@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix skos: . vivo:departmentOrSchool "Arts, Faculty of"@en, "English, Department of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Wittfoth, Monina"@en ; dcterms:issued "2011-01-10T21:00:09Z"@en, "2010"@en ; vivo:relatedDegree "Doctor of Philosophy - PhD"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """Based on the premise that modernity’s understanding of the linguistic sign has a long history dating back to ancient Greece when the linguistic mediation of knowledge preoccupied thinkers like Parmenides and Plato, this dissertation synthesizes contemporary post-structuralist and rhetorical understandings of language with like-minded findings of other fields of language study. It sees post-structuralist and deconstructive understandings of language as being congruent with the long tradition of rhetorical theory and the infamous linguistic turn in philosophy, that was initiated by the Tractatus Logico-Philosophicus, as a turn away from the actual phenomena of language towards an idealization of it. Nevertheless the thesis discovers recent findings by some of the beneficiaries of the “philosophy of language” that corroborate rhetorical theory’s insights. Inspired by both Derrida and Plato, this dissertation presents a rhetorical-deconstructive image of language that, recalling the root of the term skopevw (‘I look,’ ‘behold,’ ‘contemplate’), I characterize as sceptical. This study follows the theoretical matrix of de Man, Fish, Culler and Barbara Johnson, who are, of course, themselves following Derrida. It has a holistic attitude to language characteristic of the Bakhtin/Volsohinov approach, drawing insights from classical and contemporary rhetorical theory, post-structuralist theory, findings of systemic functional grammar, recent work in cognitive science and psychology on affect and language use, the scholarship of reported speech, and the ostensive-inferential theory of communication called Relevance. This cross border work in intellectual history, language theory and stylistics examines the interstices of theory and style in the work of figures such as Plato, Aristotle, Bacon, Nietzsche, Wittgenstein, Voloshinov, Bakhtin and Derrida. Sperber and Wilson, Scholars of Reported Speech and Halliday are central to its findings; the thinking of William James and Silvan Tomkins play supporting roles. It positions Plato as the founder of rhetorical theory and studies the voices of Plato and Derrida as language theorists. I examine both how Plato and Derrida talk about language and what theories of language underlie their styles, determining finally that their sceptical-rhetorical theories prompt their ironic-conversational stylistics."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/30556?expand=metadata"@en ; skos:note "Linguistic Returns: the currency of sceptical-rhetorical theory and its stylistic inscription in the Platonic and Derridian text by Monina Wittfoth B.A., The University of Toronto, 2000 M.A., The University of British Columbia, 2003 A DISSERTATION SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in The Faculty of Graduate Studies (English) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) October 2010 © Monina Wittfoth, 2010 ii Abstract Based on the premise that modernity’s understanding of the linguistic sign has a long history dating back to ancient Greece when the linguistic mediation of knowledge preoccupied thinkers like Parmenides and Plato, this dissertation synthesizes contemporary post-structuralist and rhetorical understandings of language with like-minded findings of other fields of language study. It sees post-structuralist and deconstructive understandings of language as being congruent with the long tradition of rhetorical theory and the infamous linguistic turn in philosophy, that was initiated by the Tractatus Logico-Philosophicus, as a turn away from the actual phenomena of language towards an idealization of it. Nevertheless the thesis discovers recent findings by some of the beneficiaries of the “philosophy of language” that corroborate rhetorical theory’s insights. Inspired by both Derrida and Plato, this dissertation presents a rhetorical-deconstructive image of language that, recalling the root of the term skopevw (‘I look,’ ‘behold,’ ‘contemplate’), I characterize as sceptical. This study follows the theoretical matrix of de Man, Fish, Culler and Barbara Johnson, who are, of course, themselves following Derrida. It has a holistic attitude to language characteristic of the Bakhtin/Volsohinov approach, drawing insights from classical and contemporary rhetorical theory, post-structuralist theory, findings of systemic functional grammar, recent work in cognitive science and psychology on affect and language use, the scholarship of reported speech, and the ostensive-inferential theory of communication called Relevance. This cross border work in intellectual history, language theory and stylistics examines the interstices of theory and style in the work of figures such as Plato, Aristotle, Bacon, Nietzsche, Wittgenstein, Voloshinov, Bakhtin and Derrida. Sperber and Wilson, Scholars of Reported Speech and Halliday are central to its findings; the thinking of William James and Silvan Tomkins play supporting roles. It positions Plato as the founder of rhetorical theory and studies the voices of Plato and Derrida as language theorists. I examine both how Plato and Derrida talk about language and what theories of language underlie their styles, determining finally that their sceptical-rhetorical theories prompt their ironic-conversational stylistics. iii Table of Contents Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii Table of Contents. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii Acknowledgements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v PART 1 1.1 paving the way. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.2 pretexts: Bacon, resistance and the style myth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.3 the road ahead. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 1.4 Wittgenstein & linguistic complexity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 1.5 the participants and other parts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 PART 2 2.1 adventures of signs in human communication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 2.2 Relevance Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 2.3 historical context: Bacon and the “plain style”. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 2.4 epistemological quagmire: affective rationality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 2.5 epistemological quagmire: analogy & knowledge. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 2.6 fuzzy logic and the biochemical image of grammar. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 2.7 epistemological quagmire: part-whole relations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 2.8 the inherent resistance to theory: de Man and rhetorical theory. . . . . . . . . . . . . . . . . . 71 PART 3 3.1 conn-texts & citation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 3.2 legacies: code-communication and the written language bias. . . . . . . . . . . . . . . . . . . . . . 86 3.3 legacies: performatives, seriousness & ordinary language. . . . . . . . . . . . . . . . . . . . . . . . . 89 3.4 intersections: language theory and reported speech. . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 3.5 scholarship of reported speech. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 3.6 verbatim’s fraudulent celebrity status. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 3.7 involvement strategies & affective interests in speech. . . . . . . . . . . . . . . . . . . . . . . . . . 102 3.8 from involvement to performativity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 3.9 Voloshinov’s sign: dancing in the caverns of our minds. . . . . . . . . . . . . . . . . . . . . . . . . 110 3.10 Nietzsche’s metaphorical turn. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 3.11 language’s bewildering complexity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 PART 4 4.1 stylistic turns. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124 4.2 the vicissitudes of “plain” style myths. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 4.3 Aristotle’s clarion call for clarity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 4.4 good writing and difficult texts: clarity and opacity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148 4.5 genre expectations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152 PART 5 5.1 “Platonism’s” voice-overs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 5.2 genre and the style of “propositional knowledge”. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163 5.3 genre and social relations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169 5.4 “Always with irony” (D 67). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175 5.5 themes, theses, and “the ability to follow the given thread” (D 63). . . . . . . . . . . . . . . 178 5.6 weaving, meaning, and le vouloir-dire. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188 5.7 reading Derrida writing about Plato . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193 iv 5.8 Derrida’s reader and the addressee of “Plato’s Pharmacy”. . . . . . . . . . . . . . . . . . . . . . 197 5.9 Poetry’s banishment re-visited. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202 5.10 some polemical mechanics of style . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210 5.11 self-referential text as discourse about discourse. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213 5.12 linguistic cues, ‘how things are said’ & meta-rhetoric. . . . . . . . . . . . . . . . . . . . . . . . . . . 220 Post Script. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237 Works Cited . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239 vAcknowledgements I would like to extend my sincere thanks to a number of friends and colleagues without whom this dissertation would not have been completed. I thank Julian Patrick and Francis Sparshott who got me started on this project way back in 1983. I am grateful to Iain Higgins, who ‘gave me a chance,’ to Nan Johnson who had confidence in me, and to Adam Frank who challenged me. I owe very special thanks to my supervisor at UBC, Janet Giltrow, for her guidance, intellectual engagement and friendship on this great adventure. I would also like to thank Gregory Renault for his early support of my academic aspirations and the ‘social-theory crowd’ who guided my first steps. I am very grateful to my family and Yukon friends for their faith and unwavering support, and to Wayne and Evelyn for lending their eyes to the task of editing. 1The notion that a theory might be extravagant is probably more controversial; but I have always seen this as a positive feature in the sense that there are more conceptual resources available than are necessitated for any particular task. This seems, perhaps, to go against the usual demand for parsimony, for “the simplest solution that is compatible with the facts”; and it is true that I see no great virtue in simplicity – I prefer the criterion of “the best tool for the job.” M.A.K. Halliday “Linguistics as Metaphor” (OL 265) Our discussion will be adequate if it has as much clearness as the subject-matter admits of, for precision is not to be sought for alike in all discussion . . . We must be content, then, in speaking of such subjects and with such premises to indicate the truth roughly and in outline, and in speaking about things which are only for the most part true and with premises of the same kind to reach conclusions that are no better. Aristotle Nicomachean Ethics (W.D Ross 1094b). 2 “The question of the sign is itself more or less, or in any event something other, than a sign of the times. To dream of1 reducing it to a sign of the times is to dream of violence. Especially when this question, an unexpectedly historical one, approaches the point at which the simple significative nature of language appears rather uncertain, partial, or inessential. It will be granted readily that the analogy between the structuralist obsession and the anxiety of language is not a chance one” (WD 3- 4). See the first section on “Ancient Rhetoric” in Friedrich Nietzsche on Rhetoric and Language. PART 1 1.1 paving the way ~ Inspired by both Derrida and Plato, this literary-language theory dissertation presents a rhetorical-deconstructive image of language that follows the theoretical matrix of Paul de Man, Stanley Fish, Jonathan Culler and Barbara Johnson, who, of course, are themselves following Derrida. However, despite Derrida’s intimation that the question of the sign is “an unexpectedly historical one,” and despite Nietzsche’s Lectures on Classical Rhetoric,the long history of this1 supposedly contemporary and post-structuralist understanding of language is not fully appreciated. Plato’s treatment of language – which is especially attentive to the epistemologically problematic but nevertheless creatively productive power of tropes and affect in language use – indexes the linguistic consciousness of his audience. Michael Naas’ account of persuasion (in Turning: From Persuasion to Philosophy) and the dance between trevpw (I turn) and peivqw (I persuade), together with its suggestive middle voice (I obey), outlines a movement which signals not only the human-inter-subjective element of linguistic interactions, but also the tropological nature of language’s reality. The linguistic trope both displaces human experience and filters and shapes our account of it. The trovpoV, or ‘turn,’ is a way, path, or direction, and a manner, fashion, or custom. It is both form and content, style and meaning, both the mechanism and the goal of persuasion, a bifurcated functionality emblematic of the linguistic complexity thematized in this thesis. The contours of the rhetorical approach to language that Naas traces in Homer’s Illiad present an image of language which may begin to seem oddly modern. The work of Naas (in philosophy), Schiappa and others, like Covino (working in the history of rhetoric) have contributed to revised histories of the birth of rhetorical thought. Rhetoric’s origins are no longer figured as American rhetorical theory has wished: neither as a response to the burgeoning democracies of antiquity, 3 See for example Leff and Welch on mis-readings of Aristotle, and Welch and Jarratt on revisions; compare with2 standard views in Kennedy; and note that Burke didn’t need to “extend” classical rhetoric nearly so much as he thought. Cornford remarkably concludes that the “underlying assumption” of Platonic “doctrine” is that “every common3 name must have a fixed meaning, which we think of when we hear the name spoken” (9). However, Gonzalez’s account of the history of Platonic scholarship suggests that this “tradition” likely has its start in the eighteenth century with the end of Neoplatonism Third (viii). Arthos’ investigation of “the distant ancient meaning of rhetoric” that Gadamer discusses at the end of his career4 speaks to this rhetorical affiliation with Plato and the epistemological focus of antiquity. Arthos observes that the “kairotic- performative rhetoric championed by Isocrates and Cicero . . . is for Gadamer . . . anchored squarely in Plato’s dialogic example” (172). Arthos argues that “the constraint upon access to knowledge disciplines metaphysics rather than rhetoric, and reorients the tradition to the modesty of its ontological competence” (172). This lines up with de Man’s account of Nietzsche’s critique of metaphysics. The contradictions in Aristotle’s linguistic discussions may attest to an interest in establishing a scientific approach5 to language despite its obviously rhetorical nature. De Rijk’s opus Aristotle: Semantics and Ontology comments on the relationship of language to thinking in Aristotle and Plato. “Considering the fact that authors like Plato and Aristotle believed language to be the key to unravelling the mysteries of our surroundings, we should approach them in a way that reflects this attitude; thus whatever they have to say about what they consider to be the case should be examined in the broader context of Language and Reality. The semantics indeed will turn out to be the mechanism of his way of thinking and explaining things, and should therefore be used as the key par excellence to understanding Aristotle’s thought” (10). Frede and Inwood’s collection highlights the special attention that was given to language because of “the interdependence between language and thought” and “the importance attributed to rhetoric and other forms of self-expression” (4). and not probably arising from the need to speak in the polis at all, but rather as part of the epistemological project of understanding the world that characterized the intellectual climate of Ancient Greece. In the days of Parmenides and Plato, when it was assumed that thinking2 entailed the use of language, language came under scrutiny as the medium of thought. Both the literariness and the insistent scrutiny of linguistic mediation in Plato’s dialogues have always interested me. The view of language inscribed throughout the Platonic text seemed to run completely counter to the understanding of language traditionally ascribed to “Platonism.” Schiappa’s conclusion that Plato coined the term rJhtorikhv (the art of rhetoric)3 prompts my consideration of Plato as the founder of rhetorical theory: not because he was advocating a new theoretical art of rhetoric, but because he coins this term to prompt discussion of language’s rhetorical nature, and because he explicitly analyzes the nature of the sign and takes great pains to show over and over again how it works (and doesn’t work). He is,4 moreover, clearly not discovering something about language that was not already known. From such a vantage Aristotle’s Rhetoric appears to be the codification of a centuries-old understanding of language. The fact then that Nietzsche’s Lectures on Ancient Rhetoric reveal5 a proto-post-structuralist understanding of language which has a lot in common with that of Plato and others becomes less surprising than it might otherwise seem. I wonder if modernity’s 4 See Halliday’s Introduction to Functional Grammar.6 convictions about progress and advancement have blinded it to the insights of previous eras. The similar strains in the Derridian and Platonic approaches to language have guided my language-theory investigations; but they have also profited from beneficiaries of analytic philosophy whose empirical, corpus-based work (although sometimes starting from code assumptions about language) led to conclusions that supported my post-structuralist image of language. My original intention to compile a brief renewed-rhetorical image of language as a lens for analyzing both the language theories and writing styles of Plato, Derrida, Burke and perhaps even Habermas anticipated a relationship between the theory of language and the theorist’s style. Austin-Grice beneficiaries like Sperber and Wilson are very helpful in this language-theory-style project; and the gesture of bringing these formerly duelling factions together seemed productive. The scholarship of reported speech was also a natural fit with Plato’s dialogues and the enigmatic and playful approach to other voices in the Derridian text. But rather than simply supporting my work with stylistic tools, these linguistic-pragmatics fields expanded its theoretical account which contracted its stylistic analysis. The swelling of the language theory section, which seems to index the fertility of bringing these diverse voices together, reduced stylistic analyses to the Platonic and Derridian texts. In addition, M.A.K. Halliday’s essay-type theoretical summaries provided me with a new and exciting image of grammar, as well as shedding light on some of the tensions within “mainstream” linguistics. To my surprise Halliday’s summarizing essays addressed in one way or another most of the main points I was advancing about how language works. Halliday’s grammatical writing had never inspired any eureka-moments for me before I encountered these theoretical essays; my experience reading his (albeit useful) cohesion analysis had been enervating. Nevertheless Halliday was considered by many to be the grammar guru of the late-6 twentieth and early-twenty-first centuries. He is the leading proponent of the now widely recognized systemic-functional grammar, although he represents himself as working within a broad network of grammarians with varied, open-ended systemic approaches. It is significant 5 John M. Ellis remarks that language theory is characterized by inconsistencies, amnesia, and a lack of rigour when it7 comes to dealing with issues that were already problematized by Wittgenstein and Saussure. for current purposes that as late as 1997 he still modestly positions against “mainstream linguistics” (OL 266). Halliday’s work endorsed my project’s speculative literary and philosophical perspective; this endorsement of the rhetorical-deconstructive image of language by a grammarian summarizing the theoretical implications of a lifetime of fine-grained, empirical studies of language at the grammatical level, calibrates an alignment of systemic- functional grammar, ancient linguistic insights, post-structuralist discourses, and the findings of postivistically oriented research into communication and reported speech. Halliday’s theoretical summaries cast a new light on an affiliation I had been making between “mainstream” linguistics and analytic philosophy, as well as qualifying my references to the term grammar. It is the mainstream designation here that is most problematic. Halliday takes Ellis’1993 Language, Thought and Logic to task precisely for treating the entire discipline of linguistics as being aligned with analytic philosophy’s rule-governed-grammar approach. Ellis mistakenly accuses “linguistics” of code-metaphor assumptions that systemic-functional grammar repudiates (OL 232-47). However, Ellis’s failure to recognize the work of systemic grammarians attests to the perplexity and diversity in language theory to which he himself is reacting. Arguing that language theory is the purview of the two disciplines philosophy and linguistics, Ellis believes that the fracturing of these into multiple fields has resulted in a reinventing-the-wheel tendency – to “begin again at the beginning” or “revolutionize” language theory (10-11). He complains that scholars working in these fields fail “to see what the history of linguistic theory had to say” about the assumptions on which their arguments are premised (5). Ellis would doubtless not make the mistake of overlooking systemic grammar now.7 Although Ellis and I have radically different approaches, the tenacious grip that the code metaphor has exerted in some fields of language theory over the past few centuries and the failure to see what was already known about language that he complains about has propelled my text as well. Halliday’s acknowledgement of Whorf’s great influence on his thinking is corroborated 6by Halliday’s work, which evinces the decidedly post-structuralist comportment that Whorf and Sapir have bequeathed to their language-theory legatees (OL 188). The contrast between the Whorf-and-Sapir orientation and the analytic-philosophy approach lines up with the distinction that Voloshinov makes between “two trends in linguistics”; Voloshinov tracks strains of linguistics amenable to post-structuralism back to Herder in the eighteenth century. Voloshinov’s albeit crude and repeatedly qualified distinction is a useful heuristic for gauging the positivist and post-structuralist orientations in linguistics. He defines the first trend, “individualist subjectivism,” with reference to the comparative-descriptive linguist Wihelm von Humboldt, who, he claims, was influenced by Johan Gottfried Herder (48). Interestingly, Herder’s belief that “thought and language are inseparable” carries on the thought-language ligature from antiquity that is frequently referenced in this dissertation (Honderich 352). Overall, the holistic and interdisciplinary approach of Voloshinov’s first trend is consistent with the approaches of both this thesis and Halliday’s summaries. Humboldt imagines language as a fluid, social and creative process, depending on laws of individual psychology (Voloshinov 49). Voloshinov attributes the second trend, “abstract objectivism,” to the rationalism of the seventeenth and eighteenth centuries, noting that “the ideas behind the second trend received their first . . . expression in Leibniz’s conception of universal Grammar” (57). He says “the second trend has profound inner connection with Cartesian thought and with the overall world view of neoclassicism and its cult of autonomous, rational, fixed form”; and he claims that “the idea of language as a system of conventional arbitrary signs of a fundamentally rational nature was propounded by representatives of the Age of the Enlightenment in the 18 century” (57-58).th Notwithstanding his own obvious admiration for Saussure and despite Saussure’s widespread influence both within linguistics and outside of it, Voloshinov identifies him with this abstract- objectivist trend. Saussure is exemplary here for having defined his “object of study” by isolating the linguistic sign from its social and psychological roots. Without diminishing the fruits of Saussure’s lucid analyzes, his gesture severs the linguistic sign from the life forces that 7 Similarly, Halliday’s esteem for Saussure does not prevent him from remarking the problem of separating langue8 and parole, and from asserting the sign’s lack of autonomy – a matter which Saussure clearly recognizes (OL 267, Saussure 1- 17). Halliday, for example, says that “language construes experience in terms of complementarities” (citing as an9 example terms which are both “bounded” and “unbounded”). For Halliday this means that language constructs reality in terms of binaries; and he considers this contradictory-binary richness part of the “extravagance of language” (OL 265). Not only is Derrida famously fascinated with language’s dependence on differentiation and binary pairs (like form and content) and the privilege ceded to one side of an opposition (like speech and writing), Halliday’s “extravagance” is very like the exorbitance or supplementariness that Derrida treats with such insistence in Of Grammatology and elsewhere. See Zappen 1989 and Lipson documentation of references linking Bacon to this style.10 create and shape it. Voloshinov perceives a genealogical linkage between Leibnizian universal8 grammar, conceptions of autonomous, rational, fixed forms, and rule-governed grammar. Rule- governed grammar – a set of logical, learnable rules that governs how language works and how we understand one another – wants the truth conditions of propositions to determine meaning. This rigid deterministic conception of grammar is radically different from the approach for which Halliday is the exemplar here. In Part 2, I will outline Halliday’s biochemical image of grammar and attempt to distinguish the (increasingly mainstream) grammarians and linguists, like Halliday, from those adhering to the rule-governed grammar that Leibniz imagines. This thesis somewhat arbitrarily refers to “mainstream linguistics” as excluding the important field of systemic grammar in order to distinguish analytic philosophy’s beneficiaries from those of Herder, Humbolt, Whorf and Sapir. Although I suspect that Halliday would side with Kenneth Burke in framing himself as a constructionist rather than a deconstructionist, Halliday nevertheless gets at many of the same kinds of issues about language that Derrida does, making him an excellent complement to the other language theorists convened in this dissertation. 9 1.2 pretexts: Bacon, resistance and the style myth ~ Francis Bacon has long and misleadingly been revered for having inaugurated the “style of science.” This style of writing, originally characterized as “plain,” has since Bacon’s time morphed into modern conceptions of simplicity, clarity and ultimately transparency: terms that not only mean entirely different things, but that are clustered together as though naturally affiliated and proclaimed as the writing ideal. Bacon’s style-of-science affiliation provides an10 opportunity to foreground two themes of interest to this thesis. On the one hand, scholars studying research writing, like Giltrow, find that this simple, clear style of science simply does 8 These scholars analyze a range of disciplinary research genre in science (like math and cell biology), social science11 and humanities disciplines. Giltrow, Hyland and Myers have analyzed in particular writing in the so called “hard sciences.” An application of Halliday’s cohesion analysis to these collocates would doubtless be a fruitful index of12 Enlightenment thinking about language. See Cohesion in English. not exist. Contrary to this imagined style of science, linguistic-pragmatics studies of research writing show that, completely apart from such obvious matters as register (or specialized lexicon), the research genres are strongly marked by disciplinary-ethos investments; they have rhetorical aims directed at such matters as disciplinary-knowledge construction and credibility, displays of respect, identification, and community-solidarity building; and they are addressed to specific and limited audiences – hence hardly “plain” or transparent to non-specialists (e.g., Hyland, Meyers Politeness, Swales, et al.). Giltrow concludes that “the language of science,” which is “seeded with social interests” designed to convey community credibility and the stance of neutrality, is anything but unstyled (Modern 173). The supposed transparent style of science11 is a myth: but a powerful value-shaping myth, and important here not only because of the prestige it claims for itself but because of the understanding of language it promotes. On the other hand, as far as Bacon himself is concerned, recent research in the history and rhetoric of science reveals that his views are not so straightforward. I believe that thinking of him in terms of this plain-style-of-science myth reduces and trivializes the linguistically insightful comments scattered throughout his writings. But the fact of his having been taken as Mr. Scientific Plain Style is evidence or measure of the force and currency of problematic conceptions of linguistic transparency, which wrongly affiliate plain, clear, simple and true.12 For example, consider the implications about language from the following passage from Novum Organum: But the Idols of the Market Place are the most troublesome of all – idols which have crept into the understanding through the alliances of words and names. For men believe that their reason governs words; but it is also true that words react on the understanding; and this it is that has rendered philosophy and the sciences sophistical and inactive. Now words, being commonly framed and applied according to the capacity of the vulgar, follow those lines of division which are most obvious to the vulgar understanding. And whenever an understanding of greater acuteness or a more diligent observation would alter those lines to suit the true divisions of nature, words stand in the way and resist the change. Whence it comes to pass that the high and formal discussions 9 Throughout this dissertation if emphasis added to citations it is noted parenthetically, as here.13 While it is in the context of Cratylus that Sallis comments that “it begins to become evident to what degree the14 entire enterprise is bound up in names, explaining names by other names, which themselves would always require to be explained by still other names, and so on without limit” (257), this discussion also appears elsewhere. See for example, Sophist 244d. The term sceptical can be taken as an adjective here, but the connections to ancient scepticism also exist, a matter to15 which we return. For the purposes of this thesis I will treat Sextus Empiricus’ (160-210 C.E.) comments as exemplary. Leo Groarke claims that Sextus Emipiricus “describes the ‘skeptic’ (from a Greek verb meaning ‘to examine carefully’) as an ‘investigator’ (a ‘zetetic’). According to Sextus, the skeptic is someone who has investigated the questions of philosophy but has ‘suspended judgment’ (practising epochê) because he is unable to resolve the differences among the contrary attitudes, opinions and arguments he found. Instead of adhering to a definite philosophical position, the skeptic is someone who continues to investigate.” Sextus particularly foregrounds “the skeptic’s open-minded attitude to the possibility of apprehending truth” – however at the same time sceptical arguments “raise deep questions about any claim to truth.” of learned men end oftentimes in disputes about words and names; with which (according to the use and wisdom of the mathematicians) it would be more prudent to begin, and so by means of definitions reduce them to order. Yet even definitions cannot cure this evil in dealing with natural and material things, since the definitions themselves consist of words, and those words beget others. So that it is necessary to recur to individual instances, and those indue series and order, as I shall say presently when I come to the method and scheme for the formation of notions and axioms. (emphasis added LIX)13 Like other thinkers from Plato to Derrida, Bacon quite typically comments here on such matters as 1) linguistic mediation, the shaping influence of words on human perception (“words react on the understanding”), 2) genre, register and audience design, the different meanings that words have in different types of discourse with different addressees (words that might be “framed and applied according to the capacity of the vulgar”), and 3) the problem of linguistic recursion mise en abîme (the open chain of signification). Bacon’s allusion to the troublesome and perpetual need to define words both echos Plato’s Cratylus and anticipates the Derridian critique:14 “definitions themselves consist of words, and those words beget [other words].” The disparity between Bacon’s rhetorical understanding of language and positivist assumptions underlying the plain style reveals certain language-theory misconceptions of interest to this thesis. Both Bacon’s views of language, and the fact that he is remembered for an opposite view, make him an exemplary site for introducing some of the features of language that are now associated with post-structuralism – but which actually have a history dating back to ancient sceptical views of language. The figuring of Bacon as Mr. Scientific Plain Style is15 paradigmatic of certain language-theory misconceptions that de Man explores in “Resistance to 10 Aristotle’s common topoi are generally understood as lines of argument, known commonplaces that are16 recognizable patterns of thought exploited by analogy to a matter at hand. While McKeon considers all topoi to be place of invention, Aristotle’s specific topoi are a subject of considerable debate. See McKeon Rhetoric, C. Miller Special Topics, Leff Topics, Leff Topoi, McAdon Signs. Theory.” Such uptake failures of rhetorical-sceptical understandings of language, which seem to gather critical mass at various historical moments of scientific over-exuberance, are of central concern to this thesis. Hence Bacon’s writing about writing and de Man’s resistance article offer useful opportunities (or sites) for raising certain issues about language and as a basis for getting started on this dissertation’s inquiry into language theory. I am borrowing classical rhetoric’s enigmatic metaphor tovpoi (‘places’) in order to describe the way that Bacon’s and de Man’s discussions about language and the plain-style myth itself are sites (or places) that carry with them useful contextualizing resources. I am using Bacon, the resistance metaphor, and the plain style as specific topoi (‘places’) for my discussion; exploiting them opportunistically, I use them as a pretext for discussing ancient linguistic insights, theory-uptake failures, problematic linguistic intuitions, Cratylian conceptions, code and conduit metaphors, etc. 16 According to de Man such discourses about language (as Bacon’s above) that theorize about linguistic signs engender a “resistance to theory” which he claims, “is a resistance to language about language” (13). (In this vein, we shall later see the way that talk-about-talk in the Platonic text has been largely overlooked.) Resistance to language theory has its roots in fantasies about human knowledge, as we shall see. It is also a resistance to thinking-about- language that arises from unexamined intuitions of linguistic transparency. De Man says, “it is a resistance to language itself [resistance to language’s tropological nature, for example] or to the possibility that language contains factors or functions that cannot be reduced to intuition” (13). Since language use seems intuitive, we assume de Man says “that when we refer to something called ‘language,’ we know what it is we are talking about” (13); and we seem to assume that we know how it works (without thinking about it). But intuitions are sometimes wrong. As we shall see, Sperber and Wilson’s inferential model of communication together with the linguistic- pragmatics scholarship on reported speech shows that only some uses of language are descriptive; and reference (a sign-to-thing correspondence) often plays an amorphous and 11 This is the conception of language parodied by Plato’s dialogue, Cratylus, named after the character in the dialogue17 who attempts unsuccessfully to defend this view. De Man’s term conventional, which somewhat correspond’s to Plato’s use of oJmologiva (agreement) in the Cratylus,18 is an alternative to Saussure’s term arbitrary. Derrida aptly observes that the sign’s “arbitrariness” is “grossly misnamed” (OG 44). He argues that only the structural “relationships between specific signifiers and specific signifieds would be regulated by arbitrariness” (OG 44). He suggests that there is a “discontinuity” in the structure, not the historical origin of signs (OG 326). Signs have a history; a sign’s meanings are governed by its history not arbitrariness. whimsical role. Completely apart from the problem of intuition’s empirical reliability, there is some question of whether “intuitions” about language are actually “natural” or whether they might be schooled and historical. The sender-to-receiver-transmission-of-information code- metaphor (kin to transparency’s conduit metaphor) may still be “intuitive” for some; but we shall see that, contrary to the hunches we might have, some operations of language work in surprising ways. This thesis aspires to complicate rather than simplify images of language, arguing for a linguistic complexity that defies being definitively codified. My dissertation aspires to promote a linguistic consciousness that defies being tied down into “a single guiding thesis,” as Derrida will say (D 7). It is on the basis of what he calls Hegel’s Cratylian conception of language – in which meaning is essentially embodied in names, names which, like onomatopoeic words, signify the essence of the thing named – that de Man indicts Hegel in “Resistance to Theory.” The17 Cratylian conception imagines a language in which “the phenomenality of the signifier, as sound, is unquestionably involved in the [referential] correspondence between the name and the thing named” (10); words denote, or refer to, things by some kind of essential tie between them, and the meaning of a term is anchored by the essence of the thing. But de Man points out that the “phenomenal” attribution is a mistake; “the [referential] relationship between the word and the thing is not phenomenal but conventional” (10). It is old news that both word and thing are18 phenomenal (that is, they are sound, text, event, concept, object), and that their referential relation to each other is conventional. De Man’s concern is to underscore that it is a mistake “to confuse the materiality of the signifier [sound/text/image] with the materiality of what it signifies [concept/thing]” – a mistake that is frequently made (11). De Man observes that the “conventional” relationship between word and thing gives language “considerable freedom from 12 referential restraint” (10). This “freedom from referential restraint” is not a complete absence of referential correspondence; it is a kind of referential promiscuity. More importantly, the problem de Man is concerned with is not language’s very real referential and descriptive functions, which to judge by their continuous service are relatively effective. Rather, twentieth-century philosophy of language’s and “mainstream” linguistics’ mistaken understandings of the status and roles of reference and description in language use are at issue. A confident attitude towards the singular importance and role of descriptive referentiality in language use cohabits with the code- metaphor intuition. Sperber and Wilson describe this sender-to-receiver-signal-transmission idea as “packing a content into words” and “sending it off to be unpacked by a recipient at the other end” (1). When the Chomskian-linguist Pinker, for instance, asks “what is the trick behind our ability to fill one another’s heads with so many different ideas?” – he makes it sound as though the idea were popping out of one mind, then scurrying over and jumping into another (1). The one-way transmission of information like a package is delivered to the hearer by the speaker; the speaker thus informs the hearer with a pre-packaged, descriptive message-content. Halliday flatly dismisses the code metaphor and the paramount status of reference: when language is reduced to the status of an (imperfect) formal system, a “code” to be measured by its correspondence (or lack of it) to some supposedly independent reality, whether material or mental, by reference to which its meanings are judged to be (or else not to be) “true,” the resulting vision is so impoverished that serious questions about language can hardly even be raised . . . (OL 237) The code metaphor depends on an essentialist, Cratylian conception of language, in which meaning inheres in the word itself – a conception that confounds our understanding of how language actually works and that Halliday thinks precludes serious questions about language. Although the de Manian account of language’s referential promiscuity may seem counterintuitive, it is consonant not only with Halliday’s perspective but also, as we have noted, with findings of linguistics and pragmatics scholars. Pyscholinguists Clark and Gerrig observe that although “most theories of language use take for granted that all language use is 13 This thesis uses the term “performative” in the broad sense of the term following Culler (1997). 19 See Richard Bett (Pyrrhonism), Cherniss, and Román Alcalá on the link between Plato and ancient scepticism, and20 see Naas on Homer’s rhetoricity. The various shades and schools of ancient and modern scepticism that are classified in ancient philosophy are not especially relevant to this discussion. My references to scepticism are to a “loose sense” (as Bett says in Nietzsche) of suspending judgement in one’s investigation – which importantly presupposes that one investigates. description,” reported speech is a type of “demonstration” which is not descriptive (764). These “demonstrative” functions of language (explored in detail in Part 3) push description and reference to the back seat, and are, I think, more aptly described by the contemporary expanded use of Austin’s term performative. In Halliday’s four metafunctions that define the dimensions19 of semantic space (experiential, interpersonal, logical and textual) it is the experiential and interpersonal functions that hold centre stage. He claims that the work of grammar is both to “construe” experience (by which he means construct) and to “enact” the interpersonal (Matter 63). Both of these metafunctional operations of grammar pertain to how language acts in the world that I consider the performative dimensions of language. De Man’s optimistic sounding “freedom from referential restraint” (or referential promiscuity) has the worrisome implication that language is epistemologically “suspect,” and its use no longer “determined by considerations of truth and falsehood” (10). These truth-status considerations pertain to language’s misunderstood relationship to logic (and thus to rule- governed grammar) as we shall see. Ellis calls this the “verification” theory of meaning based on the (Cratylian) belief that “words labelled things” that they “denoted” or to which they “referred” (4). De Man’s lighthearted characterization of referential unreliability as freedom playfully mocks the epistemological anxieties that might result when the reliability of the linguistic medium is undermined. Since the epistemological functions of language depend on description and reference, when language becomes epistemologically suspect our understanding of the world is at stake. For some this news might be disturbing and resistance-provoking. But such linguistic-mediation worries are an old story, dating back to ancient scepticism and stemming from the view of language evident in the incipient rhetoricity of the Homeric text.20 From a sceptical viewpoint our cognitive dependance on language together with language’s 14 Language’s metaphoricity – its fundamentally figurative or, as de Man calls it, tropological character – stems from21 its semiotic nature; signs are conventional phenomena (not essential emanations of things as Cratylians wish); rather they are images of images of things (and actions, relations, etc.), as we shall see in the discussion of Nietzsche. It is this shaping influence of linguistic mediation that prompts Halliday’s comments on neutrality and ideological22 inscription in “Is the Grammar Neutral?” Halliday suggests that a degree of deconstruction is always required in order to assess the perspective from which an observation is made (OL 286). fundamental metaphoricity make all human understanding linguistically mediated and flawed21 (or at least highly selective) as a result. Halliday doesn’t even acknowledge a purely referential linguistic ideal when he gleefully embraces language’s “semogenic power” – its meaning- making, construing-constructing-enacting effects (Matter 63). For sceptical language theorists the selective and shaping influence of language on human understanding (linguistic mediation) does not mean the sky is falling; it does not imply complete blindness; it simply means that humans are somewhat myopic and do not have X-ray vision. Language users need to take their vision limitations into account.22 Richard Bett’s investigation of Nietzsche’s relation to ancient scepticism observes that “Nietzsche always dissented from a certain traditional optimistic picture of the prospects for philosophy, or for inquiry more generally.” Bett describes the “traditional” picture as faith that the world is as it is objectively – that is, independently of our attempts to describe it; yet our accounts of how the world is have a real possibility of being objectively correct – that is, of describing the world as it objectively is – and we have a real possibility, when this is so, of knowing that it is so (Nietzsche 68). This “traditional” picture (which may not have begun in earnest until the Enlightenment) underwrites the Cratylian conception of language that this thesis tries to displace. Arguably, Thomas Kuhn has disposed of the fantasy that we will necessarily know when we have accurately identified something, and postmodernity has lost faith in the human potential for objectivity. How can we objectively see something like language that we are so intimately involved with? Ann Shukman characterizes the philosophy of language propounded by the French post-structuralists as coinciding with “a general skeptical attitude to language to be found in Wittgenstein and most modern thinkers since him” (563-64). It is the investigative and open-minded implications of the term scepticism associated with Sextus Empiricus’ definition of it that I wish to exploit here, and not some metaphysical 15 See footnote #15 on Sextus Empiricus’ definition of sceptic.23 Gonzalez’ introduction to The Third Way re-charts the horizon of Plato scholarship in terms of sceptical and24 doctrinal impulses dating back to Aristotle. This remapping of the terrain is especially interesting because the post-neoplatonist shift (that occurs before the eighteenth century) lines up with Enlightenment thinking and mis-uptakes of Bacon’s views of language. For example, at the end of the Protagoras after Socrates has brought Protagoras to his knees argumentatively and25 we are directed back to the opening scene and lines where Socrates had presented Protagoras as “the wisest of our generation” (Loeb 309d). Or did he really? Certainly on first reading that might be the impression one gets, but when we are directed back to the beginning we get a different impression. The ground seems to slip away, as the texts changes meaning. I have consulted the Hamilton Cairns Collected Dialogues, the Cooper Complete Works, the Loeb editions, Grube’s translation of the Republic and John Sallis’ own translations. Accordingly the citations of Plato will refer to the translator or to Loeb editions. nonsense that nothing can be known (whatever that means), nor reform positivism that wants to solve language’s shortcomings by remaking it. Recalling the root of the term sceptical –23 skopevw (I look, behold, contemplate) – I characterize as sceptical certain theorists who attend to and investigate language instead of trying to systematize it, scholars who study semiotic material instead of trying to idealize it. These are thinkers who, despite the non-ideal character of their subject matter and undaunted by the epistemological implications of language’s non- ideality, often treat it with optimism (like Bakhtin and Halliday), with humour (like Plato), and even a perverse glee (like Nietzsche and Derrida). O’Leary-Hawthorne comments on the “linguistic scepticism that runs through the [Platonic] dialogues” (168), but this sceptical view of language does not preclude investigation. As Sallis reminds us, Plato unendingly exhorts readers to use words as the medium of understanding, relentlessly advocating the pursuit of wisdom using lovgoi (‘words,’ ‘accounts,’ ‘discourses,’ ‘stories’).24 Language’s metaphoricity (which we examine through Nietzsche in Part 3) and our dependence on thinking in language and by analogy (which we examine in Part 2) means that human cognitive operations produce results of varying accuracy that are difficult to measure. And there are many characteristics of signs and their human users that contribute to language’s referential failings. Epistemological frailty that disrupts confidence in the validity of human knowledge (discussed in Part 2), self-referential writing that leaves readers uncertain about what the writer is directing our attention to (discussed in Part 5), and any number of language’s25 eccentricities and exploits (discussed throughout this thesis) might induce anxiety and thus play a role in the resistance to theory. 