diagnosis should make Eads ?deal with the fact that he is not a real man.? 24 May 2019, 1:08 p.m. Tweet. ---. ?I have known trans people who have been mocked while they were gravely ill in a hospital bed. I have known trans people who were outright turned away by doctors, or have had pharmacists refuse to fill prescriptions that were desperately needed.? 24 May 2019, 1:10 p.m. Tweet. ---. ?I need you to understand that when we say that these protections are a matter of life and death, we mean that LITERALLY. Transgender people have already died in utterly preventable and tragic ways because medical "professionals" turned their backs on us when we needed help.? 24 May 2019, 1:11 p.m. Tweet. 284 ---. ?When I was psychiatrically hospitalized, I had transphobic providers who REFUSED to give me access to my hormones, claiming that perhaps it wasn't my obsessive- compulsive disorder that was causing my mental anguish, but testosterone.? 24 May, 2019, 1:13 p.m. Tweet. ---. ?This is not about politics. This is about fundamental human rights. I want you to imagine getting into a serious car accident, and as you are literally dying before someone's eyes, they are MOCKING you when you thought they had come to help you.? 24 May, 2019, 1:14 p.m. Tweet. ---. ?I want you to imagine getting a cancer diagnosis, and going to doctor after doctor, TWENTY TIMES, to no avail. Imagine one callously remarking that maybe the cancer would teach you a lesson. Imagine the time is ticking, and no matter how much you plead, no one will help.? 24 May, 2019, 1:16 p.m. Tweet. ---. ?I want you to imagine the humiliation of hemorrhaging blood, and being made to crawl down a staircase, while two ambulance workers that you called for help refer to you as ?it? and look at you with disgust.? 24 May, 2019, 1:18 p.m. Tweet. ---. ?I don't care about your ?politics? right now. Fuck politics. This is whether or not you would cosign that level of cruelty. Whether the lives lost, the families mourning, the suffering endured, truly mean nothing to you.? 24 May, 2019, 1:20 p.m. Tweet. ---. ?And these are just the stories we know of. There are more transgender people who simply didn't call 911, didn't go to the doctor, didn't ask for help because they truly believed that no one would help them. Because it was too traumatic to take that risk again.? 24 May, 2019, 1:22 p.m. Tweet. 285 ---. ?And honestly, if the politics you abide by lead you to believe that this is in any way acceptable, your political affiliations are morally bankrupt and so are you. If your political allegiances lead you to that depth of cruelty, I have questions for you. Real fucking questions.? 24 May, 2019, 1:25 p.m. Tweet. ---. ?Do you believe that's how another human being should be treated? Do you believe, in a civil society, this is how medical professionals should behave? That someone should be left to die? Laughed at as they are helpless and afraid for their life? Would you wish that on someone?? 24 May, 2019, 1:26 p.m. Tweet. ---. ?And if you believe the stripping away of these protections is anything other than utterly barbaric, here's the reality: Your contempt for transgender people ? so much so that you would co-sign their deaths ? shows exactly which one of us is truly disgusting.? 24 May, 2019, 1:36 p.m. Tweet. Fortenbaugh, William W. Aristotle on Emotion: A Contribution to Philosophical Psychology, Rhetoric, Poetics, Politics and Ethics. 