16 e.g., Fahenstock Enriching, Fowler, and Halliday Meaning.26 1.3 the road ahead ~ Did resistance, then, play a role in the repression of Bacon’s rhetorical understanding of language? What effect did the Enlightenment’s scientific super-exuberance have on conceptions of language? The Royal Society’s interests in empirical knowledge appears to have blinkered the uptake of Bacon’s sceptical views of language. Although late modernity seemingly swallowed the Royal Society’s Bacon-story whole, what is more to the point for this thesis is the reception of Plato’s, Nietzsche’s and Derrida’s discussions of language. How did the rhetorical insights of thinkers like Plato and Nietzsche influence philosophy’s “linguistic turn” in the work of Frege, Russell, Moore and the early Wittgenstein? Has the purported “plain style of science” or the mythical “style of non-style” played a role, not only in obscuring millennia- old insights into language’s operations, but also in fostering resistance to theory? This exploration of language theory synthesizes central insights from rhetorical- deconstructive-post-structuralist lines of thought from Plato to Derrida; it shows in broad strokes how these views of language line up with findings of recent linguistic research in human-science disciplines and branches of language theory stemming from analytic philosophy. Voloshinov’s theoretical work – which melds well with not only Derrida and Halliday, but also Sperber and Wilson and the scholars studying reported speech – sets the stage for the linguistic complexity that I believe must always remain in view for fruitful, detailed analysis of language in any branch of language study. Jeanne Fahnestock’s account of how “classical and early modern attention to language” was obscured by later interests foregrounds the linguistic awareness or consciousness evident in the rhetorical and dialectical manuals of ancient and early modern times (emphasis added Cognitive 161). This attention to language (attributed to both Plato’s and Bacon’s eras) is the principle concern here. Instead of striving for a “robust theory” or “thick description” that others have called for, and lacking confidence in the utility26 of language-theory codification for non-specialists, this dissertation seeks a generalized understanding of language. Attention to language – a greater awareness or linguistic 17 Ellis describes the code metaphor’s boomerang tendency in twentieth-century language theory: repeated acceptance 27 of code after repeated deconstructions of it, and acceptance without any explanation for, or efforts to address the previous scholarship refuting it. He links the code metaphor to conceptions of communication and faults language theory for treating language as though its purpose were communication (15). Giltrow writes that “the expert voice is known and valued for its objectivity and its neutral stance. In turn, its28 authority derives from this neutrality: when experts present research, their way of doing so shows that they have no personal stake in the knowledge reported” (Modern 171-72). She believes it is a stance, however, a stylistic effect and not a quality of perception. If it were imagined as a quality of perception, pure neutrality or objectivity would require Cratylian code for neutral, objective expression of what was perceived. consciousness of both linguistic mediation and language’s bewildering complexity – seems warranted. Rather than advancing a new theory, I want to synthesize a common stream of related linguistic insights with a long history which have been frequently overlooked in the last few centuries. The conception of language implicit in various linguistic assumptions and attitudes is particularly interesting. Ellis, for instance, complains that language theory’s missteps often involve “assumptions built into what look like statements of the obvious” (15): hence Richard Robinson’s claim that “the purpose of a name is to refer us to a thing” (Criticism 334). Such obvious and facile assumptions are responsible for the Cratylian impulse: the word-thing fixation to which, Ellis believes, linguistics and philosophy continually revert under a variety of names (i.e., code, information-content, truth claims). For sceptical theorists attentive to27 language’s metaphoricity such recidivism is mystifying. Language’s powerful affective elements inevitably embroil analysis of it in social, psychological, and communicative dimensions; importantly, these matters are always in play in language’s epistemological uses. At this historical moment code-metaphor conceptions and faith in linguistic transparency, probably sponsored by late modernity’s overwhelming technological successes, continue to muddle understanding. If we can put a man on the moon, language must be working the way we assume it does. And language is working, just not the way we might think it does. If science, with its plain-style myth and ethos of objectivity and neutrality, is an idol of late modernity’s scientistic culture, it is unsurprising that code metaphors, determinate reference, and linguistic transparency have such a grip on conceptions of language. Pure neutrality (or objectivity) demands this Cratylian image of language.28 18 Notwithstanding disputes about the meaning of his terms, Austin also, of course, claims that “to perform a29 locutionary act” is “also to perform an illocutionary act” (How 98); “in general the locutionary act as much as the illocutionary is an abstraction only: every genuine speech act is both”(147). At this historical moment and in this ideological context – when code-conduit-metaphor assumptions have been eroded by post-structuralism, but nevertheless still linger in aesthetic values about writing, in assumptions about scientific knowledge, and in the reference posts used by “mainstream” linguistics – a revived and renewed image of the linguistic sign seems warranted. I am trying to displace the Cratylian-code-conduit metaphor ideas about language with an image of an affectively charged, metaphorical, semiotic material that not only mediates every aspect of human experience, but by nature both generalizes and reifies: two salient actions undertaken by linguistic signs often not taken into account. I do not intend to offer a complete picture of language, or how it works, or even of the sign itself; but by keeping social- psychological factors of linguistic mechanisms in view, together with logical-structural ones, and by foregrounding the discrete nature of epistemological and communicative aspects of language use I hope to reinforce a sense of language’s perplexing complexity. It is not just the multi-purpose nature of language that creates complexity but that multiple purposes are served at the same time. From Halliday’s perspective there are always both construing and enacting (interpersonal) elements in an utterance at the same time. Multiple mechanisms in language – describing, identifying, constructing, communicating – operate at the same time. This simultaneity makes it impossible to analyze linguistic operations correctly29 without recognizing and accounting for multiple concurrent layers. Halliday says “one of the sources of [linguistic] complexity is that there is more than one kind of meaning in a language,” and as a result “grammar is doing more than one job at once” (emphasis added Matter 63). Borrowing a metaphor from Matthiessen, Halliday fantasizes about a “theory designed to represent the multidimensional ‘architecture’ of language” (Matter 77). It is this linguistic complexity that problematizes any effort to articulate a coherent theory of language (such as Wittgenstein attempted to do with his notes when he set out to write Philosophical Investigations), and that prompts the stylistic eccentricities of some sceptical-language-theory 19 See also Nietzsche’s remarks “On the question of being understandable” (GS 381) and “We incomprehensible ones.30 – Have we ever complained because we are misunderstood, misjudged, misidentified, slandered, misheard, and not heard?” (GS 871). See also Derrida’s engagement with charges levelled against him: “logical contradiction,” “performative contradiction,” “relativist,” “nihilist,”and a non-serious philosopher (Monolingualism 2-4). The sheer volume of texts dedicated to the problem of “how to read Platonic dialogue” speaks for itself. texts (i.e., Parmenides’ poem, Platonic dialogue, Nietzsche’s prophetic voice, Derrida’s metaphors, Wittgenstein’s aphorisms). Fahnestock is interested in “the view of language implicit in rhetorical stylistics” (Cognitive 165). I am interested in the way conceptions of language prompt different types of stylistics. In particular, I want to consider the way language theories are stylistically inscribed in sceptical-language-theory texts. 1.4 Wittgenstein & linguistic complexity ~ It seems likely that linguistic complexity results in a certain codification resistance as well as a certain summary resistance of stylistically eccentric texts (i.e., Plato, Derrida Wittgenstein). Linguistic complexity also figures into such anomalies as this thesis’ use of a referentially descriptive mode to refer to potential referential failure and underwrites the challenges of taking a wide-angle view of language into account. Wittgenstein’s “Preface” to Philosophical Investigations is useful here because it stages some of the conceptual and stylistic challenges of language-theory texts central to this thesis. His prefatory comments are thus a topos that introduce thematic considerations. For example, even something as trivial as his complaint that his ideas, in both oral lectures and written publications, have been “misunderstood” and “mangled” is representative. Although any writer might make this claim, as Culler and Lamb’s Just Being Difficult: Academic Writing in the Public Arena suggests, it seems a fortiori a problem for sceptical language theory. Such misreadings and manglings of30 language-theory texts haunt this thesis from its beginning with Bacon to its ending with Plato. These mis-takes seem symptomatic of a certain resistance to theory. While de Man’s resistance metaphor signals both reading failures and the misguided intuitions and anxieties of readers, what about writers? How do worries about writing sceptical language theory figure in stylistic choices? And how much does a concern with implicitly refuting their own premises – the tendency for sceptical theories of language to undermine themselves, for instance, by referring to potential referential failure – determine what language 20 Giltrow cites Biber and Finegan’s “taxonomy of stance styles” in different genres of which “faceless is the one that31 fits research writing. Facelessness is a ‘marked absence of all stance features’” (Modern 173). The stance of facelessness seems to be what researchers learn to do when they find their research voice within in their particular disciplinary field and they seem to do so without considering its linguistic implications. If they actually imagine themselves as neutral objective observers, they will tend to resist perceiving their linguistic usage as a stance and resist recognizing their own designs on their readers. Having designs on one’s readers clearly precludes pure neutrality and is incongruent with the ethos of neutrality. theorists say and how they talk? Overtly performative uses of language might be more desirable than ostensibly referential-descriptive uses burdened by paradox and contradiction; concerns about academic writing’s objective-rational ethos could discourage writers from making observations whose self-referentiality is potentially undermining. I wonder how the language- theory subject matter determines stylistic questions (e.g., genre, mode, approach) and whether the image of a descriptive-referential academic writing that disavows the kind of footloose stylistic we find in Wittgenstein and Derrida is itself hostage to a mythical style of science. This thesis is particularly attuned to how such considerations might be evident in the stylistic choices of Derrida and Plato. When questions about the style and ethos of academic writing (with its stance of neutrality) are tied together with a theorist’s desire to provoke thought, the threads get quite tangled. Wittgenstein says “I should not like my writing to spare other people the trouble of thinking. But if possible to stimulate someone to thoughts of his own” (“jemand zu eigenen Gedanken anregen”) (Preface). This desire to prompt or provoke [anregen] another’s thinking presents a reader-writer relationship with a more intimate (playful, invasive) cast than the respectful (collegial, “objective”) character of the reader-writer relationship in academic31 writing. This thesis is interested in the subjectivity of reader and writer in language theory texts. Wittgenstein’s interest in provoking reader thinking highlights his designs on his reader, warranting consideration of his relationship to his reader in terms of Sperber and Wilson’s model of ostensive-inferential communication. The salient role of reader inference that is documented by Sperber and Wilson directs attention to both the deciphering role of the reader and the social implications of leaving things implicit. Thus, for example, irony creates the impression that the writer shares a view with the reader; it is something they have in common, a kind of intimacy (242). Eliciting this sense of shared vantage is what Tannen calls an 21 Citations from Aristotle’s Rhetoric are usually from Kennedy’s On Rhetoric. I have also used Cooper’s translation32 of Rhetoric and the Loeb editions, as well as McKeon’s Introdcution to Aristotle, and an online text for a portion of the Metaphysics translated by W.D. Ross. I will indicate the translator or Loeb edition for all translations except On Rhetoric which is Kennedy’s unless otherwise noted. “involvement strategy,” as we see in Part 3. But how “involved” are writers with readers when the stance marking their prose is neutrality? Writers’ attitudes towards readers – what readers are expected to know and recognize and how much the writer presumes to inform the reader, or engage them in play – are just a few of many ways of evaluating the profile of a text’s reader. This thesis considers the differences between reader-writer relationships in texts designed to inform, to remind, or to provoke and then explores the role of reader-writer relationships in the Derridian and Platonic texts. It tracks the kinds of implicatures Wittgenstein, Derrida and Plato expect their readers to make. The discussing-observing-concluding (rational-objective) demeanour of research writing might look askance on the provocation motive which seems too intimate, playful, and demanding (not very neutral) for academic writing. Research writing does not seem intended precisely to provoke in the arresting way that Philosophical Investigations does. Aristotle, whose text serves as primogenitor of academic writing in this thesis, advises that “what is written should generally be easy to read” (On Rhetoric 3.5). Posterior Analytics outlines a generalized method of proceeding from known to unknown because learning “proceeds from pre-existent knowledge” in the form of an already known “fact” or the “meaning of a term” (Loeb 1.1 71a). Fine-grained, linguistic-pragmatics analyses find that at a32 word-order level given information precedes new – we begin with what is known before introducing something new. Such oddities as reversing given and new, or confounding genre expectations at the particularistic, pragmatic, word-order level, can arrest reading fluidity and be off-putting to readers. Riffaterre and others suggest that stylistic anomalies catch our attention and make us stop to think. As we shall see at the end of Part 4, the expectations that bring stylistic anomalies into relief are established by genre. It seems likely that writer interests in provocation might prompt some theorists (like Derrida and Wittgenstein) to subvert stylistic norms of academic writing. Possibly a motive such as Wittgenstein’s provocation can shed light 22 Note also Derrida’s opening line to Dissemination: “This (therefore) will not have been a book”(3). We will return33 to the question of writing a coherent theory of language in the context of considering the implications of totalizing texts and Bacon’s admiration of aphorism. on the footloose stylistic of Plato, Nietzsche, or Derrida. Wittgenstein’s realization that “the nature of the investigation” prohibited turning the Investigations into a “book” (“einem Buche”) is instructive here. He says “my thoughts were33 soon crippled if I tried to force them on in any single direction against their natural inclination. – And this was, of course, connected with the very nature of the investigation. For this compels us to travel over a wide field of thought criss-cross in every direction” (Preface). The aphoristic style of Philosophical Investigations is paradigmatic of the ways that language-theory texts perform their own theories. Wittgenstein’s “remarks,” he tells us, constitute “sketches” illustrating some feature or features of a linguistic usage (Preface). The carefully choreographed dance performed by these seemingly artless remarks both comments on and, at the same time, demonstrates the diversity of kinds, functions and purposes of “names.” For example, the same name serves as different kinds, for different functions and different purposes; he queries whether or not many words name anything at all. This protean quality of names – only one dimension of linguistic complexity, albeit an important one – adapts them to the determining role that context and genre play in linguistic operations, that is, to the extent that language is determinate at all. Wittgenstein tells us he suddenly realized that he “should publish those old thoughts and the new ones together: that the latter [Philosophical Investigations] could be seen in the right light only by contrast with and against the background of [his] old way of thinking” (Preface). Context is a dimension of language that seems to dominate all levels of linguistic analysis, from the most particularized linguistic studies of phonemes to the most generalized levels of ideological activity. Context provides a place or location which situates linguistic interpretation – whether the place is a generalized, high-level argumentative structure or a very particular scene with specific discursive resources. Context has many faces. It can be viewed through many levels of magnification and produced by a variety of rhetorical-stylistic gestures. In 23 Philosophical Investigations, the need for context underwrites the use of the Tractatus as the background on which the sketches are drawn. An assumption, however apt or erroneous, from the earlier work both prompts a sketch and then becomes its context. Here what is contextualized are whole, historically-situated discourses, disciplinary lexicons, and approaches to language. Wittgenstein’s often Cratylian interlocutor presents these attitudes, summoning various positions that were pursued in early-twentieth-century philosophy of language. However, the voice of Wittgenstein’s fictional interlocutor together with the arrangement of counter-punctual remarks are stylistically distinct from the way that Sperber and Wilson, for example, contextualize their views. Wittgenstein’s style of contextualization reflects the linguistic complexity he seeks to reveal. Wittgenstein tells us and shows us that nothing in language use functions or has meaning in isolation, and he does not think he can explain how language works in a “coherent” book that will offer a codified theory, as he had earlier attempted. Language theory is “codification resistant,” as de Man says. And expert codification is too detailed to have practical utility for purposes other than specialist classification. This impracticality stems from, among other things, the copiousness and particularity of what Wittgenstein calls “language games.” Non- grammarians might be reminded of the effort of reading Halliday’s grammar, however useful it is for expert application. Halliday says at the end of his long career that “our interaction with our environment is so complex and multidimensional that there has to be a lot of ‘play,’ or indeterminacy, in the construal [of grammar] for it to be able to work at all” (Matter 63). This complexity reminds me of biochemistry, which presents an image of fantastic intricacy and complexity; the factors and forces effecting potential bonds between different elements of the periodic table are too complex and indeterminable to keep in mind for non-specialists. What de Man calls “codification resistance” lines up quite nicely with Halliday’s view on language theory’s indeterminacy (OL 266). It is not just the linguistic sign that is indeterminate, but also the grammatical mechanisms and the theories defining them. This thesis, too, may be charged with a degree of criss-crossing-in-every-direction as it 24 In a 2005 rhetoric of science collection, Fahnestock suggests that cognitive science make use of “the view of34 language implicit in rhetorical stylistics” as the basis for its neuroscience research examining the functions of language activities in the brain. Her belief that the rhetorical tradition’s “insights into language” should inform contemporary neuroscience reflects interestingly on the twentieth century’s “philosophy of language” failure to recognition the insights of classical rhetorical theory (Cognitive 164-65). gathers together discourses that recommend a certain scepticism about linguistically mediated human understanding. I want to remind readers of the troubling epistemological implications and multiple opportunities for error resulting from analogical-cognitive operations that are mediated by a linguistic semiotic. My interest in a renewed linguistic consciousness coincides with the linguistic concerns of the evolutionary biologist Richard Lewontin, who worries about the misleading implications of the metaphors used by science in public discourses. But the linguistic-mediation problem is also a problem for science itself. Since language mediates even the terminology of hypotheses, linguistically entailed assumptions may occlude important evidence or produce false results. It is not just a question of empirical observation; language both construes and enacts, producing what we see. An increased consciousness of linguistic mediation and increased attention to language is warranted in all types of research. It is a question of considering the implications and assumptions underlying the actual terms in which research projects are imagined and the cognitive analogies on which selection of terms is based. 34 1.5 the participants and other parts ~ Bacon appears in this dissertation’s cast of characters as exemplary of certain linguistic matters, not the least of which is his role as an intermediary between ancient and contemporary views. Nietzsche enters as the theorist who specifically links classical rhetorical theory to his own view of language’s metaphoricity. Derrida’s and Plato’s sceptical-rhetorical views of language and footloose stylistic have prompted this investigation as objects of study. The cursory stylistic analysis of their texts concluding this dissertation is the destination for its four preceding Parts. What I am calling the sceptical view of language (where sceptical means eyes- wide-open) is elaborated through Voloshinov’s Marxism and the Philosophy of Language, with assistance from the findings of Sperber and Wilson’s model of communication and the surprising results of the research on reported speech. Linguistic-pragmatics scholarship, and 25 Sperber and Wilson in particular, provide a lexicon and an understanding of linguistic communication that illuminate the Derridian and Platonic stylistics. Halliday appears periodically as a language expert; and Wittgenstein’s exquisite, particularized examples of often obscure, abstract features of language use warrant his continued reappearance. While the authority of these scholars speaks for itself, Aristotle’s role here is a bit trickier. He might appear as an authority on language’s rhetorical operations; but he also proves to be someone whose views of language don’t seem to square. The gap between the linguistic implications of his utterances (what he says) and their style (how he says it) seems to indicative of his scientific motives. He serves as an object lesson in this thesis’ discussion of style. Part 2 presents some of the factors and participants featured in this thesis. Bacon’s “plain style” continues to serve as a pretext, while some of Voloshinov’s views come into contact with Halliday’s grammar and Sperber and Wilson’s ostensive-inferential model of communication. Part 2 ventures out on a sightseeing tour into several abstract dimensions of linguistic mediation like the role of affect in cognition, the subjective nature of the human perspective, and linguistically mediated reasoning processes. De Man’s notion of resistance continues to stage some of the difficulties of writing about language theory as our journey takes us to such occult locations as Silvan Tomkins’ theory of affect, William James’ “Sentiment of Rationality,” the cognitive use of analogy, Van Dijk’s analysis of the cognitive functions of specification and generalization, ancient maxims about knowledge and recognition, and the totalizing implications of the term theory itself. The leading role for language in these high-level and abstract epistemological conflicts makes each component of the mix a relevant consideration of how language works or doesn’t work. While most of these matters stem from ancient philosophical ruminations and will be familiar, even obvious, they seem to be frequently overlooked when it comes to thinking about language. Part 2 gathers together insights affiliated with the problems of linguistically mediated human reason and the resulting implications about language. Part 3’s account of late-twentieth-century language discussions have Nietzsche’s 26 important earlier insights as their destination. Part 3 exploits the late-70s controversy resulting from Searle’s response to the publication of Derrida’s talk “Signature, Event, Context” as a way of bringing Derrida’s ideas about context and citation into contact with the scholarship of reported speech. We witness a meeting of minds regarding important aspects of language use between different, formerly antagonistic, disciplinary inheritances in philosophy. Some of Derrida’s ideas about language mesh with conclusions of scholars investigating reported speech in various language-theory disciplines and departments (i.e., sociolinguistics, psycholinguistics, anthropological linguistics, grammarians, linguistic-pragmatics, discourse analysis and literary theory). As we shall see, these finding are also consonant with Voloshinov’s theorizing and Sperber and Wilson’s theory of communication. We witness the role of the promiscuous linguistic sign in human interactions, involvement strategies, and affective interests in speech. An interesting network of relations appears between, i) the Tomkins, James and experimental- psychology discussions of affect in Part 2, ii) Fahnestock’s reporting of recent neuroscience research, iii) and the Voloshinov-Halliday view about the relationship between the sign and the psyche. This interdisciplinary work reinforces the ancient thinking-in-language concept and sets the stage for Nietzsche’s critique of language’s tropological nature. In Part 4 Nietzsche’s metaphorical turn leads to questions of style and clarity. Hume’s famous simplification of his Treatise, documented in Culler and Lamb’s Just Being Difficult: Academic Writing in the Public Arena, illustrates the vicissitudes of “the plain style” in Enlightenment philosophy; and various circumstances are called to testify to the role that the style-of-science myth has played in analytic philosophy. Aristotle’s clarion call for clarity provides an object lesson on style. When the performativity of direct discourse that we experienced in Part 3 meets the clarity of Aristotle’s analysis (or what has always seemed to be analysis) questions about just what style is blur. Style itself is part of the performative dimension of language use and the style-performativity-genre bond both prompts different reader responses and is motivated by different writer interests. Style’s intimate relationship with genre, which is style’s domestic home, helps clarify the role that genre expectations play in how 27 language works. Part 4 closes with a discussion of genre in anticipation of the Platonic and Derridian texts. Part 5’s analysis of Plato and Derrida concludes this dissertation. This study of the intertwined linguistic theories and stylistics of the Platonic and Derridian texts illustrates the image of language outlined in the previous Parts. This rhetorical-sceptical view of language results in what I call Plato’s and Derrida’s conversational stylistics. Part 5 takes stock of the history of mis-readings of the Platonic text, gauges the interpretive challenge its genre poses, and then explores the way that each writer exploits genre expectations to produce their distinct, ironic-conversational stylistics. Glancing back at assumptions about “propositions” from the “philosophy of language” we consider the unique ways that themes and theses are woven into the Platonic and Derridian texts; we observe the implicitness that characterizes these texts and the reader involvement they elicit. After gauging the way things are said in Derrida and Plato, the way that Derrida re-performs Plato’s conversational style, and the way that their texts “read us,” as Halperin says, we head towards the dissertation’s concluding analysis of what Plato says about language, and the way that his astonishing rhetorical mastery configures his sceptical- rhetorical views of language. 28 i.e., rhetoric’s appeals to the ethos (‘character’ or ‘habit’) of both speaker and audience, to the pathos (‘experience’35 or ‘suffering’) of human life, and the logos (‘account’ or ‘story’) like the cause and effect topoi, etc. PART 2 2.1 adventures of signs in human communication ~ The translator’s introduction to Marxism and the Philosophy of Language records Ann Shukman’s observation that “Voloshinov’s extreme contextualism leads him to a semiotic theory that is primarily sociological, and to a theory of language that emphasizes process rather than system, function rather than essence” (viii). Upcoming sections reveal the way that Voloshinov’s interest in linguistic processes, like the reporting of speech, lines up with the focus on actual language use from the practical advice of rhetorical theory’s ancient-handbook tradition to contemporary discourse analysis. Concerning “function,” the trans-disciplinary anatomy of Volshinov’s theorizing – semiotic, sociological and psychological – offers a glimpse into the manifold functions of language comprehended in his approach. The social origins of linguistic-semiotic phenomena shape and problematize language’s epistemological functions: complications arising from linguistic mediation central to Parts 2 and 3 of this dissertation. Voloshinov claims that the sign’s social origins have profound psycho-ontological impacts. These matters are addressed in this section by Tomkins’ and James’ reflections on the ties between affect and language, as well as the discussions of logic, reasoning-by-analogy, and part-whole relations. It is the social origins of signs that give rise to the complexity and multi- functionality evident in the work of linguistic-pragmatics and discourse-analyst scholars presented in Parts 2 and 3, and in the overlapping and intertwined behaviour of classical rhetoric’s appeals that are discussed in Part 4. 35 This thesis is especially attentive to the implications of Volshinov’s “extreme contextualism” for genre and style. I have already alluded to the way that Wittgenstein’s text states and performs the “contextualism” that Shukman refers to: the way that the intellectual atmosphere and conversations of the 1920s that resulted in the Tractatus Logico-Philosophicus becomes the context for the “remarks” of Philosophical Investigations. On the other hand, Plato’s fictional dialogues provide situated conversations – in physical and social settings – 29 Glendinning observes that Derrida (like Voloshinov) defends “a particularly radical form of contextualism” (318).36 For instance: “The outer sign originates from this sea of inner signs and continues to abide there, since its life is a37 process of renewal as something to be understood, experienced, and assimilated, i.e., its life consists in its being engaged ever anew into the inner context” (33). The translator’s selection of comments, from Jameson’s review in Style and Jakobson’s Letters and Notes, reflect38 the text’s singular linguistic-semiotic trajectory. Jameson says “quite simply one of the best general introductions to linguistic study as a whole” (“Translator’s Preface” viii). characters with distinct personalities, and discussion of particular topics. Each ingredient of the dialogue contextualizes every word uttered. Alternately, Derrida’s famous parasitic writings, like Wittgenstein’s Investigations, grow out of other texts, though not primarily his own. Rather than attempt to inscribe codification-resistant language theory on a blank page (that is, without context), Derrida contextualizes his writings in the palimpsest of others’. We note here that the36 concept of context is itself extremely amorphous. It has a different look in different instances: depending on the size of its playing field, the level of language use under scrutiny, and the style of contextualism. If we imagine the copiousness and ponderousness that might have resulted had Wittgenstein attempted to summarize all his particular “remarks”– to codify, organize, and index their interconnections into a coherent book – we can see why these thinkers might have fashioned specific habitats (or “places”) for contextualizing their linguistic ideas. Voloshinov too finds a stage for his linguistic theory by setting it in the ideological domain: a “place” or location which is figured as the anthropomorphized universe of the sign. Like Bakhtin, he imbues signs with life and charges ideology with the animated energies of this enlivened sign. It is important to observe that Voloshinov’s ideology imparts the original37 broad range of perspectives in Marx’s definition, and not the narrowly political implications twenty-first-century readers might expect. Since in The German Ideology the term ideology represents “the production of ideas, conceptions of consciousness,” all that “men say, imagine, conceive,” and includes such spheres as “politics, laws, morality, religion, metaphysics, etc.,” ideologies are clusters of affiliated conceptions (47). The specifically political implications anticipated from both the title, Marxism and the Philosophy of Language, and its leading role for “ideology” are strikingly absent from the text. Though there are infrequent references to38 base, superstructure, and class struggle and doubtless political implications could be (and have 30 Some of the political interpretations are referred to in the “Translator’s Preface.”39 Halliday explains this two-sidedness of the sign when he says that language “has to create a universe of its own, a40 parallel reality that is, as it were, made of meaning; [and] it is this ‘textual’ potential that enables language to be at the same time both a part of human experience and a model of it” (OL 244). been) inferred from this book, I suggest that Voloshinov’s exploitation of ideology might be conceptually and stylistically opportunistic. What looks like political diffidence might be a39 rhetorical strategy. Admittedly the title’s reference to Marxist Ideology could also have provided Voloshinov (or Bakhtin) cover from the political persecution that Strauss suggests has produced stylistic “indirection” throughout the ages. But however much such indirection might have motivated the writer, at the same time ideology conceptualizes and images a semiotic site par excellence. It is a conceit or image of linguistic phenomena illustrating the twisted journey and adventures of the sign, and at the same time serves as a conceptual pretext for mapping the interdependence of idea and sign: a relation that crucially illustrates the linguistic mediation of human conception that preoccupies Nietzsche. “Without signs,” Voloshinov says, “there is no ideology” (9); “consciousness itself can arise and become a viable fact only in the material embodiment of signs” (9); the “sign is not only a reflection, a shadow, of reality but is also itself a material segment of that very reality” (11). The material reality of ideology is linguistic; and its phenomenal appearance has the form of signifiers, audio and visual objects in the world. Most importantly, the sign sits in reciprocal relation to the idea; the sign conditions the concept that conditions the sign. Signs and concepts are the part of reality that constitute our understanding of it. Modernity’s failure to recognize40 this ancient insight can be indexed by North America’s hysterical reaction to Whorf’s evidence that words shape thought and the absurd accusation of linguistic determinism levelled against him. The Whorf panic imagines a terrifying imprisoning sign whose rigid-determining capacity manacles conceptual creativity. This reaction bears the imprint of a Cratylian conception: that is, the code-metaphor model, the conviction that names refer to a determined referent, a blindness to language change, and the tendency to confuse the phenomenality of the sign with the phenomenality of the thing named. Thought must be independent from language in a Cratylian universe; otherwise, cognitive operations would be restricted by a finite lexicon. But 31 in Language as Symbolic Action.41 I am not suggesting that the entire audience is committed to code-referential, truth-condition determinations, a fact42 evinced by the emergence of the field of pragmatics itself. Code-referential-truth conditionality has posed some prime questions about language, and Sperber and Wilson try to take up these questions, sometimes answering them, but more often showing that they are the wrong questions. See Grice’s Studies in the Way of Words for an indication of this lineage and the model of inference he works with. for Voloshinov “the meaning of a word is determined entirely by its context” and “there are as many meanings of a word as there are contexts of its usage” (79). The sign’s thoroughly social existence and ancestry embroils it in subjective human affective life. In Voloshinov’s view, its social gestation fosters the sign’s vitality and mutability and its refracting and distorting aptitude. He says, quite typically, “a sign does not simply exist as part of a reality – it reflects and refracts another reality. Therefore, it may distort that reality or be true to it, or may perceive it from a special point of view, and so forth” (10). This promiscuous-reciprocal-referential relation of sign to reality is also thematic in Halliday, Nietzsche, Derrida, and, as Part 5 suggests, in Plato. It is famously explored as “lenses” in Kenneth Burke’s “Terministic Screens.” Wittgenstein uses the related metaphor “spectacles”41 when he troubles over the way that logic becomes a lens for the philosophy of language (§103). For Voloshinov signs and concepts are inter-illuminating and in constant flux: characteristics of the human-linguistic-semiotic sphere thematized in Bakhtin and Plato. Words – whose entire realities are absorbed in their functions as signs – are a sensitive, mutable, context-dependent medium unsuited for the signal-type transmission entailed in code-metaphor conceptions of language. 2.2 Relevance Theory ~ On the other hand, Sperber and Wilson address an audience accustomed to (a-contextual, referentially-determinate) code metaphor models: an audience which imagines communication as descriptive-information exchange, and focusses on the truth conditions of propositions and speaker intentions as criteria for determining communication success. Their work in contemporary pragmatics – linguistics’ disciplinary response to the work of philosophers like the early Wittgenstein, Austin and Grice – disrupts many of their audience’s Cratylian assumptions about language. By adapting and expanding Grice’s notion of inference and his42 32 All references to Sperber and Wilson in this dissertation are to their Relevance: Communication and Cognition43 except as noted. I use the term explicit in its common sense meaning here, and elsewhere until Part 5, when it becomes necessary to44 acknowledge that the concept of verbal explicitness entails determinate reference and has the same fraudulent status as verbatim explored in Part 3. maxim of “relevance,” their Relevance Theory model of ostensive-inferential communication43 undoes habitual code-model notions of communication as sending information from one mind to another, and shows instead that hearers derive “implicatures” from utterances by making “inferences.” Sperber and Wilson play multiple roles in this thesis. On the one hand, their model serves here to dismantle the code metaphor and the primacy of truth-evaluated reference in communication. On the other hand, their theory is useful for understanding the workings of broadly conceived poetic or rhetorical effects and thus helpful in producing implicatures from the Derridian and Platonic texts, as well as being theoretically instructive about how these writers may have conceived what they were doing with words. The ubiquity of implicitness in the ostensive-inferential model and the important role for hearer inference illuminate how language works in sceptical (eyes-wide-open) language theory. I focus crudely on several salient elements in relevance theory’s ostensive-inferential model of communication useful for a very basic understanding of it. Sperber and Wilson demonstrate that: 1) linguistic communication is an interpretive-inferential phenomenon, and 2) some relative of logic (which is a red herring in the code account) plays a limited though crucial role in producing propositional implicatures used in the interpretive-inferential processes; 3) referentiality (another red herring) plays a less significant role than say, the importance of hearers’ participation in what are often vague communicative efforts. In other words, reference is clearly subordinate to the importance of hearers making their own inferences. And 4) communicators are little interested in being informed by means of explicit reference. Actual44 language use is all about implicitness. Rather than communication being the simple transfer of a message from one mind to another (a passive recipient), hearers are active participants. The idea of informing someone of a message has shifted ground and metamorphosed into hinting at the sort of thing a speaker and hearer might want to discuss. 33 In the ostensive-inferential model of communication, meaning-making is a mutual affair, with speakers who point at something and hearers who first determine what is being pointed at and then interpret what about it the speaker finds interesting enough to point out to them. The ostensive-inferential model includes a kind of deductive (i.e. logical) function performed by the hearer (along with other heuristics) in the inference process. Utterances are enriched or expanded to propositional form (by inference) according to intuitive logical principles (and other heuristics) based on the context; and what the hearer already knows (a contingent factor) are premises in the enriching inference. Each inferentially enriched utterance, in this model, serves as a bit of the material that hearers must use in subsequent inferential activities – activities which are highly interpretive despite deductive workings. This inferential process culminates with the most relevant implicature (an estimation not a deduction) that the hearer can derive. Not only are speakers “presumed to aim at optimal relevance not at literal truth,” but the guesswork involved in enriching utterances to propositions and the range of things that the hearer already knows makes the inferential activities in this model relatively indeterminate from the outset (Relevance 233). Relevance drives truth-conditions and reference to the sidelines. This fuzzy process of inferential guesswork, repeatedly performed at several levels of the communicative process (by active hearers/readers), has salience in the ostensive-inferential model of communication. For Sperber and Wilson “human intentional communication is never a mere matter of coding and decoding” (174). This is in part because human “languages do not encode the kind of information that humans are interested in communicating”; instead language users are interested in what languages do, and what they can do with them (174). This observation about what we are interested in communicating reminds us that communication processes are shaped by their primarily social motivation: human interests in interacting and participating. Voloshinov too emphasizes the social conditioning of signs and linguistic expression. He argues that human linguistic signs and the speech genres that stylistically shape utterances have completely social origins (20). The inferential model of communication discovers that speakers 34 Note, Sperber and Wilson say that “assumptions are unlike knowledge in that they need not be true” (40). 45 are not informing (passive) hearers of a message content, but pointing to or hinting at the kinds of things they are interested in discussing and drawing attention to certain things so that the (active) hearers can make inferences. Sperber and Wilson describe this process as “making assumptions manifest” and bringing assumptions into the “cognitive environment.” So, for45 instance, even when an explicit interest in closing the window motivates communication, the question, “is it cold in here?”, might be a preferred method of expressing the interest. And the speaker’s preference for the question itself can trigger a chain of implicatures. Such preferences are a fortiori the case when the communicative intentions are less certain, as Sperber and Wilson suggest they frequently are; they say that the effects of most forms of human communication are “vague” (55). When other factors like being polite, just having something to say, or just wanting to respond are at stake, communicative aims may be much more general, not nearly as determined as “closing the window.” Implicatures are not explicitly stated by speakers but inferred by hearers (or readers) based on what is said (and not said). Hearers (or readers) attempt to determine how the implicatures of an utterance achieve relevance: they gauge the relevance of implicatures – not just what is said, but also the implications of its being said, of how it is said, and its being said in this particular context by this particular person. In Sperber and Wilson’s terminology, hearers seek optimally relevant implications from the assumptions that have been made manifest. The role of the hearers in making inferences from the assumptions that have come into the cognitive environment (including what the hearer already knows) is paramount in this social and interpretive activity of communication. In fact, Sperber and Wilson find that people don’t enjoy being told things that they already know: A speaker aiming at optimal relevance will leave implicit everything her hearer can be trusted to supply with less effort than would be needed to process an explicit prompt. The more information she leaves implicit, the greater the degree of mutual understanding she makes it manifest that she takes to exist between her and hearer. Of course, if she overestimates this degree of mutual understanding, there is a risk of making her utterance harder or even impossible to understand. It 35 is not always easy to strike the correct balance: even a slight mismatch between speaker’s estimate and hearer’s abilities may make what was merely intended to be helpful seem patronising or positively offensive to the hearer. (218) It is better to repair a misunderstanding than to over-inform: so much for code metaphors of message delivery, sociality is the organizing principle here. The speaker hints and the hearer guesses, an interaction that looks a lot like a “game.” One of Halliday’s pillars is that language is a social semiotic: that it always has both interpersonal and ideational/experiential meaning. As we shall see in Part 3, Tannen has collated research demonstrating that “involvement” and “solidarity” – entirely social-affective motives – are what is at stake in conversational discourse. Social (affective) motives, according to Voloshinov, shape all semiotic components and structures. Finally I wish to draw attention to the distinction that Sperber and Wilson observe between a communicative intention and an informative intention (which, however problematic, is nonetheless useful). Communicative intentions, which pertain to a social sphere, can be made manifest ostensively: that is, by non-linguistic demonstration or showing, like pointing, or looking at something and making a face. Sperber and Wilson conclude that human communication is conducted in this ostensive-inferential sphere; communication pertains to showing or demonstrating, making assumptions manifest so that a hearer can look and try to determine what implicature is most relevant in the context (49). This speaker-hearer activity squares precisely with the findings of research on reported speech, as we shall see. It is here that we witness the final collapse of the code metaphor because the utterance may signify something (like “it’s cold”) entirely other than its referent (“close the window”). The distinction between the referentially-descriptive and the ostensive-inferential is parallel to the distinction between the constative and the performative in Austin. Since language often performs multiple functions (at the same time), frequently with divergent aims, and since it is often used not descriptively but as a pointing device (or a demonstration, as we shall see with reported speech), it cannot be taken at face value (or read as description), as the style of non-style insists on doing. Language use in communication has more to do with pointing, inference and mutual 36 interest than message transfer. In this context, Sperber and Wilson determine that language is needed for the cognitive function of information processing rather than communicative functions strictly speaking (173). “Languages,” they say, “are indispensable not for communication, but for information processing; this is their essential function” (172). Information and what it might be are other problems that we don’t have time to explore here. Information processing clearly is employed in the inferential processes of linguistic communication (e.g., she is asking is it cold in here and looking at the open window, perhaps she wants me to close the window, or maybe she’s worried about the dust coming in). If we imagine “information processing” as a linguistically dependent function using a socially conditioned semiotic material we end up with an account that resonates with the semiotic milieu of both Halliday’s “construal” of experience (OL 276) and Voloshinov’s “consciousness” and “psyche” (9, 26). Sperber and Wilson’s findings suggest that it is not specifically (social) communicative impulses that depend on specifically linguistic signs, but thinking itself that requires them. Linguistic communication is a complex interface of social semiotic and affective human motivations that we explore shortly through the ideas of Tomkins and James. The assumptions that flow from unexamined conceptions of communication have doubtless bedevilled language theory. Ellis considers “the assumption that the purpose of language is communication” to have been a major misstep in twentieth-century language theory (15). He associates conceptions of communication with code metaphors, pointing out that Whorf observed the error of “supposing the function of language to be only the COMMUNICATION of thought” (20). Halliday adds that both Firth and Hjelmslev “rejected the notion of language as a system of communication; they saw language as constructing reality not passively reflecting it; and they distrusted the pseudo-science and the spurious rigour of models from formal logic (‘rigor mortis’ was Firth’s way of characterizing it)” (OL 242). Halliday says that his mentor, Firth, “explicitly rejected the definition of language as a means of communication and was highly critical of those who imported into linguistics a conception of language derived from communication theory” (OL 242-43). Sperber and Wilson have remade 37 communication theory in a way that matches the insights of linguists like Whorf and Halliday. Of course, none of these scholars are disputing language’s use in linguistic communication. But communication is not all language is used for. It is used for thinking, identification and understanding the world – epistemological purposes. And it is unexamined notions of communication, intuitions leading to code and conduit conceptions that are so confounding to language theory. Nevertheless it is, according to Voloshinov, language’s sociality – its social conditioning and social origins – that shapes linguistic-semiotic phenomena and governs the operations of the entire linguistic sphere. This means that epistemic operations employ socially conditioned, socially fabricated, and socially saturated semiotic materials. Perhaps the disjunction between language’s social origins and its epistemological-thinking- information-processing functions is one of the sources of its complexity and codification resistance. Code does not survive Sperber and Wilson’s re-figuring of communication (though the boomerang tendencies Ellis observes cannot be overstated). Reference survives by adapting to indeterminacy; and description moves over to make room for other types of linguistic operation. The epistemological uses of language is of particular interest, as it relates to Bacon and the vicissitudes of the style of science. However consistent or inconsistent Bacon’s views of language may or may not have been, they were eventually displaced by superficial code-conduit models of transparency. To the extent that Bacon’s rich understanding of language was reframed in this simplistic manner, the seventeenth century’s exuberance about natural history, the freezing point of mercury, and the transit of Venus seem to have drown out Bacon’s linguistic insights. This cursory consideration of Bacon raises questions about what kind of “advancement” regarding the linguistic paradigm took place during the Enlightenment and whether a Kuhnian backsliding might not have occurred in linguistic consciousness. Voloshinov complains about the “mechanistic conception[s] of linguistic necessity” that had become standard in the (“mainstream”) linguistics of the early 1920s (81). Did they get their start in the seventeenth century’s knowledge quest when the aspirations to secure knowledge of the world 38 See Graham’s The Performance of Conviction and Stanley Fish’s “Epilogue: The Plain Style Question” for46 indications of the way that these trends have been analyzed. and to theorize about language seemed to veer in opposite directions? Perhaps anxiety about language’s epistemological frailty prompted the Royal Society’s resistance to Bacon’s linguistic theories and millennia-old rhetorical insights in favour of the excitements of new knowledge. 