2nd ed., Bristol Classical Press, 2002. Fraser, Bruce. ?Pragmatic Competence: The Case of Hedging.? New Approaches to Hedging, edited by Gunther Kaltenbock, Wiltrud Mihatsch, and Stefan Schneider, Brill, 2010, pp. 15?34. Gibbs, Raymond W., Jr. ?Metaphoric Imagery as Embodied Cognition.? Journal of Mental Imagery, vol. 35, 2011, pp. 42?48. Giving USA: The Annual Report on Philanthropy for the Year 2017. Giving USA Foundation, 2018. Goffman, Erving. The Presentation of Self in Everyday Life. Doubleday, 1959. 286 Gorgias. ?Encomium of Helen.? The Greek Sophists, translated by John Dillon and Tania Gergel, Penguin, 2003. Grice, Paul. ?Logic and Conversation.? Studies in the Way of Words, Harvard UP, 1989, pp. 22?40. Gross, Daniel. The Secret History of Emotion: From Aristotle?s Rhetoric to Modern Brain Science. U Chicago P, 2006. Guggenheim, Davis. An Inconvenient Truth. Paramount Classics, 2006. Hawhee, Debra. ?Looking into Aristotle?s Eyes: Toward a Theory of Rhetorical Vision.? Advances in the History of Rhetoric, vol. 14, no. 2, 2011, pp. 139?65. ---. Rhetoric in Tooth and Claw: Animals, Language, and Sensation. Chicago UP, 2016. ---. ?Rhetoric?s Sensorium.? Quarterly Journal of Speech, vol. 101, no. 1, 2015, pp. 2?17. Holcomb, Chris, and M. Jimmie Killingsworth. Performing Prose: The Study and Practice of Style in Composition. Southern Illinois UP, 2010. Holmes, Janet. Gendered Talk at Work: Constructing Gender Identity through Workplace Discourse. Blackwell, 2006. ---. Women, Men and Politeness. Addison, Wesley, Longman, 1995. Hyland, Ken. ?Hedges, Boosters and Lexical Invisibility: Noticing Modifiers in Academic Texts.? Language Awareness, vol. 9, no. 4, 2000, pp. 179?97. Ihara, Rachel. ?Student Perceptions on Self-Assessment: Insights and Implications.? Teaching English in the Two-Year College, vol. 41, no. 3, 2014, pp. 223?38. Izard, Carroll E. ?The Many Meanings/Aspects of Emotion: Definitions, Functions, Activations, and Regulations.? Emotion Review, vol. 2, no. 4, 2010, pp. 363?70. 287 Jackson, Brian. ?Neuroscience and the Urgency of Emotional Appeals.? JAC, vol. 31, no. 3?4, 2011, pp. 473?96. Jakobson, Roman. ?Closing Statement: Linguistics and Poetics.? Style in Language, edited by Thomas A. Sebeok, MIT Press, 1960. Jensen, Tim. ?Textbook Pathos: Tracing a Through-Line of Emotion in Composition Textbooks.? Composition Forum, vol. 34, Summer 2016. Jespersen, Otto. The Philosophy of Grammar. Allen & Unwin, 1929. Johansson Falck, Marlene, and Raymond W. Gibbs Jr. ?Embodied Motivations for Metaphoric Meaning.? Cognitive Linguistics, vol. 23, 2012, pp. 251?72. Johnson-Laird, Philip N. Mental Models: Towards a Cognitive Science of Language, Inference, and Consciousness. Harvard UP, 1986. Johnstone, Barbara. ?Stance, Style, and the Linguistic Individual.? Stance: Sociolinguistic Perspectives, edited by Alexandra Jaffe, Oxford UP, 2009, pp. 29?52. Jones, Van. Election Night Coverage. CNN, 9 Nov. 2016. Kahneman, Daniel. Thinking, Fast and Slow. Farrar, Straus and Giroux, 2011. Kahneman, Daniel, and Amos Tversky. ?The Psychology of Preferences.? Scientific American, vol. 246, no. 1, 1982, pp. 160?73. Kennedy, George A. Comparative Rhetoric: An Historical and Cross-Cultural Introduction. Oxford UP, 1997. Kessler, Glenn, and Mihcelle Ye Hee Lee. ?Fact-Checking Donald Trump?s Acceptance Speech at The 2016 RNC Convention.? The Washington Post, 22 July 2016, https://www.washingtonpost.com/news/fact-checker/wp/2016/07/22/fact-checking- donald-trumps-acceptance-speech-at-the-2016-rnc/. 288 Kilgarriff, Adam. ?Simple Maths for Keywords.? Proceedings, Corpus Linguistics, edited by M Mahlberg, V Gonzalez-Diaz, and C Smith, 2009, https://www.kilgarriff.co.uk/Publications/2009-K-CLLiverpool-SimpleMaths.doc. K?vecses, Zolt?n. Metaphor and Emotion: Language, Culture, and Body in Human Feeling. Cambridge UP, 2000. Lakoff, George. Don?t Think of an Elephant. Chelsea Green Publishing, 2004. ---. ?Hedges: A Study in Meaning Criteria and the Logic of Fuzzy Concepts.? Papers from the Eighth Regional Meeting, edited by Paul Peranteau, Judith Levi, and Gloria Phares, Chicago Linguistics Society, 1972, pp. 183?228. ---. Moral Politics. U of Chicago P, 1996. ---. Women, Fire, and Dangerous Things. U Chicago P, 1987. ---. Women, Fire, and Dangerous Things: What Categories Reveal about the Mind. Chicago UP, 1987. Lakoff, George, and Mark Johnson. Metaphors We Live By. Chicago UP, 1980. Lakoff, Robin. Language and Woman?s Place: Text and Commentaries. Edited by Mary Bucholtz, Revised Edition, Oxford UP, 1975. ---. ?The Logic of Politeness, or: Minding Your P?s and Q?s.? Papers from the Regional Meeting of the Chicago Linguistic Society 9, edited by Claudia Corcum, T. Cedric Smith-Stark, and Ann Weiser, Chicago Linguistics Society, 1973, pp. 292?305. Langacker, Ronald. ?An Introduction to Cognitive Grammar.? Cognitive Science, vol. 10, 1986, pp. 1?40. ---. Cognitive Grammar: A Basic Introduction. Oxford UP, 2008. ---. Concept, Image, and Symbol: The Cognitive Basis of Grammar. De Gruyter, 1991. 289 ---. ?Constructing The Meanings of Personal Pronouns.? Aspects of Meaning Construction, edited by G?nter Radden, Klaus-Michael K?pcke, and Peter Siemund, John Benjamins, 2007, pp. 171?87. ---. Essentials of Cognitive Grammar. Oxford UP, 2013. ---. Foundations of Cognitive Grammar: Descriptive Application. Vol. 2, Stanford UP, 1991. ---. Foundations of Cognitive Grammar: Theoretical Prerequisites. Vol. 1, Stanford UP, 1987. ---. Grammar and Conceptualization. Mouton de Gruyter, 2000. ---. Investigations in Cognitive Grammar. De Gruyter, 2007. ---. ?The English Present Tense.? English Language and Linguistics, vol. 5, 2001, pp. 251?73. Lausberg, Heinrich. Handbook of Literary Rhetoric. Edited by David E. Orton and R. Dean Anderson, Brill, 1998. LeDoux, Joseph E. ?Feelings: What Are They & How Does the Brain Make Them?? Daedalus, vol. 144, 2015, pp. 96?111. Levinson, Stephen C. Pragmatics. Cambridge UP, 1983. Lindenman, Heather, Martin Camper, Lindsay Dunne Jacoby, and Jessica Enoch. ?Revision and Reflection: A Study of (Dis)Connections between Writing Knowledge and Writing Practice.? College Composition and Communication, vol. 69, no. 4, 2018, pp. 581?611. Littlemore, Jeannette. Metonymy: Hidden Shortcuts in Language, Thought, and Communication. Cambridge UP, 2015. 290 [Longinus]. On Great Writing (On the Sublime). Translated by G.M.A Grube, Hackett Publishing, 1991. L?pez, Ian Haney. Dog Whistle Politics: How Coded Racial Appeals Have Reinvented Racism and Wrecked the Middle Class. Oxford UP, 2015. Lunde, Ingunn. ?Rhetorical Enargeia and Linguistic Pragmatics: On Speech-Reporting Strategies in East Slavic Medieval Hagiography and Homiletics.? Journal of Historical Pragmatics, vol. 5, no. 1, 2004, pp. 49?80. Matlock, Teenie. ?Fictive Motion and Cognitive Simulation.? Memory and Cognition, vol. 32, 2004, pp. 1389?400. McCabe, Deborah Brown, and Stephen M. Nowlis. ?The Effect of Examining Actual Products or Product Descriptions on Consumer Preference.? Journal of Consumer Psychology, vol. 13, no. 4, 2003, pp. 431?39. McEnery, Tony, Mark McGlashan, and Robbie Love. ?Press and Social Media Reaction to Ideologically Inspired Murder: The Case of Lee Rigby.? Discourse & Communication, vol. 9, no. 2, 2015, pp. 237?259. Mendelberg, Tali. The Race Card: Campaign Strategy, Implicit Messages, and the Norm of Equality. Princeton UP, 2001. Micciche, Laura R. Doing Emotion: Rhetoric, Writing, Teaching. Boynton/Cook, 2007. Mieszkowski, Katharine. ?Did Al Get the Science Right?? Der Spiegel, 21 July 2006. Miller, Susan. Trust in Texts: A Different History of Rhetoric. Southern Illinois UP, 2008. Mirhady, David. ?Aristotle?s Enthymeme, Thymos, and Plato.? Influences on Peripatetic Rhetoric: Essays in Honor of William W. Forenbaugh, edited by David Mirhady, Brill, 2007, pp. 53?64. 