2.3 historical context: Bacon and the “plain style”~ Zappen’s work on Bacon’s rhetorical theory documents an affiliation that history and rhetoric of science scholars (circa 1920-1950) make between Bacon and the “plain style” of “positivistic science” (Historiography 75). The positivists simply seem to lay claim to Bacon as the father of science and ideologically reify his purported plain style as the style-of-non-style. But the twentieth-century story changes after the positivists. Zappen observes that as early as the 1960s this view was rejected by historians of science and scientific rhetoric, and replaced by an understanding of the complexity of Bacon’s theory of style. These later historians observe that Bacon’s theory “is actually complex, that it embraces several methods of presentation and several styles, not just the plain style” (Historiography 76-77). However, it is not clear that Bacon advocated such a style at all, let alone practised it. Zappen’s reference to “the plain style” seems to stem from inferences based on Bacon’s critique of ornament and expressed admiration for “plainness” rather than a prescription advocating any particular style. Interestingly, until Lipson, the question of what the “plain style” might actually be does not seem to arise in these history and rhetoric of sciences discourses. The prose practices and intellectual climate of the early Renaissance together with the contemporary teaching of writing or rhetoric would provide a context for understanding what prompts Thomas Sprat, in the History of the Royal Society, to conclude “that eloquence ought to be banished out of all civil societies as a thing fatal to peace and good manners” (111). The competing linguistic discussions of the period that Kenneth Graham catalogues suggest that there were considerable contemporary debates and disagreements about trends in manners of discourse. Sprat’s prognostication on the ill effects of ornaments – “the myths and46 uncertainties” that “specious tropes and figures have brought on our knowledge” – contribute to 39 While it is not clear what definition of “ornament” might have been circulating – and there were likely as many47 different ideas of what ornament was as there were persons thinking about it – even Sprat admits that ornaments “were at first, no doubt, an admirable instrument in the hands of wise men” (112). Fahnestock equates “ornamentation” and “forcefulness”; she states that “[early modern] rhetorical manuals placed the figures of speech under this heading, for the figures were seen as sources of distinction or ‘ornatus,’ that vexed term usually translated as ‘ornament,’ rather than, as Vickers and others have argued, something closer to “armament [ornamentum], the weapons and gear that make a warrior or speaker effective” (Stylistics 218). In an ironic oversimplifying gesture, Fish argues that the distinction distinguishing the wide variety of the era’s48 intellectual trends is an epistemological one: those who, as he puts it, “see through a glass darkly” and those for whom the “glass has been made clear by reason” (Epilogue 381). The distinction is framed as “an assumption of the mind.” He places Bacon on both sides of the assumptions-of-the-mind divide apparently on the basis of comments made in Advancement. a substantial contemporary debate (112). Whatever strains of rhetoric, whatever religious and scientific movements, and whatever overlapping and divergent views (spawned in this burgeoning print-literacy era) that occasioned Sprat’s reaction against “eloquence” and “ornament,” it is clear that while Bacon sometimes expresses concern about undue47 ornamentation, he also explicitly approves of and uses rhetorical figures in his text, as Lipson observes. Somewhere along the road the determinate-reference-code model is projected onto notions of “plainness.” Although Bacon (infrequently but admiringly) employs the term plain when talking about speech, it is usually in the context of hoping that his own efforts will be “set forth plainly and perspicuously.” But it sometimes appears in prescriptions that seem more concerned with “honesty” and “openness” than “simplicity” or “clarity.” For example, “plainly” seems to mean “openly and “honestly” when Bacon advises that “if in any statement there be anything doubtful or questionable, I would by no means have it suppressed or passed in silence, but plainly and perspicuously set down by way of note or admonition” (IX Preparative Towards a Natural and Experimental History). Lipson concludes that “confusion over whether Francis Bacon was the father of plain prose probably stems from the different possible meanings of plain style” (144). Graham finds that there are competing views of “plainness” in the Renaissance: 1) an anti- rhetorical plainness associated with “a certain claim to truth” characterized by an “insistent demand for certainty,” and 2) a rhetorical plainness which seems to constitute a call for a common (or democratic) language and entails a sceptical-rhetorical and socially negotiated understanding of truth (14-17). 48 While Bacon’s views on language seem to fit him in the latter class, as Lipson indicates, 40 Graham claims that “in Neostoic doctrine, right reason replaces Scripture as the principle source of plain truth,”49 hence “the reformed claims about the authority of Scripture parallel the claims made on behalf of the authority of reason in Neostoic doctrine” (6,8). The anti-rhetorical reform and Neostoic claim to truth and call to plain speech are thus distinct from the humanist rhetorical tradition’s dialogical plainness. The humanist call for plainness was a call for common language as a mechanism of communication: if Bacon fits in this category, it reflects interestingly on his conception of empirical science. what is germane here are the varying meanings of plain. The meaning of plain is, according to Graham, “as plain as the meaning of ‘true,’ which is in fact its closest synonym in the sixteenth and seventeenth centuries” (1). The valence of “true” in “plain speech” helps us understand, as Graham observes, how “even such Shakespearean wordsmiths as Hotspur and Othello appear to believe that their language is plain” (2). This dimension of the term “plain” helps make sense of the “proliferation of [Renaissance ] plain styles” which are “catalogued by modern critics” and which are, arguably, not very plain to a twenty-first-century ear: “native plain style, classical plain style, Puritan plain style, Anglican plain style, Bacon’s plain style, Herbert’s plain style, Restoration plain style” (Performance 2). “Every claim to truth,” according to Graham, “is likely to call itself plain at some point” (2). Both Fish and Graham also highlight the “making evident” valence of “plainness” which Graham calls “a texture of obviousness” that he affiliates with repeated claims about scriptural “meanyng” (12). The career of the scientific plain style might begin here in its prior affiliation with truth and the discovery of (purportedly obvious) truth. However, truth is not self-evident, neither simple, clear nor obvious: and neither is the meaning of plain style. While it is not clear what the “plain style” is (or was), it has become affiliated with conceptions of determinate reference and description, and possibly a style that pretends not to perform. In Part 3 we will see the way that reported speech is paradigmatic for language’s performative dimensions, what Halliday calls its “semogenic” powers that combine “construing” and “interpersonal” functions. While Bacon’s empirical interests may have motivated his aspirations to speak “plainly and perspicuously,”what is interesting is that his many comments on language which have important epistemological consequences do not survive his purported prescriptions. Instead,49 his remarks about openness and his aspirations for the success of his own writing have morphed into a mythical style of science involving simplicity, clarity and transparency. But Bacon 41 resolutely presents the epistemological problem of linguistic mediation: The syllogism consists of propositions – propositions of words; and words are the tokens and signs of notions. Now if the very notions of the mind (which are as the soul of words and the basis of the whole structure) be improperly and overhastily abstracted from facts, vague, not sufficiently definite, faulty – in short, in many ways, the whole edifice tumbles. (The Great Instauration) Here the definition problematic (linguistic recursion mise en abîme) together with the interplay between ideas or conceptions and words radically undermines syllogism (a frequent brunt of Bacon’s derision). Isn’t it interesting that the linguistic implications of this rejection of syllogism – the implications that referential failure is probable and our understanding faulty – were overlooked? Fish’s “The Plain Style Question” makes clear that the diversity of categories of styles prevailing in the seventeenth century have confounded repeated classification efforts, and that there is no agreement whether science or religion played the biggest role in stylistic prescriptions. In a light-hearted ironic tone, Fish offers yet another pair of seventeenth-century styles – this time in terms of reading experiences: “the experience of a prose that leads the auditor or reader step-by-step, in a logical and orderly manner, to a point of certainty and clarity” and “the experience of a prose that undermines certainty and moves away from clarity, complicating what had at first seemed perfectly simple, raising more problems than it solves” (378). (We will return to this affiliation of certainty and clarity.) Overall Fish observes that, as a prescription for writing, officially “the plain style wins the day” over a prose carefully qualifying its own certainty (379). Notably, since this prescription (as Graham’s catalogue indicates) means nothing in particular and so many possible things, the “plain style” becomes an empty sign. Perhaps this is how the scientific plain style becomes mythologized. The naivete with which the idea of a plain style has been received seems to go together with code-model intuitions. We can see that it might have been here that the materiality of the signifier started to become transparent, and the sign become confused with the materiality of what it signified – a Cratylian dream, where the sign is essentially alloyed to its unique referent. 42 Parmenides’ epistemological quagmire concerns “deceitful” words which arbitrarily divide what is one into many. 50 He says, “thinking and the thought that it is are the same; for you will not find thought apart from what is, in relation to which it is uttered” (8.34-35, Robinson 110). Gallop remarks that “‘speech’ and ‘thought’ are coupled continually throughout” Parmenides’ text (“Introduction” 31n). This language-thought union which I believe was commonplace in the intellectual climate of ancient Greece appears in Plato a century later: for example (see Theat 190a, Soph 263e, Phaed 266b, see also the proto-Lacanian account in Phil 38e-39a). 2.4 epistemological quagmire: affective rationality ~ Having gauged the sign’s promiscuity (its freedom from referential restraint) and the Enlightenment-bred resistance to this insight (despite Bacon’s contemplations about the referential instability of words), we turn now to other epistemological implications of linguistic mediation, and consider the influence of commonplace assumptions about language’s rationality. We begin by contemplating language’s role in cognitive functions and the influence of the sign’s affective sociality on supposedly “rational,” “logical” mechanisms. Bacon’s above noted concern that potentially “faulty” “notions of the mind” are the “soul of words” places him in a long line of language sceptics aware of the worrisome epistemological implications of linguistic mediation (from Parmenides to Derrida). Fahnestock explores this word-mind intimacy in her work on topoi (that we discuss shortly) and in her work on recent neuroscience findings (discussed in Part 3); this intimacy is also apparent in Voloshinov’s endorsement of the ancient insight that we think in language (or at least some kind of semiotic). He observes that50 “the reality of the inner psyche is the same reality as that of the sign. Outside the material of signs there is no psyche; there are physiological processes, processes in the nervous system but no subjective psyche” (26). Interestingly, psychologist Silvan Tomkins will even link some of these physiological, nervous-system processes with the psyche and language. The biological responses, that Tomkins calls “affect,” and that constitute “the motivation system,” reveal an interplay between language, affect and thinking, intolerant of code-metaphor models and crucial to understanding how language actually works. It is relevant to this dissertation that recent discussion of this psyche-language union also recalls classical rhetoric’s appeals – socially-inscribed affective responses that speakers tap into for producing persuasive effects. These include, for instance, appeals to ethos which in order to persuade count on establishing the esteem of the speaker or identification between speaker and 43 audience. Jennifer Richards’ Routledge handbook Rhetoric identifies affect as the defining feature of rhetoric: “rhetoric is concerned with the affective power of language and with describing and classifying the devices that produce emotion, or develop a logical proof, and so sway the judgement of an audience” (121). It should not be surprising then that Voloshinov’s linguistic theory includes a chapter titled “The Philosophy of Language and Objective Psychology.” His purpose is not to analyze the individual psyche, but rather to trace the life cycle of the (social) sign – the way that it is shaped by its travels within an individual psyche, out into the world, and back in again – the socio-psycho gestation of the sign. This journey charts the sign’s subjective and inter-subjective social conditioning and accretions, providing a companion account to the absence of centre, the adventures of the trace and linguistic supplementarity that Derrida treats. For Voloshinov the individual “consciousness” is dependant on the social semiotic medium of language. In his account, the role of the human psyche and its social network in shaping the sign also indexes the psyche’s use of the sign. This reciprocal language-psyche connection signals the relation of language to an affective realm of subjective and inter- subjective desires, interests, feelings and beliefs – a language-psyche-affect conjunction. Halliday’s views fit perfectly with Voloshinov’s here, as they do with Nietzsche’s. For Halliday “language structures all of our experience, it enacts all of our interpersonal processes” (OL 251). “Language is the essential condition of knowing, the process by which experience becomes knowledge” he claims; and “we might interpret learning as something that is inherently a semiotic process” (Learning 94). It is worth taking time to absorb the radical claims that Halliday is making about the linguistic nature of consciousness, as well as the way this linguistic transformation of experience into meaning not only construes or constructs a referential reality, but also acts in it. Language is simultaneously a mode of acting and a mode of thinking: Language does not ‘reflect’ our thought processes; it actively creates (‘construes’) the categories and relations with which we think. Here ‘transforming matter into information’ means transforming our experience into 44 McIwain categorizes Tomkins’ affects more specifically as six basic affects (interest-excitement, enjoyment-joy,51 surprise-startle, fear-terror, distress-anguish and anger-rage), one affect-auxiliary (shame) and two drive-auxiliaries (disgust and dissmell) (500). meaning: both what we experience as ‘out there,’ in the world that lies round about us, and what we experience as ‘in here,’ as going on inside our own bodies and our own consciousness. Hence a theory of grammar (a ‘grammatics’) is a theory of how language does this. If that was all that the grammar achieved, we might say that it was just another system of information. But this is only one part of what the grammar does. As well as construing experience, the grammar also enacts our interpersonal relationships; and here it is not a mode of thinking – it is a mode of acting. When human beings evolved the resource of language, it served from the beginning not only as our means of information (the ideational function) but also as our means of interaction (the interpersonal function); and the two are inseparably bonded – you cannot have one without the other. Every act of meaning is both construal and enactment at one and the same time. (OL 276) Halliday here helps us see the potential choreography between the inter-subjective desires of the psyche and the social interpersonal functions of language that roll out at the same time as the construal of what we might imagine as an objective-referential reality that is out there, that we might think we are describing. Halliday thus highlights the thinking element already implicit in Voloshinov’s language-psyche-affect bond, which, as noted above, has a corollary in psychologist Silvan Tomkins’ script theory. Tomkins’ seminal studies of affect, which result in his theory of the affect (or motivation) system, have roots in Darwin’s pioneering work in The Expression of the Emotions in Man and Animals. Tomkins’ complex, hard-wired, biological-feedback-system model of affective responses both modifies and displaces Freudian drive theory. It foregrounds the discrete role played by the physiological-psychical, nervous-system responses he calls “affects” and their crucial contribution to human motivation. Briefly put, human affects are physiological responses with a psychic impact like terror, anger, shame, distress, startle, joy, and excitement. They are the responses that motivate human behaviour. The objects and aims of affects, unlike51 those of drives, are mobile and flexible; they respond to contingent factors impinging on them, and are governed by reciprocity. An affect can change, displace and replace its objects, attach to a new object, and join with other affects to act in concert. Tomkins describes them as multiform 45 I am thinking in particular of Daniel Kahneman’s Nobel prize in economics for examining the foundational-52 economics assumption that human behaviour is governed by self-interest and that humans are capable of “rational” decision-making. Kahneman’s findings about the non-rational basis of human, economic decision-making have since been repeatedly corroborated and become commonplace in public discourses, particularly in the aftermath of the 2008 stock market collapse. Components of Tomkins’ theorizing have become current (e.g., script theory, features of affect and its discreet53 motivational role) whether or not his authorship is recognized. Researchers in psychology and psychiatry familiar with Tomkins’ work observe that he has not yet had the full impact on current scholarship that he is likely to have in the future (de St. Aubin, McIlwain). Discourses on the relationships between linguistically formulated ideas, affect and cognition are commonplace in contemporary, psychiatry, psychology and sociology. But as McIlwain notes Tomkins is under-acknowledged, many don’t cite him where one would expect and few have read him in the original (500-01). In the context of my work it is ironic that his “dense and difficult to follow” writing is partly blamed for this uptake delay (de St. Aubin 1). Like any linguistically formulated understanding memory scripts can take place in varying relations to54 consciousness. Some may be unconscious like Tomkins’ example of what he calls an anti-toxic script: “a child who suffers massive intimidation at the hands of an alcoholic father who often beats him may retreat to a defensive introversion, seeking to hide from and to avoid or escape dreaded violence. In an antitoxic script the child could become very withdrawn to the point of catatonia” (190). An “addictive” script which transforms “a sedative into an end in itself,” on the other hand, might be quite conscious: “the cigarette addict exaggerates the necessity of smoking a cigarette for his well-being as the miser inflates the necessity of money” (192). and multifunctional because they can play different roles depending on incidental interactions and on whether they join forces with other affects and/or drives. The stellar role of affect as the primary source of human motivation and social development gives an added eminence in human cognitive processes to what has been considered passion or emotion – non-rational spheres. It creates a new picture of an affectively conditioned human rationality. In Tomkins the cognition and motivation systems are intertwined providing a novel image of what human reason is that is consistent with recent psychology studies (examining, for example, economic assumptions about rational decision-making).52 Tomkins’ studies on the motivation system lead to the discovery of a relation between linguistically formulated “scripts” or stories and human affective responses to situations and events. A script is a linguistically (or semiotically) remembered account in which affectively laden scenes undergo psychological magnification (a semiotic-cognitive interaction) connecting them to other affectively laden scenes. A script then becomes “a set of rules for predicting,53 interpreting, responding to and controlling a magnified set of scenes” or a series of events (Shame 179-80). A script constitutes cause-and-effect relationships between scenes or situations, which determine how an individual will respond to and interpret a situation. As54 predictors and determiners of human affective responses, Tomkins’ linguistic scripts, according to Ed de St. Aubin, anticipate the recent and much celebrated “emergence of narrative 46 Narrative psychology research into the way stories shape lives “the way human beings deal with experience by55 constructing stories and listening to the stories of others” challenges “the notion that human activity and experience are filled with ‘meaning’ and that stories, rather than logical arguments or lawful formulations, are the vehicle by which that meaning is communicated. . . . Sarbin (1986) proposes that ‘narrative’ becomes a root metaphor for psychology to replace the mechanistic and organic metaphors which shaped so much theory and research in the discipline over the past century” (Hevern). psychology” (1). Script theory thus links affective (biological) responses to linguistic55 formulations (accounts or stories) operative in the psyche. The reciprocal self-validating and sometimes self-fulfilling role that these linguistic scripts play between the motivation system and the psyche seems to corroborate Voloshinov’s conviction about the sign/psyche relation. One might, for example, experience an instantaneous, semi-conscious perception of an affective response (like excitement, shame, or distress) that triggers a kind of movie-like-memory that runs something like “this always happens when . . . .” This would be the surfacing of a Tomkins-script, a set of scenes which have been carefully edited and linked by aggregate affective pressures, and then stored to direct future interpretations of experience. The interplay between a person’s affective response and their (linguistic) interpretation of that situation reveals the way in which feelings or sentiments permeate what are thought of as rational cognitive functions. We might think that an “account” (linguistically formulated interpretation of a cause-and-effect relation) is rational, but if it is affectively directed and produced, we have to ask what “rational” means. What criteria distinguish rational from non-rational interpretations of experience? Not only does the abstract concept of “rationality” become amorphous, but what we think of as rational – cause and effect relations, patterns of thought – also acquire a nebulous, logical status. Like any linguistic account scripts are subject to constant revision; if linguistic scripts determine what is rational, then rationality too is constantly revised. Rationality is the signature quality of human reason. According to the Oxford Companion to Philosophy reason is the faculty of truth-seeking or problem-solving, and logic is sound reasoning (Honderich, ed.). However, Tomkins’ script theory suggests that the rational and logical live in a never-imagined, close proximity to the affective realm. The rational and logical are always motivated by human affective responses. Tomkins complicates naive 47 He says, for example, that “inasmuch as the neuron is the most elementary organized unit of the information56 processing mechanisms, it serves well to illustrate some of the general features of cognition, as well as the degree to which the simplest parts of a cognitive system share in the characteristics of the more complex cognitive mechanisms, including those of the total cognitive system” (Cognition 34). idealizations of rationality or reason, as such. In the “Introduction” to Cognition, he identifies the problematic role that the idealization of “reason” as a “divine spark” plays in the landscape of psychology (2). For Tomkins everything is biological and evolutionary. He believes that there are a very wide range of types of knowing – from the genetic and neuronal to complex human physical and mental problem solving. But “knowing” is widely varied even within the “restricted case of thinking that is oriented toward problem solving, there are critical differences between solving mathematical problems and learning how to hit a golf ball effectively” (6). Tomkins’ cognitive system is firstly intertwined with the motivational system (apparently he refers to the cognitive system as a discrete system provisionally for the purposes of analyses). The cognitive and the motivational systems sit in complex relations to each other: related by partial dependence, independence and interdependence that varies according to contingent factors and “the specific state of the whole system at any one moment” (7). Rather than some generalized and unexamined conception of rationality or reason (as some kind of abstract-logical deductive ability), Tomkins life-long empirical studies of human motivation in psychology lead to a conception of cognition as a kind of networking function fed by a variety of specific biological mechanisms that is interdependent with affective mechanisms. Cognition is not just operational in abstract cogitations about human ethics or how language works, but also in the complex motor-perception-memory mechanisms required to hit the golf ball and in the operations of the nervous system. For Tomkins cognition is a kind of56 organizing, connecting, information processing that is built into various biological mechanisms and always intertwined with the motivational (affective) system. Against unexamined ethereal conceptions of rationality or reason dominating common-sense understandings of human behaviour and language use, Tomkins’ affective responses activate and shape the “interpretation” that produces an account (whether that “account” is the conclusion of a scientific experiment, an estimation of how far you have hit the ball, or the explanation for your 48 For the currency of this connection between language, thought and cognition see a 2004 article on inner speech by57 Akira Miyake, et al. They write “language plays an important role in everyday living. Although not the only medium, the central role of language in human communication is incontrovertible. The utility of language is not restricted to interpersonal communication, however. It can also shape and support intrapersonal thinking processes in various ways (Carruthers, 2002; Gentner & Goldin-Meadow, 2003; Sokolov, 1972; Vygotsky, 1934/1986). One such intrapersonal cognitive function attributed to language is the regulation of thought and action (Barkley, 1997; Luria, 1961). In particular, at least among adults, inner (or subvocal) speech is considered to play some role in regulating one’s behavior (Carlson, 1997; Sokolov, 1972; see also Gruber & Goschke, 2004). In this sense, inner speech (and, more generally, language) can be viewed as a contributing factor to efficient executive control” (124). backache). The role of language in storing the interpretation of affectively shaped accounts in Tomkins’ theory, together with the increasing recognition of Tomkins-like views in psychology, corroborates Voloshinov’s views about the psyche being a semiotic domain (the zone of contact between the sign and the psyche’s emotions). The emergence of linguistic scripts from “the57 totality” of cognitive and affective mechanisms presents an image of cognition that is a complex of interdependent networks. These frequently involuntary, automatic mechanisms undermine common-sense assumptions about what “rational” means (Cognition 9). Motivation’s affective nature lingers ubiquitously in our interpretation of experience, complicating rational accounts or explanations and communication’s fuzzy inferential operations. De Man’s metaphor of resistance is helpful because it signals repression, something recognized as a “non-rational” element in human cognitive operations. The temptation to assume that especially scholarly judgement is reasonable, “logical,” and consistent tends to overlook the role of affect in cognition. On one hand, at a high level of generality, this assumption overlooks the way that the attachment to ideas is an affective response. It overlooks the way ideological attachments – like faith in the unmediated access to empirical knowledge – influence how we interpret things, what we see to start with, and what we think language is. On the other hand, at a microscopic level, the assumption that judgements are rational and logical overlooks the way that what we consider rational (a cause-and-effect script) is affectively shaped. If all linguistic presentations are susceptible to affect, the domain of the rational needs to be re-thought. We need to make room for the salient role of motivation in rationality. This re-thinking affects a whole series of interrelated assumptions about thought, language, and logic: that human thought is rational (without consideration of linguistic mediation and the role of motivation), that language is a transparent and impartial epistemological medium (without 49 De Man’s observations about the logic-grammar conjunction in the early Enlightenment are congruent with the58 uptake failure of Bacon’s sceptical views of language and Fish’s comments on the hegemony of the positivistic style. “Seventeenth-century epistemology, for instance, at the moment when the relationship between philosophy and mathematics is particularly close, holds up the language of what it calls geometry (mos geometricus) . . . as the sole model of coherence and economy. Reasoning more geometico is said to be ‘almost the only mode of reasoning that is infallible, because it is the only one to adhere to the true method, whereas all other ones are by natural necessity in a degree of confusion of which only geometrical minds can be aware.’ This is a clear instance of the interconnection between a science of the phenomenal worlds and a science of language conceived as definitional logic, the pre-condition for a correct axiomatic-deductive, synthetic reasoning. . . This articulation of the sciences of language with the mathematical sciences represents a particularly compelling version of a continuity between a theory of language, as logic, and the knowledge of the phenomenal world to which mathematics gives access” ( 13-14). consideration of its suitability for “accurate” linguistic identification and accounts), that language functions according to a coherent, logical, rule-governed grammar (without consideration of its socially conditioned, affectively motivated origins). De Man explores the affiliation of the ideas of (common-sense notions of rule-governed) grammar and logic in “Resistance to Theory.” He notes that there has been a “persistent symbiosis between grammar and logic,” that “the grammatical and the logical function of language are co-extensive,” and in fact “grammar is an isotope of logic” (14). Grammar, from58 this perspective, would be a type of logic – a confusion of precise mathematical operations with linguistic ones. The affectively motivated logics of linguistic scripts taint idealized images of a logical linguistic grammar. In fact it might be observed that the affectively laden linguistic sign makes a game out of “logical” grammar – an approach to language taken up by both Wittgenstein and Derrida. The role of logical operations in human language is usefully qualified and delimited in Sperber and Wilson’s theory as we shall shortly see. As well, we will consult Halliday’s views of what grammar actually is, in order to better assess this rule-governed- grammar-logic affiliation. What is at issue here, is the imprudent severing of affect and cognition (and of passion and reason) in our conception of rational thought. This divorce disrupts apparently complex interdependent networks of “logical” and affective functions (that lack simple binary polarity). We might do well to recall that human reason is never imagined as devoid of passion in Plato. Recall, also, that Aristotle’s rhetorical appeals are to ethos, pathos and logos; about logos he says “persuasion occurs through the arguments (logoi) when we show the truth or the apparent truth” (On Rhetoric 1.2.6). Logos persuades not only if logical, but if it appears to be logical. If the persuasive power of logical linguistic argument were recognized as 50 Bowers, et al., define intuition as “a preliminary perception of coherence (pattern, meaning, structure) that is at first59 not consciously represented, but which nevertheless guides thought and inquiry toward a hunch or hypothesis about the nature of the coherence in question” (74). Following this model Baumann and Kuhl “define intuition in terms of the perception of coherence” – the “spontaneous or preliminary perception of coherence that is not consciously represented” (1213). This definition together with the studies of the primary role of affect in intuition corroborate James’ original insight into the sentiment of rationality as well as directing attention to the question of part whole relations imbricated in cognition (for Van Dijk and Tomkins) and classification (for Plato and Wittgenstein) as we shall see. Both cognitive and classificatory uses of part- whole relations are linguistically (or semiotically) indexed. its allure, we might understand reason quite differently. In this logic-affect vein, Bacon highlights the charms of logical deduction: “Syllogism, as it is a thing most agreeable to the mind of man, so it hath been vehemently and excellently laboured” (AOL XIV). The pleasure of logic again recalls Aristotle’s appeal to feelings of confidence and fittingness. William James’ corroboration of the pleasures of logic helps to further disarticulate some of the linkages between language, logic and rational thought. In a gesture which seems to anticipate Tomkins’ views on the role of affect in cognition, James’ “The Sentiment of Rationality” argues that rationality is a sentiment – an affective response, “a strong feeling of ease, peace, rest.” A recently published study in experimental psychology over a century later echos James’ language when it identifies a “fluency-triggered positive affect” experienced “as a cognitive feeling of ease” (Topolinksi and Strack 40). The affective response identified in these experiments is provoked by intuitions of coherence very like the sentiment of rationality that James describes. James says that rationality is the experience of things fitting together, a “sufficiency of the present moment” and “absence of all need to explain” or to account for something (James 58). He says that “the transition from a state of puzzle and perplexity to rational comprehension is full of lively relief and pleasure” and suggests that there are good reasons for upholding the view that rationality is “the absence of any feeling of irrationality” (57). Coherence, for James, is experienced as a feeling; the sentiment of rationality is a feeling of cognitive “fluency” (58). Coherence and fluency – both associated with intuition in contemporary experimental psychology studies – show that affect plays a variety of roles in cognitive processes. For59 example, Topolinski and Strack list the range of studies in experimental psychology which have established that long-term, positive-mood affect facilitates “semantic spread,” increasing both 51 Baumann and Kuhl conclude that “access to coherence producing cognitive processing . . . depends on affect60 regulation” (1221). According to Topolinksi and Strack this and other Kuhl studies advance an “affect modulation hypothesis” in which a general and lasting mood affect promotes or inhibits semantic spread and thus conceptual fluency (49). See Topolinksi and Strack’s extensive list of studies regarding mood affect and certainty judgements, risk taking, etc. (48). In most of the studies referred to here semantic coherence was tested using word triads. This makes these studies61 particularly apt for consideration of the relationship of mathematical logic to language use. accuracy and fluency in cognition. Their own efforts to refine a model of intuition determines that short-term, positive “core affect” contributes to conclusions of coherence (whether or not coherence exists). These findings, which both substantiate Tomkins ideas about the60 involvement of affect in cognition, and summons James’ wordings, further challenge the characterization of human thought – often called “reason” – as logical (or rational) . James’ treatment of rationality – as a kind of confidence (that things fit together) and a kind of feeling – is corroborated by these experimental-psychology studies which find that high fluency intuitions of coherence trigger positive affect. Topolinksi and Strack determine that61 the short-term “core affect” is the outcome of the intuitive chain triggered by fluency and, most significantly, that this core-affect response can be manipulated to produce counter-to-fact intuitions (that is, incoherent things are perceived as coherent) (58). In other words, the properly timed production of affect can make the logical appear unlogical and vice-versa – notwithstanding that the term “logical” (applied to linguistic usage), like its counterpart “rational” has lost some of its lustre in the course of this discussion. Rhetorical theory has known for a long time that when it comes to linguistic arguments logical appeal is not the syllogism, not the long train of reasoning, but the appearance or feeling of logic (see, Rhetoric 1.2.14). Perhaps an orator’s success is knowing how to trigger intuitions of coherence; and perhaps it is just such knowledge that provoke Socrates’ worries (in the Gorgias) that a medical patient might be more easily persuaded by a rhetor than a physician. Aristotle’s shrewd advice that a speaker target the audience to whom he speaks has some relation to James’ view that rational sentiments are stimulated differently in different individuals. Crudely stated, James suggests that some individuals are attracted to certainty like simple religious or logical/scientific accounts – whereas others crave excitement and are drawn to puzzles, difficulty and irrationality. While James is concerned with individual subjectivities, 52 the malleability of what is considered rational or coherent is also cultural and thus applicable to any community to whom Aristotle’s orator might be speaking (e.g., a religious community, a legal community, an ethnic community, an adolescent community). But James’ individual focus offers particular insight into the resistance to theory. Since the sentiment of rationality is produced by a complex of dispositions and these dispositions are different in different individuals, what in some, produces anxiety and creates an obstacle, for others stimulates curiosity and acts as enticement. If affective elements cannot be severed from coherence, rational thought must be both affectively modulated and guided by subjective disposition. Affect not only plays a part in a sign’s history, but language use itself is affectively motivated according to both Tomkins and the findings of scholarship on reported speech. Notions of explicit reference, logically presented description, and coded communication are now completely muddled, not only by the whimsies of the sign, but by the role of affect in human thought and motivation. Language is not a sterling instrument awaiting its proper use: a clean tool, incisive and clear-cutting. As a product of affectively conditioned experience, it is itself a musty medium. 2.5 epistemological quagmire: analogy & knowledge ~ as Plato saith, Whoseover seeketh, knoweth that which he seeketh for in a general notion; else how shall he know it when he hath found it? (Bacon AOL 48) While the wobbly condition of the sign renders linguistic mediation problematic in its own right, the role of affect in human rationality and the intercourse between affect and sign compound its already debilitated condition. The question of what constitutes rationality in the first place – what kind of mental operations are in play when we attempt to understand the world – might be asked. According to James, one powerful player is analogy, the cognitive mechanism through which identification takes place. When identification (using a linguistic sign) takes place by means of analogy to another thing, then the history of the sign (of the other thing) comes along for the ride. That “analogy lies at the core of human cognition” is a 53 Holyoak cited in Leech, Mareschal, & Cooper. See “Analogy as relational priming: A developmental and62 computational perspective on the origins of a complex cognitive skill,” and responses in Behaviour and Brain Sciences. Recall Foucault’s critique of classifying diseases according to symptoms in Birth of the Clinic as though symptoms63 indicate absolute genus-species categories. Under the heading “The Principle of Analogy” he discusses both the process of classification in order to identify diseases to start with and then the function of analogy within that process. He says, “it is the analogy of these relations that makes it possible to identify a disease in a series of diseases” (100). commonplace of psychology and cognitive science. Green et al., say that “analogical thinking62 is a fundamental cognitive tool for learning, understanding and generating novel ideas” (1005). James’ illustration of how analogy stirs the sentiment of rationality highlights both the recognition element of analogical reasoning and the affective processes entailed in it. Citing Bain’s comments James notes that: “‘A difficulty is solved, a mystery unriddled, when it can be shown to resemble something else; to be an example of a fact already known.’” He adds that “the craving for rationality is appeased by the identification of one thing with another” (62-63). Here, the identification of a thing by analogy to another thing produces the sentiment of rationality: the feeling of satisfaction that comes from identification provokes coherence intuitions. If this is what human reason is (a feeling of things fitting together), and logic in linguistic argument is what we feel follows (what seems coherent), we might want to be circumspect about what kind of identification results from an affectively motivated analogy between thing A and thing B – especially when thing A is identified (not by a unique sign “A” but) by a sign over which it has no propriety in the first place, a sign any other thing can use. It is not just a question of affective involvement in rational cognitive functions, but a question of linguistically mediated identification (and classification) which makes all classified things caught up in the endless chain of signification. And it is not just the role of affect and semiotic promiscuity that are problematic, but the complex and error-prone nature of the cognitive use of analogy. French’s observations about the cognitive uses of analogy63 underscore the selectivity involved in linguistically-mediated, analogical identifications. He highlights that there are a variety of uses for analogical reasoning (that is, different types of cognitive functions like the generalizing-classifying function, or the unique-identification function). French’s list of analogy-uses includes, for example, 1) emphasis of particular selected 54 French states that analogy making “encompasses our ability to explain new concepts in terms of already-familiar64 ones, to emphasize particular aspects of situations, to generalize, to characterize, to explain or describe new phenomena, to serve as a basis for how to act in unfamiliar surroundings, and to understand many types of humour” (200). features or parts of a thing, 2) generalizing a thing, 3) specifying a thing. But observe that64 reductive (generalizing) interests, particularising interests and interests in only one part, are not the same and will lead to different analogies for the same thing (that is, A, B, and C will all be used to identify D). Furthermore all of the analogies will be characterized by varying (and unmeasurable) degrees of inaccuracy depending on the context. Such comparisons are vague at best. French’s example of an American mistaking a European bidet for a toilet is a typical (if embarrassing) identification error. One wonders what kind of identification this analogical process could possibly produce and thus how “accurate” referentially descriptive code could possibly be – were it even remotely possible for the human brain to remember the number of terms that would be required to label everything determinately. Insufficient knowledge of the old (the toilet) or the new (the bidet) signals the limits of human perception and knowledge, leading to category mistakes and false identifications. “Analogy-making,” according to French, “is typically conceived of as involving a ‘mapping’ between two domains” (200). Analogy is a metaphorical operation, (meta-fevrw) a carrying over or transfer from one sphere to another (as Nietzsche says), neither precise nor absolute. What is classified as one thing might later be “discovered” to be something else as Kuhnian caveats warn and Plato’s Timaeus illustrates. Linguistically mediated identification and classification’s dependence on analogy renders these cognitive operations vague and fuzzy at best. Insight failures, misunderstandings, false assumptions make reasoning by analogy error prone and tending to category mistakes. But the expression “category mistake” itself is problematic; it wrongly implies the potential for “correct” identification. The term “correct” isn’t really applicable in reasoning by analogy. In his analysis of insurance claims, for example, Whorf observes the hazards posed by gasoline drums that were erroneously deemed “safe” because labelled “empty” – when instead they should have been deemed potentially more volatile than full ones (135). But the term “empty” elicits the category mistake “safe” with novice gasoline 55 handlers. It seems to mean that the dangerous substance is gone and the danger absent, whereas it actually means the opposite, that the drum might now be “full” of an air/gasoline mix that is much more volatile. French refers to several abstract processes necessary for analogical reasoning including recognition of a source and target, mapping between the two, and transfer from source to target (200). These mapping and transfer processes are the essence of metaphor, of the tropological, so it is not just the linguistic sign but human reasoning itself that is tropological. “Recognition” (of a source and target), however, requires knowledge of a thing and careful inspection – not the overhasty and vague generalizing that Bacon worries about. In French’s example, the drain on the perceived “toilet” needed to be carefully apprised. Note also that the sign “toilet” was known by French, but the sign “bidet” was likely not known. James has suggested that rationality is triggered by “the identification of one thing with another” (like a bidet) as “an example of a fact already known” (like a toilet) (63). This recognition is conditioned by the paradoxical knowledge-recognition paradigm – namely, that one cannot recognize what one does not know – a salient insight, as Bacon reminds us, of Plato’s Meno. Wittgenstein too says: “One has already to know (or to be able to do) something in order to be capable of asking a thing’s name” (§30). One must be able to identify or delimit the boundaries of a thing in order to inquire of its name. Analogical thinking depends on recognition, which depends on (linguistically mediated) “knowledge” – a slippery commodity at best. And we can never know whether it is correct or incorrect, only effective or not effective; “accuracy” can only be estimated from empirical results. The cognitive use of analogy, and identification by recognition imbricate fuzzy, knowledge-dependent cognitive functions with the mediation of the linguistic sign. These fuzzy metaphorical operations call into question the degree of precision that is possible with a socially-conditioned and promiscuous linguistic sign. A finite lexicon together with limitations on human memory and perspective further alienate prospects of linguistic precision. But whence comes the prestige of precision? According to Wittgenstein, it is logic, not language use or 56 communication, that demands precision. He thematizes the often vague uses of language in terms of “exactness” and concept-delimiting “boundaries.” In a discussion of the whimsical linguistic operations of unregulatable language games, he queries “does it take that [a clearly delimited boundary] to make the concept usable? Not at all!” (§69). His Cratylian interlocutor later asks “But is a blurred concept a concept at all?” implying that precision failure means that the thing referred to does not exist. Wittgenstein dismisses this suggestion out of hand with the rhetorical questions: “Is an indistinct photograph a picture of a person at all? Is it even an advantage to replace an indistinct picture by a sharp one? Isn’t the indistinct one often exactly what we need?” (§71). “[Inexact],” he asserts, “does not mean ‘unusable’” (§87). And it seems possible that a generalized understanding of how language works might be more broadly usable than a precise one: a fuzzy picture of how language works, might be more serviceable for keeping linguistic mediation in mind, for keeping linguistic consciousness at the intellectual scene. Querying the prestige status bestowed by precision (and required only for logic) Wittgenstein goes on to discuss “the use of words with games and calculi” (§81): [I]f you say that our languages only approximate to such calculi you are standing on the very brink of a misunderstanding. For then it may look as if what we were talking about were an ideal language. As if our logic were, so to speak, a logic for a vacuum. – Whereas logic does not treat of language – or of thought – in the sense in which a natural science treats of a natural phenomenon . . . (§81) Wittgenstein argues that the precision of logic is not applicable to the kind of (non-ideal) language that we actually use (“ordinary language”) and which, he is at pains to point out, is perfectly orderly without emendation to propositional form. Halliday likewise observes that “the ‘grammatical logic’ of a natural language is different from mathematical logic” (285). Wittgenstein’s move – displacing the paramount role that had formerly been reserved for mathematical logic in linguistic communication – is significant for analytic philosophy. The codification-resistant system of language games Wittgenstein uncovers cannot be regulated in 57 We see the way that the Tractatus provides context to the Investigations when Wittgenstein says: “it will then also65 become clear what may lead us (and did lead me) to think that if anyone utters a sentence and means or understands it he is operating a calculus according to definite rules” (§81). They say: “human spontaneous non-demonstrative inference is not, overall, a logical process” (69).66 the determinate way that some philosophers (like Wittgenstein himself) had once hoped.65 2.6 fuzzy logic and the biochemical image of grammar ~ Whatever its failings the state of scientific technology is a persuasive indication that language works. Human reasoning processes use some kind of relatively “logical” process despite linguistic-analogically-produced fuzziness and even if we cannot delimit the precise extent to which intuitive-logical principles or affective responses govern specific processes and their outcomes. Sperber and Wilson’s conception of the role that deduction plays in non- demonstrative inference (that is, in inferences where conclusions are not necessitated by premises) seems also applicable to what we more broadly think of as rational thought. Recall that “logical” is one of Halliday’s metafunctions associated with ideational processes and parallel to the experiential dimension which is construed by grammar. In the ostensive inferential model, deductive functions are limited by a more or less secondary (but crucial) role played in producing contextual implications from utterances (the effect produced from the combination of what is already known and the assumptions that have been made manifest). This “deductive device” explicates the contents of assumptions according to a set of formal elimination rules together with other non-deductive heuristics to produce contextual effects. Contextual effects are then gauged according to relevance: a measure of their size and their processing effort that is a kind of informed-intuitive hunch. Sperber and Wilson speculate that traditional, logical, elimination rules (modus ponens, modus tollens, conjunction, disjunction, etc.) spontaneously performed in non-demonstrative inference “might be less a logical process than a form of suitably constrained guesswork” (69). Although the entire enterprise would be66 qualified by the un-delimited role of affect (that is, of motivation) and the fuzziness of linguistic-analogical-reasoning, results can be quite “logical” in a fuzzy sort of way, just not as precisely so as the logical positivists and analytic philosophers may have wished. While Sperber and Wilson’s speculations about the role of a deductive device in 58 See Remarks on the Foundations of Mathematics (§ 131). He adds “the propositions of logic are ‘laws of thought,’67 ‘because they bring out the essence of human thinking’—to put it more correctly: because they bring out, or shew, the essence, the technique of thinking. They shew what thinking is and also shew kinds of thinking” (§133). In Figures she analyzes the arguments of Kekule, Mendel, Lavoisier and Harvey who write on matters falling into68 modern disciplines of chemistry, genetics, palaeontology, medicine. producing contextual implications from utterances are focussed specifically on the mechanisms of communication, they assert that generally “humans are rather good at non-demonstrative reasoning; otherwise the species would be extinct” (68). This takes us back to the question about what motivates reason. Since the