291 Moon, Gretchen Flesher. ?The Pathos of Pathos: The Treatment of Emotion in Contemporary Composition Textbooks.? A Way to Move: Rhetorics of Emotion and Composition Studies, edited by Dale Jacobs and Laura R. Micciche, Boynton/Cook, 2003, pp. 33?42. Mulderrig, Jane. ?Using Keywords Analysis in CDA: Evolving Discourses of the Knowledge of Economy in Education.? Education and the Knowledge-Based Economy in Europe, edited by Bob Jessop, Norman Fairclough, and Ruth Wodak, Brill, 2008, pp. 147?69. Myers, Marshall. ?The Use of Pathos in Charity Letters: Some Notes Toward A Theory and Analysis.? Journal of Technical Writing and Communication, vol. 37, no. 1, 2007, pp. 3?16. Neal, Michael. ?The Perils of Standing Alone: Reflective Writing in Relationship to Other Texts.? A Rhetoric of Reflection, edited by Kathleen Blake Yancey, UP of Colorado, 2016, pp. 64?83. Oakley, Todd. From Attention to Meaning: Explorations in Semiotics, Linguistics, and Rhetoric. Peter Lang, 2009. ---. ?The New Rhetoric and the Construction of Value: Presence, the Universal Audience, and Beckett?s ?Three Dialogues.?? RSQ, vol. 27, no. 1, 1997, pp. 47?68. O?Connell, Peter A. ?Enargeia, Persuasion, and the Vividness Effect in Athenian Forensic Oratory.? Advances in the History of Rhetoric, vol. 20, no. 3, 2017, pp. 225?251. Oliver, John. 2016 Republican National Convention. 77, HBO, 24 July 2016. Panksepp, Jaak. Affective Neuroscience: The Foundations of Human and Animal Emotions. Oxford UP, 1998. 292 Panther, Klaus-Uwe, and Linda L. Thornburg. ?A Cognitive Approach to Inferencing in Conversation.? Journal of Pragmatics, vol. 30, 1998, pp. 755?69. ---. Metonymies as Natural Inference and Activation Schemas: The Case of Dependent Clauses as Independent Speech Acts. Edited by Klaus-Uwe Panther and Linda L. Thornburg, John Benjamins, 2003, pp. 127?48. Pascual, Esther. Fictive Interaction: The Conversation Frame in Thought, Language, and Discourse. John Benjamins, 2014. Peck, Joann, and Terry L. Childers. ?To Have and to Hold: The Influence of Haptic Information on Product Judgments.? Journal of Marketing, vol. 67, no. 2, 2003, pp. 35?48. Penman, William, and Doug Cloud. ?How People Make Sense of Trump and Why It Matters for Racial Justice.? Journal of Contemporary Rhetoric, vol. 8, no. 1/2, 2018, pp. 107?46. Pilkington, Ed. ?Feel The Love, Feel The Hate ? My Week In The Cauldron of Trump?s Wild Rallies.? The Guardian, 1 Nov. 2018, https://www.theguardian.com/us- news/2018/nov/01/trump-rallies-america-midterms-white-house. Plett, Heinrich. Enargeia in Classical Antiquity and the Early Modern Age. Brill, 2012. Poster, Carol. ?A Historical Recontextualization of the Enthymeme.? Rhetoric Society Quarterly, vol. 22, no. 2, 1994, pp. 45?65. Prince, Ellen F., Joel Frader, and Charles Bosk. ?On Hedging in Physician-Physician Discourse.? Linguistics in the Professions, edited by Robert di Pietro, Ablex, 1982, pp. 83?97. 