premises, for Sperber and Wilson, do not necessitate the conclusions, non-demonstrative reasoning might look a lot like narrative. Whatever logic and rationality are exactly, it seems clear that commonsense assumptions about them need to be questioned. We need to keep in mind reasoning’s cognitive dependance on analogy (or metaphor) and rationality’s motivation and (affect-laden) linguistic ties. Rather than telling us something about how language works, Wittgenstein claims that logic is an indicator of human mental habits. He says “the laws of logic are indeed the expression of ‘thinking habits’ but also of the habit of thinking. That is to say they can be said to shew: how human beings think, and also what human beings call ‘thinking.’” This suggestion resonates interestingly with67 Halliday’s conception of grammar as we shall shortly see. The questions arising here include what criteria determine the logic of these thinking habits, and what produces linguistically formulated patterns of logical entailment. Fahnestock takes up the question of rational “argument” in the writings of early modern science. She analyzes these texts to determine how their writers thought about science and concludes that the models of reasoning evident in the writings of Enlightenment science were produced by the rhetorical-dialectical training of the day. She compares the exhaustive forms of argument catalogued in sixteenth-century-humanist dialectic manuals (which were a centrepiece of early modern education applicable in all disciplines) to the science writing of the Enlightenment. These manuals provided named rhetorical figures and patterns of argument,68 memorized and practised by students; “the pattern itself,” Fahnestock claims, was considered “the source of its ‘reasonableness’” (Stylistics 226). Inscribed in the reasoning processes of 59 Fahnestock cites Melanchthon’s early sixteenth century comment: “So great is the resemblance between Dialectics69 and Rhetoric that scarcely any distinction can be observed between them” (Figures 122). See for instance McAdon Signs, Rhetoric, Leff Topics, Topoi For related discussion of invention see Miller Special70 Topics, Novelty. Leff says that “the number of classical conceptions of topics rivals the Babel of modern scholarship, since classical71 rhetoricians are not very scrupulous in defining this key term. Aristotle, for example, makes the topics one of the central elements in Rhetoric without ever explicitly defining what they are” (Topics, 23). Enlightenment-scientific texts she discovers the contours of dialectical arguments from the training manuals. Arguing that the correspondence between the reasoning and the composing practices of these writers stemmed from their dialectical training, she concludes that what passes for rational thought was shaped by Renaissance dialectical pedagogy. Jardine’s analysis of Bacon’s art of discourse makes clear that what theorists like de Man call rhetoric is what Bacon treats as dialectic. Fahnestock clarifies that during this period dialectic and rhetoric were in such close proximity as to be barely distinguishable. Thus both terms are associated with the69 practice of argumentation that Fahnestock analyzes as well as the verbal patterns she calls “figures of argument.” It seems very likely that Wittgenstein’s “thinking habits” that constitute the “laws of logic” might line up nicely with the verbal patterns inscribed in dialectical training manuals. Fahnestock refers to the (now lost) “intimate connection” between figures and topics in the training manuals’ hybrid disciplinary approach (Stylistics 228, 224). The argumentative strategies Fahnestock researches (which are adumbrated under both style and invention categories in rhetoric-dialectic manuals) include classical rhetoric’s topoi: generally understood as lines of argument (and hence invention). Perhaps this style-argument ambiguity explains why contemporary theorists have found Aristotle’s topoi to be “confused” and why there is no agreement about exactly what Aristotle means by “topos.” According to Leff the ambiguity70 about what topoi are remains constant throughout antiquity. Leff observes that the “term71 ‘topic’ incorporates a bewildering diversity of meanings” and that “among modern authors we find conceptions of the topics ranging from recurrent themes in literature, to heuristic devices that encourage the innovation of ideas, to regions of experience from which one draws the substance of an argument” (Topics 23). Whether as recurrent themes, invention heuristics, or 60 patterns and materials of pre-existing argument, curiously enough there is widespread agreement on the topoi’s strictly pedagogical purposes – a view which Quintillian makes unquestionably clear (see Leff Topics 34). Topoi were designed for practice exercises not as ends in themselves; they taught students how to think, that is, they taught verbal patterns of reasonableness. Leff says that “just as gymnastic training prepares the body for physical performance, so can exercises in topical reasoning prepare the mind for verbal performance” (Topoi 208). The suggestion that what is “logical” or rational (perhaps stemming as Wittgenstein suggests from thinking habits) might be tied to the verbal patterns of linguistic usage lines up with Halliday’s conception of grammar. There is no question that for Halliday thinking habits are embedded in the grammar of language, in what he calls the “wordings,” representations that organize meanings in lexicogrammar. Halliday believes that our “extraordinarily rich and varied” human experience is “mediated through the senses on various levels” and that “what the grammar does is to transform this experience into meaning” (Matter 63). Another way of saying that “grammar ‘transforms experience into meaning’” is to say that grammar is a theory of human experience. The lexicogrammatical representations or wordings (I should make clear that ‘grammar’ includes vocabulary, the more specific elements of the wording) – that is, the bits of discourse that we recognize as speech, or as writing, construe fragments of experience, real or imaginary, in ways which relate each instance to this overall theory. (Matter 63-64) The wordings or bits of discourse theorize how the world works. The construal of experience constitutes a “theorizing [of experience] in the form that we call understanding” and sets up “logical-semantic relationships between one clausal unit and another” (OL16-17). These (grammatically) “logical” relationships learned through language are indeed an expression of human thinking habits, as Wittgenstein says. Halliday’s account suggests that grammar is syntax and vocabulary, learned patterns of speech, wordings, bits of discourse: that is, very like Fahnestock’s verbal patterns derived from rhetorical figures – the “figural logic” of verbal topoi, received, recognizable lines of argument, more like narrative than (deductive- 61 mathematical) rules. Recall de Man’s comments affiliating “logic” and rule-governed grammar. Commonsense notions of grammar circulate in code-models of communication and conceptions of rational thought: where logic is imagined as symbolic/mathematical formula, and grammar as logic’s mirror image in linguistic signs. In this commonsense image, grammar is a set of logical rules to be followed. Halliday’s radically different understanding of what grammar is specifically excludes such conceptions: grammar is “not, of course, the image of ‘grammar’ that many people carry around with them” which is “the grammar of primary school, where they were taught that a language is a set of rules to be obeyed” (Matter 62). For Halliday, who claims that the organizing concepts of his thought emerged as a by-product of struggling with particular problems in language, “a language is a system of meaning – a semiotic system” (OL 1-2). “A language is almost certainly the most complicated semiotic system we have,” he says; “it is also a very fuzzy one, both in the sense that its own limits are unclear and in the sense that its internal organization is full of indeterminacy” (OL 2). He writes that “the grammar of any language can be represented as a very large network of systems” but that “the network is openended” (OL 180). For Halliday systemic theory is “like language itself”: a “system whose stability lies in its variation”; it is “a ‘metastable’ system,” which “persists because it is constantly in flux” (OL 192). “A lexicogrammar is not a closed, determinate system” and clearly has little in common with rigid conceptions of logic as an absolute grid of entailments according to determinate rules (OL 194). Halliday’s “five critical features of the grammar of a natural language” present an entirely other image of grammar than the notably-absent adjective “rule-governed” conveys; highlighting the flexibility and fluidity of language, systemic grammar is said to be “comprehensive, extravagant, indeterminate, non-autonomous and variable” (OL 250). The construal of experience in language not only creates meaning, but for Halliday (as for Whorf) “grammar models reality” (OL188). Language is used to “construe experience, where construe means ‘construct semiotically’ – that is transform experience into meaning”; Halliday explains 62 that “experience [here] relates the process of construal to the experiencer rather than the ‘real world’” (OL 244). We will return to these shaping, constraining/enabling features of grammar on our interpretation of the world in Part 3, but for now, it is important to recognize the two- sided, symbiotic role of lexicogrammar. For Halliday language both is meanings (semiotic) and makes meanings (semogenic). He says the usual way we talk about language is by saying that language ‘expresses’ meaning, as if the meanings were already there – already existing, in some formation or other, and waiting for language to transpose them to sound, or into some kind of visible symbols. But meaning is brought about by language; and the energy by which this is achieved, the source of its semogenic power is grammar. (Matter 63) Halliday here echoes Foucault’s notion that language produces what it seems to describe and his grammar, figured here as energy, resembles the complex dynamics of biochemistry more than any set of “logical” rules we could use to analyze truth conditions of propositions. Fahnestock seems to be thinking along the same lines as Wittgenstein and Halliday when she suggests that “reasoning and [verbal]-pattern completion” are in close proximity (Figures 124). When he asks “if language is a code, then where is the pre-coded message?” Halliday is also thinking along the same lines as Voloshinov of the liveliness of semiotic sphere and of grammar (OL 243). We are reminded that thinking takes place in living language not abstract mathematical symbols. Halliday’s innovative view of grammar as wordings, also resonates with Voloshinov’s account of the fragmentary nature of inner speech and human linguistic thought. In Voloshinov’s account thinking, or inner speech, employs signs or words, not in sentences, but in fragmentary utterance that are “joined together” and “alternate” not according to received commonsense notions of the “laws of grammar or logic, but according to the laws of evaluative (emotive) correspondence”: The units of which inner speech is constituted are certain whole entities somewhat resembling a passage of monologic speech or whole utterances. But most of all, they resemble the alternating lines of a dialogue. There was good reason why thinkers in ancient times should have conceived of inner speech as inner dialogue. These whole entities of inner speech are not resolvable into grammatical elements (or are resolvable only with considerable qualifications) 63 “Remembered prior texts” is a stock phrase in Becker: “the actual a-priori of any language event – the real deep72 structure – is an accumulation of remembered prior texts” (86). and have in force between them, just as in the case of the alternating lines of dialogue, not grammatical connections but connections of a different kind. These units of inner speech, these total impressions of utterances, are joined with one another and alternate with one another not according to the laws of grammar or logic but according to the laws of evaluative (emotive) correspondence, dialogical deployment, etc., in close dependence on the historical conditions of the social situation and the whole pragmatic run of life. (38) Alternating lines of dialogue are not logical propositions, but bits and pieces – like Halliday’s “bits of discourse.” As we shall see in Part 3, the transcripts of actual conversational dialogue used in studies of reported speech show that communication is often effected ostensively in gestures – using single words of support, repeating what another has just said, play acting someone else’s utterance – rarely in the whole sentences of a rule-governed grammar. Again, as if he were in a conversation with both Fahnestock and Halliday, Bolinger asks “is grammar something where speakers ‘produce’ (i.e. originate) constructions, or where they ‘reach for’ them, from a preestablished inventory, when the occasion presents itself?” (381).72 Voloshinov is mindful of the historicity of the sign with its constant re-accentings and accretions when he draws attention to attendant stylistic shifts that effect what he calls gramaticalization and degrammaticalization. He says that “stylistic individualization of language in concrete utterance is historical and creatively productive. It is here precisely that language is generated, later to solidify into grammatical forms: everything that becomes a fact of grammar had once been a fact of style” (51). Voloshinov seems here to imagine bits of discourse or wordings that begin life as a stylistic innovation and then become established norms. He says later that “the borderline [between style and grammar] is fluid because of the very mode of existence of language, in which, simultaneously, some forms are undergoing grammaticalization while others are undergoing degrammaticization” (126). Recall Fahnestock’s account of the overlapping classifications of argument and style: fluid boundaries bridging style, syntax, patterns of speech and argument she calls figures of argument. With his mind on rules established after a stylistic innovation has become an established norm, 64 A “system,” says Halliday, “is not the same thing as ‘systematic’” (OL 180). 73 Wittgenstein says “The grammar of the word ‘knows’ is evidently closely related to that of ‘can,’ ‘is able to.’ But74 also closely related to that of ‘understands’”(§150) Voloshinov denounces the philological focus of early-twentieth-century linguistic study, claiming that “systematizing grammatical thought could have developed to its full scope and power only on the material of an alien, dead language” (78). Such systematizing (rule-73 governed) grammatical thought he says, “interpret[s] living language as if it were already perfected and ready-made and thus must look upon any sort of innovation in language with hostility” (78). Voloshinov suggests that beneath views of “language as a system of normatively identical forms lies a practical and theoretical focus of attention on the study of defunct, alien languages preserved in written monuments” (78,71). Halliday too observes this “historical association of linguistics with writing”; he observes that “linguistics begins when a language is written down, and so made accessible to conscious attention”; and he comments on the distorting influence of what others have called the written language bias, to which we return in Part 3 (OL 193). Voloshinov’s focus on the sign’s historicity and his celebration of the sign’s fluidity recognize that linguistic semiotic materials, their grammatical forms and stylistics, are fluid, constantly reacting, responding and changing. Halliday offers a new way of thinking about grammar, that welcomes the caprices of Voloshinov’s living historical sign, and an image that allows us to glimpse the complex biochemistry of grammar – too multifaceted, flexible and vast to be a set of memorisable rules. This image of grammar invites reconsideration of rational thought as verbal patterns and syntax of language that we learn from our parents and society – or perhaps from Renaissance-dialectic manuals. Halliday’s notion of grammar has fuzzy logic: it is too complex to keep track of as a set of “rules” to be followed and the grammar of individual words is often unique. Grammar is mutable like language itself. Such fuzzy74 linguistic logic may have natural ties to the standard elimination rules that Sperber and Wilson describe. Though it may have less prestige than formerly attributed to the mythical-precise linguistic logic, their non-demonstrative deductive guesswork seems to work quite well. 65 Nevertheless, Cratylian code-metaphor convictions of a rule-governed grammar still prevail in accounts of how language works. Sperber and Wilson complain that “despite the widespread scepticism about the role of deductive reasoning in comprehension, many existing pragmatic theories, especially those built on Gricean lines, seem to be based on informal systems of just this type” (93). 2.7 epistemological quagmire: part-whole relations ~ The human language-thought ligature together with rational thought’s affective components and analogical cognitive operations are further complicated by the subjective vantage of human language users. When it comes to epistemological identification and classification efforts, our perspective limitations compound the problems already amassed by the delinquent habits of the sign. To begin with, Cratylian conceptions of rule-governed grammar do not explain how human memory could manage the number of terms necessary for the improbable-impossible undertaking of uniquely identifying every aspect of reality. However, Van Dijk’s work on the cognitive functions of classification and specification helps us understand how human memory makes do with a finite lexicon and linguistic signs. James has anticipated the trajectory of Van Dijk’s cognitive functions when he distinguishes a reductive “passion for parsimony, [and] for economy of means in thought” (which is a focus on generalization or abstraction) from a “passion for distinguishing” which is an “impulse to be acquainted with the parts rather than to comprehend the whole” (58-59). Identification of a part and its classification according to category are specification and generalizing techniques that Van Dijk identifies in his work in linguistics, cognitive psychology, and discourse theory. Considering foregoing meditations on affective motivation and sentiment in rational thought, and linguistic complexity more generally, it should not be surprising that Van Dijk’s explorations of the workings of language found it necessary to resort to cognitive-functional explanations. His analyses determine that specification and generalization techniques are fundamental, paired cognitive functions used for memory and cognition. He describes these discursive 66 Gonzalez raises this vision-beholding interpretation of qewrei:n in connection with Plato: “what one gains through75 entering the world of dialogue is not a ‘theory’ in the sense of a ‘doctrine,’ but a ‘theory’ in the sense of a ‘vision’” (Third 18). See also Statkiewicz (2). cognitive operations (of specification and generalization) as global structures that organize local information for storage, retrieval and comprehension. These generalizing (or abstracting) and specification (or particularizing) cognitive operations depend on “an intuitive primitive” function of part-whole relations which Van Dijk asserts “cannot be analyzed into more basic cognitive notions” (3). Generalizing and specifying are then basic cognitive operations conducted in linguistic signs. An epistemological problem is posed by identification and classification however: linguistically framed identifications by analogy depend on being able to see, in the figurative sense (that is, knowing), what part fits into what whole – and there are many wholes in which a given part might fit. This brings us back to the question of context. Of course such fitting is a human passion; fitting parts together into a whole induces the sentiment of rationality. But how can we know that any given thing fits into any particular context, and how that context fits into the larger context? When the different contexts into which a thing seems to fit are mutually exclusive (a problem guaranteed by the linguistic sign’s promiscuity), when the same word is used to describe radically different things (that cannot be classified together) or to describe things that can only be classified together in limited ways (but are otherwise too different), the limited perspective of human vantage interferes with epistemological identifications. Arguably the total knowledge-vision (omniscience) required to know where to draw these lines is superhuman and requires divine speech. The Greek term qewrei:n, the etymological root for ‘to theorize,’ means ‘to see,’ ‘look at,’ ‘view,’ ‘behold’ (often in a public or sacred context). Recall that when he is describing the75 requirements of the art of rhetoric Socrates says “I myself am a lover of these divisions and collections, that I may gain the power to speak and to think (emphasis added), and whenever I deem another man able to discern an objective unity and plurality, I follow in his footstep where he leadeth as a god . . . for those who have this ability [and if Socrates doesn’t who does?] . . . I 67 Aristotle’s Posterior Analytics attempts to deny this problematic: “In defining by division there is no need to know76 all the facts. Some, however, maintain that it is impossible to know the differentiae between each thing and the rest without knowing each thing severally, and impossible to know each thing severally without knowing the differentiae” (Post. An. 2.13, 97a). According to Schiappa, dialektikhv (the art of speech) is one of several terms Plato coined by adding the term77 tevknh to an existing root including rJhtorikhv or the art of rhetoric (skill, art, craft) (15). These two terms can have a very similar meaning and so called “good” rhetoric and dialectic are aligned in Phaedrus. The ability of the dialectician is what is needed for the “good” rhetorician; the “bad” rhetorician does not have this ability/knowledge. Socrates says: “Then the art of contention in speech in not confined to courts and political gatherings, but apparently, if it is an art at all, it would be one and the same in all kinds of speaking, the art by which a man will be able to produce a resemblance between all things between which it can be produced, and to bring to the light the resemblances produced and disguised by anyone else” (Loeb 261e). call dialecticians” (Hackforth Phaedrus 266b). Dialectician, here, has nothing to do with76 Hegelian dialectic (thesis or argument); the root terms for Platonic dialectic and dialectical are built on diav (through) and levgomai (‘speak,’ ‘say’); the parts of dialectic add up to something like ‘through speech.’ Plato’s dialectician is s/he who is skilled in speaking (and thinking) and77 s/he who can through speech discern a collective unity and plurality. It is this impossible knowledge and ability that are needed to make accurate identifications and classifications – impossible knowledge and skill that warrant the Socratic stance of ignorance. This inaccessible qualification is required for speaking and thinking in Plato’s art of rhetoric. Only omniscient divine speech (with infinite memory) could be code. Wittgenstein points out that one must be able to determine the “simple” from the “composite”– an always shifting part-whole determination that is thematized in Philosophical Investigations. Lacking divine speech and the vision of a dialectician, humans range from parts to wholes (never knowing when they have been misled by the promiscuities of the sign). The problem of where one whole becomes a part and what delimits the parts and the wholes – the interminable question of where to draw the line that preoccupies Burke – always blinkers identification. Wittgenstein asks, “if it had been laid down that the visual image of a tree was to be called ‘composite’ if one saw not just a single trunk, but also branches, then the question: ‘Is the visual image of the tree simple or composite?’ and the question ‘What are its simple component parts’ would have a clear sense” (§47). Implicitly, it depends on what constitutes a whole (in this instance, an image of trunk and branches) and what constitutes a part (in this instance he refers only to the trunk and branches). If the scope is predetermined then the question about what the composite whole is and what parts it has “would have clear sense.” Distinguishing always 68 “Further, this admiration of men for knowledges and arts — an admiration in itself weak enough, and well-nigh78 childish — has been increased by the craft and artifices of those who have handled and transmitted sciences. For they set them forth with such ambition and parade, and bring them into the view of the world so fashioned and masked as if they were complete in all parts and finished. For if you look at the method of them and the divisions, they seem to embrace and comprise everything which can belong to the subject. And although these divisions are ill filled out and are but as empty cases, still to the common mind they present the form and plan of a perfect science. But the first and most ancient seekers after truth were wont, with better faith and better fortune, too, to throw the knowledge which they gathered from the contemplation of things, and which they meant to store up for use, into aphorisms; that is, into short and scattered sentences, not linked together by an artificial method; and did not pretend or profess to embrace the entire art. But as the matter now is, it is nothing strange if men do not seek to advance in things delivered to them as long since perfect and complete” (Novum Organum LXXXVI). Parmenides’ goddess characterizes her own language as duplicitous when she advises that we will learn by79 “listening to the deceitful ordering of [her] words” (8.52). The words are deceitful precisely because they are words, signs – distinguishing one thing from another. When language carves reality up into words (naming two forms where there is only one reality), it misrepresents what is one as many: Here I end my trustworthy account and thought concerning truth. Learn hence forth the beliefs of mortals, harkening to the deceitful ordering of my words. For they have made up their minds to name two forms, one of which it is not right to name – here is where they have gone astray – and have distinguished them as opposite in bodily form and assigned to them marks distinguishing them one from another. (emphasis added 8.50-56, Robinson 118) recursive part-whole relations at any particular moment (in any particular context) is the unending and inevitable challenge in any identification and classification project, the Parmenidean question of where to draw the line. Bacon addresses this matter in terms of style – a fact of some consequence to this thesis’ stylistic focus. He complains about those who attempt to transmit sciences “as if they were78 complete in all parts and finished,” suggesting that their method and divisions pretend to “present the form and plan of a perfect science.” But he suggests that “their divisions are ill filled out and are but as empty cases.” The part-whole, identification-and-classification difficulties presented by these empty cases are a reminder of the post-modern concern with the violence of the line with its long history dating back to Parmenides’ views about the unity of reality. Where ever the line is drawn it necessarily covers up what is under it, often arbitrarily79 slicing up what is otherwise unified, and imposing a mark which becomes a lens through which it is seen. De Man suggests that insight is always attended by a degree of blindness; Parmenides’ monism prompts epistemological worries about the arbitrary lines drawn by linguistic signs; and Foucault complains in Birth of the Clinic about how the medical gaze carves up the scene. Broaching the same totalization problem that Bacon does about filling in the gaps Foucault observes that “total description . . . is much more the dream of a thought than a basic conceptual structure” (115). Halliday too acknowledges the problem of a language and a 69 grammar that necessarily makes a fluid world sit still – that it is “we who cut it up with our verbs and nouns into processes and things” (OL 244). Drawing together both the part-whole- comprehension problem (a vision-knowledge handicap) and the analogy problematic Halliday argues both that one could say there are “no natural classes in the perceptual world” and also that “there are far too many natural classes; phenomena resemble each other in indefinitely many ways, hence there are innumerable ways of categorizing, leading to variation not only between one language and another but also within one and the same language” (OL 244). Again it is a question of human motivation and what motivates where the line is drawn. One might pretend that they can describe everything in words – but since no one is omniscient, the ability to comprehend or see the whole and the parts is inaccessible. When it comes to looking at language this problematic is exponentially multiplied. Halliday writes: “the categories we set up to explain how language works are almost all inherently fuzzy”; the descriptive categories, themselves “fluid and unstable” constitute the “fuzzy sets” of “the categories of language itself”; these “categories themselves are inescapably the product of compromise, since the different perspectives locate the boundaries between categories at different places” (OL 28, 266). Thus systemic theory adopts what Halliday calls “trinocular vision” – at least a threefold perspective on linguistic categorization. He observes: Indeterminacy is bound to arise in language because the grammar is constantly juggling with conflicting categorizations, accommodating them so as to construe a multidimensional meaning space, highly elastic and receptive to new meanings. In doing so, the grammar adopts a kind of trinocular vision, giving it a threefold perspective on the categories and their configurations. (OL 254) I suggest that this category problem in language theory is paradigmatic of any classification undertaking because of the linguistic mediation of epistemological endeavours. Furthermore, just as logical and affective reasoning processes might be impossible to delimit and disentangle, some things might not be verbally analyzable in a useful or productive way. It is really a question of the limits of what we can see (i.e., the level of detail and the distance), what we can comprehend in a single glance (i.e., the penetration of human understanding and the scope of cognitive abilities), and the subjective vantage from which we 70 Halliday states: “the value of a theory lies in the use that can be made of it, and I have always considered a theory of80 language to be essentially consumer-oriented. In many instances the theorist is himself also and at the same time a consumer, designing a theory for application to his own task . . . Since there are so many tasks for which one needs a theory of language, any particular theory is likely to be, or very quickly become, a family of theories – still on speaking terms, one hopes, and with a personal rather than positional family role system. This is why there is no orthodox or ‘received’ version of systemic theory, such as may arise with self-contained systems that are impervious to influences from outside” (OL 192). look. Perhaps a picture or image is worth more than a thousand words. As Wittgenstein says: Compare knowing and saying. how many feet high Mont Blanc is – how the word “game’ is used – how a clarinet sounds. If you are surprised that one can know something and not be able to say it, you are perhaps thinking of a case like the first. Certainly not one like the second. (§78) But what is of most interest here is the second case: try saying “how the word ‘game’ is used.” Arguably one could say many things about Wittgenstein’s language games. Indeed the list would be so long, complex, the elements interfused and at times contradictory, that arguably, apart from scholarly classification, their expression would be too copious and paradoxical for practical applications. Systemic grammar includes a family of related analyses – a complex network whose affinities and interrelations are doubtless only accessible to grammar specialists. This copiousness-complexity problem characterizes the generalized expressions of80 sceptical (eyes-wide-open) language theory. Quintillian’s comments on the list of topoi he devises as a culturally-specific training apparatus (with a best-before date) also speaks to the grammatical-codification-of-linguistic-operations problematic. According to Leff, Quintilian “disclaims any intention to devise a fully rigorous and exhaustive topical system. The effort would be futile and useless; it would cause the theorist to become prolix without being complete” (Topics 33). The interfused nature of the operations of language, the multiple aims of a single utterance, salient factors in linguistic complexity, recall the metaphor of the stream at the root of rhetoricity (rJevw – ‘flow’). What is of most interest to this thesis is how this aspect of linguistic complexity – language’s multifunctional, interfused nature, its ability to do several things at the same time – might determine genre and style choices of sceptical language theorists if only in 71 Halliday observes that “the extravagance of modelling that same domain of experience in more than one way leads81 to a richer and more life-supporting account. Of course, this will always leave room for what Claude Hagège (1997) referred to as “unheeded contradictions,” the leftover bits of language-building materials that continue to lie around; but the principle of contradictory construal is intrinsically a productive one” (OL 253). order to evade copiousness and paradox (the onerous risks of contradiction). Leaving81 completeness (totalization) to the divine order, Bacon, like Wittgenstein, opts for the stylistic practice of the “ancient seekers after truth” who wrote aphorisms; that is, “short and scattered sentences, not linked together by an artificial method,” and not pretending or professing “to embrace the entire art.” 2.8 the inherent resistance to theory: de Man and rhetorical theory ~ Referring to rule-governed grammar, de Man argues in “The Resistance to Theory” that “no grammatical decoding, however refined, could claim to reach the determining figural dimensions of a text” (16). In other words, it is figural-rhetorical features that finally qualify meaning and often, then, only in an indeterminate way; and this is not only because of the role of affective human motivation in language use and the sign’s tumultuous history, but rather language’s logic-resistance stems from its fundamentally tropological character. “Tropes, unlike grammar,” de Man asserts, “pertain primordially to language” (15). De Man illustrates the tropological character of language by producing a paradox in his own text. In order to show or perform the figurative and indeterminate nature of language he transforms the Cratylian resisting subject of his article into a sceptical resisting theorist: thus resistance is transformed from blindness to insight. Resistance had been thematized from the paper’s outset as a sign of psychological, ideological repression: a response to writing about writing by those whose Cratylian anxiety about uncertainty or whose vision-failure leads them to overlook sceptical language theory from Plato to Derrida. De Man claims that resistance to theory is a resistance to the rhetorical or tropological dimension of language: Rhetoric, by its actively negative relationship to grammar and to logic, certainly undoes the claims of the trivium (and by extension, of language) to be an epistemologically stable construct. The resistance to theory is, a dimension which is perhaps more explicitly in the foreground in literature (broadly conceived) than in other verbal manifestations or – to be somewhat less vague – which can be revealed in any verbal event when it is read textually. (17) 72 Recall Halliday’s comment that “grammar is a theory of human experience” (Matter 63)82 “Literariness,” de Man says, is “the use of language that foregrounds the rhetorical over the grammatical and the logical function” (14). It is this awareness of the “literary” or rhetorical effects of language that creates language scepticism and sceptical (eyes-wide-open) theories of the linguistic sign. Thus de Man performs a figurative transformation (a turning) himself: resistance morphs from a psychological neurosis into a liberation front. Recognizing language’s inherent rhetoricity, resistance defends the embattled sign’s rights to promiscuity. Anxiety- driven resistance is displaced by its opposite, an inherent resistence to theory. No longer a sign of repression resistance to theory becomes a sign of awareness: inherent in the idea of theory and its enterprise, inherent in the attitude of a sceptical-language theorist. What exactly is theory then that makes sceptical language theorists resist it on principle? We will return to the inherent inexactness of a linguistic semiotic that is fundamentally tropological with respect to Nietzsche in Part 3. But the question of inexactness, already raised by Wittgenstein, prompts consideration of whether a generalized language “theory” would be more useful than a precise one. Querying the precision required of ideal logical propositions Wittgenstein writes: “you still owe me a definition of exactness”(§69). Generalization (as a cognitive classification function) is, after all, a very useful and important linguistic cognitive function. One, albeit vague and abstract, definition for theory is “a conception or mental scheme of something to be done, or of the method of doing it; a systematic statement of rules or principles to be followed” (O.E.D.). By definition theory is general and conceptual. The Oxford English Dictionary definitions of “theory” include the collocates “scheme,” “method,” “systematic,” “rules or principles.” If we were performing coherence tests here we might think “grammar” (in sense of “rules”– even the arrangements of discourse that Hallidays’ systemic theory treats can also be described as variable and flexible rules, lexico-specific patterns). There is a part-whole relation between “theory” and “grammar” as rules. Grammar rules are general rules for particular instances of language use. A grammar is a kind of theory. 82 And sceptical theorists – whose theories are codification resistant – are also theory- 73 Recall, Voloshinov says that “a sign does not simply exist as part of a reality – it reflects and refracts another83 reality. Therefore, it may distort that reality or be true to it, or may perceive it from a special point of view, and so forth” (10). resistant because, according to Wittgenstein, theories can impede or blinker understanding. In his discussion of “exactness” (§88) – the (rule-governed) grammatical investigation of the “essence of language” (§92), and the “purest crystal” of logic “beneath the surface of language” (§97, 92) – Wittgenstein complains that “the crystalline purity of logic was, of course, not a result of investigation: it was a requirement” (§107). It was a requirement and an ideal that preceded investigation, acting as a lens. “Where does this idea [of the necessarily logical structure of propositions] come from?” he asks (§103). “It is like a pair of glasses on our nose through which we see whatever we look at. It never occurs to us to take them off” (§103). “One thinks that one is tracing the outline of the thing’s nature over and over again, and one is merely tracing round the frame though which we look at it” (§114). The idea of propositions having to be (mathematically) logical is an example of the way that a conception and a theory becomes a lens. This commonplace rhetorical insight – already commented on by Voloshinov, explicitly83 explored in Kenneth Burke’s article “Terministic Screens,” and treated in Halliday as the constraining and enabling operations of language – has its roots in Parmenides’ ancient complaint about the deceitful nature of words that divide a unified reality with the arbitrary marks of linguistic signs. Cratylian, code-metaphor theories of language were a requirement for the positive science that became the hallmark of the Enlightenment, thus becoming the organizing lens through which language was viewed. But Wittgenstein wants to empirically investigate how ordinary language actually works – not to project some idealized theory onto it. And de Man suggests that literary theorists want to know how textual effects are produced that “no grammatical decoding, however refined, could claim to reach” (16). If theorizing about how language works means finding a schema or overview, a systematic method for the operations of rules for how language works, then it is necessary to understand how parts are related to wholes. In the absence of divine memory and divine vision, the biochemistry of Halliday’s grammar presents an image of grammatology in which the rules are too copious and too complex to keep 74 track of. When Wittgenstein empirically investigates how language works without the occluding lens of ideal propositions, he encounters a very messy and fuzzy linguistic complexity and a myriad of singular peculiarities with codification-resistant logics and topsy-turvy part-whole relations. What would it mean to systematize them? He claims not to be able to. And are not just a few of his examples worth at least a thousand words? Without being able to comprehend the collective unity and plurality (required of the dialectician for speaking and thinking), without the omniscience it takes to see all the part- whole relations in which linguistic semiotic materials are enmeshed, a systematic schema of how language works will always be as Halliday argues, indeterminate, open-ended, and in constant flux. Our intimacy with the linguistic sign doubtless contributes to our misunderstandings of it. The restricted selective human perspective cannot “behold” all the part- whole relations needed to trace the schema at any given moment. In the face of such immense complexity, the awareness prerequisites of one’s own vision limitations, semiotic dependencies, and cognitive deficiencies warrant the Socratic stance of ignorance. If theorizing language means to map the coordinates of how language works in a coherent, self-consistent treatise, knowledge of only isolated parts means that attempts to string them together will be necessarily “artificial” and totalizing – forced through a particular lens. Such a project will always be suspect, leading to an “inherent resistance to theory” in “literary” language theory. De Man claims that: [Rhetorical readings expose the structures and functions of language] but are an unreliable process of knowledge production . . . they are . . . consistently defective models of language’s impossibility to be a model language. They are, always in theory, the most elastic theoretical and dialectical model to end all models and they can rightly claim to contain within their own defective selves all the other defective models of reading avoidance, referential, semiological, grammatical, performative, logical or whatever. They are theory and not theory at the same time, the universal theory of the impossibility of theory. . . . Nothing can overcome this resistance to theory since theory is itself this resistance. (20) Even rhetorical readings which do indeed string parts together into an apparently consistent whole de Man claims are “totalizing (and potentially totalitarian)” (20) – an echo of Bacon’s 75 worries about presentations of science which are “fashioned and masked as if they were complete in all parts and finished” (Novum Organum LXXXVI). Arguably there are a host of good reasons why, in order to express insights implicitly that might be difficult or impossible to explain in more explicit consistent ways, sceptical language “theorists” would be prompted to employ stylistic indirection. Despite the imperfections of language and various insufficiencies of linguistically mediated human thought, as de Man observes, “to claim that this would be a sufficient reason not to envisage doing literary theory would be like rejecting anatomy because it has failed to cure mortality” (12). Not only do we not need x-ray vision, it is okay to wear glasses, or even contact lenses. Traces of this epistemologically inspired language scepticism can be found in Parmenides’ fragments and it is inscribed throughout the Platonic text. De Man argues that resistance to theory: can be read in the text of literary theory at all times, at whatever historical moment one wishes to select. One of the main achievements of the present theoretical trends is to have restored some awareness of this fact. Classical medieval and Renaissance literary theory is now often being read in a way that knows enough about what it is doing not to wish to call itself “modern.” (13) Recall that Derrida too sees “the question of the sign” as “an unexpectedly historical one,” and as if to corroborate certainty compulsions and theory-resistance tendencies observes that the “analogy between the structuralist obsession and the anxiety of language is not a chance one” (WD 3-4). Since the long-standing sceptical approach to language, which seems to have suffered from a certain resistance, is still insightful, this thesis underscores the affinities between the views of different thinkers from Plato to Derrida. A theory that embraces the historicity of the linguistic sign, that escapes determinate systematization, and that is observant rather than oblivious, might resist the totalizing-systematizing look of theory itself, might not want to call itself “theory” at all. Such an interest doubtless motivates Derrida’s expression “theoretical 76 In Of Grammatology Derrida outlines a “theoretical matrix” offered not as a “new method” but rather as an attempt84 to expose “the problems of critical reading” (lxxxix). He says grammatology is “a science of writing [that] runs the risk of never being established as such and with that name. Of never being able to define the unity of its project or its object. Of not being able either to write its discourse on method or to describe the limits of its field” (4). matrix.” Inasmuch as its encounter with linguistic complexity deters the enterprise of84 “beholding” the contours of how the parts of language work as a self-consistent whole, sceptical (eyes-wide-open) language theory is itself theory resistant. Since it might not look like “theory,” the form of such a language “theory” begs to be investigated. In Part 5 we catch a glimpse of what a sceptical-theory-resistant-language theory might look like, and what its styles and genres can be. 77 The term “rhetorical situation” was coined by Bitzer but equally addresses what prompts Burke’s “pentad” – act,85 scene, agent, agency, purpose. PART 3 3.1 conn-texts & citation ~ whence the fascination for titles, which often function in a rather unsaturated context (Bennington 87) Part 3 of this dissertation explores aspects of various contemporary language discourses that set the stage for Nietzsche, whose primary linguistic insights were overlooked as the positivist impulses of analytic philosophy gathered steam in the early twentieth century. In the mid-twentieth century the so called “philosophy of language” was consolidating its power, superceding sceptical-rhetorical theories. Whorf had been spurned as heretical; the rich rhetorical tradition had been relegated to an outpost postion and reduced to the study of political oratory; and the view that language was all about communication had been centralized. The Searle-Derrida dispute that resulted from the publication of a talk Derrida delivered to the 1971 Congrès international des Sociétés de philosophie de langue francaise in Montreal offers a window into the language-theory scene just as the sceptical-rhetorical theorists started demonstrating against analytic philosophy’s idealization of language. Although the very significant conference theme for this meeting of the French Language Philosophy Societies was “communication,” the rhetorical situation of the talk sparking the controversy – its context, its85 theme and its audience – seems to have got lost in the crossfire. Years after the conference, Derrida’s talk, “Signature, Event, Context” (SEC), was published. Then the fracas broke out with Searle’s pompous response “Reiterating the Differences: a Reply to Derrida.” But Searle’s view of language may not, at the time, have considered rhetorical situation especially important when intention-governed meaning was still current in the “philosophy of language.” Quite apart from the specifics of the audience, and apart from the talk’s inclusion of Austin on the same stage with other prestigious thinkers, Condillac and Husserl, Searle didn’t seem to notice the communication problematic in the published version of SEC. Derrida’s talk, though remembered for its discussion of context and citation, challenged the communication and code-metaphor assumptions underlying the philosophic texts of various 78 Citations from “Signature, Event, Context” are from Weber’s Limited Inc. translation except where I have preferred86 Alan Bass’ translation in Margins of Philosophy. thinkers, including Condillac, Husserl and Austin. However, perhaps because the talk was wrongly perceived as an attack on Austin, Searle didn’t recognize the importance of the communication theme. Or perhaps he just didn’t get the problem; perhaps he didn’t recognize the communication issue because he assumed that the purpose of language was communication, and thus it didn’t seem relevant. This failure evinces not only the recognition problems that haunt sceptical-language theory, but Sperber and Wilson’s findings that reader inferences determine relevance – not code-models of intention-governed meaning. Recall also Ellis’critique that (facile) code-communication models have dominated language theory; and Halliday’s announcement that Whorfian-type linguists had long since rejected communication as a model for how language works. Analytic philosophy’s idealizing tendency seems to have bundled logic, rule-governed grammar and communication-success together and come up with code-metaphor models of how language works; and whatever breakthroughs Austin had made, the model of intention-governed communication was still prevalent. In SEC Derrida sets out to disrupt “the authority of the code as a finite system of rules” and to question code- communication assumptions that had directed investigations in early-twentieth-century “philosophy of language” (8). Querying the core of these assumptions, the opening line of SEC asks whether it is “certain that there corresponds to the word communication a unique, univocal concept, a concept that can be rigorously grasped, and transmitted: a communicable concept?” (Margins 309). Derrida’s opening line thus alludes to commonsense notions of communication: notions86 that fail to take into account the problematic nature of consciousness, intention, and the signifying structure of the sign. By this time, these were standard features of Derrida’s critique of writing previously explored in some detail in La Voix et le Phénomène that Searle doesn’t seem to know about. He was likely not an audience member of the French Language Philosophy Societies’ meeting in 1971. Derrida’s critique of the 1960s (like those of Burke, Whorf, and Kuhn) directed attention to problems in language – problems that in the end led to the demise of 79 the famous “linguistic turn [that] Wittgenstein initiated in the Tractatus” (Bergmann 63). This turn – once celebrated as the inauguration of analytic philosophy – was crucially already recanted by Wittgenstein himself in Philosophical Investigations. It was many years after Philosophical Investigations’ recantation, and a number of years after the 1971 talk itself when, in 1977, “Signature, Event, Context” was published in Glyph, and Searle replied, that the now infamous spat occurred. The arrogance of Searle’s “Reiterating the Differences: a Reply to Derrida” (which gratuitously defends Austin and also indexes the hegemony of analytic philosophy at the time) now smacks of irony. Speaking of communication gaps – not only did Searle not seem to recognize or grasp the implications of the communication question with which the talk opened, he seems to have missed Derrida’s expressed admiration for Austin altogether. Actually, Searle really didn’t seem to understand what Derrida was saying at all: he seems to have miss-taken what Derrida intended to communicate, specifically the critique of intention-governed, code-communication assumptions about language use. The question of being able to recognize sceptical-language theory thus presents itself in this infamous spat. On the one hand, Searle’s failure to recognize the important elements of Derrida’s critique may look like theory resistance, or seem to have been prompted by affective responses, or a bit of both. One might wonder whether Searle’s attachment to Austin blinded him to the Derridian critique the way the Royal Society’s science-project enthusiasm blinkered the uptake of Bacon’s views of language, or whether Searle just found the Derridian critique epistemologically worrying (likely both). On the other hand, quite apart from his apparent lack of effort, Searle might have an excuse for not understanding Derrida. He is clearly not Derrida’s addressee, not his target audience: a disciplinary-genre matter, an audience-design question that we discuss in Part 4’s discussion of style and that arises in Part 5 when we consider the reader- subjectivity of Derrida’s text. In Part 4, when we examine the purported “plain style” of Searle’s disciplinary field, we see how the style of science and code-communication models get confused with conceptions of “ordinary language,” leading to dubious stylistic assumptions about clarity and simplicity. Most importantly the interest in citation and context in Derrida’s 80 Austin for example assumes a communicative model in the following: “When is a statement true? The temptation is87 to answer (at least if we confine ourselves to ‘straightforward’ statements): ‘When it corresponds to the facts.’ And as a piece of standard English this can hardly be wrong. Indeed, I must confess I do not really think it is wrong at all: the theory of truth is a series of truisms. Still, it can at least be misleading. If there is to be communication of the sort that we achieve by language at all, there must be a stock of symbols of some kind which a communicator (‘the speaker’) can produce ‘at will’ and which a communicatee (‘the audience’) can observe: these may be called the ‘words,’ though, of course, they need not be anything very like what we should normally call words—they might be signal flags, &c.” (Philosophical Papers, 121) talk is particularly germain to the scholarship on reported speech which dominates Part 3 making SEC not only an excellent segue-way but also a point of alignment between Derrida’s philosophising and various fields of discourse analysis. Human motives – such as those prompting Searle’s vituperation – re-appear on the stage in the analysis of reported speech. And the language thought ligature in contemporary neuroscience comes into view before we conclude Part 3 with Nietzsche’s proto-post-structuralist understanding of language. Searle’s recognition failure of the code-communication problem recalls the addressee of Sperber and Wilson’s relevance theory accustomed to code models which they importantly refute. Recall also their conclusion that language is required for the “information processing” aspect of linguistic interactions and not “communication” itself: here the beneficiaries of analytic philosophy are moving away from the core assumptions that triggered the linguistic turn to start with. Austin appears, as Derrida observes, “to consider speech acts only as acts of communication” (13) where sceptical-language theory does not think about language primarily through the lens of communication at all. But the problem here is not necessarily the use of87 communication to study how language works: there is nothing wrong with studying instances of linguistic communication in order to understand the operations of language as Sperber and Wilson’s insights indicate and as we shall see when we examine reported speech. The problem is unexamined assumptions about communication – assumptions unquestioningly linked to conceptions of rule-governed grammar that result in code metaphors. This confusion is, I suspect, largely a result of the “written language bias” that we examine shortly. Derrida’s talk, of course, pre-dates Sperber and Wilson’s bi-lateral idea of communication – an image of communication more like partnered improvisational dance than unidirectional-code transmission – and one that does not arrive on the scene for over a decade. Derrida’s critique is 81 directed toward code’s naive notion of active speakers informing passive hearers – a quite different image of communication than Sperber and Wilson’s ostension and inference, alternations between pointing and probing, which correlates so closely with findings on reported speech. By 1971 – in the wake of La Voix et le Phénomène , De la Grammatologie and La Dissémination – SEC’s query about the meaning of the term communication, questioning whether the concept of communication has a structured centre, anchoring its meaning was standard Derridian fare. Presuppositions (that don’t seem to get scrutinized) about coded communication as a model for how language works prompt Derrida’s announcement that “one must first of all ask oneself whether or not the word or signifier ‘communication’ communicates a determinate content, an identifiable meaning, or a describable value” (1). What communication means is still a question, hearkening back to de Man’s complaints that people assume they know what language is. And we still need to ask whether or not we know what the term communication means, and what conception of it governs assumptions about language. Arguably in 2009 the sens «propre» of “communication” includes everything from social interaction (or sharing) to one way signal transmission between computers. Its habitual contemporary use over this vast range – as though it had a single determinate meaning, that we don’t need to think about – makes communication something of an empty term (a lot like democracy). We assume that we know what it means. But do we? It is not just that, like other signs, it lacks determinate, delimitable referents, but that it is so wanton that it can be all things to all people. It is what Burke calls an “ultimate term” – there is no discussion about what it means but it embraces a cacophony of perspectives and meanings (ROM 186-197). The indeterminate nature of the linguistic sign here obliquely (and playfully) alluded to 82 “Graphematic” from grafhv or writing displaces “linguistic semiotic” for Derrida’s strategic purposes. The clause88 can be construed in this context as “the linguistic semiotic structure of the locution.” While admittedly, writing is the primary academic medium, and it has a positive status for its role in “tradition” in Husserl, and the “graphematic structure” is in writing, about writing, based on writing, and is nevertheless a critique that applies equally to the non-written linguistic sign – despite (and because of) all the factors that may have prompted Derrida to focus on writing rather the speech – I believe the gesture is largely strategic and opportunistic. In Of Grammatology Derrida claims that “everything that for at least some twenty centuries tended toward and finally succeeded in being gathered under the name of language is beginning to let itself be transferred to, or at least summarized under, the name of writing” (6). This move displaces prior conceptions of language with a concept of writing (writ large) I believe because a traditional conception of writing, as a derivative form of speech, offers fecund material for his critique. This contrivance foregrounds the opposition between the full plenitude of speech and its shadow (the secondary, inadequate writing), inviting the deconstruction of the dualistic assumptions on which western metaphysics is based. Presence is implicated in the code metaphor for communication which has resulted in so much confusion. His gesture of displacing “language” with “writing” provides a convenient site (or topos) for Derrida, although it has been frequently misinterpreted as the privileging of writing over speech. “Firth consistently maintained that all study of language was a study of meaning; and meaning was a function in a89 context, where the ‘context’ was located both in various strata of language itself and in its situational and cultural environment” (OL 243). in Derrida’s reference to the “graphematic structure” of the locution is also of concern to88 Voloshinov, who not only claims that “multiplicity of meanings is the constitutive feature of word” (101), but also that, as a result, “the meaning of a word is determined entirely by its context” (79). Halliday too takes for granted Firth’s notions about the primacy of context and the indeterminacies of linguistic meaning. Not surprisingly, context is another star player in89 Derrida’s analysis of the linguistic-semiotic structure of the locution. Commenting again on the amorphous meaning of the term communication, Derrida observes that “the ambiguous field of the word ‘communication’ can be massively reduced by the limits of what is called a context” (2). Hence his talk, SEC, on the theme of communication, addressing the “graphematic” (linguistic-semiotic) structure of the locution itself is “concerned with the problem of context and with the question of determining exactly how writing [i.e. language] relates to context in general”(2). SEC’s examination of the relationship between the linguistic-semiotic structure of the locution and the idea of contextualization reveals the identifying role of citationality in the structure of the sign itself. This citation-focus together with other disseminations from Derrida’s talk (like, somewhat ironically, the written-language bias) make SEC a fortuitous topos from which to launch this section of this dissertation. By taking the spat back a step to the Austin- Derrida encounter in SEC, Glendinning’s “Inheriting ‘Philosophy:’ The Case of Austin and Derrida Revisited” offers the opportunity to explore confusions surrounding “non-seriousness” and “ordinary language” in the scholarship of reported speech: confusions that so interestingly 83 bring Austin beneficiaries into harmony with Derrida. The fundamental requirement that a sign be recognizable and repeatable leads to Derrida’s determination that one of its defining features (whether visual or audio) is its iterability as a mark. “The possibility of disengagement and citational graft” Derrida claims, “belongs to the structure of every mark, spoken or written” (12). “Every sign, linguistic or non- linguistic, spoken or written . . . can be cited, put between quotation marks” (12). Citationality, or reproducibility, here is an important identifying, structural feature of the sign. Co-incident to the importance Derrida attaches to citationality in the analysis of the sign, examinations of reported speech have become the focus of more empirically oriented research – perhaps prompted by Voloshinov’s view that reported speech is paradigmatically “speech within speech, utterance within utterance, and at the same time also speech about speech, utterance about utterance” (115). The research of the scholars working on reported speech examines a variety of genres: academic, literary, and various civic (i.e., legal and political) types. The findings of this work tread on the sacred ground of verbatim accounts: leaving open the question about what happens – in the legal system, the news media, academia – when the cherished notion of faithful-verbatim reporting loses credibility. The extreme import of faithfulness in reported speech – that is, verbatim, word-for- word, exact duplications of utterances – in the high-stakes situations of news reporting, academic research and court-room testimony prompts Short, Semino and Wynne to develop a “context-sensitive account of the role of faithfulness in discourse presentation” (353). This futile gesture – perhaps interesting from a psychoanalytic perspective for hints of anxiety and even resistance – bears the imprint (once again) of Cratylian-code-metaphor models of language. Short, Semino and Wynne are responding to the verbatim-deconstruction heresies of scholars like Sternberg, Mayes, Tannen, and Clark and Gerrig, the latter of whom flatly state “the verbatim assumption is incorrect” (799). These scholars’ analyses of actual conversation, civic genres (like legal proceedings and news reports) and traditionally-defined literature successfully refute commonsense verbatim assumptions and determine that the concept of verbatim itself is a 84 Bennington’s text presents various currents in Derrida’s thinking (“if not the totality of J.D.’s thought, then at least90 the general system of that thought”) in Bennington’s Oxford-schooled language playfully styled as his “pedagogical and logical norms” (1). fallacy. They challenge not only faith that the Cratylian-code-metaphor, word-for-word reproduction of an utterance conveys the original speaker’s intentions, but also whether verbatim is anything but an idealizing concept when citation in fact involves “tearing a piece of discourse from its original habitat and recontextualizing it within a new network of relations [that] cannot but interfere with its effect” as Sternberg observes (108). Derrida too argues that the fact that every sign can be cited means that “it can break with every given context, engendering an infinity of new contexts in a manner which is absolutely illimitable” (12). This is hardly a surprise. Being “quoted out of context” is a well polished phrase – a common complaint – widely recognized as a way of misconstruing someone’s words. Margerie Garber’s “Quotation Marks” observes the obvious when she notes “that every quotation is a quotation out of context” (669). Similarly Geoffrey Bennington states: “One quotes by definition out of context” (85). Since notions of context have taken up the burden of the code- metaphor – as Short, Semino and Wynne’s interest in a “context-sensitive account of the role of faithfulness” reveals – context now performs the labour of determining meaning. But how determinate is it? The importance of context does not alter the fact that it is a slippery and recursive creature itself. Bennington’s Derridabase states, first of all, that the idea of the sign90 is inoperable without context; and secondly, context is recursive, contexts exist within contexts – like the sign, context is subject to the determination and delimitation problem of recursion mise en abîme: We [Bennington and Derrida] say “quote in another context” rather than “quote out of context” to mark the fact that there are always contexts: the logic of the trace makes the idea of a sign or statement outside any context unthinkable, while making possible an exploitation of very open contexts (whence the fascination for titles, which often function in a rather unsaturated context). (emphasis added 87) Every element of the context is itself a text with its context. There are only contexts, and one cannot proceed to make the usual text/context distinction unless one has already taken the text in itself, out of “its” context, before 85 This is the name analytic philosophy seems to have given to those it wanted to exclude, that is, the name given to 91 those who did not follow the “linguistic turn” but instead continued to challenge historical philosophic traditions. See Grondin’s “Continental or Hermeneutic Philosophy.” Austin’s questioning of his own tradition’s assumptions – much as he works within them – is what attracted Derrida92 to his work to begin with. Derrida says that Austin’s “critique of linguisticism and of the authority of the code, a critique based on an analysis of language,” was what “most interested and convinced [him] in Austin’s undertaking”(emphasis added SEC 19). The assumption that language use is descriptive goes with code conceptions. The performative dimension of93 language is precisely an effect, a force of language unwarranted by its descriptive-referential meaning. Observe this undermining trajectory on the opening page of How to do things with words: “It was for too long the assumption of philosophers that the business of a ‘statement’ can only be to ‘describe’ some state of affairs, or to ‘state some fact’, which it must do either truly or falsely. Grammarians, indeed, have regularly pointed out that not all ‘sentences’ are (used in making) statements: there are, traditionally, besides (grammarians’) statements, also questions and exclamations, and sentences expressing commands or wishes or concessions. And doubtless philosophers have not intended to deny this, despite some loose use of ‘sentence’ for ‘statement’” (1). demanding that it be placed back in that context. And if one accepts what has already been said about consciousness and intention, which rules out the possibility of ever deciding rigorously what a text “means,” then we must accept that every demand to put things back in context is already interested and cannot be neutral. (emphasis added 90) Thus Derrida sets out in “Signature, Event, Context” – the talk delivered to philosophers on the theme of communication – not only to disrupt “the authority of the code as a finite system of rules” but also to disavow abuses of the concept of “context as the protocol of code” (8). Derrida’s suggestion here that “context” is being exploited as if it were code explains how scholars who know very well that citations are always taken out of context can nevertheless cling to the importance of verbatim faithfulness – that the exact words will duplicate the intentions of the (original) speaker – thus to code models of communication. One of the larger contexts of Derrida’s paper then might be described as a conversation between analytic and “continental” philosophy. Searle’s miss-take of Derrida’s critique –91 taking the critique of the assumptions to which Austin is heir as dismissive of Austin himself – is festooned with ironies beyond Searle’s inability to understand Derrida’s intentions. Searle’s92 attack on Derrida and the subsequent hullabaloo and misunderstandings make it ironic that it is Austin who figures in “Signature, Event, Context” (SEC). It is ironic because Austin’s work also undoes code-conceptions of language (or at any rate, it starts this work) and it subsequently motivates linguistic pragmatics researchers who end up reaching conclusions a lot like Derrida’s. Derrida’s texts are always parasitic, always opportunistic (that is, always emerging93 from carefully selected topoi), always writing about writing: thus Austin’s text is very useful to 86 Glendinning uses this phrase to temper gross distinctions between these two “traditions” in philosophy, in order to94 defuse the hostilities (308). Derrida refers typically to someone’s “style” of philosophy or thinking. Holquist observes that Bakhtin’s “concept of heteroglossia is a happy redaction of the conditions otherwise so95 gloomily charted by Derrida’s epigones as ‘differance’” (“Introduction” xxxii). Whether Bakhtin is Voloshinov or simply a very close colleague sharing many of Bakhtin’s views is not relevant here. The proximity of Derrida’s and Voloshinov’s understandings of language is clear. Similarly, Halliday’s Whorf connection makes him an easy bedfellow with Derrida- affiliates. the SEC critique-of-code-communication-model enterprise. This is not only (but significantly) because Derrida admires Austin’s active, flexible, rigorous thinking (depicted as “patient, open, aporetical, in constant transformation . . . fruitful in the acknowledgement of its impasses”), and not only because Austin participates in a “way of going on” in philosophy analyzing language94 as code-communication, but also because Austin’s insights into “performative” verbs offer Derrida a fruitful opportunity – a topos or place – to explore, among other things, the interdependence between the iterative nature of the sign as a mark (or citationality) and context (14). Apart from all the ironies, Austin’s work is itself part of the context of Derrida’s talk at (at least) two levels: as both a site of analysis – the “place” giving rise to Derrida’s query – and as contributor to the larger scholarly conversation about language into which Derrida speaks. 3.2 legacies: code-communication and the written language bias ~ Admittedly the speakers gathered together in this thesis might not happily share such close quarters, given the sometimes considerable differences between them. That Voloshinov’s and Halliday’s thinking lines up so well with that of Derrida might be less surprising than that95 the “mainstream” linguistics-related work on communication and reported speech finds such close affiliation with Derrida’s. Scholarly affiliations and lineages, characterized as “inheritance” in Glendinning’s “Inheriting ‘Philosophy:’ The Case of Austin and Derrida Revisited” thematizes different “ways of going on” in philosophy – not broadly with the terms “analytic” and “continental,” but in more nuanced terms of style, clarity, obscurity and ordinariness. Much “mainstream” linguistics seems heir to Austin’s legacy, Austin himself being heir to code-communication-model assumptions about the truth status of propositions – precisely the assumptions Derrida questions in SEC. However much Austin’s own work actually undermines the code-communication orientation to which it is heir – assumptions treating code- communication as the model for how language works persist in his thinking. What Glendinning 87 It seems unlikely that writing – famous for communication failures and producing a plurality of interpretations –96 would have prompted notions of communication success. As Socrates notoriously observes “if you question [written texts], wishing to know about their saying, they always say only one and the same thing. And every word, when once it is written, is bandied about, alike among those who understand and those who have no interest in it, and it knows not to whom to speak or not to speak” (Loeb Phaedrus 275e). calls “truth-evaluable content” seems to act as a reference-post not only in Austin’s work but in subsequent generations of scholars following him (i.e., philosophers like Grice, psycholinguists like Clark and Gerrig, sociologists of language like Goffman, linguistic-pragmatics scholars like Sperber and Wilson). Although I doubt the presumed success of oral communication (an intuition that may never have benefited from transcription verification), I suspect the recidivism of code, that Ellis observes, stems from just such unexamined conceptions of communicative success: and that these unexamined conceptions are then incautiously affiliated with the logical-truth-status-of- propositions. Unexamined intuitions about communication working so well seem to be96 superimposed on beliefs about the role grammar plays; communication is perceived through the lens of rule-governed grammar (as the logic of language); and then without pausing for breath rule-governed grammar is perceived back through the lens of communication. Communication works, grammar is a kind of logic, therefore communication must depend on logical deduction. Wittgenstein’s insight that the “requirement” of “the crystalline purity of logic” was like a “pair of glasses” through which to see – a pair of logic goggles – makes it easy to see how these communication intuitions might have prompted logic-bespectacled philosophers to fantasize conceptions of rule-governed grammar, truth-evaluable content and propositions in order to account for assumptions about communication success. Voloshinov’s complaints that such rule-governed grammars stem from analysis of written texts direct us to some speech and writing confusions embedded in this code-logic approach. The model of communication in which meaning determination results from the truth status of propositional content depends on sentences, artifacts of writing not speech. When we look at transcripts of actual speech (as we will with Holt and Tannen) it becomes clear that people only occasionally use whole sentences when they speak to one another. If philosophers 88 Per Linell’s introduction opens: “In this book the modern linguist's view on language is discussed from a rather97 unusual point of view. It is argued that our conception of language is deeply influenced by a long tradition of analyzing only written language, and that modern linguistic theory, including psycholinguistics and sociolinguistics, approaches the structures and mechanisms of spoken language with a conceptual apparatus, which - upon closer scrutiny - turns out to be more apt for written language in surprisingly many and fundamental aspects. I will refer to this situation as the written language bias in linguistics.” While it is clear that Austin works with code-communication and truth-conditional reference posts, this critique98 does not appear to be directed towards his the very initial and tentative work How to Do Things With Words. Rather Linell and Marková’s critique is directed to “speech act theory as developed by John Searle” (173). Interestingly Linell and Marková criticize “the written language bias” in “Chomskyan generative grammar, Grice’s 99 theory of conversational implicature, Sperber and Wilson’s theory of relevance, and Labov and Fanshel’s model for analyzing real talk” (175). For example, Clark employs the following example of supposedly oral conversation: “Ann, to Charles, in front of100 Barabara: Charles, I insist that Barbara tell you who we met at the museum today.” (Arenas 225). Regardless of what Clark’s point is here, the example reinforces the myth that people normally speak in sentences like this, Ann is more likely to have said “ask Barbara who we met at the museum today.” supposedly focussed on “ordinary language,” like speech-acts theorists, partnered intuitions about oral-communication success with written sentences – as though speech saw an image of writing when it looked in the mirror – then we can see how Per Linell’s The Written Language Bias in Linguistics gets its start.97 Writing within the field of linguistic pragmatics, Linell and Marková fault speech-act theory precisely for its “written language bias.” They complain that the “written language bias98 in linguistics” works on sentences as autonomous units and made up examples (instead of actual discourse), and that it “accepts the idea that the task of semantics or pragmatics is to map onto99 each possible expression a unique semantic interpretation” (“a determinate content,” as Derrida says, with “a concept that is unique, univocal, [and] rigorously controllable”) (174). Linell100 and Marková claim that speech act theory is more of “a supplement” to, than a replacement for, “traditional sentence-based semantics,” which is concerned with “truth conditions and propositional content” (174). Not only was speech presumed to act grammatically like writing but, as Linell and Marková remark, the supposedly oral sentences analyzed were often inventions of the philosophers studying them – despite the radical changes in context between most written and spoken utterances. Successful communications – whether written or spoken – do not depend on truth-conditional rules as the code model imagines. It is much more likely that they depend on relevance “rules” – if you can call them rules when they look more like dances or games. Derrida was doubtless justified in detecting a myriad of unexamined and muddled 89 Glendinning states that “the problematic of inheritance is itself a central and inescapable issue” of the topic he has101 chosen to write on. assumptions surrounding the term “communication” in the speech-act-theory discourses. 3.3 legacies: performatives, seriousness & ordinary language ~ Nevertheless Austin’s ideas about the way that actions are performed by language use (language acting rather than referentially describing) represented by his term performative did rejuvenate ways of thinking about how language works within analytic philosophy and its “mainstream” linguistics beneficiaries. Glendinning’s “Inheriting Philosophy,” which adjudicates “the case of Austin and Derrida,” outlines the arguments in Austin’s work on “performative communication” that exclude “non-serious” speech events from “serious” ones – a problematic and unfortunately named distinction. Glendinning is a particularly useful guide here because, as an analytically trained philosopher, he shares Austin’s heritage, but he also reads Derrida. It is easy to see how Austin (the alert thinker that attracts Derrida’s attention) got stuck in problematic code-communication assumptions when Glendinning traces the serious/non-serious distinction back to Frege. He shows that Austin’s use of this nebulous and intention-governed distinction is a Frege legacy – a bit of discourse, available as a ready-to- hand assumption whose implications did not require assessing. Since we must start with assumptions – a double bind Plato, Nietzsche and Derrida have all wanted to foreground – being an intellectual beneficiary is doubtless always complicated. The vicissitudes of heritage are101 manifest when we find this distinction from Frege reappearing in Austin legatees Clark and Gerrig who (ironically) link it to citation in order to argue, as Derrida does, for the ordinariness of the “non-serious” in the ubiquity of quotation. When Austin writes How to Do Things with Words serious/non-serious seems to mean something like logical/unlogical, literal/non-literal, real/unreal. While I doubt the usefulness of such distinctions, (especially the literal/non-literal which depends on a Cratylian rather than a tropological conception of language), Glendinning shows that Austin himself was suspicious of such dualistic metaphysical assumptions. Citing Sense and Sensibilia, he observes that (with regard to perception) Austin claims to “‘abandon old habits of Gleichschaltung, the deeply 90 ingrained worship of tidy-looking dichotomies’” (311). Gleichschaltung, according to Glendinning, means bringing “something into line” making “something conform to a certain standard by force”; he also elaborates on the term’s Nazi political affiliations (illustrating the way that concerns for “justice” are implicated in the Derridian critique) (312). The term Gleichschaltung and Austin’s phrase “tidy-looking dichotomies” will prove useful to (later) considerations of Aristotelian stylistics; however for current purposes Glendinning cites Austin’s non-serious exceptions as being guilty of the same Gleichschaltung and tidy-looking dichotomies that Austin criticizes. But perhaps Austin just reached for this terminology (criteria) without pause (as Bolinger and Becker have suggested that we reach for a phrase, “a remembered prior text”) as a convenient category to narrow down his field of study. Glendinning observes that Austin’s exclusion of “non-serious” speech acts from the analysis of performative verbs calls into question the concept of ordinary language. He says that “Austin’s concept of ordinary is, as Derrida puts it, ‘marked’ by this exclusion” (326). Summarizing Derrida’s objection that Austin is “insufficiently sensitive” to the possible risk of things “going wrong” in speech acts (irony or sarcasm counted among the non-serious “infelicities,” for example) Glendinning writes: Instead of pursuing an investigation of the structure of locutionary acts which shows why this ‘risk’, qua possibility, is essential to its being such an act, [Austin’s] procedure passes over the general logic of locutions and positions it, qua eventuality, as something that transgresses the ‘ordinary circumstances’ of language use – something that can, and therefore should, be pared away or excluded from an analysis of the ‘normal use’ of words in ‘ordinary circumstances.’ (327) It is the very idea of excluding “ordinary” types from analysis of “the ordinary” that is at issue. It is not that Austin’s insights about the behaviour of certain verbs were not useful or interesting; it is that “ordinary” and “normal uses” of language are sidelined by assumptions about communication that exclude more fundamental insights about the semiotic-linguistic structure of the locution. Of course for Derrida the impossibility of determining and delimiting intentions – on which so much of the non-serious category depends (along with its Cratylian 91 Who can determine and delimit even their own intentions? As Socrates says: “our soul at any one moment teems102 with countless . . . contradictions” (Republic 603d Loeb) There is an irony in this identification of citation as non-serious, since, as Derrida points out in SEC, the original103 list of performative verbs were citation par excellence. implications) – makes the infelicities problematic to start with. But Glendinning focuses on102 the fact that Derrida does not believe that the exceptions can be ruled out in the investigation of the ordinary. He continues: Unless analytical philosophy is to be characterised by a lack of concern for the structure of ‘ordinary language’ this [ruling out of exceptions] cannot be so. For the reason why Derrida is convinced that consideration of ‘the realm of the “nonserious”’ cannot be put off until later can only be understood in terms of such a concern. Specifically, his claim is that because it is internal to the structure of ‘the realm of the “serious”’ (the realm which Austin’s analysis aims, here and now, to consider) that it possesses features held by Austin to define the excluded realm . . .” (327). As heir to problematic-communication assumptions, when Austin reaches for the ready-to-hand label “non-serious,” he fails to perceive that non-serious uses are internal to the structure of serious uses. Perhaps it isn’t just a coincidence that Derrida’s talk – claiming that citationality is structurally necessary identifying feature of the sign and finding that so-called non-serious uses of the locution are internal to the structure of serious ones – jives so well with the findings of scholars studying reported speech, the scholars studying citation. Clark and Gerrig’s analysis of reported speech determines that quotations, what they call non-serious “demonstrations, are performed as parts of serious activities” (766). Arguing that citations are “demonstrations,” a103 type of “showing” likened to Wierzbicka’s description of quotations as “imaginary speech performances,” and (following Goffman) that “demonstrations” are “non-serious” (meaning “not literal”) they conclude that reported speech constitutes a substantial component of language use, i.e., these “non-serious” performances are ordinary, normal usage (801, 766). The footprints of the dubious “serious/non-serious” distinction lead from Frege to Clark and Gerrig. Clark and Gerrig attempt to distinguish non-serious demonstrations, like quotation (which they claim depicts reality), from serious, descriptive uses of language. I believe that these categories are unsustainable: the terms serious and non-serious bring with them their own 92 histories of association and affiliations – especially code conceptions of communication and literalness. Though Clark and Gerrig attempt to provide descriptive classification (which from their perspective is “serious,” non-demonstrative language use) Foucault might remind them that description is inevitably prescriptive – prescription being a by-product of descriptive classification. Halliday observes language contrues reality – that is, constructs reality – and enacts it at the same time. As Austin’s own analysis begins to reveal (when he extends illocutionary force to all speech acts and adds describing to the list of “performative” verbs) and as any language usage manual clearly illustrates, descriptions act in the world, producing what they seem to describe – both reifying (constructing) and prescribing (enacting). Austin’s inquiry into the actions of a very limited number of performative verbs lodges a speech-as-action wedge into description-referential code-communication models of the structure of the locution dominating analytic philosophy. This wedge opens up a treasure trove of linguistic insights, within analytic philosophy, about language’s attributes. These pragmatic insights are congruent with ancient and contemporary sceptical views of language, and lead to this thesis’ use of the term performativity. Hoping that a fuzzy picture is exactly what we need, language’s performativity gathers together diverse effects of language in which it seems to act: figurative effects, labelling/categorizing/reifying effects, (interpersonal) enacting effects, and so on. For example, the Foucauldian productive-creating-prescribing-type acting in the world – that is, language that both produces what it seems to describe (i.e., mental or physical disorders), and that then disciplines by pre-and proscribing – illustrates two ways that language performs actions. As it turns out, citation, which constitutes a much larger portion of every day language use than the initially-identified performative verbs, is paradigmatic of language’s performative disposition. 3.4 intersections: language theory and reported speech ~ Since Wittgenstein’s turn to “ordinary” or “everyday” language is a turn to how we actually use language – in everyday conversation, in news reports, in academic writing – arguably scholars of reported speech really did turn back to “ordinary language.” Among the 93 curiosities arising from the scholarship on reported speech, one that speaks to its “ordinariness,” is that results from analysis of (what would be considered) “literary” texts line up with analysis of a wide range of everyday language genres like transcripts of actual conversational discourse, courtroom testimony, legal documents, and journalism. Tannen’s analysis of conversational talk leads to her observation that the strategies used in everyday talk are the same ones that she had previously considered “quintessentially literary” (30). This overlap between the strategies of conversational talk and literary writings make spontaneous talk an exemplary site for analyzing the mechanics of language use. Meyers too comments on the link between “work on spoken language” and “literary stylistics” in studies of reported speech: prompting the observation that “everyday talk is in its way very artful” (Reported 378). Whether this is life imitating art, or art imitating life, (or perhaps the art of linguistic life), what seems clear is that reporting other’s speech is so fundamental, so consistent and persistent in language use, that it simply occurs where language is used. As Derrida has reminded us it wouldn’t be linguistic semiotic phenomena if it were not iterable, repeatable, re-sayable. Citation really is ordinary. Analysis of reported speech produces unexpected and counterintuitive results in which descriptive reference once again leaves the scene: citation seems to do other types of work more in line with Austin’s actions. Recall that de Man defines “literariness” as “the use of language that foregrounds the rhetorical over the grammatical and the logical function” (14). Examination reveals that reported speech is a rhetorical device employed for a variety of strategic purposes (often functioning at multiple layers at the same time). Thus Fludernik writes that “so-called verbatim representations of utterances do of course, exist,” however, counter to expectation so- called verbatim representations are not “the most mimetic, natural, unmarked form of speech report,” rather they are “the most highly artificial, marked and formulaic type of reported discourse” (2). Direct reported speech – that seems the most “natural,” “transparent,” unmarked, “faithful” type of report – is the most contrived. The same kind of findings lead Holt to argue that “direct reported speech is a complex device characterized by contradiction” (427). Oddly, some kinds of language use seem designed to deceive; what seems most mimetic is the most 94 artificial. Whether these contradictory and counter-to-expectation characteristics of direct report indexes intuitions about language’s descriptive-referential nature (that may be schooled and historical) or whether they are just another manifestation of linguistic complexity is not clear. The convergence of literary, philosophical and linguistic scholars whose work is coalescing here around citation covers a broad range. Derrida brings a philosophic approach to the question of linguistic mediation which is epistemological and structural; and Voloshinov (whose advocacy of studying reported speech prompts much of the pragmatics work presented here) offers a Bakhtin-Circle-“literary” type theorizing, which celebrates reported speech’s active reception of another’s words together with one’s own evaluation of them. On the other hand, most scholars studying reported speech empirically analyze instances of actual reporting in texts often from corpus studies. For example, Deborah Tannen’s “discourse analysis,” gathers together research mainly in anthropological linguistics supplemented with socio- and psycho- linguistics and her own background in literature and literary theory (5). This work on conversational discourse observes the ubiquity of repetition (not just in the iterative nature of the sign) but more broadly in the human mimetic impulse (what Nietzsche calls the drive to imitate). Mimesis is also a fundamental analytic concept in Sternberg’s seminal literary- theoretical, discourse-studies type exploration. Sternberg’s analysis of the five attributes of reported speech (affect, detail, realism, distinctiveness, reproductivity) in texts (ranging from Plato to Dick Francis) prompts him to assert that all reported discourse, regardless of form (or directness), is a mimesis of discourse (107). His much cited “Proteus in Quotation-Land” underwrites Fludernik’s exhaustive book-length study of forms of reported speech (most especially free indirect discourse). Her approach in The Fictions of Language and the Languages of Fiction situates itself squarely “between a linguistic and a literary analysis of narrative prose,” much like other “discourse analysts” cited here including Meyers, Calsamiglia and Ferrero, and Holt. (12). Reviewing samples of actual recorded conversation makes the sociality of spoken communication immediately clear. Holt’s transcription protocol, which captures non-verbal 95 vocalizations and linguistic dialect, implicitly indexes just how much communication depends on body language and non-verbal gestures: a reminder of what is edited out when supposedly “verbatim” accounts (such as news reports) normalize dialectical usage and accent, and omit various expressive features like pauses and tone of voice. For example, Holt’s transcript below of a daughter narrating an incident to her mother shows the importance of the mother’s active signs of interest, comprehension, and concurrence, evinced by various verbal and non-verbal vocal gestures. Holt is especially interested in the implicit evaluation in the direct report of Mrs. Field’s utterance “oh, I thought Lesley had given me these” (ll 6-7). While this statement seems banal enough out of context, within the context it acquires sharp edges, expressing Lesley’s clear distaste for something about the reported speaker. Lesley is reporting looking for some books. 1 Lesley: and em I couldn’t hfind them ‘n then hsuddenly 2 we realized th’t hI’d mi- I: lent’n to Missiz 3 Field (0.2) hyears ago. 4 Mum: Oh:. 5 Lesley: •hhh An’ hMark a:sked her f-back ifor th’m this 6 g evening ‘n she said •hh hOh: I thought Lesley 7 g had given me the:se 8 (0.3) 9 Mum: iOh:.uh hhe -:-:h 10 Lesley: .NO h: such lu:ck ihh hhuh 11 Mum: .Mm-i:::- 12 hHm:. 13 (0.6) 14 Lesley:When you isee ‘er you’d better iwa:rn her. (Holt 444) It is easy to see that Lesley’s report of Mrs. Field’s utterance is a negative assessment of Mrs. Field when it is re-contextualizated within the new narrative. The apparently banal words are not interesting until the evaluative tone is added. Sperber and Wilson’s suggestion that “human external languages do not encode the kind of information that humans are interested in communicating” resonates with this evaluative use of language (174). The referentially 96 descriptive facts hardly seem worth repeating: that Lesley looked for some books, realized that she had loaned them to someone and discovered that the borrower had thought the books were a gift (perhaps a harmless misunderstanding). Without the evaluative tones describing such vapid events would likely elicit wonder about why they were worth paying attention to. Although it is easy to see how the direct reported utterance (“oh, I thought Lesley had given me these”) is acting out or performing someone else’s speech, it is not clear exactly whose perspective it enacts. Is it acting out the perspective of the reportee, Mrs. Field, whose words are repeated or that of the reporter, Lesley? Mrs. Field’s words seem to be “re-accented” and somewhat subsumed by Lesley’s reporting of them. From Voloshinov’s (and Bakhtin’s) perspective there are two subjectivities in direct report, that of the reporter and that of the reportee, which makes reports of someone else’s speech “dialogic” by definition. There is an unstable, oscillating, back and forth movement of speaker intentions. Derrida uses the term disseminate to describe the uncontrollable leakage of meaning expressed in Bakhtin’s term “dialogic”; but Bakhtin is also expressing intersubjective perspective by this term. The oscillations between Mrs. Field’s meaning, and Lesley’s, the latter’s evaluation of Mrs. Field and her disdain with regard to this particular incident is quite clear from the reported line. In the performance of another person’s speech we witness what Voloshinov calls the “evaluative reception of another’s utterance”: a portent of the linguistic complexity that can be anticipated in reported speech where descriptive reference seem to have been relegated to the margins (118). Holt’s commentary on “Mum’s” response to the report: “‘oh,’ followed by particles that communicate her surprise and amusement” seems to indicate “Mum’s” recognition of, or concurrence/tacit agreement with Lesley’s reaction to Mrs. Field (444). Holt also observes that “both speakers produce concurrent responses” that overlap with one another which are not explicit assessments “but reactions to the reported speech that convey their point of view” (444). This moment of recognition/concurrence might be the objective of the communication and the purpose of the story: a shared understanding or attitude. I am reminded of Derrida’s 97 communication question and whether or not we know what communication is; it seems vague, not the precise milieu required for logic. Lesley and her mother’s overlapping series of vocalizations (indexing agreed evaluation) are clearly not the model of communication analytic philosophy had in mind when it sought to determine the truth status of propositions for determinations of meaning. In relevance theory terms Lesley has implicitly pointed at Mrs. Field’s character (or something about her), not by descriptive reference but by how she mimics (or performs) Mrs. Field’s words and the way that they are contextualized. From this hinting her mother seems to have inferred what Lesley’s reaction was. Not only does this illustrate the way hearers make their own inferences, it corroborates Sperber and Wilson’s view that communicative effects are often vague. A fuzzy picture might be all that we need, or want. Whatever was communicated between Lesley and her mother – distaste or displeasure of some sort – is not very exact. Perhaps this vagueness of informative purpose stems from communication’s social nature and objectives. It is the sharing of attitudes, shared evaluation – not informing someone of a series of banal events – that seems to be what is important here. 3.5 scholarship of reported speech ~ Much of the scholarship on reported speech considers the linguistic forms of report – ranging on a continuum from “direct,” through “free direct,” and “free indirect,” to “indirect.” Sternberg identifies the following five representational poles of “the most mimetic” speech reports: affective, specific, realistic, distinctive and reproductive (111). While the latter are associated with directness, as it turns out, the forms (direct, indirect, etc.) can mimic one another: producing effects opposite to those that they are supposed to produce. Hence, Sternberg defines the “Proteus Principle” that governs reported speech – a shape-shifting ability guiding “the interplay of unity and variety in quotation”; he observes the “many-to-many correspondences between linguistic form [direct, indirect, free indirect, etc.] and representational function” (112). So despite Voloshinov’s identification of the “analytic” disposition of indirectness, which ostensibly removes the “affective-emotive features of speech” 98 because “analysis is the heart and soul of indirect discourse,” Sternberg shows that direct forms can be manipulated to produce the same analytic impression and vise versa (Voloshinov128- 29). If artful deployment can reverse inherent tendencies in these forms, their protean quality underscores the essential rhetoricity of language. More profoundly, these forms’ (of reported- speech) ability to mime each other suggests that spontaneous directness might be a window on all forms: the characteristics of direct discourse reveal reported speech’s true colours. The previously mentioned overlap between the strategies of everyday talk and literary discourse, together with this mimicking ability of different forms of reported speech suggest that functions of direct report can be considered paradigmatic of certain fundamental (performative) aspects of language use. There is general agreement that vividness is one of the striking attributes of the most mimetic instances of reported speech; and it is a quality that can be produced by any combination of the five representational poles (affect, detail, realism, distinctiveness, reproductivity). Tannen describes this vividness as creating a scene (107), Holt refers to direct discourse’s “theatrical” nature (430), and Clark and Gerrig suggest that “when we hear an event it is as if we directly experience” it: direct discourse demonstrates an event by reenacting it, revivifying it, making the addressee re-experience it (793-94). From a narrative perspective it is easy to see how the vividness of direct discourse, as a dramatic “showing,” might be employed where emphasis is required or desired. Exploiting direct discourse’s vividness is a rhetorical strategy: a setting of things before the eyes, in Aristotelian terms. Holt observes its regular use as the “punch line” of an amusing anecdote, “at the climax of a story,” or for “ utterances that are the focus of a telling” (447, 430). Clark and Gerrig see it as a rhetorical marker. Rather than being the most transparent, reproductive and innocent form of reporting, direct discourse is an astute and motivated artifice: a rhetorical marker in conversation that hints, enacts and underlines. The punch line, the point of the story or (section of narrative), is not spelled out (in referentially descriptive terms) but rather it is implied. Lesley’s direct report of Mrs. Field’s 99 “oh, I thought Lesley had given me these” not only underscores the point of the story (her evaluation of Mrs. Field) but conveys it implicitly, inviting her mother to participate in the process and to share her point of view. Holt notes not only the evaluation or assessment expressed by direct discourse implicitly, but also the collaborative, participatory nature of the shared response. Her analyses of conversations found it possible to identify recurrent responses on the part of the recipients that make use of clues given by tellers in prior utterances. Not only do these correspond with the tellers’ implicit assessment, but, in some cases, the tellers join in with the recipients by concurrently producing the “same” kind of response at the “same” time. (426) Implicitness invites participation and involvement. By rhetorically marking the point of the story directness marks both the speaker’s evaluation or assessment of the reportee and the invitation to make inferences. Clark and Gerrig see direct discourse as a strategy that highlights inferencing opportunities for hearers eliciting their involvement in both the communicative process and the speaker’s opinions. Verbatim-citation’s referential status dissolves here where implicitness and evaluation seem to reign. Perhaps the evaluation and assessment that reporting scholars frequently observe indicate the kind of thing that humans are interested in communicating: the kind of vague things that people share in communicative processes. The goal of Lesley’s communication looks more like sharing an opinion than delivering a package of information and I wonder how often confirming a shared (and vague) opinion is communication’s objective. The research findings of Voloshinov, Tannen, Meyers, and Holt all emphatically corroborate the active role of the hearer in conversational discourse: active hearer and speaker roles that are defining features of communication in Sperber and Wilson. Hearer participation, implicitness, and evaluation are interfused in conversational discourse and structurally entailed in direct reported speech. These three interfused thematics reflect the nature of communication outlined in Sperber and Wilson’s ostensive inferential model and both the sociality of semiotic-linguistic materials in Voloshinov’s account, and Halliday’s interpersonal-enacting metafunction. These rhetorically- strategic, socially-collaborative and participatory uses of direct discourse have a paradoxical, 100 contradictory relation to the concept of verbatim duplication: the pretense of faithful reproduction of another person’s (referential) meaning. Rhetorically marked implicitness, participation and evaluation in direct discourse a fortiori evince language’s complexity when considered together with other “reasons” for using reported speech: reasons that bear varying relations to Sternberg’s five representational poles (affect, detail, realism, distinctiveness, reproductivity). There are emphasis-oriented uses that Meyers calls “intensifying” functions (i.e., vividness); he also points to “typifying” functions (i.e., characterizing situations by typical responses). Clark and Gerrig mention “dissociation or detachment” (a speaker’s interest in not owning the utterance), ineffability (direct report of what cannot be stated), and politeness, modesty or saying something that would be awkward to say (when, for instance, providing an account of someone else’s “bad” language, or in the reporting of praise that might create the appearance of boasting). Holt and Mayes both remark on the status that direct report has as “evidence”; and Calsamiglia and Ferrero observe that citation “confers authority,” underscoring latent Cratylian code assumptions about exact words conveying speaker-intentions. Importantly all of these “reasons” for using direct discourse are utterly rhetorical; they are motivated-contrived-strategic moves with specific designs on hearers and readers. They are artifice as Fludernik has noted: the “complex device characterized by contradiction” that Holt has observed. A rhetorical understanding of language (that embraces diverse linguistic operations, not just reference) recognizes that language use exploits various strategies at the same time. For instance, however dissimulating each might be, the authority and the evidentiary status that direct citation confers are not mutually exclusive, but actually work very well together (a fact that, as this thesis shows, I find irresistible). Vividness can be deployed together with a speaker’s interest in disowning a comment or wanting to typify it or both. An assessment delivered by direct report builds solidarity (by supporting another person’s perspective) as well as providing inferencing opportunities to other participants. Tannen, Holt, and Clark and Gerrig all comment on the ability of reported speech to perform several actions at the same time: a 101 feature of linguistic complexity inimical to code. Reported speech like other features of language use is mulitpurpose and mulitfunctional; various aspects of language use are interfused; they are “inseparable,” as Tannen observes (31). As Sternberg says when certain features of a text are selected and foregrounded for the benefit of a new context, they are torn out of their original habitat (108). He suggests verbatim citation – which might look, to Cratylians, like evidence – might be “less a mimetic luxury” than a “a shrewd rhetorical weapon” (131). 