293 Quandahl, Ellen. ?A Feeling for Aristotle: Emotion in the Sphere of Ethics.? A Way to Move: Rhetorics of Emotion and Composition Studies, edited by Dale Jacobs and Laura R. Micciche, Boynton/Cook, 2003, pp. 11?22. Quintilian. Institutio Oratoria. Translated by Harold Edgeworth Butler, Loeb Classical Library, vol. 1, Harvard UP, 1921. ---. Institutio Oratoria. Translated by Harold Edgeworth Butler, Loeb Classical Library, vol. 2, Harvard UP, 1921. ---. Institutio Oratoria. Translated by Harold Edgeworth Butler, Loeb Classical Library, vol. 3, Harvard UP, 1922. ---. Institutio Oratoria. Translated by Harold Edgeworth Butler, Loeb Classical Library, vol. 4, Harvard UP, 1922. Quirk, Randolph, Sidney Greenbaum, Geoffrey Leech, and Jan Svartvik. A Comprehensive Grammar of the English Language. Longman, 1985. Rakoczy, Hannes, Michael Tomasello, and Tricia Striano. ?Young Children Know That Trying Is Not Pretending: A Test of the ?Behaving-As-If? Construal of Children?s Early Concept of Pretense.? Developmental Psychology, vol. 40, no. 3, 2004, pp. 388?99. Renon, Luis Vega. ?Aristotle?s Endoxa and Pluasible Argumentation.? Argumentation, vol. 12, no. 1, 1998, pp. 95?113. Roberts-Miller, Patricia. Demagoguery and Democracy. The Experiment, 2017. Russell, James A., and Lisa Feldman Barrett. ?Core Affect, Prototypical Emotional Episodes, and Other THings Called Emotion: Dissecting the Elephant.? Journal of Personality and Social Psychology, vol. 76, no. 5, 1999, pp. 805?19. 294 Sch?n, Donald. The Reflective Practitioner: How Professionals Think in Action. Basic Books, 1983. Schrauf, Robert W., and Julia Sanchez. ?The Preponderance of Negative Emotion Words in the Emotion Lexicon: A Cross-Generational and Cross-Linguistic Study.? Journal of Multilingual and Multicultural Development, vol. 25, no. 2?3, 2004, pp. 266?84. Scott, Mike. ?Comparing Corpora and Identifying Key Words, Collocations, and Frequency Distributions Through the WordSmith Tools Suite of Computer Programs.? Small Corpus Studies and ELT: Theory and Practice, edited by Mohsen Ghadessy, Alex Henry, and Robert L. Roseberry, John Benjamins, 2001, pp. 47?67. ---. Wordsmith Tools. Version 7, Lexical Analysis Software, 2017. Semple, Kirk. ?Blistered Feet and Sleepless Nights: A Fraying Exodus Presses On.? The New York TImes, 26 Oct. 2018, p. A7. Siegel, Ethan. ?Newt Gingrich Exemplifies Just How Unscientific America Is.? Forbes, Aug. 2016, https://www.forbes.com/sites/startswithabang/2016/08/05/newt-gingrich- exemplifies-just-how-unscientific-america-is/#63441e815e47. Silver, Naomi. ?Refletive Pedagogies and the Metacognitive Turn in College Teaching.? Using Reflection and Metacognition to Improve Student Learning: Across the Disciplines, Across the Academy, edited by Matthew Kaplan, Naomi Silver, Danielle LaVaque-Manty, and Deborah Meizlish, Stylus, 2013, pp. 1?18. Sketch Engine. Lexical Computing, 2018, http://www.sketchengine.eu. Smith, Adam. The Theory of Moral Sentiments. Edited by Knud Haakonssen, Cambridge UP, 2009. 295 Sommers, Jeff. ?Problematizing Reflection: COnflicted Motives in the Writer?s Memo.? A Rhetoric of Reflection, edited by Kathleen Blake Yancey, UP of Colorado, 2016, pp. 271?87. Spahn, Claus. ?Irre Funkspr?che Aus Dem All.? Die Zeit, 40th ed., 27 Sept. 2001. Sperber, Dan, and Deidre Wilson. Relevance: Communication and Cognition. Blackwell, 1995. Stenberg, Shari. ?Teaching and (Re)Learning the Rhetoric of Emotion.? Pedagogy, vol. 11, no. 2, 2011, pp. 349?69. Talmy, Leonard. ?Force Dynamics in Language and Cognition.? Cognitive Science, vol. 12, no. 1, pp. 49?100. Tomasello, Michael. A Natural History of Human Thinking. Harvard UP, 2014. Tommasini, Anthony. ?The Devil Made Him Do It.? The New York Times, Sept. 2001, https://www.nytimes.com/2001/09/30/arts/music-the-devil-made-him-do-it.html. Trump, Donald J (@realdonaldtrump). ?Border Patrol Agents are not allowed to do their job at the Border because of ridiculous liberal (Democrat) laws like Catch & Release. Getting more dangerous. "Caravans" coming. Republicans must go to NOW Option to pass tough laws. NO MORE DACA DEAL!? 1 April, 2018, 7:56 a.m. Twitter. Tsuruta, Chikako. ?Trust in Trump in Translation.? Contemporary French and Francophone Studies, vol. 21, no. 5, 2017, pp. 592?97. Turner, Mark. The Literary Mind. Oxford UP, 1996. Tversky, Amos, and Daniel Kahneman. ?Availability: A Heuristic for Judging Frequency and Probability.? Cognitive Psychology, vol. 5, 1973, pp. 207?32. 