3.6 verbatim’s fraudulent celebrity status ~ Sternberg’s identification of the parts of reported speech – the utterance reported as the inset and the surrounding context as the frame – underscores citation’s uprooting of an utterance from its original context and its re-contextualization. He calls this a mediating or interfering activity, because of course it is always motivated (108). Although this contextualization applies as much to the indirect form of summary as to the direct form of citation, it is only direct citation that has an unrivalled reputation for fidelity, a hang-over from Cratylian descriptive- reference and code obsessions. The idea of verbatim – that what was said can be reproduced by reproducing the series of words uttered (even if also reproducing the context of the utterance) – depends on determinate-descriptive reference. If what was said were not the intended (delimitable and determinate) meaning of the “original” speaker, verbatim would have no status. Since context gives the promiscuous sign its only hope of signification and context itself can never be absolutely delimited, verbatim’s celebrity status depends on Cratylian-code conceptions of language. When it comes to verbatim reporting, scholars of reported speech generally recognize that the concept of duplicating the intended meaning of a prior utterance by chopping it out of its original context and plunking it into a new one simply doesn’t work. The stature of verbatim is a fraud to start with. But there are other clues to this imposture. For instance, there is a real problem with getting the actual words right: and if the words themselves are so hard to remember, so much the more so the recovery of an original context. According to Mayes, memory research shows 102 Gauged by internal and situational evidence, plausible situations are ones in which it is unlikely that the speaker104 would remember the exact words spoken as indicated because too trivial and mundane, e.g., “And then he said, ‘well, we’ll just wait a little while’”; improbable are when the quote is so long or syntactically complex that it is doubtful the speaker could have remembered; highly improbable include such examples as habitual e.g., “And every time I’d like, tell my advisor what I wanted to do, he’d go ‘no no no no no’ you know”; and impossible include hypothetical situations speculating on what might have been said or wasn’t quite said, etc. (332-36). that people simply do not remember the exact wording of entire utterances, much less entire conversations (331). Clark and Gerrig note “empirical research shows that people can’t recall an utterance word for word, even after a few seconds, without taking special pains to rehearse and memorize it” (796). In addition, as Clark and Gerrig observe, people do not actually try to duplicate the exact words; when asked, they try to give an accurate account of the original in their own words (795-96). They find that speakers don’t reproduce an utterance verbatim even when they have memorized the text and could reproduce it word for word (797). Furthermore, Mayes determines that the direct quotes in at least fifty percent of her data do not represent actual previous utterances: they are something like what was said (the gist) or something that might or would be said (hypothetical). She categorizes these into four categories – plausible, improbable, highly improbable and impossible. She concludes that quotations are frequently104 invented: an observation that resonates with Tannen’s notion that direct reported speech is “constructed dialogue.” 3.7 involvement strategies & affective interests in speech ~ The beguiling-dissimulating quality of verbatim citation – its complex of strategic- linguistic and social-affective communicative interests – is clearly part of what makes it “a shrewd rhetorical weapon.” Holt argues that “showing” or performing an utterance allows the hearer “to arrive independently at an assessment” of it and thus be involved in the meaning making (444). Directness’ implicit invitation to participate in the process of communication, to become involved in meaning production and to share opinions, seems to draw on human affective interests in sharing (that Tomkins’ identifies as communion). The fact that rhetorically-strategic and social factors are tangled up together in what Chafe calls “involvement strategies” (strategies that involve participants in communicative processes), doubtless contributes to the effectiveness of direct report. Labov’s “concept of internal 103 Holt repeatedly highlights the participative-affiliative aspects of social communication observing that a recipient105 may seem to be reacting independently to a reported utterance, but the teller’s preferred response has already been made clear. She notes that when responses overlap with similar ideas, “it appears as though both are reacting in the same way independently, resulting in a highly collaborative and affiliative sequence” (427). evaluation” prompts Tannen to suggest that “hearers and readers who provide interpretations of events . . . are convinced by their own interpretations” because they are more involved in the thinking (136). Meyers talks about the way participants in a focus group discussion “never just say what their opinion is, but take a stance in an ongoing discussion” (396). This means that they demonstrate their own interest both in participating in the group (in getting along with others) and (because a stance leaves a person’s perspective implicit) in giving other participants the opportunity to make inferences and investigate.105 Tannen collates the views of Gumperz, Goodwin, Merrit and Chafe, using Chafe’s terminology and his three types of involvement: “self involvement of the speaker, interpersonal involvement between speaker and hearer, and involvement of the speaker with what is being talked about” (137). These involvement strategies include all the previously mentioned motivations: social, communicative, participatory interests elaborated in the discussion of Lesley’s conversation with her mother, in addition to some more visceral and basic affective responses to prosody and repeated expressions. Tannen’s term “constructed dialogue” for directly reported speech indicates its duplicitous character: its constructed, artful, pretence as a disinterested, neutral reproduction, rather than a mechanism of enticement that interpolates a range of affective and social interests. She writes: Constructing dialogue creates involvement by both its rhythmic, sonorous effect and its internally evaluative effect. Dialogue is not a general report; it is particular, and the particular enables listeners (or readers) to create their understanding by drawing on their own history of associations. By giving voice to characters, dialogue makes story into drama and listeners into an interpreting audience to the drama. This active participation in sense making contributes to the creation of involvement. Thus understanding in discourse is in part emotional. (132) The focus here on internal evaluation, participants using their own background knowledge to draw their own inferences, and actively participating in sense making are all features of direct 104 reports that create hearer-reader involvement. The affective social needs that prompt such enticements results in reported speech being characterized as an “involvement strategy.” Here we focus more on other involvement strategies of everyday talk, ones that constitute more basic affective dimensions of prosody and the repetition of others’ words. Drawing on Nietzsche’s suggestion that we have a drive to imitate, Tannen proposes that humans experience pleasure in the familiar and the repetitious (98-99). This mimetic impulse – which was considered self evident in antiquity both as pleasure and as a learning mechanism – leads to a fundamental pleasure in repetition. Drawing on the anthropological linguistic insights of Becker and Bateson about repetitions of prior text, with supporting arguments from Halliday and Hassan on cohesive devices, and the research of neurolinguists and linguists like Bolinger, Gibbs and Gonzales on pre-patterning, syntactic combination and processing efficiency, Tannen explains the astonishing ubiquity of repetition (both individual words and phrases) in conversational discourse as an automatic impulse. She argues that “speakers repeat, rephrase, and echo (or shadow) others’ words in conversation without stopping to think,” that it is an automatic, interactive, involvement, coherence-building response (98-101). Tannen supplies sample after sample of such banal repetition as the following: (The dialogue comes from a discussion of the Whorfian hypothesis.) (4) 1 DEBORAH You know who else talks about that? 2 Did you ever read R.D. Laing? 3 The Divided Self? 4 CHAD Yeah, But I don’t /??/. 5 DEBORAH He talks about that too. 6 CHAD He talks about it too. (Tannen 68) Here the whole phrase is repeated but in many of her samples (of conversations between well educated friends discussing things like the Whorfian Hypothesis) the profusion of repeated single words is also remarkable. The interactional uses of repetition Tannen lists include: 105 “getting or keeping the floor, showing listernership, providing back-channel response, stalling, gearing up to answer or speak, humour and play, savouring and showing appreciation . . . linking one speaker’s ideas to another’s, ratifying another’s contributions” (61). Repetitions are clearly not descriptive referential transmissions of information. Tannen concludes that “active participation in sense making contributes to the creation of involvement” and that “understanding in discourse is in part emotional” (132): a view that lines up with James’ and Tomkins’ assumptions about motivation and thinking, as well as the recent-experimental- psychology findings on the “inseparability of emotion and cognition” that we have discussed (161). Tomkins’ ideas about the social motives for speech embrace both of these involvement enticements: affective responses to prosody and repetition (sound, speech) and affective interests in participation (communion, sharing). Repetition might create meaning both by producing it as well as building comprehension. Tannen suggests that in addition to their interactive functions, involvement strategies might also be coherence-building responses (61). Recalling James’ sentiment of rationality as a feeling of coherence and the role of affect in determinations of coherence in recent experimental psychology makes me wonder how much sheer repetition influences conceptions of rationality. Halliday and Hasan claim that repetition builds coherence in texts. It appears that Chad’s repetition of Deborah’s “he talks about that too” reinforces the interpretation of “that” (something about the Whorfian hypothesis): reinforcing the idea that they share this interpretation of Laing like a shared evaluation. I wonder if the repetition itself can make ideas seem coherent, the way repeated patterns of the humanist-dialectic manuals’ topoi was the source of their reasonableness. There is an uncanny homology between Derrida’s identification of citationality as the identifying feature of the linguistic sign and Tannen’s notion that “repetition is at the heart not only of how a particular discourse is created, but how discourse itself is created” (3): ruminations giving new meaning to Bakhtin’s observation that “our speech is filled to overflowing with other people’s words” (337). 106 Fahnestock footnotes Dubin’s interesting findings that “prosodic differences are not the same as the tonal106 differences involved in a language that uses pitch differences to make semantic distinctions, such as Chinese. A Chinese speaker distinguishing among three variants of a phoneme on the basis of pitch in order to construe the meaning of a word uses the same area of the left hemisphere involved in phonemic distinctions” (167). In addition, Fahnestock reports that the areas of the right hemisphere involved in prosodic construal are the same as107 those identified in social monkeys whose system of alarm calls for various dangers – suggesting an evolutionary origin for the brain's ability to decode sound patterns for their affective content (167). Fahnestock observes that “the separate manipulation of prosody was thoroughly appreciated in classical and early108 modern rhetorical stylistics, but its importance in human language has been neglected by most linguists (Dwight Bolinger being a distinct exception) until recently when the results of brain imaging studies have forced its consideration” (167). Some very basic physiological responses seem to link social and cognitive operations. Tannen suggests that we have an affective response to sonorous, rhythmic poetic effects of discourse. Fahnestock’s rhetoric-of-science research discovers some neuroscience findings on brain hemisphere activity during language processing that seem to corroborate Tannen’s observations. Fahnestock’s sources indicate that linguistic communication activates multiple sites mostly in the (“logic” oriented) left hemisphere of the brain (167). On the other hand, the right hemisphere (associated with affective valences of language) “has the special role of analyzing the prosody of an utterance, the ‘tone’ that is produced by variables of pitch, duration, and loudness”; Fahnestock reports that “the right hemisphere’s analysis of prosody, of stressed syllables and rising and falling intonation, has been tied to decoding the emotional value of an utterance” (167). According to Fahnestock’s sources the special role of the right hemisphere106 in the emotional signification of speech has been repeatedly confirmed (167). This107 neuroscience research seems to verify Tannen’s idea that “poetic” effects of speech elicit affective responses, and confirm Tomkins’ view that we have a physiological affective response to speech itself; if we have a physiological affective response to the physical act of speaking, Tannen’s idea that affective responses compel linguistic repetition is not surprising; like the non-verbal vocalizations in Holt’s transcripts, repetition might often be a lot like singing, auto- affection, participation and self-stimulation combined. Linguistically mediated understanding is conjoined with affective motivations in language use that is neither descriptive nor referential.108 3.8 from involvement to performativity ~ Tannen claims that “casting ideas as dialogue rather than statements is a discourse strategy for framing information in a way that communicates effectively and creates 107 involvement” (112). Arguing that specific detail triggers images, Tannen links two of Sternberg’s attributes of mimetic speech – detail and distinctiveness – to affective responses that create involvement. As in a Tomkins-like script specific and distinctive details, images and scenes involve and engage the hearer in the discourse because details trigger memories which trigger emotions. Tannen suggests that the evocation of images and scenes that an individual associates with specific details contributes to the background knowledge and context used for the comprehension of utterances – a possibility that lines up with the role of background knowledge used for inference in Sperber and Wilson’s model of communication. Detail both provides context and creates involvement. Clark and Gerrig also comment on the involvement produced by vivid representation when they argue that direct reported speech is a demonstration – a “showing” rather than a description. “On the speaker’s side,” they observe, “to demonstrate an event is to reenact or revivify it. On the addressee’s side, to become engrossed in an event is to re-experience it vividly” (794). The vividness imparted by detail, image and scene together with other linguistic-involvement enticements leads Tannen to conclude that “constructed dialogue” involves an audience in interpretation as onlookers to a drama (124). Direct discourse is the performance of (usually) another’s speech. This play-acting dimension of direct discourse, together with citation’s other strategic features (evaluation, emphasis, selectivity, involvement, etc.) undermine confidence in the sterling character of reported speech as reproducing an original utterance. Like code, verbatim’s prestige depends on descriptive reference: verbatim’s demise is the straw that finally breaks the camel’s back for Clark and Gerrig. They write, But almost every argument we have adduced for the demonstration theory is also an argument against the verbatim assumption. By our account, what speakers commit themselves to in a quotation is the depiction of selected aspects of the referent. Verbatim reproduction per se has nothing to do with it. The verbatim assumption has three principal defects. It rests on an undefined notion of verbatim reproduction. Whatever the definition, it doesn’t allow for the selective depiction that is characteristic of quotation. And, more fundamentally, it is expressed in terms of linguistic structures (e.g. ‘the actual words spoken’), and that is a category error. (795) 108 Dörge’s Illocutionary Acts:Austin's account and what Searle made out of it, taken together with the wide range of109 the senses in which Austin’s terms are used, and my own reading of Austin persuades me that Austin’s work was so tentative – thinking in progress – that attempting to pin down relatively determinate consistent meanings for his terms might be utterly futile. Clearly Austin’s use of the term performative to describe the function of a limited number of “conventional” usages was already transformed into a question about the illocutionary force effected by a much larger number of verbs by the end of How to do Things with Words. He acknowledges that “in general the locutionary act as much as the illocutionary is an abstraction only: every genuine speech act is both” (147). Culler’s conclusion about the “performative dimension of all speech acts” seems quite consistent with Austin’s own: he decides that Austin’s account leads to the implication that constatives are simply a type of performative (505-06). Culler suggests that the shifts in the concept of performativity (from Austin’s to his own) “pose questions about how to think about the constitutive force of language” and “the nature of discursive events” (503). The failure of delimitable-determinate-descriptive reference that takes verbatim down, also takes down code. But code’s boomerang tendencies are still evident in Clark and Gerrig’s disciplinary-field’s reference posts: they state that “language use is generally investigated as if it were entirely description,” however, “if quotations are demonstrations, then demonstrating is an essential part of language use” (802). Demonstrations – mini-speech performances – are part of ordinary language: “non-serious” uses are internal to the structure of “serious” ones, as Derrida has observed. It is easy to see that Lesley’s report of Mrs. Field’s utterance acts out Lesley’s evaluation without stating it: just as irony in a text is neither stated nor described but is performed, produced from differences (e.g., between what is said, or what seems to be said, and what is otherwise made apparent, or seems to be said, elsewhere). Similarly, Tannen’s “repetitions” of single words are rhetorical mechanisms that do things: they act out or perform solidarity, thoughtfulness, interest, having more to say, etc. Such rhetorical dimensions of language use prompt Culler to describe literature “as fundamentally performative” (Performative 503). He suggests that “literary criticism involves attending to what literary109 language does as much as to what it says” (506). That speech reports are mini-performances – that they happen to be like a little stage performance – is a minor dimension of their performative edge, but it reveals the secondariness of descriptive reference. In a fuzzy or general way, the term “performative” recognizes various ways that language can be seen to act or perform rather than refer or describe. Reporting’s performativity stems from the way speech reports act, what they do (evaluate, selectively emphasise, invite participation and involvement, etc.). 109 Sternberg claims that “quotation is a microcosm not only of all mimesis but of artistic strategy in human110 communication as a whole” (148). Rhetorical mechanisms can be imagined as linguistic gestures that absorb or take into account the whole range of “reasons” or motivations for speech: i.e, human affective impulses (like interests in sharing, or communion), and social interests (like conferring authority, or jockeying for position to gain the floor), together with other slightly more referential uses (like using a metaphor to set things before the eyes, or analogy to compare apples and oranges, to offer a list of persuasions, or to exploit the miracle of syllogism). All of these factors draw on the actions performed by language. Recall that Halliday says that grammar both construes and enacts; but construing is itself a type of action. Various types of linguistic activities can be seen as performative: a multifaceted performativity in which several operations might take place, at the same time. Rhetorical and specifically performative dimensions of language use are exemplified by citation and its dependence on context. If descriptive-referential uses were contrasted to performative dimensions of language use, reported speech would be paradigmatic of the way that language acts rather than describes (which, of course, it also does). Citation is paradigmatic of the performative, rhetorical aspects of language. Words are always other people’s, speech110 always “filled to overflowing with other people’s words,” as Bakhtin says (337). The scholarship on reported speech affirms the demise of descriptive-reference’s hegemony; and it signals the failure of code conceptions of communication and the termination of verbatim’s prestige. It confirms the Tomkins/James affect vector of language, and reporting’s implicitness performs a whole host of social operations that are paradigmatic of ostensive-inferential communication. The social gestation of the sign and its structure of iterability engenders its rhetoricity and performativity. As Derrida observes: Every sign, linguistic or nonlinguistic, spoken or written . . . can be put between quotation marks; in so doing it can break with every given context, engendering an infinity of new contexts in a manner which is absolutely illimitiable. This does not imply that the mark is valid outside of a context, but on the contrary that there are only contexts without any centre or absolute anchorage. This citationality, this duplication or duplicity, this iterability of the mark is neither an 110 Recall that Van dijk has already implicated the complicity of terminology in memory storage and retrieval.111 accident, it is that (normal/abnormal) without which a mark could not even have a function called “normal.” (Limited Inc. 12) 3.9 Voloshinov’s sign: dancing in the caverns of our minds ~ When other people’s words start looking like actors performing different roles depending on the play, we might want to dig deeper into the lifeworld of the linguistic sign. Voloshinov sets out in Marxism and the Philosophy of Language to resolve “the actual mode of existence of linguistic phenomena” (xv); as noted earlier, he observes that “the reality of the inner psyche is the same reality as that of the sign ” (26); and that “psychic experience is something inner that becomes outer and the ideological sign something outer that becomes inner” (39). Signs are shaped by their experiences and transformations from inner to outer and back to inner signs. In Voloshinov’s semiotic psyche experience itself is shaped by the sign, mediated, and filtered through linguistic cognitive processes – a view that ties in neatly with Halliday’s (previously noted) understanding of the way that our inner and outer experience is shaped by lexicogrammar. Like Halliday, Voloshinov claims that “expression organizes existence” and that “there is no such thing as experience outside of embodiment in signs” (85). In other words the psyche employs signs to organize and to understand experience: a two-sided role for the sign as both filter and mouthpiece of perception that contributes to its moebius-strip-like character. Voloshinov’s semiotic conception of the psyche – in which expression organizes the experience of existence – is also similar to what has been suggested by Fahnestock, Becker and Bolinger. Bolinger talks about reaching for bits of text and Becker asks “what if language extends down into the very ways we shape our experiences, store them, and retrieve them?” (8). Halliday111 directs attention to perceptual shaping of human understanding when he writes: Because the grammar sets limits on what can be meant (even though those limits are constantly being extended), it constrains the ways in which experience comes to be construed – “to be transformed into meaning” . . . Our typical clausal grammars construe experience in the contexts of daily life; in doing so, they constrain our understanding, precluding us from becoming more deeply aware of 111 the nature of everyday phenomena. (OL 257) Halliday’s concern with the shaping influence of language’s lexicogrammar and its perceptual limitations lines up with Fahnestock’s observations about the influence of Renaissance dialectic manuals on the patterns of thought we call “reasoning.” These shaping, filtering semiotic functions raise epistemological questions that do not involve “communication” per se; they involve the way we perceive the world through signs – signs that Voloshinov is always reminding us are socially shaped, that is, shaped for and by their social-communicative functions. He imagines the sign (or the word whose “entire reality is absorbed in its function of being a sign”) as “a little arena for the clash and cris-crossing of differently oriented social accents” (14, 41). A word, he says, is oriented toward an addressee, a “word is a two-sided act. It is determined equally by whose word it is and for whom it is meant” (86). The sign’s ability to absorb its social surroundings (whether that be the psyche of an individual or a public-social world) is what makes it dialogic (that is, accented by more than one subjectivity – hence heteroglossic), and what makes it evaluative. Observing that a “word’s evaluative accent” is different in different contexts, he says that “the problem of multiaccentuality ought to be closely associated with the problem of the multiplicity of meaning” (80-81). Voloshinov, like Derrida, is interested in the sign’s shape shifting abilities, its inclination to gain and loose meaning; Voloshinov additionally highlights the sign’s relationship to thinking, its role in human understanding. Voloshinov argues that “introspection is a kind of understanding”– a semiotic cognitive process that is importantly both active and linguistic (37). Self-observation (introspection) is the understanding of one’s own inner sign. In this respect it is distinguished from observation of a physical object or some physical process. We do not see or feel an experience – we understand it. This means that in the process of introspection we engage our experience into a context made up of other signs we understand. A sign can be illuminated only with the help of another sign. (36) With observations which line up with the ostensive-inferential model of communication and the findings of reported speech, Voloshinov thus takes aim at the “false notion of passive 112 understanding, the kind of understanding of a word that excludes active response” (73). His account of active understanding as supplying a “counter word” has interesting correlates in several other fields of research: To understand another person’s utterance means to orient oneself with respect to it, to find the proper place for it in the corresponding context. For each word of the utterance that we are in process of understanding, we, as it were, lay down a set of our own answering words. The greater their number and weight, the deeper and more substantial our understanding will be. Thus each of the distinguishable significative elements of an utterance and the entire utterance as a whole entity are translated in our minds into another, active and responsive, context. Any true understanding is dialogic in nature. Understanding is to utterance as one line of dialogue is to the next. Understanding strives to match the speaker’s word with a counter word. Only understanding a word in a foreign tongue is the attempt made to match it with the “same” word in one’s own language. (102) Laying down a counter-word in the understanding processes translates an utterance into one’s own words, as it were. If this is the case, then it is easy to see why verbatim is so unlikely as the terms of an utterance will already have undergone translation from outer to inner words: that we don’t even try to remember the original words confirms Clark and Gerrig’s memory research. Voloshinov’s counter-word-translation idea also squares with Van Dijk’s account of linguistic generalizing functions for memory storage and retrieval. Some kind of generalizing translation/summary likely takes place in order to remember utterances. This counter-word idea also has interesting intersections with Fahnestock’s rhetoric-of-science research on language and human understanding. Drawing attention to the relationship between language and comprehension, Fahnestock reports that recent brain imaging studies show “reading has an aural and even an oral dimension” and parts of the brain believed to be “dedicated only to language production are activated during comprehension” (170). Reading silently activates the same regions of the brain as are activated in speech; and comprehension activates speaking zones. Fahnestock cites Dubin’s findings that when listening to speech “silent, covert subvocalization was occurring as part of comprehension. That is, in trying to understand the words being heard, the person was 113 This important qualification underscores the need to consider the different types of semiotic systems that exist for112 those excluded from language. What kinds of semiotics enable their cognitive functions? Consider also Wittgenstein’s meditation on “reading” which begins with queries about “understanding” (and the113 grammar of the verb “knows”) (§145ff). Wittgenstein poses the question of whether someone uttering memorized words out loud while looking at the printed text is “reading.” rehearsing the speaking of those words without being aware of doing so” (170). It seems possible that this subvocalization is like the laying-down-of-counter-words that Voloshinov describes. Subvocalization might line up with the translation of outer to inner words: translations that we then typify (or generalize) in order to remember them. Again drawing on Dubin’s research, Fahnestock writes: [imaging studies have indicated that] verbal working memory for letters, words and names utilized a strategy of silent, nonconscious rehearsal that involves some of the same parts of the brain as actually speaking these items. Studies showed activation of cerebellar regions that would normally be involved in the motor speech task, even though no actual speech occurred. (171) Recalling words seems to draw on the same resources as uttering them. These comprehension mapping studies seem to provide results that line up with Voloshinov’s speculations about the semiotic processes entailed in understanding. Wittgenstein too writes that “if I so much as look at a German printed word, there occurs a peculiar process, that of hearing the sound inwardly” and when the eye passes over the printed text “at the same time involuntary speech goes on in the imagination” (§165, 168). Fahnestock concludes that for non-deaf humans even in silent reading “language as revealed in brain112 imaging studies is always a residually aural phenomenon” (171). Summarizing current theories of “verbal working memory” originally developed by A.D. Bradley, Fahnestock says neuroscientists now postulate an “articulatory loop” as “part of comprehension” (171). Lieberman, she says, explains that the articulatory loop means that while a sentence is being parsed “words are subvocally maintained using neuroanatomical structures that regulate speech production” (171). This sounds a lot like thinking in language. Fahnestock finally concludes that “listening for comprehension involves a simulated speaking of what is heard” (171). Comprehension doesn’t seem to involve passively absorbing an utterance but actively mentally reproducing it or perhaps rendering it in one’s own words. Voloshinov’s account of113 114 inner speech or thought in which utterances “resemble the alternating lines of a dialogue” and his comment that “there was good reason why thinkers in ancient times should have conceived of inner speech as inner dialogue” resonates with what I call Derrida and Plato’s conversational stylistic, discussed in Part 5 (38). The Sophist’s Stranger says “thinking and discourse [lovgoV] are the same thing, except that what we call thinking is, precisely, the inward dialogue carried on by the mind with itself without spoken sound” (Cornford 263e). On the other hand code models of communication tend to imagine speakers as encoding intentions that exist outside of language. For example, Hacker’s “The Rise of Twentieth Century Analytic Philosophy” cites Frege’s view that “the task of the philosopher is to break the power of the word over the human mind, to free thought ‘from that which only the nature of the linguistic means of expression attaches to it’” (245). But there likely is no thought that exists free from linguistic (or some kind of semiotic) expression. As previously noted, Halliday asks “if language is a code, then where is the pre-coded message?” (OL 243). I wonder if non-semiotic processes should qualify as thought at all: affective responses might be extra-linguistic, but even the perception of an affective response becomes linguistically mediated as soon as it is consciously recognized and labelled. 3.10 Nietzsche’s metaphorical turn ~ Nietzsche’s crucial insights into the linguistic nature of thought – specifically the epistemological implications of linguistic mediation – of course predate Voloshinov, Wittgenstein, Derrida and the scholars studying reported speech. His metaphorical account of language’s metaphoricity marks this dissertation’s transition to questions of style, examined in Part 4 where, the significance of his poetic register and tropic mode of writing are foregrounded. Here what is pressing is language’s metaphoricity itself. Because Nietzsche’s tropic mode of writing condenses this dissertation’s previous meditations on semiotic-analogical-human thinking, on partial-limited-human vantage, and our linguistically-mediated-conceptual world, his theoretical remarks are a superb summary of the sceptical-language-theory this thesis presents. Nietzsche’s language-theory lectures on classical rhetoric connect the dots between 115 ancient rhetorical theory and post-structuralist critiques of language: “language is rhetoric” he writes (Rhetoric 23). Given rhetorical theory’s longevity and currency it might be surprising that Nietzsche’s views about language seem to have been overlooked by analytic philosophy’s linguistic turn. Perhaps Wittgenstein’s logic specs (the “pair of glasses” requiring “the crystalline purity of logic”) occluded Nietzsche’s logic-annulling, proto-post-structuralist critique of reference and they just couldn’t see it. Unlike Tractatus Logico-Philosophicus that launched the linguistic turn, the image of language in Philosophical Investigations bears a close family resemblance to that of Nietzsche. Epistemology’s hankering after determinate reference, its interest in description that is more than just relatively effective, one that is accurate and exact that satisfies the demands of “logic,” doubtless prompted a degree of resistence when positivism was reaching its apex. But Nietzsche’s prophetic voice and quasi-mythical register might also have been off-putting to philosophers seeking “facts” (“Tatsachen”) (Tractatus 31): “truths are illusions . . . they are metaphors,” he famously announces (Truth 84). His oracular style may have discouraged analytic philosophy’s interest: it may not have looked like language theory when viewed through logic lenses. Nietzsche’s critique of our language-dependent, conceptual world is founded on his views of the subjective, anthropocentric nature of human perception and the metaphorical nature of the linguistic sign itself. Like the star of his own opera, his melodramatic voice is incredulous at pretensions to objectivity. Dismissing out of hand the idea that we can accurately perceive things to start with, he denounces the notion that “the essence of things ‘appears’ in the empirical world” (Truth 86). We are not omniscient and cannot pretend to objectivity, not only are we completely submerged in the reality we seek to discover but we are part of it, already interested, motivated and involved. Because “the genesis of language does not proceed logically,” the language we use to understand the world is referentially (and thus epistemologically) impaired at its root (Truth 83). And since it does not begin “logically” (as Cratylus dreams) it is always already either arbitrary or haphazard: “and all the material within and with which the man of truth, the scientist, and the philosopher later work and build, if not 116 Perhaps drawing on the same Platonic motifs, Voloshinov comments that signs are “reflections” or “shadows of114 reality” paradoxically both part of material reality and unique in it. He says, “every ideological sign is not only a reflection, a shadow, of reality but is also itself a material segment of that very reality” (11). derived from never-never land, is at least not derived from the essence of things” (Truth 83). Since human beings cannot perceive essences of things (but only a subjective and shadowy image/appearance reminiscent of those in Plato’s cave) Nietzsche claims that even “the criterion of correct perception” of essences of things “is not available” to subjective, anthropocentric intellects (Truth 86). 114 Since our perception of the empirical world is faulty – “the full essence of things will never be grasped” – and because of the partial, subjective nature of human perception and understanding, there is no reason at all to assume an accurate referential word-thing relationship (Rhetoric 23). Nietzsche’s insight into the metaphoricity of language – its alterity to the world it inhabits, its lack of essential ties to its referents – instigates his rhetorical theory of language and his quasi-mythical, allegorical, tropic mode of writing. He understands words as distorted and limited anthropocentric perceptions of the world whose ontological condition makes them fundamentally unsuited for the very epistemological endeavours we depend on them for. Nietzsche considers words “arbitrary assignments” and “arbitrary differentiations” which display “one-sided preferences” for this or that property of a thing (Truth 82), and further that “language never expresses something completely, but displays only a characteristic which appears to be prominent” (Rhetoric 23). Like reported speech, like any kind of analogy, features are always being selected: the selective nature of perceptions – what one chooses to focus on – and thus the analogies or metaphors that are selected to express those perceptions are crucially imbricated in all language use. For Nietzsche there is no competent representation of the world that would be prior to linguistic metaphoricity; there is no literal; reference is always myopic, interested and incomplete. Words, metaphors, and conventional signs are selected on the basis of an already involved and restricted perspective, rather than essential or logical relations. But it is the metaphorical nature of the linguistic sign that completely confuses linguistically mediated understanding. De Man writes that for Nietzsche “the trope is not a 117 As noted earlier, this linguistic quicksand has a parallel in Plato: as Socrates and Hermogenes discover behind115 names all we find is more names (Crat 421e, see also Soph 244d). See also de Man’s “Epistemology of Metaphor.” derived, marginal or aberrant form of language but the linguistic paradigm par excellence. The figurative structure is not one linguistic mode among others but it characterizes language as such” (Nietzsche 35). “To begin with,” Nietzsche says, “a nerve stimulus is transferred into an image: first metaphor. The image, in turn, is imitated in a sound: second metaphor. And each time,” at each remove, Nietzsche writes, “there is a complete overleaping of one sphere, right into the middle of an entirely new and different one” (Truth 82). A visual or other sensory image is displaced by a sound image/word/linguistic sign – both metaphors (meta-fevrw) a carrying over or transfer from one sphere (the essence of a thing, i.e., a biological object) to another (a human perception, i.e., a visual image) to another (a linguistic semiotic phenomenon, i.e., the word “tree”). We skip from the realm of things to a realm of human perception to a realm of semiotics. Linguistic signs are fundamentally tropological– metaphors of things, which are already metaphors. De Man writes that “Nietzsche’s critique of metaphysics . . . lies in the rhetorical model of the trope” – the metaphoricity of language sabotages metaphysics, there is no referent to centre or anchor a concept outside of a metaphor of a metaphor, multiply-displaced (38). We make our limited and motivated selections from an endless chain of metaphorical displacements. “Truths are illusions which we have forgotten are illusions; they are metaphors that have become worn out and have been drained of sensuous force, coins which have lost their embossing and are now considered as metal and no longer as coins” (Truth 84). De Man adds that “what is being forgotten in this false literalism is precisely the rhetorical, symbolic quality of all language” (Nietzsche 39). Our failure to keep the metaphorical nature of language in mind is what Kofman calls the forgetting of metaphor. De Man explains that “the degradation of metaphor into literal meaning is not condemned because it is the forgetting of truth but much rather because it forgets the un-truth, the lie that the metaphor was in the first place. It is a naive belief in the proper meaning of the metaphor without awareness of the problematic nature of its factual, referential foundation” (Nietzsche 39). Language’s metaphoricity dispels Cratylian115 118 Recall the part-whole-comprehension problem (a vision-knowledge handicap) and the analogy problematic that116 prompt Halliday to argue both that one could say there are “no natural classes in the perceptual world” and also that “there are far too many natural classes; phenomena resemble each other in indefinitely many ways, hence there are innumerable ways of categorizing, leading to variation not only between one language and another but also within one and the same language” (OL 244). fantasies about the essential referential meaning of terms and draws into relief the strange confidence in language’s logical-descriptive-referential mechanisms – the unaccountable assumption that language use is “entirely description” that Clark and Gerrig reference. Since Clark and Gerrig’s terminology “non-serious” (for reported speech) implies non-literal, Nietzsche’s insights into language’s fundamentally tropological nature underscores the wrongheadedness that directed the selection of the term non-serious (a legacy through Goffman and Austin from Frege). Nietzsche lays the foundations for Derrida’s critique of centre when he foregrounds the obliteration of uniqueness that occurs with generalizations and the whimsical nature of human concepts. He writes about the generalizing tendency of terminology itself: Every word instantly becomes a concept precisely insofar as it is not supposed to serve as a reminder of the unique and entirely individual original experience to which it owes its origin; but rather a word becomes a concept insofar as it simultaneously has to fit countless more or less similar cases – which means, purely and simply, cases which are never equal and thus altogether unequal. Every concept arises from the equation of unequal things. (Truth 83) Since words are already idealized generalizations – paradigms of the generalizing functions used for memory storage and retrieval – the need to use the same word “to fit countless more or less similar cases” is a referential-identification problem for epistemological endeavours. Since metaphors are always already “arbitrary assignments” and “arbitrary differentiations” which display “one-sided preferences” for this or that property of a thing (Truth 82), never expressing something completely, but displaying “only a characteristic which appears to be prominent” (Rhetoric 23), and since metaphors are always already analogies in which prominent features are selected by restricted-anthropocentric perceptions, the particular term selected often reflects whatever properties of a thing that happened to seem important to a particular individual in a particular context. A metaphor selected for an object at any one moment comes without116 119 i.e., as a term that stands in for something else – a comparison to something else; Aristotlean, “to use a word in a117 changed sense” (LSM). Fahnestock observes that the use of metaphor is an argument by analogy (Figures 129). Tannen cites Friedrich’s118 view that “metaphor is only one kind of analogy and part of a much larger context of analogical devices and associational thinking” (38). guaranteed connection to a metaphor someone else might use to describe the same object, and a metaphor describing an object in one context might be used to describe an entirely unrelated object in another. Nietzsche is concerned with the epistemological problems discussed in Part 2: identification by metaphor is equivalent to reasoning by analogy; as a type of analogy metaphor selection (in the common, naive or Aristotlean sense) is an argument by analogy. Since117 118 “every concept arises from the equation of unequal things” the earlier-noted “category-mistake” problematic wrongly implies the potential for “correct” identification; thus the term “correct” isn’t really applicable in reasoning by analogy in which identification depends on what features are selected for consideration. Nietzsche’s views here line up with both Wittgenstein’s analogy about the way that “logic” becomes a pair of eye-glasses through which to see, and Burke’s analysis of the way that words act as “terministic screens.” The specific metaphor (word or analogy) used to identify a thing predetermines both what we see and how we see it. But we cannot see without metaphors and analogies. As Nietzsche shows we must use metaphors and analogies to identify things and to understand the world. Nietzsche is concerned with the shaping effect of metaphor that prompts evolutionary biologist Richard Lewontin to lament the willy-nilly application of the (biological) evolutionary model by analogy onto other spheres of inquiry, i.e., the projection of evolutionary biology onto say aspects of behavioural psychology, or the economic market, or (most recently!) literary genres, for some examples. The important epistemological questions Nietzsche raises, also implicated in Halliday’s account of language’s enabling and constraining abilities, link sceptical language theorists from Plato to Derrida. Halliday argues that language allows us to ossify actions into objects. The verb “transform” becomes objectified as a process – a “transformation” – something that occurs over time gets scooped-up and summarized as a particular instance. 120 The evolution of the nominalized grammar of scientific discourse enabled the founders of experimental science to reason and to theorize – to construct technical taxonomies and chains of logical argument; but it locked them, or their successors, into a world that was objectified (modelled as objects) and determinate, a construal in terms of abstract and virtual things remote from the positive disorder of everyday life – and as it turned out rather less helpful for conceptualizing the fluid and indeterminate “reality” required by quantum physics. (OL 257-8) Halliday is, of course, not suggesting that nominalizations are used only to objectify actions as static entities, but rather that a fluid reality is more difficult to study when using terminologies that often transform dynamic motion into an observable arrested object because the terminologies shape what we see to start with, determining how we imagine. Nietzsche’s attention to the obliteration of uniqueness and generalization that attends when the word-instantly-becomes-a-concept addresses not only this shaping influence of language that Burke, Leowntin and Halliday are discussing, but the idealized metaphorical nature of the linguistic medium. Nietzsche uses the example of a leaf to demonstrate how an idealized concept of a leaf stands in for all types of leaf: every imaginable type of leaf is generalized into a stylized image. But no such leaf exists and such an idealized leaf may not display features of any particular leaf. Such idealization is a fortiori integral to ideas or concepts: phenomena that have no essential existence outside of the human imaginary. The homogenizing effect of words which joins forces with metaphoricity to disable determinate reference in the “natural” world makes human conceptualization – and the whole ideological realm – utterly un-centred, unanchored and unstable. The semiotic ideological world (that Voloshinov too has highlighted) is the conceptual reality that we live in. Lewontin says “we think science is objective” but scientists “view nature through a lens that has been moulded by their social experience” (3). The social-semiotic-conceptual world we take for granted directs our understanding of the world and thus our lives. Nietzsche laments our failure to direct and shape our own ideological world and the influence we allow such undirected conceptualizations to have on human behaviour: a view also discernable in Marxism and the Philosophy of Language, which, like Philosophical 121 Voloshinov too uses a building metaphor– the “edifice of ideological signs” to discuss the socially negotiated119 conceptual superstructure: “individual consciousness is not the architect of the ideological superstructure, but only a tenant lodging in the social edifice of ideological signs” (13). Investigations, frequently resonates with a Nietzschean understanding of language. Nietzsche says we place our “behaviour under the control of abstractions” – “perceptual metaphors” that have been transmuted into “a schema” – metaphors which “dissolve an image into a concept” even though they are “merely the residue of a metaphor” (Truth 84-85). The entirely semiotic nature of the conceptual world in Voloshinov – “without signs there is no ideology” (9) – also suggests an affiliation between Marx and Nietzsche, since both recognize the influence of ideology on human behaviour. Nietzsche figures the human conceptual world as a building constructed on streams of “running water”: these streams are the “unstable foundations” on which we have constructed our conceptual edifices, out of metaphors of metaphors of metaphors (Truth 85).119 3.11 language’s bewildering complexity ~ Since everything becomes metaphorical, there is no longer any literal meaning and, hence, no longer any metaphor either. Since everything becomes metonymical, the part being each time greater than the whole and the whole smaller than the part, how could one arrest a metonymy of a synecdoche? How could one fix the margins of any rhetoric? Derrida “Double Session” (258) Nietzsche’s presentation of classical rhetoric as congruent with his own understanding of language makes him an exemplary sceptical-rhetorical theorist. Plato’s stylistic choices and the sceptical linguistic approach underpinning his style are sources of affinity between these two philosophers. In Beyond Good and Evil Nietzsche depicts Plato as an “interpreter” who uses Socrates as a “popular tune from the streets”and then, like Bach or Beethoven, submits these themes “to infinite and impossible variations” (BGE 190). Kofman’s account of music’s paramount status in the Nietzschean hierarchy of symbolic spheres provides a glimpse of the esteem for the Platonic text recorded in this musical-variation metaphor (Metaphor 7-11). In Part 5 when we examine sceptical-rhetorical theory in action, we will see why the way that Plato does things with words might inspire such admiration. I suspect that language’s metaphoricity was taken for granted in the intellectual climate of ancient Greece, as we shall see 122 Facile notions of imitation need to be re-thought. Halliwell catalogues five categories of pre-Platonic mimesis:120 visual resemblance (including figurative works of art), behavioural emulation/imitation, impersonation including dramatic enactment, vocal or musical production of significant expressive structures of sound, and metaphysical conformity (such as the belief that the material world is a mimesis of the immaterial domain which Aristotle attributes to Pythagoreanism) (15). See McKeon’s “Literary Criticism and the Concept of Imitation in Antiquity” for Plato’s varied uses of the term mimesis. McKeon writes “the word ‘imitation,’ as Plato uses it, is at no time established in a literal meaning or delimited to a specific subject matter. . . . Like most of the terms that figure prominently in the dialogues, ‘imitation’ as a term is left universal in scope and indeterminate in application (3). He argues that “to require Plato to conform to an Aristotelian conception of definitions or terms in which words are assigned univocal meanings would be to distort his inquiry and make nonsense of much of his dialectic. It is invalid criticism to point out that a term like ‘imitation’ has many meanings in Plato” (4). Sternberg states that “in his insistence on the rhetorical (psychological, affective) effects of letting the characters speak for themselves [Plato] anticipates by a good two thousand years the novelistic theories of empathy and control” (114). in Part 4 when we discuss antiquity’s interesting attitude towards historical “facts.” The sceptical image of language as a both a metaphorical and mimetic medium – an understanding of language with a long history – indicates why language’s descriptive-referential functions might not be as central as the positivists had wished when the linguistic turn in analytic philosophy got started. Language is an imitation in the most productively-creative and celebrated sense of the term which makes it useful and convincing in its most descriptive and referential roles (as modern technology makes clear). Nevertheless the linguistic sign’s120 fundamentally analogical relationship to its referent predicts its freedom from referential restraint. The foregoing sections of this dissertation have highlighted various features of the linguistic semiotic medium which contribute to its productive-creative potential: its promiscuity and instability are factors guaranteeing its superabundant capacity and seemingly endless flexibility. A variety of cognitive aspects of language use which explain some of language’s epistemological limitations for human users have come into view: the interdependence of affect and cognition, our cognitive dependency on analogy, the knowledge restrictions of human vantages (for both determining part-whole relations and for recognition). That we think using a semiotic system also reflects on both the social gestation of the linguistic sign and on the sociality of human communication: its social-affective motives straining towards the often vague aims of involvement, evaluation, and sharing. The linguistic mediation of human thought and the limitations of human memory likely influence the role played in cognitive operations by generalization and particularization. Language’s codification resistance – that Halliday’s 123 account of the numerous ways of classifying grammatical operations predicts – stems from the bewildering linguistic complexity resulting from a sceptical-rhetorical image of language. The reusability of words and their mulitipurpose nature – words that are already metaphors of metaphors – would have guaranteed its indeterminacy to begin with; when its tropological nature joins forces with human sociality language is used to act in the world, and it acquires a performative character. In Part 4 we will see the way that the need for the linguistic sign to be re-useable, that makes it so context dependent has resulted in language’ genre-dependence, yet another aspect of its complexity. This dialogical aspect of language guarantees that the linguistic semiotic realm is always in flux, always fluid, ultimately indeterminate and undecideable. Ancient rhetorical theory’s attention to various dimensions of the epistemological quagmire recognizes not only our subjective human vantage but also the interdependence of pathos and logos permitting our complex-interfused-linguistic motivations to retain their conjugal relations. Recall again Socrates’ comment that “our soul at any one moment teems with countless . . . contradictions” (Republic 603d Loeb). Naas’s exploration of Homer’s rhetoricity traces the rhetorical linguistic consciousness in ancient Greek literature back to what he calls the topos of persuasion: a topos that links persuasion to turning, trope and metaphor. The topos of persuasion is a fundamentally ambivalent turning – oscillating between activity and passivity, between its common-middle-voice meaning, obedience, and its active-voice meaning, to turn or bend the mind of another. “To persuade is to lead someone to obedience; to obey is to let oneself be persuaded” and “‘changing the mind’ of another involved being changed oneself’” (Turning 20, 22). Persuasion is cast here as a kind of involvement; language’s rhetoricity stems from its metaphoricity and its social and affective origins. Perhaps the matted and tangled threads that produces this bewildering linguistic complexity is the linguistic- semiotic condition of human thought – a condition that prompts a sceptical (eyes-wide-open) language theory . 124 Silva Rhetoricae considers “sorites” to be a figure or reasoning (a type of figure of speech) appealing to logos: that121 is, “a chain of claims and reasons which build upon one another, concatenated enthymemes.” PART 4 4.1 stylistic turns ~ We pause, before Part 5’s examination of the way that Plato’s and Derrida’s views of language are stylistically inscribed in their texts, to consider what it means to talk about style. We have heard Fahnestock talking about a “rhetorical stylistics” and the formative role played by stylistics in the construction-of-argument advice of the humanist-dialectic-training manuals (Stylistics 215). Her studies of the early modern period indicate that the humanist-dialectic scholar Melanchthon considers dialectic and rhetoric to be so similar that “scarcely any distinction can be observed between them;” she says his manuals emphasize “the interconnectedness of argument and style” (Stylistics 224). The elusive boundary between style and argument – a form-content delimitation that will never be simply manifest – is the focus of attention in Part 4. Fahnestock’s review of various kinds of the humanist tradition’s hybrid topoi, like sorites (which she says could be expanded so that they were “understood as a121 series of compressed or overlapping syllogisms”) finds, as we have noted, that the patterns themselves “should be seen as the source of their ‘reasonableness.’” Their apparent logic, their persuasiveness, stems from the fact that they predict and fulfill patterns. “The logic is figural” she writes (Stylistics 226). The citationality (or the repeatability, reuseability, quotability) that is presented as the identifying feature of the sign in “Signature, Event, Context,” and the phenomenon of repetition in conversation has left lingering questions about what repetition’s coherence-producing cognitive effects might be. The potential coherence-building properties of the conversational compulsion-to-repeat might have its origins in the way that repetition produces weak implicatures and contributes to “background information” in the production of relevance, that we discuss in Part 5. But there is a real question about what “reasonableness,” “rationality” and “logic” are, and how much these terms, which mean quite different things but are often treated as though interchangeable, are related to each other. The affiliations in this collocation have 125 The characterization of Socratic dialogue (for details of this genre see also Charles Kahn) in Bakhtin’s chapter on122 genre in Dostoevsky’s Poetics suggest that Plato’s “freely creative” use of this “serio-comic” genre may have influenced Bakhtin’s terminology dialogic. He says “at the base of the genre lies the Socratic notion of the dialogic nature of truth, and the dialogic nature of human thinking about truth.” Bakhtin refers to “the Socratic method of dialogically revealing truth” (109- 110). The dialogic nature of truth reveals the link between linguistic scepticism and dialogism: what truth is depends on which face of a thing one looks at or beholds, aware of a multiplicity of facets to be seen and a multiplicity of vantages from which to see. See de Man’s “Dialogue and Dialogsim” and Cossutta for different commentary on dialogism’s relation to dialogue and the dialogic character of the Platonic text. See also Tejera. interesting overlaps with “plain,” “clear,” and finally “transparent,” as we shall shortly see. Fahnestock asks us to consider whether the humanist tradition’s topoi are coherent because they are repeated patterns. Her account suggests that rational sentiments – stirred by the recognition- of-familiar-patterns (which constitute the topoi) – is what makes them “logical” and hence why they appear so prominently in the writings of early modern science. Fahnestock’s conception “rhetorical stylistics” calls into question just where the line between style and argument might be, and whether there really is one. If the topoi are repeated patterns, then are they style or argument? Perhaps they are styles of thinking. We might re-think the interconnectedness of style and argument as styles of thought: styles of thought lining up with conceptions of the world (with ideologies), and not the least, styles of thought lining up with understandings of language. In this style-argument territory Nietzsche is again exemplary: on the one hand his metaphorical turn is an epistemological critique pertaining to the ontological condition of the linguistic semiotic milieu, and on the other hand, this epistemological critique of language (which is also a critique of metaphysics) is stylistically inscribed in his text. In other words, his metaphorical turn is stylistically heralded by his own turn to turns: his quasi-mythical and allegorical tropic mode of writing. His thinking about language’s metaphoricity is reflected in his style. If Giltrow’s view – that we can read truly “dialogic” writers’ styles for their theory of language – is right, there may be a natural affinity between language scepticism and Bakhtin’s conception of dialogism. The styles of other sceptical-theory-resistant-language theorists122 comes to mind: Plato, with his image-greedy Socrates, shadowy caves and winged horses, or Derrida, whose “style,” Garver observes sympathetically, is “full of metaphors” and “absurdities,” not to mention the “florid language that sometimes leaves one mystified” (xxvi). 126 We might even think of Parmenides’ goddess, who directs the seeker to follow the path of Peitho (Persuasion) because she attends upon truth (Fragment 2.4). The common countenance of these texts which also express sceptical views of language suggests that a kind of footloose stylistic attitude easily affiliates with language scepticism. Wittgenstein’s shift from the treatise style of the Tractatus to the aphoristic mode of Philosophical Investigations may trace an exemplary trajectory from idealizing language positivism to (eyes-wide-open) language scepticism. Recall Wittgenstein’s prefatory comments about the codification resistence of his “remarks,” Bacon’s denunciation of the totalizing pretense of the treatise (with its implicit claim to omniscience), and his praise for the isolated insight-potential of aphorism. Such reflections and their implications might make us wonder whether the style-of-science myth has shaped genre practices like that of the treatise. I suspect that the great-grandfather of science, Aristotle himself, played a role in the style-of-science myth. And, as we shall see when we come to discuss genre, since style is a major player in genre, it too plays a significant role in meaning-making and how we understand what we hear or read. If a certain scepticism about language is indexed by a text’s stylistics, Nietzsche’s quasi- mythical, allegorical, tropic mode of writing can be seen to reflect the precarious ontological condition of language (its affective social dimensions, its multiple interfused purposes and functionality, its performativity). This relationship suggests that stylistic choices can be meaningful in manifold ways. But the question of just what style is lurks problematically through this discourse. Since style is the property (of a text) which ostensibly subsumes (or has typically subsumed) “metaphor,” Nietzsche’s stylistically-heralded metaphorical turn seems to invite investigation of style. But what is style? Is it “how speakers write or say whatever it is that they say” (Abrams), or the more old fashioned “arrangement of words in a manner which at once best expresses the individuality of the author and the idea and intent in his mind” (Holman)? These definitions make style sound like a transportation vessel: the outer vehicle of the inner content or meaning . Aristotle, who treats style with varying degrees of disregard in 127 This is the view that Derrida pokes fun of in Spurs: Nietzsche’s Style where he never explicitly discusses123 Nietzsche’s style. It is also happens to be the metaphor that Aristotle uses to demonstrate the appropriate selection of metaphor: “one should consider what suits an old man just as a scarlet cloak is right for a young one; for the same clothes are not right [for both]” (On Rhetoric 3.2.9). Rhetoric, says that the virtue of levxiV (‘saying,’ ‘expression,’ implicitly ‘selection of words’) is to be clear without being commonplace (Poetics 1458a). The Aristotelian term “metaphor” pertains to how people say things, the selection and arrangement of words, the right way of saying things. These images of style are kin to conceptions of style as the garb of language’s content – how it is dressed, the clothes it is wearing. But Judith Butler’s paradigm of gender’s123 relation to performance is closer to argument’s relation to style. Style is more aptly thought of as the performance that wins the Oscar than the dress worn to the ceremony. And language’s metaphoricity will lead inevitably to the brink of a style/argument precipice, and its homology – the endlessly problematic form/content pair. In Part 4 we contemplate Nietzsche’s metaphorical turn, his argument for the tropological (metaphorical) nature of language that is also a stylistic turn; and we explore the implications of metaphoricity for language theory. We review Valenza and Bender’s contribution to Culler and Lamb’s Just Being Difficult: Academic Writing in the Public Arena that examines the landmark-stylistic shift in philosophy made by Hume in order to accommodate the reading aptitude of a broadly educated public: a move out of the academy into public discourses. After considering the myth of the plain style, the “plain,” “clear,” “transparent” collocation in philosophy, we assess Aristotle’s clarion call to clarity, including the contradiction between his style of writing and what he says about it, finally returning to the relationship of reader and writer motivations in difficult texts. We conclude Part 4 with a quick glance at genre from a theoretical perspective in order to keep its paramount role in meaning making in mind as we move into the examinations of the curious generic qualities of the Platonic and Derridian texts. First of all, language’s metaphoricity contaminates Cratylian dreams of literality from the outset: the term metaphor is itself already a metaphor – metafevrw, to carry over or transfer, to change or alter, and (most famously and metaphorically, from Aristotle) “to use a word in a 128 Derrida’s critique of metaphor (“White Mythology”) dwells extensively on the Aristotelian treatment of metaphor.124 He chidingly observes that Aristotle’s definition of metaphor is itself full of metaphors: “if we went back to each term in the definition proposed by the Poetics, we could recognize in it the mark of a figure (metaphora or epiphora is also a movement of spatial translation; eidos is also a visible figure, a contour and a form, the space of an aspect or of a species; genos is also an affiliation, the base of a birth, of an origin, of a family, etc.). All that these tropes maintain and sediment in the entangling of their roots is apparent” (252-53). Aristotle says, for example: “It is a great thing, indeed, to make a proper use of these poetical forms, as also of125 compounds and strange words. But the greatest thing by far is to be a master of metaphor. It is the one thing that cannot be learnt from others; and it is also a sign of genius, since a good metaphor implies an intuitive perception of the similarity in dissimilars” (Poetics, 1459a in McKeon). changed sense” (LSM). As Derrida points out metaphorikos still designates in modern Greek “that which concerns means of transportation” – fevrw, I carry (a verb used, already in the ancient lexicon, in multiple metaphorical ways) (Retrait 102, LSM). When Voloshinov suggests “multiplicity of meanings is the constitutive feature of word” the common-sense, Aristotelian use of the term metaphor is made lame (101). Since only the first referential assignment of a term would be “non-metaphorical,” it is impossible to rigorously distinguish the metaphorical from the “non-metaphorical.” In order to determine what is metaphor we would need to know a name’s original unique referent. This would be a futile effort – not only because of a lack of124 knowledge of historical origins, but because of the impossibility of accurate identification to start with – epistemological impediments like identification by analogy and the omniscience needed for accurate part-whole classification which together make the term “category mistake” a misnomer. These are problems that Nietzsche has treated as: subjective-anthropocentric human-vantage, the generalizing, homogenizing and idealizing effects of words and conceptions, and the displacement effected by language’s essential metaphoricity. Hence the sticky etymological task of trying to classify terms as either “literal” or “metaphorical” leads nowhere: begging the question of what “literal” actually means. One could never determine which was the “literal” meaning to begin with. Metaphor selection decides not only what selected features one sees in a thing, but also what kind of a thing it is, what it is similar to. If, for example, we notice that it is something to sit on, a bidet might seem more similar to a toilet than a sink: ajnalogiva or analogy means proportion in ancient Greek, having to do with structural pattern or shape or form. Metaphor125 is at once both a means of emphasis (setting before the eyes, as Aristotle says) and an argument 129 about what a thing is (or is like): both a rhetorical-stylistic mechanism and an argument at the same time. Metaphor selection is a decision that oscillates between “style” and “argument,” it is both form and content. Perhaps this is why the claims Aristotle puts forward about metaphor- selection bear such close resemblance to those dealing with what is understood as invention or argument. How much of Lesley’s direct report of Mrs. Field’s utterance is stylistic and how much argumentative? With the linguistic form of direct report it seems impossible to sever stylistic strategy from argumentative strategy. Reported speech and its forms of directness and indirectness depend conspicuously on implicitness – implicitness and forms of report may seem like a stylistic decision (being both interpersonal and performative) but the direct report is also the point of the conversation (the implicit what of the discussion, a shared evaluation). As Halliday reminds us, language does more than one thing at the same time. The interstices between style and argument, form and content are interfused – brackish waters whose limits cannot be determined. We cannot disentangle the referential from the performative, because they might not come apart. When language is used entirely as ostension, how could we delimit and quantify the descriptively-referential element (about closing the window) of the question “is it cold in here?” 4.2 the vicissitudes of “plain” style myths ~ Aristotle’s advice on clarity and style – “let the virtue of style be defined as ‘to be clear’” – treats the use of metaphor as a privileged strategy for setting things “before the eyes,” a matter to which we shortly return (On Rhetoric 3.11, 3.2). Like other qualities that champion determinate-reference in modernity, “clarity” is linked to the “plain style” which itself has morphed from Bacon’s meaning “true” or “open” to the ubiquitous modern-linguistic ideals of clarity, simplicity and transparency. Hume’s exemplary stylistic shift, outlined by Valenza and Bender in their contribution to Culler and Lamb’s recent collection about stylistic difficulty, sheds light on the vicissitudes of the plain style. Their observation that Hume’s interest in a “scientific method” that would “allow him to achieve a revolution” in philosophy analogous to “the one Newton achieved in physics” helps us to see how the plain style ends up acquiring 130 affiliation with unexamined views about broadly accessible “plain” language (30). Recall Fish’s claim that conceptions of “plain style” (however empty or “ultimate” the term) had won the day by the end of the seventeenth century. According to Valenza and Bender a host of eighteenth- century writers from John Locke to Hugh Blair championed a “belief in the power of common language to address philosophical questions” (35). Valenza and Bender narrate the eighteenth century’s consuming interest in public-sphere discourse: in moving “specialized, disciplinary knowledge into the public-sphere,” a shift accomplished by using language that was “readily available to an ever-broadening, nontechnical readership” (31). Their analysis traces Hume’s stylistic transition from the “highly learned” Treatise of Human Nature (written in “the idiom of university-trained specialists”) to the “conversable” Enquiry Concerning Human Understanding (written in “the conversational vernacular of a broadly educated public”) (29). Valenza and Bender’s reading depicts the Hume story as paradigmatic of the road-to- code: the crossroads where “plain’s” new meanings become attached to determinate reference. The mysteries of these linguistic interminglings and the kind of coherence that emerges from their collocations is worth some attention (and would benefit from a Halliday-type cohesion analysis that would be too detailed to undertake here). Simple, clear and transparent – each term with its own distinct meanings nevertheless seem to rally around and bolster literal, determinate-reference and code – naive intuitions that seem attracted to one another. If words were literal, determinate references to unique things – i.e., words imagined with invisible umbilical chords to things, a genuinely Cratylianian universe – it might be possible to use the “correct” term, which would facilitate mechanical code transfers of clear information. Observe that if it were possible it might be simple to do (matching word to thing), but not likely very simple to read: a distinction that might have been overlooked in the rush to celebrate clarity and simplicity. Valenza and Bender claim that in the eighteenth century, the broadly accessible implication of “common speech” gets affiliated with plain style and lack of ornament – that is literal, determinate reference. But neither literalness nor determinate reference facilitate comprehension or ease of reading. The word-thing fantasy does nothing for reading, nothing to 131 make writing easy to read: if it did and were possible, Aristotle might never have felt impelled to write Rhetoric to start with. The collocations (clusters of repeated words) surrounding the plain style need to be viewed with suspicion. Valenz and Bender claim “rhetorical clarity” was the “hallmark” of eighteenth-century- empiricist philosophy, fostering the maxim that “when clarity was not possible, inquiry was inadvisable”; and they argue that fields of inquiry were knowingly restricted by this view (35). Their account of the eighteenth-century belief that “when language is held strictly accountable to the ideas underlying it, both language and ideas will be transparent to their readers” demonstrates an emerging confidence in both determinate-reference and linguistic transparency (35). Hume’s difficult Treatise was roundly criticized when it was published because of its inaccessibility to public-sphere readers. According to Valenza and Bender, the contemporary belief that common speech embodies common sense and that anything worth saying can and should be said in broadly accessible terms is at the heart of Thomas Reid’s eighteenth-century critique of Hume’s Treatise (36). We can see the simplicity motif creeping in here. Such appeals to commonsense involve one of the most disingenuous and yet persuasive figures of argument: expertise is discredited in favour of what is simple, clear and obvious to everyone (like the fact that the earth is flat). The anti-intellectual notion of commonsense – that human experience is (or should be) simple, obvious and accessible – seems, in Hume’s day, to have become allied with (democratizing) conceptions of common speech and Enlightenment aims more generally. Centuries old rhetorical-dialectical approaches to language have moved offstage as notions of determinate reference and transparency come dancing into the scene. In wake of the public-sphere failure of his Treatise Hume turned to a more conversational style in philosophic writing to make philosophy more accessible for “well-bred company” (30); Valenza and Bender claim that he moved “from philosophical discourse to more readily available language” which meant that he had to abandon the most difficult and technical aspects of his philosophical writing; and when the technical language was cast away “he also cast away the set of ideas not responsive to treatment in the language of the essay, 132 [which was] the genre of the public sphere” (33-34). “Critics have remarked,” they observe, “almost since Hume’s own time that in order to fit the Treatise’s ideas into the more conversational Enquiries he left out much of what has most interested professional philosophers in his own time and ours” (38). He sacrifices the most valued of his philosophic contributions – in order to appease the interests of broad public appeal, in order to make his complex text more simple, in order to stay within the purview of commonsense and common speech – a step that looks to me like the dumbing-down of philosophy. Valenza and Bender determine that ultimately the achievement of “the Treatise is to illustrate that our faith in the alignment between fact and its representations is itself a problem – [one] riddled with difficulties implicit in the act of representing” (36). In other words the Treatise is concerned with the important epistemological question of alignment between fact and its representations (i.e., referential fact and linguistic identification). Thus Hume’s shift (to the conversational Enquiries) abandons his discourse on representation. Commonsense, which looks a lot like resistance and not much like sense, together with its ally determinate reference wins the day. Commonsense advocates a simpler, easy-to-read philosophy. Valenza and Bender conclude that Hume’s belief in “the rhetorical transparency of common sense embodied in ordinary language” was simply a mistake: commons sense and common language have nothing to do with rhetorical transparency, a matter to which we return shortly (36). The phrase “ordinary language” (and Valenza and Bender’s miss-take of here) is a late-comer to this discourse about simplicity, clarity and transparency; Valenza and Bender’s miss-take of the phrase introduces a new partner to the plain style’s collocations. Observe that simplicity, clarity and transparency pertain to the style of science myth, whereas Wittgenstein’s or Austin’s attention to “ordinary language” use was something else altogether – a focus on actual speech, as a source of how language works. Whether we examine conversation, ceremonial practices or academic writing – the overlaps tell us something about ordinariness in language use. But there are reasons for Valenza and Bender’s misunderstanding about “ordinary language.” It is possible that the style-of-science myth (with its impromptu collocates) had 133 Williams is not alone in this characterization, as the opening remarks of Glendinning’s paper indicate. In analytic126 philosophy’s recent identity crisis discourses, others have made the same argument: Sorell suggest that “a very high degree of clarity and precision of formulation and argument” is a defining feature (1), Preston cites several sources who conclude analytic philosophy “is characterized by style rather than substantive philosophic views” (9), and Føllesdal claims that analytic philosophy is concerned with “the validity of a line of argument,” truth status, clearness and cohesion (200-02). Hacker says that Frege’s seminal influence is premised on the view that “logic is the science of the laws of thought” (245). For the history of analytic philosophy and its dance with the linguistic turn see also Akehurst, Biletzki and Matar, and Skorupski. already done its work on the linguistic-turn philosophers: it may have already taken hold in Wittgenstein and Austin. Glendinning observes that Austin made a style of thinking (or as Austin himself put it a “fashion” of thinking), otherwise known as Oxford philosophy, available to a wider audience (314-315). In this light Oxford philosophy evinces the blending of style and argument. It is not hard to imagine an empty or “ultimate” term like the “plain style” of science getting confused with a term like “ordinary language” – an affiliation that taints the latter term. It seems likely that Oxford philosophers were devotees of the mythical “plain” style of science, and confusing it with ordinariness, were practitioners of a broadly-accessible “Hume style” (that is, a common-sense-and-content-limiting style rather than anything like what natural scientists actually use in scholarly writing). For instance, Glendinning refers to Williams’ “characterization of analytic philosophy in terms of its ‘plain style’” (309). Williams’ view that “analytic philosophy can be distinguished ‘from other contemporary philosophy’ by the fact that it tries to employ ‘moderately plain speech’” is widely accepted. This identification makes analytic philosophy’s distinction (as a school or tradition) “a matter not so much of its content but of its style” (10). Implying that the style might be content-limiting, Williams observes that126 the style of analytic philosophy “must to some extent determine the subject matter” (310). Like the Hume-type-plain style it somewhat restricts itself to simpler, more familiar subject matters, more disposed to reductive clarity. The question is whether the face of linguistic complexity is able to make an appearance at all in such a Hume-type style. Williams’ admission, that the supposed clarity “possessed by the sciences” may be a virtue, illustrates both the widespread uptake of the style-of-science myth (as a taken-for- granted fact) outside of science, and its role as an ideal within philosophy; but then Williams asserts that he does not see any advantage in “attempts to make philosophy simply look like a science” (310). Even though arguing against philosophy’s use of it, Williams too buys the myth: 134 the plain style itself has become a kind of figure of argument; premised on fantasies of determinate reference, it argues for a mythical linguistic transparency. While it is supposedly okay for science (which never actually uses it), Glendinning promotes Williams’ disavowal of the Hume-type plain style for philosophical investigation: [William’s purpose] is to question whether the minimally expressive style of the sciences, and their quest for theoretical simplicity, can do justice to the phenomena that philosophy . . . takes as its subject matter. In short, it raises the question whether the plain style with its supposedly clear content can harbour its own lack of clarity concerning its way of going on. (310) Questioning the benefit that such a style might offer to philosophy – how much it might occlude important questions – Glendinning here casts doubt on the mythical “plain” style’s ideality. The hegemony of the plain-style myth – advocating notions of transparency and clarity as the peerless ideal of writing across the disciplines – continues to obtain. It is taken-for-granted. The myth transcends actual-scientific-article-reading-experience that would otherwise be instructive about science’s (difficult, technical) style. Analytic philosophy’s Hume-type heritance of the style myth initiates the infamous “bad writing award” which the journal Philosophy and Literature presented to unsuspecting and influential scholars engaged in theory for several years (1). Culler and Lamb’s provocatively titled collection dealing with style, Just Being Difficult? Academic Writing in the Public Arena, takes inspiration from the idea of the award itself, “bad writing,” and the remarkable public controversy inflamed by it. “What is it?” they wanted to know, about writing that would warrant this award for badness. Intrigued that the “accusation of bad writing” did not seem to “require explanation or demonstration” as if all one had to do was quote a sentence out of context, Culler and Lamb are quite startled by the fact “that the [journal] editor asked only to see a sentence or two” on which to base judgement (43). They are also surprised that (across disciplines and sectors) everyone is an expert (as “commonsense” implies), qualified to gauge a text’s writing and “instantly recognize how awful it is” (1-2). This (interdisciplinary and) public-sphere authority seems to evince the rise of the Hume-type-plain style (accessible to a broadly- educated public) under the hegemony of the style-of-science myth. Why should the award- 135 winners’ language-and-cultural-theory texts be accessible outside of their own academic discipline, let alone to a broad public-sphere audience? And who invited these outsiders into such specific college corridors anyway? There are a host of confused conceptions of language mixed up here: about how language and genre work, not to mention the specific nature of expert registers. One curious and telling notion about the image of language underlying the award’s conception is that a sentence out-of-context was adjudicable: assessable for something as vague as goodness or badness in writing. Since a sentence out of context might not be meaningful at all, this means that the quality of writing can be judged without even understanding the text. Apparently meaningfulness was not a criterion for assessing goodness and badness in writing. Since they only needed to see “a sentence or two,” and thus didn’t need to understand what was being said, I think this means that it was the style that was under scrutiny, that the adjudicators might have thought they were judging the style of the texts, and that the style could somehow be separated from the “arguments”– that the form could be distinguished from the content. Culler and Lamb conclude that the language-theory texts of the award recipients were denounced as bad writing because obscure and implicitly meaningless. “The allegation of bad writing works,” they say, “through an appeal to transparency that assigns badness to opacity” (2). Opacity seems to be the evil other of clarity. “In this model of transparency, clarity is what provokes immediate recognition, bad writing” they say “might be above all merely unfamiliar” (2). The journal was not considering, for instance, chemistry articles which would likely have been completely obscure and estranging to those who enthusiastically weighed in on the theory discussion. Observe that embedded in a text’s (odious) obscurity and (estranging) unfamiliarity, we again stumble on the familiarity-recognition thematic. Recall that familiarity has a part in linguistic involvement strategies (the complex social and affective interests entailed in speech). The obscure theory texts haven’t offered the castigating readers anything: they were not invited, not addressed. The failure to involve a reader by offering them something they recognize risks a kind of alienation that might prompt reader resistance. The Meno’s recognition paradigm 136 reminds us that one cannot recognize what one does not know. Culler and Lamb eventually conclude that comprehension failures may be responsible for the uninvited journal editor’s indignation: but he would never have been offended if he hadn’t crashed the party to start with. This anecdote seems to show how difficult, unfamiliar, or not immediately recognizable ideas can prompt suspicion if not outright rejection and discourage new and/or different ideas: the (strange) ideas of others, or (complex) difficult ideas, or (demanding) thought-provoking ideas. Such thinking-about-new-ideas and recognition-failure difficulties are the focus of Hobson’s epigraph for the introduction to Jacques Derrida: Opening Lines (a summary of Kant’s Prolegomena to Any future Metaphysics): A new science, which is wholly isolated and the only one of its kind, may be approached with the prejudice that it can be judged by means of the supposed knowledge that one already possesses, even though it is the reality of this very knowledge which must first be wholly doubted. To do this only produces the belief that what is seen on all sides is what was already known before, perhaps because the terms sound rather familiar.Yet everything must seem extremely distorted, nonsensical and like gibberish, because it is not the thoughts of the author that are being taken as the basis, but only one’s own way of thinking, which by long habit has become second nature. (Hobson 1) This prelude to the Derridian text – foregrounding the knowledge/recognition problem, the question of familiarity and the other that lurks in sceptical language theory – alerts us to the potential for difficulty, heteroglossia-turned gibberish, and the way that established bits of discourse become our habits of thinking. Adorno’s compressed, aphoristic comment in “Morality and Style” on thinking failure – “[the] shoddiness that drifts with the flow of familiar speech is taken as a sign of relevance and contact”– exhibits the same familiarity motif. Here, questions of effortless recognition and epistemological significance are tangled up with speech’s social interest in contact (perhaps in shared evaluation). Hume’s troubling acceptance of the commonsense view that ideas are unworthy if not simple and accessible calls into question how language’s complexity can be addressed, if at all. As Austin remarks about complexity “life and truth and things do tend to be complicated” (PP 252). The assumption that complex and difficult notions can be rendered in a transparent, ever- 137 clear language is curious. That conceptual difficulty and complexity can provoke strongly negative affective responses and resistance to certain types or styles of theory (with the demands or impositions of writers challenging their readers) seems straightforward enough; but what conception of language guarantees that complex and difficult matters can be rendered in simple, clear prose? Derrida has harsh comments to make about the wish to simplify at all costs: One shouldn’t complicate things for the pleasure of complicating, but one should also never simplify or pretend to be sure of such simplicity where there is none. If things were simple, word would have gotten around, as you say in English . . . Those who wish to simplify at all costs and who raise a hue and cry about obscurity because they do not recognize the unclarity of their good old Aufklärung are in my eyes dangerous dogmatists and tedious obscurantists. No less dangerous (for instance, in politics) are those who wish to purify at all costs. (Limited 119) The conception of language that imagines clarity and transparency as its sovereign signs – the one conveyed by the Hume-type-plain style – might have troubling political implications and questionable social aims. The collocates that have gathered around science’s plain-style myth tell us something about linguistic intimacies and how they work – the way that meanings affiliate and rub off on their dancing partners. We need to ask what kinds of “content” might be occluded by Austin’s fashion of thinking – or whether in fact it is just a style. 4.3 Aristotle’s clarion call for clarity ~ Keeping in mind the ambivalent role that Aristotle plays in this thesis, we turn here to his clarion call for clarity. Unlike imaginings of determinate reference that underlie the Hume- type plain style, Aristotle’s stylistic advice – “let the virtue of style be defined as ‘to be clear’” – did not begin life in conceptions of determinate reference, linguistic transparency or faith that everything was simple (On Rhetoric 3.2). Transparency and simplicity are rhetorical art, requiring artfulness, largely the art of metaphor or analogy (which requires knowledge and insight). Just as it is not logic but the appearance of logic that will be convincing – by extension it is not simplicity, but its appearance that one strives for to comply with Aristotle’s enjoinder to set things “before the eyes” using metaphor. Aristotelian metaphor is the premiere art involved 138 Recall Fahnestock’s characterization of ancient and early modern “attention to language.” Ancient linguistic127 consciousness has become a subject of recent interest; for instance, Frede and Inwood’s study of classical Greek and Latin understandings of language observes that Greek and Roman culture “was intensely language conscious” (1). As noted earlier in de Rijk: Aristotle’s theory of “semantics indeed will turn out to be the mechanism of his way of128 thinking and explaining things, and should therefore be used as the key par excellence to understanding Aristotle’s thought” (10). in producing clarity (3.11). Conjuring such illusions of transparency and simplicity employs the centuries old language arts (lovgwn tevcnh) that Aristotle codifies in Rhetoric and that he claims are a faculty or ability not just a craft/art. These ancient lovgwn tevcnh seem to stem from a culture in which language (rather than being transparent) was understood as having an almost visceral phenomenality and tangible force.127 Aristotle is an excellent example in this dissertation for numerous and incongruous reasons: not only because he codifies the sceptical language theories that permeated the intellectual atmosphere of ancient Greece, and not only because of his clarion call for clarity, but because there is a contradiction between what he seems to say in Rhetoric and how he writes, that is his “argument” and his “style.” In particular, he is exemplary in the way that his scientific method (mostly outlined in what is called the Organon) seems to be a style of thinking: a possibility immanent in both Austin’s “fashion of thinking,” and Fahnestock’s claims about the rhetorical stylistics embedded in the “arguments” of early-modern science. Aristotle’s analytic process or method can be broadly described as a variation of Plato’s collection and division (complemented by definition) – a linguistically mediated method a fortiori determined by the selection of words. Plato’s collection and division specifically takes stock of the linguistic mediation of thought, as Socrates makes clear when he says: “I myself am a lover of these divisions and collections, that I may gain the power to speak and to think (emphasis added Hackforth Phaedrus 269b). Aristotle’s Rhetoric demonstrates how choice of128 words and manner of expression can signify a determinate meaning (in conflict with what it otherwise seems to say). His overall style shows how reported speech can perform the ethos of the scholar (however disingenuous or contrived), and how compelling reductive clarity can be 139 The theory of how language works implicit in Rhetoric is quite at odds with the style of Rhetoric. But this style is129 entirely consistent with what Aristotle seems to say elsewhere (i.e., the six works comprising the texts known as the Organon) about language. The long history of scholarship about the Organon’s difficult and apparently contradictory texts highlights the complexity and possibly wishful, exploratory nature of Aristotle’s thinking about language. The facts that his thinking arises in various texts outside of the Organon, and that a range of subjects are addressed, may account for some of its paradoxes. However, in the end, his engagement with different aspects of language’s epistemological conundrums seem to have underwritten countless misunderstandings about language in western thought. The role of metaphor in the part of Rhetoric dealing with “style” reveals insights into the nature of language that would seem to complicate the quest for determinate scientific reference needed in the scientific project. For instance, Aristotle says “proper (to; kuvrion) and appropriate (to; oijkei:on) words and metaphors are alone to be employed in the style of prose; this is shown by the fact that no one employs anything but these. For all use metaphors in conversation as well as proper and appropriate words” (Rhetoric 3.2, Loeb. Despite Aristotle’s awareness of the ubiquity of “metaphor” in language use, when he analyzes ways to criticize arguments (arguably a rhetorical gesture) in his treatment of reasoning (Topica) Aristotle states that an argument can be refuted if its description uses a metaphor because “metaphorical expressions are always obscure” (VI, ii, Loeb. The radical disjunction between the clarity of setting things before the eyes using metaphor and metaphor creating obscurity epitomizes the problem with Aristotle’s views on language. It is ironic that Aristotle seems to have coined this metaphorical use of “metaphor”: a linguistic phenomenon he seems to think of as naturally pleasant and perhaps a kind of indulgence “metaphor especially has clarity and sweetness” (On Rhetoric 3.2.8). He also observes, for example that “to learn is naturally pleasant to all people, and words signify something, so whatever words creates knowledge in us are the pleasantest. . . . metaphor most brings about learning” (On Rhetoric 3.10.2). Given the status of learning in ancient Greek culture this image of metaphor hardly seems to square with its obscuring disposition in Topica. (however artfully manufactured). 