296 ---. ?The Framing of Decisions and the Psychology of Choice.? Science, vol. 211, no. 30, 1981, pp. 453?57. United States Postal Service. A Marketer?s How-To (and Why-To) Guide to Using Direct Mail. 2012, https://www.usps.com/business/promotions/direct-mail-brochure.pdf. van Eemeren, Frans H., and Rob Grootendorst. A Systematic Theory of Argumentation: The Pragma-Dialectical Approach. Cambridge UP, 2004. Walker, Jeffrey. ?Enthymemes of Anger in Cicero and Thomas Paine.? Constructing Rhetorical Education, edited by Marie Secor and Davida Charney, Southern Illinois UP, 1992, pp. 357?81. ---. ?Pathos and Katharsis in ?Aristotelian? Rhetoric: Some Implications.? Rereading Aristotle?s Rhetoric, edited by Alan Gross and Arthur Walzer, Southern Illinois UP, 2000, pp. 74?92. ---. ?The Body of Persuasion: A Theory of the Enthymeme.? College English, vol. 56, no. 1, 1994, pp. 46?65. Walton, Douglas. Appeal to Pity: Argumentum Ad Misericordiam. SUNY P, 1997. ---. Appeal to Pity: Argumentum Ad Misericordiam. SUNY P, 1997. ---. Goal-Based Reasoning for Argumentation. Cambridge UP, 2015. ---. Scare Tactics: Arguments That Appeal to Fear and Threats. Kluwer Academic Publishers, 2000. ---. The Place of Emotion in Argument. Penn State UP, 1992. Warren, Beatrice. ?Aspects of Referential Metonymy.? Metonymy in Language and Thought, edited by Klaus-Uwe Panther and G?nter Radden, John Benjamins, 1999, pp. 121?37. 297 ?Was Donald Trump?s Speech Too Dark?? New Day, CNN, 22 July 2016, http://transcripts.cnn.com/TRANSCRIPTS/1607/22/nday.06.html. Webb, Ruth. Ekphrasis, Imagination and Persuasion in Ancient Rhetorical Theory and Practice. Ashgate, 2009. Westin, Monica. ?Aristotle?s Rhetorical Energeia: An Extended Note.? Advances in the History of Rhetoric, vol. 20, no. 3, 2017, pp. 252?61. White, Geoffrey. ?Disciplining Emotion.? Emotion Review, vol. 2, no. 4, 2012, pp. 375? 76. Wierzbicka, Anna. Emotions across Languages and Cultures: Diversity and Universals. Cambridge UP, 1999. ---. ?On Emotions and on Definitions: A Response to Izard.? Emotion Review, vol. 2, no. 4, 2012, pp. 379?80. Worsham, Lynn. ?Going Postal: Pedagogic Violence and the Schooling of Emotion.? JAC, vol. 18, no. 2, 1998, pp. 213?45. ---. ?Moving Beyond a Sentimental Education.? A Way to Move: Rhetorics of Emotion & Composition Studies, edited by Dale Jacobs and Laura R. Micciche, Boynton/Cook, 2003, pp. 161?63. Yancey, Kathleen Blake. ?Defining Reflection: The Rhetorical Nature and Qualities of Reflection.? A Rhetoric of Reflection, edited by Kathleen Blake Yancey, UP of Colorado, 2016, pp. 303?20. ---. ?Introduction: Contextualzing Reflection.? A Rhetoric of Reflection, edited by Kathleen Blake Yancey, UP of Colorado, 2016, pp. 3?22. ---. Reflection in the Writing Classroom. Utah State UP, 1998. 298 Yancey, Kathleen Blake, and Irwin Weiser. Situating Portfolios. Utah State UP, 1997. Zadvydas v. Davis, 533 U.S. 678. No.99-7791, 28 June 2001. Zwann, Rolf, Carol J. Madden, Richard H. Yaxley, and Mark E. Aveyard. ?Moving Words: Dynamic Representations in Language Comprehension.? Cognitive Science, vol. 28, no. 4, 2004, pp. 611?19. 299