129 On the one hand, Aristotle’s Rhetoric embraces multiple human social interests pertaining to speech. Its list-like, genus-species, definition-type analysis (which makes it relatively impractical as a how-to handbook and more like a language treatise) outlines various means of persuasion which focus on (affective) appeals that make people feel confidence or trust towards an utterance. These are appeals to ethos (‘habit,’ ‘character’), pathos (‘experience,’ ‘suffering’), logos (‘accounts,’ ‘arguments’), and to topoi or lines of thought which are persuasive because familiar and recognizable – recognition and familiarity that stir people’s sentiments of rationality. The topoi (lines of thought or figures of argument) are ready- made patterns of speech or bits of discourse that speakers reach for, exploiting recognizable beliefs: notions of common sense, authority, what people think is good, what they think is bad, etc. Aristotle stresses the strictly social importance of the audience/speaker relation – the audience’s view of the speaker depends on the ethos that the speaker presents, how the audience is figured in relation to the speaker (i.e., is the speaker a young man addressing an audience of his elders?) and how the audience is emotionally engaged or (affectively) primed to be 140 Aristotle discusses the way certain tokens can elicit pity for example: “For this reason signs and actions [contribute130 to pity]; for example, the clothes of those who have suffered and any other such things, and words and nay other such things of those in suffering; for example of those on their death bed; for all such things, through their appearing near, make pity greater. And most pitiable is for good people to be in such extremities, since one who is unworthy [is suffering] and the suffering is evident before our eyes” (On Rhetoric 2.9, square brackets are those of the translator Kennedy). Ackrill’s “In Defence of Platonic Division” comments that “Platonic division covers a variety of types of analysis,131 and gets applied to a variety of types of term. No less than the word ‘analysis’ itself, which has been a modern slogan in philosophy, ‘division’ is dangerously imprecise but philosophically suggestive” (Essays 107). Deslaurier argues that though Aristotle is critical of Plato’s method of division, Aristotle’s own approach is similar (203). She notes that Aristotle acknowledges the “limited usefulness of division” in his philosophic method if it is practised according to certain rules and constraints which she outlines (204, 213). Aristotle’s constraint against arbitrariness does not prevent his use of it, as I shortly suggest about his analysis of rhetoric. Deslaurier too comments that Aristotle clearly breaks his own rule about excluding privative terms unless it is further qualified (214). Regardless of the “limited” usefulness Aristotle “says” division has, its ubiquity in his text belies what is said. persuaded (Rhetoric 2.1).130 On the other hand, Aristotle’s (scientific) method of analysis seems to be a kind of literal-minded uptake of Plato’s collection and division called diaeresis. The Platonic text illustrates the way that this subjective analysis (unloosening, distinguishing, and identifying the various parts) depends on what angle or vantage you approach the whole entity from; and the results are determined by a tension between the degree of an individual’s concentration and what terms or words happen to come to mind in the process. Plato takes pains to show how131 easily mistakes are made in diaresis. While some ancient philosophy scholars (like Deslaurier) consider Aristotle’s approach to division as being similar to that of Plato, arguably Aristotle’s actual use of this method differs radically. Furthermore, because it is such a language-dependent process, Aristotle’s literal-minded application of the method produces determinate classifications which are at odds with the affective and promiscuous ways that language works in Rhetoric. For Plato determinate classification using diaeresis would require the dialectician’s omniscient perspective (he whose theorizing beholds a collective unity and plurality); and classification is thus never determinately achieved in the dialogues (recall the Socratic stance of ignorance). In the Platonic text diaeresis is a tool of analysis: what is important is the process of analysis – the exploration – more than the end product or result achieved. As Cherniss argues, Plato does not intend diaeresis “to produce an ontological hierarchy” but rather it is a heuristic method “to facilitate the search for a definite idea, the distinction of that idea from other ideas, and its implications” (54). Cherniss says that Plato did not imagine diaeresis 141 Round brackets are mine, square brackets are the translator Kennedy’s. Note also that the term usually translated as132 proof but which Kennedy leaves untranslated in his text, pisteis means that which give confidence. It is reminiscent of James’ sentiment of rationality that is a kind of confidence that things fit together. Observe that Aristotle begins with a privative division (non-artistic/artistic) – one of his supposed constraints that133 Deslaurier identifies (214). to be a description of the “construction” of the idea, its derivation, or its constituent elements. [In other words, it is not an absolute genus species classification.] He [Plato] describes it as a useful means of narrowing the field of search; but the formal method alone may lead one to any number of definitions of the same thing unless one has the additional power of recognizing the essential nature that is being sought. [And knowledge of the essence of things is never achieved in the Platonic text.] In short, diaeresis appears to be only an aid to reminiscence of the ideas. (54-55) The famous lack of resolution in Platonic dialogues and the “contradictions,” between different accounts of “ideas” for instance, supports the view of collection and division as a linguistic process or exploration – a means of investigating apparent differences (using lovgoi) between one thing and another that are never finally determined outside of unhelpfully-obscure abstractions like “the good.” Cherniss argues that Aristotle erroneously “took diaeresis to be meant as a universal ontological scheme in which the more specific ideas are derived from the more general” ones (58). Such an ontological scheme depends on determinate reference – on being able to absolutely determine and identify each category and thing with a unique label of some sort. Determinate reference is necessary for Aristotle’s rational-scientific thought and his conception of “knowledge.” Thus Aristotle’s analysis of lovgwn tevcnh or language arts (which is focussed on the product of diaeresis) is classificatory, division into determinate parts: of the pisteis (that which convinces or gives faith), some are atechnic [‘nonartistic’], some entechnic [“embodied in art, artistic”] . . . Of the pisteis provided through speech there are three species: for some are in the character [ethos] of the speaker, and some in disposing the listener in some way (i.e., pathos), and some in argument [logos] (1.2).132 Hence Aristotle proceeds through this linguistically mediated analytic method of dividing up the matter into its determinate component parts: his classificatory ends embed his scientific method in his style. His text presents a style/argument/thinking conjunction which I don’t think comes133 142 For example: “The species [eidç] of rhetoric are three in number; for such is the number [of classes] to which the134 hearers of speeches belong. A speech [situation] consists of three things: a speaker and as subject matter on which he speaks and someone addressed” (1.3, square brackets are translator Kennedy’s). Admittedly Aristotle is also following Plato’s apparent admonition to begin with the most general divisions, see Ackrill (Defense 103). See Mitscherling’s “Plato’s Misquotation of the Poets. ”135 Observe the way that recognition and familiarity are conjoined in what Naas calls “turning the mind.” 136 apart. The selection of terms will always be a kind of argument – here his scientific thinking (both a method and a style) results in determinate classification. Observe however the way that the above passage evinces (what Aristotle readers will recognize as) his fondness for both binary pairs (i.e., the privative artistic/non-artistic) and groupings of three (i.e., ethos, pathos, logos) – the latter having the benefit of being list-like (i.e., lots) but still simple and clear enough to grasp (set before the eyes). These signal134 features of the Aristotelian text appear to be determinate divisions, even though Rhetoric’s classifications are always shifting (as Halliday’s observations about linguistic categories would anticipate). Neither should we have confidence in their supposedly “analytic” quality: notice for instance, the extremely suspect category of non-artistic which includes the use of laws and witness, here treated as pre-existing inartistic givens. Testimony includes both actual witnesses and literary texts, i.e., Homer. Both literary texts and witnesses surely involve as much artistic rhetorical exploitation as any other component of rhetoric! Citing Homer is surely not a pre- existing given (as Plato’s notorious citations should make clear). It is not clear whether this135 first dichotomous division (artistic/non-artistic) was “analytic” or whether a third or fourth category might have been more apt. It is possible that the division artistic/non-artistic itself is simply a rhetorical move – an enthememe from opposites, a way to foreground the artfulness of rhetoric – rather than an accurate ontological distinction between general categories of types of pisteis. Aristotle may have been persuaded himself. As Naas claims: to persuade is to lead someone to obedience; to obey is to let oneself be persuaded” and “‘changing the mind’ of another involved being changed oneself’” (Naas 22). Certainly, a binary opposition like136 artistic/inartistic can be used to emphasise what something is not: a line indicating what is being excluded. It is a way laying bare, or setting before the eyes; and it is the first general topos of Rhetoric, enthymemes from opposites – a technique otherwise often used by scholars, like 143 Logos is a difficult polyvalent term in the Ancient Greek that can mean a wide range of things (and not just word,137 account, myth etc.). While it may be that “argument” is a better translation of logos in Aristotle’s text, Sallis argues that “account” is more apt in the Platonic text, in more recent writing Sallis uses the term “discourse.” Indeed, “logical arguments” (if they exist at all) are difficult to find in Plato’s dialogues. P. Christopher Smith takes for granted that topoi are affective appeals rather than deductions (Apodeiknusthai). Aristotle’s view in De Interpretatione 1, 16a considered the root of his semantic theory “words spoken are symbols138 or signs of affections or impression of the soul” prompts Modrak to consider the relations between paqhvmata (pathema: the affections or impressions) and images, the imagination and logoi. Words are signs of affections/impressions – a mental state. She says that “Aristotle identifies the mental state with a logos” in the sense of logos being a meaning or its specification in a definition (84). Although not forwarding arguments about contemporary conceptions of cognition and affect nor commenting on the distinction that Aristotle is making here in Rhetoric, her comments nevertheless indicate the relations between pathos and logos, feeling and thinking. “‘Pathos’ is the term Aristotle standardly uses for the affective and cognitive states of ensouled individuals. Thinking, perceiving, loving, hating, and remembering are all included under pavqh (pathe)”(221). See also Johnson’s discussion equating theories of reader-affective response with rhetorical theory’s pathos in “Reader-Response and the Pathos Principle.” Aristotle, when they summarize a contrary view in order to set their own in relief. Similarly, what is the probability that ethos, pathos and logos are the only types or species of pisteis (that which gives confidence)? Their prosodic equality may make them appear to be a natural tripartite division (two syllable, rhyming second-declension, masculine-noun- “os” endings: ethos, pathos, logos); but lest we be too easily lulled by them, acquiescing to a “natural” affiliation their sound seems to bequeath them, Part 3’s discussions on prosody and affect might prompt consideration of the affective nature of their prosodic effects. Despite their definitive, clear articulation, these categories are neither distinct, nor of comparable scope – they might be accused of a certain arbitrariness (one of Aristotle’s constraints). Arguably as subjects of discussion ethos plays a much narrower role than either pathos or logos in Aristotle’s Rhetoric, which is more focussed on “argument.” Furthermore, all three types are completely intertwined. Ethos is a consideration for both speaker and audience, as well as being imbricated in pathos. Equally, it is difficult to distinguish pathos (experience/suffering/emotion) from many of the topoi or lines of argument. Topoi – which depend on the affectively charged familiarity (e.g., induction, past precedent, consequences) – might be a type of pathos or a type of logos. Everything is a type of logos on some level. And topoi might be considered a fourth137 species of pisteis. If lines of argument are a preestablished inventory of patterns of discourse (familiar from experience) that we reach for, and if rationality is a sentiment, what we feel follows, then logos may be a type of pathos. Ethos overlaps with the lines of argument and138 the appeals to emotion; ethos and pathos could be part of logos, or logos a type of pathos, or 144 topos might be a species in its own right. Regardless of how useful these distinctions might be, the three “species” are not only not equal (qualitatively or quantitatively) but their uses are completely intertwined, interfused and overlapping – attributes that many readers of Rhetoric recognize about most of its categories. Despite his rhetorical advice (whose affective appeals seem to undermine code conceptions and determinate reference), Aristotle treats diaeresis as though its end-product were its purpose and as though classifications could be definitively determined. His own analysis acts as though his linguistic classifications were definitive – his style performing determinate reference, apparently contradicting affective operations involved in how language works in Rhetoric. So while Aristotle seems to understand the artfulness of rhetoric, he himself treats language as though he can use it to make clear and determinate distinctions. And his twos and threes certainly provide clarity – recall Austin’s denunciation of “old habits of Gleichschaltung, the deeply ingrained worship of tidy-looking dichotomies” and Glendinning’s explanation that Gleichschaltung means bringing “something into line, to make something conform to a certain standard by force” (311-12). Austin suggests that such reductive division merely represents a “tendency to ‘simplify or pretend to be sure of such simplicity where there is none’” (Glendinning 312). Aristotle’s rhetorical categories (which still have currency) set things before the eyes in a style that looks like analysis, that looks like determinate classification. What could be more convincing than these neat divisions into distinct, discrete categories: a Linnaeus- genus-species-type classification, which looks just like science? While I think the view of language implicated in this stylistic is problematic, it would probably be wrong to assume that Aristotle is trying to deceive us. Rather I suspect that he is trying to get on with the scientific project. Determinate reference, so important for epistemological endeavours, is exactly what Aristotle wishes to establish for scientific inquiry. Aristotle may himself have been persuaded: but the contradictions and the category slippages in Rhetoric undermine the ontological certainty or “objectivity” of his classification. His confidence in truth and his sincerity do not preclude the persuasive effects of his reductive 145 While it seems clear that the “stylistic” features commented on above are in Aristotle’s text, it must be139 acknowledged here that his own Greek writing is considered notoriously “difficult” and “obscure.” This problem may have to do with the cursory nature of the original texts which are now often presumed to have been lecture notes (composed by himself or his students). This together with their tumultuous transmission and editorial history means that his manuscripts have doubtless been deeply edited by scribes and colleagues reconstructing damaged and/or never completed documents (McAdon Signs 229-231). clarity and scientific-sounding style – which is at once thinking/argument and style. The139 profoundly linguistic nature of diaeresis makes Aristotle’s analytic (scientific) method, his style of writing and his style of thinking completely interfused. Furthermore, Plato’s collection, division and definition are not only what Aristotle performs stylistically and argumentatively, they are also what he says, they appear as topoi in his rhetorical advice – the topoi of definition, logical division, and parts to whole (On Rhetoric 2.23, 8-13). Though not the image of language portrayed in Rhetoric, his scientific project of diaresis is exemplified (or performed) by his style – testifying to the union of style and argument. The relation between style and method of argument in Aristotle (which arises because of the linguistic mediation of thought) provides a glimpse into a way that style can signify – and since style is a fundamental part of genre, the role that genre plays in linguistic meaning. Kennedy’s observation about the first few subsections of Book One of On Rhetoric might provide additional insight into Aristotle’s astonishing rhetorical success: The first chapter is one of the earliest examples of an introduction to the study of a discipline (the beginning of the Topics is another) and is thus an antecedent of the Greek prolegomenon or Latin accessus commonly found at the beginning of technical works in later antiquity and the Middle Ages (28). Curiously enough, Kennedy’s remarks about the introduction to Rhetoric and Topics ring true with the impression created by the opening sections of Metaphysics (I, iii-v) where Aristotle summarizes previous thinkers in a way that sounds like he is providing a “review of previous research”: Anaximenes and Diogenes make air prior to water, and the most primary of the simple bodies, while Hippasus of Metapontium and Heraclitus of Ephesus say this of fire, and Empedocles says it of the four elements (adding a fourth-earth-to those which have been named); for these, he says, always remain and do not come to be, except that they come to be more or fewer, being aggregated into one and segregated out of one. 146 See Leff’s Ariadne, and Contemporary, McAdon’s Purpose, Welch and Kennedy’s New History for a sense of the140 different up-takes of Aristotle in American Rhetorical Theory. Aristotle’s style has been taken at its word. Anaxagoras of Clazomenae, who, though older than Empedocles, was later in his philosophical activity, says the principles are infinite in number; for he says almost all the things that are made of parts like themselves, in the manner of water or fire, are generated and destroyed in this way, only by aggregation and segregation, and are not in any other sense generated or destroyed, but remain eternally. From these facts one might think that the only cause is the so-called material cause; but as men thus advanced, the very facts opened the way for them and joined in forcing them to investigate the subject. However true it may be that all generation and destruction proceed from some one or (for that matter) from more elements, why does this happen and what is the cause? For at least the substratum itself does not make itself change; e.g. neither the wood nor the bronze causes the change of either of them, nor does the wood manufacture a bed and the bronze a statue, but something else is the cause of the change. . . .Of those who said the universe was one, then none succeeded in discovering a cause of this sort, except perhaps Parmenides, and he only inasmuch as he supposes that there is not only one but also in some sense two causes. But for those who make more elements it is more possible to state the second cause, e.g. for those who make hot and cold, or fire and earth, the elements; for they treat fire as having a nature which fits it to move things, and water and earth and such things they treat in the contrary way (W.D. Ross I, iii). Aristotle seems simply to be summarizing the previous research and discussing it. It sounds like the academic-writing genre – what scholars do. But recall that direct reports show reported speech’s true colours; recall their strategic artfulness and contradictory nature. Aristotle’s summaries are speech reports and thus able to perform all kinds of interesting feats: here they claim authority for Aristotle’s work, collegially showing how his own ideas contribute to the work of others – at the same time he is constructing a history for his own ideas, showing them as the teleological end for what has come before.140 Without wishing to diminish Aristotle’s towering and enduring contributions to western philosophy – an historian (in the modern sense) he is not, and his summaries show little interest in fidelity. His style performs a scholarly ethos: laying the groundwork for the style of academic writing, an analytical scholarly style conflating disciplinary ethos, method and argument. But Aristotle’s summaries do not qualify as scholarly research by contemporary standards. 147 See Shrimpton’s work on Herodotus and Thucydides, as well as the collection Lies and Fiction in the Ancient141 World examining history, philosophy and literature. Historians and philosophers of ancient Greece have long recognized that Aristotle’s accounts of other scholars are unreliable. In his defence, his approach to other texts and authorities may not have been untypical; the intellectual climate seems to have had a different, perhaps more metaphorical, understanding of truth – and one which might warrant polemical treatments of other scholars in the interests of getting at the truth. The extremely derivative nature of ancient historical texts is well known: various types of details and stories are picked up from texts (like Aristotle’s) by contemporary “historians” (like second-century C.E. Diogenes Laertius, whose “history” of philosophy is often cited as a work of historiography). Scholars of Herodotus and Thucydides, like Shrimpton and Moles, comment on the derivative legend-building in these texts presented as iJstoriva (inquiry). Philosopher Charles Kahn expresses consternation when modern scholars who claim to be writing history “follow in their footsteps” (3). In recent decades, scholars of classical Greek literature, philosophy and history have investigated the conceptions of truth and fiction that prompt this legend-building. They conclude that the idea of truth might have meant something different in the ancient world than what it does now. It is141 possible that the ancient world already understood language’s metaphoricity (its referential promiscuity), its ability to artfully re-present some selected feature of a thing, or of an event (whether actual or fabricated); such artful presentations might have portrayed what was felt to be an enduring “truth.” Aristotle’s frequent summaries of other philosophers are now taken by specialists with a “large measure of salt,” as Kahn puts it (xvii). A.E. Taylor claims that when it comes to Plato “Aristotle’s references are all polemical, and Aristotle is a controversialist who is not unduly anxious to be ‘sympathetic’”; he adds that “we must carefully avoid the nineteenth-century mistake of treating the statements described by Aristotle under the name of the ‘doctrine’ (pragmateiva) of Plato” (emphasis added 503-04). Schiappa comments on Aristotle’s penchant for re-writing history, noting Havelock’s concern that “Aristotle sometimes invented anticipations of his own philosophy by stretching what was known about his predecessors” 148 Bett’s “On the Pre-history of Pyrrhonism” finally rules out views of so-called opponents that Aristotle claims to be142 addressing (in a later section of Metaphysics than that cited above). He cites Aristotle’s declaration that “there are some people who claim that the same thing can be and not be” and observes that “who exactly these opponents are is not easy to say” but Betts does not bother to question whether Aristotle might have invented the opponents for polemical purposes as he is only interested in establishing that there is no warrant to interpret Aristotle’s remark as indicative of the views of Pyrrhonists (145). (Beginnings 39). Kahn similarly argues that Aristotle’s surveys of thought are “explicitly motivated by the desire to see how far they have anticipated his own” and that “the essentially non-historical character of Aristotle’s account is demonstrated both by what it omits and what it includes” (79, 82). Just as citations look deceptively like what people actually said, and tidy- looking dichotomies look clear and understandable, summarizing previous work looks a lot like scholarship, even if you make it up. Nevertheless Aristotle’s use of language and the treatise142 genre performs his scholarly ethos. He brings the voices of other scholars to bolster his own contributions in the pursuit of wisdom and what he thinks is true. His style performs his method of analysis (like Lesley, demonstrating/performing, not referring). In the linguistically- conscious climate of ancient Greece, perhaps language was an old problem – something that had to be dealt with in order to establish the rules for rational thought, in order to make way for the real work of science, politics and ethics – an attitude not unlike what starts the Enlightenment “plain style.” 4.4 good writing and difficult texts: clarity and opacity ~ Let us return here briefly to Nietzsche’s metaphoricity and his tropic mode of writing and gauge how it stacks up against Aristotle’s advice on the use of metaphor as a means of clarity, to set things before the eyes, to vividly re-present. As a contributor to the public discussion of the bad-writing awards, James Miller, discusses the twentieth-century’s own exemplar of Aristotle’s clarion call, Orwell. Citing Orwell’s famous window-pane image – the exemplary transparency metaphor – Miller writes: ‘Good prose is like a window pane,’ [Orwell] once declared. True to this motto, he never stopped looking for the right frame for the right kinds of concrete images and turns of phrase, artfully enough rendered to conjure an illusion of perfect transparency (emphasis added Lingua Franca). Miller aptly characterizes Aristotelan clarity as artfully striving for the illusion of perfect transparency. Aristotle says, “if one composes well, there will be an unfamiliar quality and it 149 escapes notice and will be clear” (On Rhetoric 3.2.6). Clarity (and transparency) are artful and anything but common speech. This is not simplicity, not clarity produced by finding the correct determinate reference: but rather the best available means of conjuring a likeness in language, the rhetorical mechanisms designed obviate the need to stop to think about the language and what it means. However, such a motive does not prompt Nietzsche’s stylistic-metaphorical turn, his tropic mode of attention to language’s metaphoricity. Arguably Nietzsche’s use of the term “metaphor” to characterize the linguistic semiotic is designed to attract attention to itself (clouding the window glass), designed to make readers stop to think and grapple with something new. To point out that the perceptual image is an Aristotelian “metaphor” re-vitalizes the ancestral roots of the term metaphor (a carrying-over): the perceptual image (what we see, or hear, or feel) is not the thing itself but an analogy, our perception or mental reproduction of the thing. Nietzsche reminds us and keeps reminding us – until the idea becomes familiar – that the linguistic sign is a carrying-over from one sphere to another. By so radically displacing the Aristotelian metaphor in favour of its ancient roots he aims not for transparency, but for opacity. The metaphor “metaphor” is a reminder of linguistic matters that we don’t think about when language use achieves transparency. Nietzsche’s quasi- mythical and allegorical tropic mode of writing, as Kofman observes “attaches novel and unheard-of metaphors to habitual ones” (Metaphor 60) and sometimes these metaphors are opaque, calling attention to themselves, arresting and demanding thought. Rather than striving for setting-before-the-eyes clarity and transparency (a metaphor designed to make a claim without being noticed, a stylistic so transparent that it hardly warrants counter-words, that doesn’t require thinking at all) Nietzsche metaphorical use of the term ‘metaphor’ reminds us precisely about what we don’t think about – what we have forgotten (as he says metaphorically). Transparency implies instantaneous understanding. Adorno complains about this irony: “only what they do not need first to understand, they consider understandable” (101). Nietzsche’s novel and unheard of metaphors are often not designed for instant agreement, acceptance or recognition, but rather to provoke thought: delaying our rational sentiments, 150 Both Nozick and Cavell are supposedly “analytic philosophers,” though their writings challenge the category in143 numerous ways. See Cavell’s “Must we Mean What we Say”, “The Availability of Wittgenstein’s Later Philosophy” and his collection Philosophical Passages: Wittgenstein, Emerson, Austin, Derrida for their interesting consonance with this dissertation. making us work for them. Niezsche’s metaphor of language’s metaphoricity makes us think about the otherness of the semiotic realm: the way in which the semiotic sphere is not essentially tied to things outside of individual psyches. He has put a fly on Orwell’s window because he wants us to think about the windowpane and the way it colours, shapes, and obscures what we see; he wants us to think about language, its nature and meditative role. Recall Wittgenstein’s reader-provocations interests. Culler cites a parallel provocation interest from the opening page Robert Nozick’s Philosophical Explanations: I, too, seek an unreadable book: urgent thoughts to grapple with in agitation and excitement, revelations to be transformed or to transform, a book incapable of being read straight through, a book even to bring reading to a stop. I have not found that book, or attempted it. Still, I wrote and thought in awareness of it, in the hope that this book would bask in its light. (44-45) Culler observes that this is “prose that basks in the light of the hope of unreadability” (45). Stanley Cavell also writes with designs on engaging his reader in active thought rather than the easy-to-follow familiar tracks; his is language use calling attention to itself, as Culler says of it: This is writing that, first and foremost, calls attention to itself as writing. The sentences do that, with a coyness one can certainly find irritating . . . Why would philosophy call attention to itself as writing? Philosophy is writing not only because that is the form in which we generally encounter it but, most important, because the fundamental philosophical question for Cavell, is how we understand each other and ourselves. (emphasis added 50)143 Wittgenstein’s interest in arresting his reader is not singular. As we have observed, James implies that not all rational sentiments are triggered in the same way: some are attracted to certainty and others enticed by enigmas. And there are a variety of factors that influence reader/writer relations and textual interests. Some writers are not seeking Orwellian stylistic transparency: writers who want readers to grapple haltingly, or writers who might have auto- affective interests in their own writing, their own creative engagement with writing’s plasticity that is bound to infiltrate their text’s style. Writers may be motivated by exigencies of efficiency 151 and getting the job done, as though there were a plain text to be read, prompted by understandings of having something to say, some kind of pre-formed text. Or writers may imagine the process of writing as the act of understanding itself: just as the Platonic text imagines logoi as the means of exploration, the process of investigation itself. Pitting Orwell’s “hopelessly naive” view of “a pre-theoretical world of facts just waiting to be described” (that goes hand in hand with his transparency convictions about style) against Adorno’s unapologetic (but mostly warranted and informed) stylistically inscribed elitism, Miller suggests that different audiences, with different interests and world views prompt interest in these differing styles. Adorno runs the risk of having no audience, Orwell the risk of an anti- intellectual one. This stylistic opposition recalls different subjectivities’ sentiments of rationality, different reader profiles. According to Riffaterre it is the unpredictable or surprising elements of a text that produce stylistic effects. He observes that while “predictability may result in superficial reading, unpredictability will compel attention” (Criteria 158). Fludernik focusses on they way that Riffaterre’s studies “have centred on semantic and syntactic anomalies [i.e., novel and unheard of metaphors] or peculiarities of the text which invite interpretation” (338). As we shall shortly see, inviting interpretation is a kind of involvement strategy: one that both the Derridian and Platonic texts perform, and one that indexes their readers’ and writers’ subjectivities. But regardless of reading disposition, different situations create different expectations. In some situations gleaning information is exigent (like reading cleaning-product instructions on the back of a can, where one does not want to have to contemplate meanings); in other situations, one expects to be led through a series of arguments (like in reading the reasons for a legal decision). Different situations or contexts produce different genres. In Riffaterre’s analysis the pattern of the familiar – established by the context – is disrupted by an unexpected element. This is a figure-ground phenomenon: there must be a familiar, expected pattern for the unexpected element to be recognizable as unexpected. An unexpected stylistic gesture requires a stylistic context, a genre: bringing us back to the role that context and genre play in meaning-making where the conjoined style and argument play a 152 Surprisingly, de Rijk argues that Aristotle does just that: intentionally exploiting ambiguity (60-66).144 leading role. 4.5 genre expectations ~ Recall Derrida and Voloshinov’s extreme contextualism, and Bennington’s announcement that “there are always contexts”; consider too the importance of context in Sperber and Wilson’s account of communication, and in the discourse of reported speech, what happens when words are chopped out of one context and plunked into another. However provisional its stability may be, it is context that stabilizes the sign’s meanings. Arguing that “literal” means different things in different contexts, Fish comments on the expectations that are elicited by different situations or circumstances – he says “a normal context is just the special context you happen to be in”– like other words “‘normal’ is context specific” (Circumstances 640). Stylistically the pattern of the familiar is disrupted by an unexpected element. Irony and ambiguity, for example, “are functions of the expectations with which we approach” a text: functions arguably meant to be avoided in the Hume-type-plain style (631). On the other hand “if we expect a text to be ambiguous, we will, in the act of reading it, imagine situations in which it means first one thing and then another” (631). The measure of the way expectations are established by genre can be gauged by the failure to anticipate that Aristotle’s treatise genre would exploit an ambiguity of meaning (i.e., that his scientific analysis would “intentionally” foster two different readings of the same text). Thus some readers approach the Platonic text144 as a serio-comic genre anticipating irony: while others read the dialogues as though they were not dialogues at all, but a treatise with remarkable literal-mindedness given the image-greedy protagonist and myriad of other well-known rhetorical exploits. If unrecognized ambiguity problematizes reading, unrecognized irony so much the more so. It is a question of gauging whether or not a genre might exploit irony, and to what extent. Context-determined-expectation is a principle of how genre works. Recall Bitzer’s “rhetorical situation”: he explores the way that context, situation or circumstance, with all their attendant psycho-social dynamics, prompt a certain type of utterance. Giltrow’s genre- 153 theorizing imagines genre as a blend of Bitzer’s rhetorical situation and form: but the priority lands so much on the psycho-social dynamics of situation that Giltrow will say “genre is context, genre is rhetorical situation.” Indeed the concept of genre might be a different way of thinking about context: since the context-dependency of the promiscuous linguistic sign is what necessitates genre’s contextualizing power. From a Bakhtinian perspective words are always uttered in “speech genres” and it is genre that makes them meaningful. Bakhtin believes that genre contextualizes and/or organizes meaning: If speech genres did not exist and we had not mastered them, if we had to originate them during the speech process and construct each utterance at will for the first time, speech communication would be almost impossible. (Speech 79) This novel way of thinking about context in terms of genre entails a considerable re-mapping of the concept of genre. A “speech genre” in Bakhtin is an utterance in which thematic content, composition structure, and style are inextricably linked to the whole utterance – interfused features that operate a lot like the linguistic sign itself (Speech 60). These content-structure-style factors are determined by both the nature and sphere of the communication: the situation and everything it encompasses like the participants, the reasons for the utterance, the available forms and so on. Although “each separate utterance is individual” (unique as Nietzsche would say), it is because “each sphere in which language is used develops its own relatively stable types of these utterances” that we can communicate at all (60). Without speech genres we would not understand one another. Speech genres are mini-contexts that we can reach for in the appropriate situation, just as we reach for bits of discourse and figures of argument; and like the sign they are relatively (un)stable types of utterances. Bakhtin is not thinking genre at a high level of generalization that Aristotlean categories might anticipate: “speech genres” are not drama, poetry, prose, and not book, public discourse, conversation, whether or not any of these types might have its own generic-type features. Giltrow thinks that we lose sight of genre at high levels of generality. Expanding and re-figuring Bakhtin, Giltrow locates genre at a much lower level of generality than has been traditionally done, like “on-line medical advice, the live TV exchange, the guest lecture, or the corporate 154 blog” (Difference). She notes Bakhtin’s attention to “the extreme ‘heterogeneity’ of speech genres – from ‘the single-word rejoinder’ to ‘the multivolume novel’” (Heteroglossia 11). Bakhtin says: The wealth and diversity of speech genres are boundless because the various possibilities of human activity are inexhaustible, and because each sphere of activity contains an entire repertoire of speech genres that differentiate and grow as the particular sphere develops and becomes more complex. Special emphasis should be placed on the extreme heterogeneity of speech genres (oral and written). (60) Giltrow’s uptake and remodelling of Bakhtin might, for example, imagine the genre of a bus- stop conversation between friends and perhaps a different bus-stop genre for strangers – because the types of things eligible for discussion would be different. Giltrow’s genres are “dynamic sites of social activity – rather than lifeless forms or places for propriety” (Heteroglossia 2). They are particular to participants’ sociality and the context in which they take place; motivated by the situation – they are variable and adaptable. She says we can think of genre in terms of multiple motivations that spring from social experience in specific situations: we need to keep in mind genre’s participation in multiple scenes of activity (Gentlemen 15,21). Genre is re- useable and can move around from site to site, between speakers and hearers in different social relations; it can become easily re-accented, modified and changed. Bakhtin distinguishes primary (simple) from secondary (complex) genres of which “the vast majority of literary genres are secondary, complex genres composed of various transformed primary genres” (Speech 98). Hence literary genres inherit the varied properties of primary (speech) genres in a completely new context and mixture. Like the linguistic sign genres change, they are relatively stable (or unstable), and they can be re-used for new purposes and amalgamated with other types. Bakhtin says that “each speech genre in each area of speech communication has its own typical conceptions of the addressee, and this defines it as a genre” (95). Think of the different social relations in the genre of the family-dinner-table conversation, the genre of law-firm discourse, the genre of the text message. The typical conception of the addressee inscribed in the genre – the typical relationship between speaker and addressee and 155 See Bazerman, Berkenkotter and Huckin, Coe, et al., Freedman and Medway, Jamieson, Kohrs, et al., and Bitzer for145 how these views of rhetorical genre theory have taken shape since the 1970s. Also see Derrida’s “Law of Genre” for its congruence with this thinking. attitudes towards one another, typical conceptions implicit in the genre itself – this is what Bakhtin imagines as distinguishing one speech genre from another. Imagine, for example, how the addressee in a tele-marketing phone call is conceived! Giltrow gathers these speaker/hearer images into different speaking situations or circumstances with different available forms: attuned to social relationships like the office memo between colleagues in different hierarchical relations, the once popular visiting card between friends and acquaintances of certain social classes, and academia’s ancient-philosophy article with its collegial familiarity: where everyone seems to know everyone else, know everything they have each written, and it often sounds as though they are fresh from an argument in the college corridors. Such a conception of genre is no longer exactly new, but it is radically different than standard traditional categories. Giltrow’s Baktinian conception of genre – a tension between145 context and form – offers deeper insight into how Halliday’s biochemical grammar and bits of discourse work. Genres have a lot in common with linguistic signs: they contextualize, that is they provide a circumstance or situation that make a certain style appropriate, that anticipate certain reader/writer relations and certain topics or subjects. Like the socially gestated linguistic sign Bakhtin imagines speech genres as extremely “flexible, plastic and free,” always being modified, deliberately mixed and re-accentuated to reflect individual speaker and the specific context of the utterance (79). Recall that the sign’s identifying citationality entails re-use – signs are reusable or recyclable by definition, so too genres, reused and re-accented. Giltrow sees genre as an engine of variation and heteroglossia; like the sign it is a site of differentiation rather than the sameness that the traditional view suggests (Heteroglossia). “Genre works as a multiplier of difference,” she says, but it is “situation impinging on form to perform local functions indigenous to a social scene” (Difference 9). The unendingly problematic concept of form is inescapably part of the picture and interestingly it includes the equally slippery concept of style. In the context of genre form might include the physical 156 manifestation of cartes de visite, office memo, book, traffic signs, academic article, cash register receipt, thank-you card, though some have no physical manifestation apart from oral speech like a bus stop conversation. These physical manifestations formally constrain and guide genre selection and use. But all forms include language and thus register, style, size/quantity, level of detail/quality, shape/social dynamics, and content. And here again inescapably, form includes content. Pure form seems inevitably caught in its own impossibility by the exigency to include identifying content in order to be recognizable at all. Just as metaphor selection is both style and argument, and just as form comes attached to content, Bakhtin states that “where there is style there is genre” (66). Style is tangled up with content and argument, and it permeates genre: orienting the gaze, shaping the types of utterance, determining subject matter, expressing attitudes towards readers. Genre is the place where sociality disciplines both what is said and how it is said. Bakhtin explores what the ancients called the spoudogevloion – “serious-smiling” genres – that Bakhtin calls serio-comic; not the least of these is Socratic dialogue. It is worth keeping in mind that the Socratic dialogue is a site of serious play, and as we turn to our discussion of Derrida and Plato it is worth keeping in mind the way that the voice of Socrates conditions the Republic which opens: “I went down to the Piraeus yesterday with Glaucon, the son of Ariston. I intended to say a prayer to the goddess . . .” (Grube 327a). 157 PART 5 5.1 “Platonism’s” voice-overs ~ Some of Derrida’s readers might be surprised to hear that back at the very beginning of his publishing career, in his 1962 Introduction to Husserl’s Origin of Geometry, he was already ruminating about “a Plato muted by Platonism” (144). This Plato-“Platonism”differentiation – a distinction between the text of the dialogues and received views of the western tradition of philosophy – attests to Derrida’s early attention to the distortions of the Platonic text. Interestingly, Derrida’s comparison of Plato and Husserl highlights the literal-mindedness (in the interpretation of myths, metaphors, and even Plato’s pedagogy), which is one of the signal problems with “Platonism”: “there is in Husserl what perhaps there was not even in Plato (except in the literalness of his myths and pedagogy), namely, a ‘Platonism’ of the eidos or the Idea” (144). Derrida thus begins the oft-used practice of putting “Platonism” in quotation marks in order to highlight the interpretation problems that have haunted the Platonic text – problems not unlike those the Derridian text itself presents. This interpretation difficulty that Plato and Derrida share is central to Part 5 where we explore the futility of voices, deaf to these texts’ stylistics, trying to say “exactly what Plato or Derrida meant.” Although Derrida suggests in “Plato’s Pharmacy” that Plato is himself at least in part to blame for the risks he took in crafting his signature stylistic, so too is Derrida at least in part to blame for the misreadings of his own high-jinx-stylistic antics. Part 5’s stylistic analysis of the Derridian and Platonic texts reveals the relationship of these writers’ often challenging styles to their sceptical views of language, how their texts foreground language’s vexing undecidability, and why they are worth the effort. The pinnacle of Plato-interpretation problems in the English-speaking world can be gauged by F.M. Cornford’s statement (circa 1935) that the “underlying assumption” of Platonic “doctrine” is that “every common name must have a fixed meaning, which we think of when we hear the name spoken” (9). This curious blend of literal-mindedness and metaphysical fantasy in the “Platonism” of Cornford’s era bequeaths to Plato a Cratylian view of language that is completely out of compass with the linguistic insights displayed by the writer of the Cratylus 158 As noted earlier, the dialogue makes clear, for example, the fundamental problem that behind every name all we146 find is more names: a sentiment also developed in other dialogues. I will later summarize the list of problems with language raised in the Cratylus that no one without a deep understanding of how language works would be able to ask. The questions themselves obviate the Cornford interpretation. See Sallis’ analysis of the dialogue’s thematization of logos, the logos about logos (Being 183-311). Mitscherling says that “Plato’s dialogues . . . exhibit an unparalleled mastery of rhetoric” (281). Benitez observes147 that “the greatest rhetorical display of . . . [the] Protagoras is . . . the dialogue as a whole” (Ancient 222). Grote lists “rhetor” as one of a list of Plato’s masteries (214). And according to Murray Plato’s reputation as a poetic literary artist par excellence dates back to antiquity (13). (9). The understanding of language required to conduct the exploration that takes place in that dialogue makes nonsense of the positivist view of language that Cornford here attributes to Plato. This mis-take of Plato’s view of language – which is inferred from the supposed146 “argumentative” level of the dialogues – is only the tip of the iceberg when it comes to the absurdity of attributing such a view of language to the writer of the Platonic text: a writer widely acknowledged as an unmatched metaphor-master, an irony-and-myth-loving wordsmith, a rhetorical genius whose poetic and literary excellence has been celebrated since antiquity.147 This dissertation tries to displace Cornford’s Cratylianian mis-reading about Plato’s attitude to language with more promising and likely affiliations between language scepticism (that has its eyes wide open) and language virtuosity. Cornford, however, appears on the scene just as Enlightenment-positivism nears its apotheosis, when analytic philosophy launches its radical project to discover the logic of rule- governed meaning in communication. At the end of Part 2 questions were raised about what a sceptical, theory-resistant, language theory would look like: questions about style and genre, about, for example, what compels the genre of Wittgenstein’s Philosophical Investigations with its intermittent dialogue and inference-oriented, questioning approach. There is too that niggling question about recognition and what one needs to already know in order to recognize something; there is a very real question about how a reader’s attitude to language (like Cornford’s or his Oxford colleague R. Robinson’s) might prevent them from seeing a differing theory of language (like Plato’s or Derrida’s). There is a question about whether language positivists seeking images of rule-governed meaning would recognize sceptical language theory 159 R. Robinson is mystified by Plato’s concern with language’s epistemological-identification purposes, and as a148 result does not recognize Plato’s discussion of language as such. Robinson typifies the analytic disposition when he remarks on Plato’s failure to understand that “the purpose of names is to refer” (334), and that “statements are true or false because they describe and assert” (335). Arguably his conclusion that Plato’s treatment of language is inadequately differentiated stems from the fact that he himself has an inadequately differentiated view of language. Part 1 observes the diversity of views represented by the sceptical tide.149 when they saw it. Before we begin Part 5’s stylistic analysis of Derrida and Plato, we will148 briefly review how this “Platonism” (or rather multiple “Platonisms” given the lack of agreement about the so-called Platonic “doctrines”) invites Derrida’s intervention. The problem of recognizing what Plato was saying about language is a piece of the “Platonism” puzzle; and it is a problem with a long history, dating back, as we noted in Part 4, to Aristotle’s interested, polemical “interpretations.” But while Aristotle casts the longest shadow over the interpretation of Plato (as Kahn remarks), in the history of interpreting the dialogues multiplicity is the norm. The direct inheritors of Plato’s school, the so-called “first academy” – whom Cherniss suggests were contemporaries of Plato who might have been less followers of Plato and more like colleagues – expounded their own views on the dialogues’ meanings. Then, in a radical departure from the “Platonism” of the “first academy,” the sceptical “second academy” began reinterpreting the dialogues with a specifically sceptical understanding of language; this revisionist move would later receive Cicero’s influential endorsement and produce various interpretive projects (some of which are kindred to my own). The differing religious affiliations of subsequent Neoplatonists brought their own149 varied slants to the Platonic text (most notably the Christian); but however doctrinal the Neoplatonists might have tended to be, their interpretations were allegorical (and thus open to the now widely recognized importance of the striking “literary” elements of the dialogues). And then in the eighteenth century – the heyday of the Enlightenment’s science-project zeal – Platonic scholarship takes a systematizing turn in a complete perversion of Bacon’s aphoristic sensibility. In his brief survey of the history of Plato studies, Gonzalez finds that by the 1700s “when the Neoplatonic interpretation comes to be rejected . . . Platonic studies had in essence to begin again. Plato’s philosophy was now interpreted according to the paradigm of ‘systematic’ philosophy of the kind found in Descartes, Kant, or even Hegel” (Third viii). Thus 160 Sparshott observes that Plato and Aristotle “were pre-systematic thinkers” who gave “dynamic accounts of ways of150 problem-solving which are still exciting to read.” As an Aristotle expert, Sparshott argues that “Aristotle is traditionally read as more systematic than he is, his occasional programmatic statements being interpreted as authoritative for all his writings. Why should they be?” (150). Ryle does not hide his interestedness. He writes, for example, “What I vainly wish [Plato] had said explicitly is151 this” (438). For example, Gallop made short work of Ryle’s wishful interpretation that Plato’s alphabetic analogies referred152 primarily to “uttered syllables” (Alphabet 364). And Ackrill made kindred correctives to Demos’ similarly interested search for affiliation, beginning with the trenchant reminder that “the Cratylus is a dialogue and not a treatise, and that it is therefore perfectly possible that views tried out at one place are refuted at another” (Comments 610). the already existing diversity of interpretation gets revamped and becomes a dogmatic system (in the sense of dovgma or opinion). Intent on establishing the dogma or doctrines of Plato,150 scholarly commentary takes a cut-and-paste approach to the text of the dialogues and, in the interest of a consistent idealist system, discards most of the text as superfluous, literary and dramatic elements. Despite this selective process nothing like agreement is ever achieved probably, as Gonzalez observes, because the attempt to systematize Plato’s aporetic, paradoxical dialogues produces countless contradictions. This final systematizing stage of the twisted history of Plato interpretation – which in the end leaves the texts of the dialogues behind – provokes Derrida’s opposition between Plato and “Platonism.” Cornford’s inapt determination doubtless reflects conversations going on with his Oxford friends during his own tenure at Cambridge. In the mid-twentieth century, Ryle and other Oxford-styled philosophers had quite tendentiously sought to discover the seeds of the analytic tradition in Plato, despite Robinson’s already thoroughgoing and insightful exploration of the Cratylus arguing to the contrary. The more tendentious arguments in this discourse151 provoked some acerbic remarks from the Ryle group’s mid-twentieth-century colleagues. The152 misleading intuitions about communication that get analytic philosophy off on the wrong foot to start with are evident in Demos’ observation that “the reader is struck by Plato’s omission (in his formal definition) of the use of language for communication with one another”; and it is a reminder of just what these philosophers thought they would find when they went looking for a philosophy of language (597). More recently, Christopher Rowe has commented on O’Leary- Hawthorne’s use of “the Sophist to offer some correctives to the idea of Plato as a ‘philosopher of language.’” Analytic philosophy’s aspirations to rule-governed grammar and logical 161 As noted in Part 1, Schiappa claims that Plato coined not only rJhtorikhv, but any number of other “arts” or “skills”153 as denoted by the suffix. Contrary to the received view of Plato as “anti-rhetoric,” as we have noted, this dissertation imagines Plato as the founder of rhetorical theory because he coined the term rJhtorikhv (the art of rhetoric) to label a centuries-old lovgwn tevcnh (or language art). Also see the opening pages of Part 1 where Gadamers’ “distant and ancient meaning of rhetoric” and Naas’ account of persuasion in the Illiad are noted.. Explicitness is the same kind of fraud as verbatim: from a philosophic perspective it is something of a linguistic154 illusion, much like determinate reference. Sperber and Wilson claim that explicitness is relative: “an explicature is explicit to a greater or lesser degree” (182). Even the most referentially descriptive language uses depend on pragmatic inferences; the most explicit-seeming utterances require contextual interpretation. Sperber and Wilson more recently have commented that “the central problem for pragmatics is that sentence meaning vastly underdetermines speaker’s meaning” (Pragmatics 3). The idea of explicitness is nevertheless useful here in a negative sense: that is, to show what something is not. In this dissertation I am using explicit to indicate a writer’s feeling that something stated is made so manifest that it cannot be avoided (explicit statement); and when readers feel that something has been made so manifest that they cannot avoid it I call it explicature. The paragon instance of explicitness here is Aristotlean writings, an explicitness so ineffectual that debate over his texts still rages. argument had an interesting effect (yet to be analyzed) on Platonic studies (perhaps even on Gonzalez’s otherwise insightful work). On the one hand, Rowe rightly endorses O’Leary- Hawthorne’s observations that analytic philosophy’s discourse “takes insufficient account of [Plato’s] ‘apparently life-long skepticism about language’” (86). On the other hand, we see the hegemony of analytic philosophy’s ideas about language precluding Plato’s language scepticism from consideration as theory. Plato’s sceptical-rhetorical theory is not “philosophy of language” to Rowe and O’Leary-Hawthorne. It doesn’t look like a theory of logic-governed, propositional meaning, so they don’t recognize it. However, analytic philosophy does not have copyright on153 the term philosopher, which surely belongs to Plato if anyone. And there is no reason to put the term “philosophy of language” into the grave along with analytic philosophy’s failed language- theory ambitions. Since Derrida and Plato are both philosophers theorizing about language, why shouldn’t they be styled as “philosophers of language?” Analytic philosophy’s influence on “Platonism” brings the problems of style and genre into the limelight. In Part 4 we discussed the way that style figures into the ethos of analytic philosophy and the guiding role that the plain-style-of-science myth has played in its post- linguistic-turn identity. Analytic philosophy’s style looks like logically sound argument: but Orwell’s windowpane metaphor and Miller’s comments on “artfully produced transparency” are a caveat about the style of non-style, with its rhetorically produced stance of neutrality. In contrast to analytic philosophy’s (scholarly) style of logical-proposition and “explicit” statement, the stylistic of the Platonic and Derridian texts are saturated with human voices;154 162 The structure of supplementarity is “the play of substitution [which] fills and marks a determined lack”; something155 added always entails a loss or an absence to be filled (OG. 157). they are awash with conversation’s attitude and implicitness as Socrates’ opening line illustrates: “I went down to the Piraeus yesterday with Glaucon, the son of Ariston. I intended to say a prayer to the goddess . . . .” When Derrida begins Dissemination with the enigmatic line “This (therefore) will not have been a book,” he begins a lively and playful conversation with his readers. As we shall see, Derrida’s conversational tone creates the impression that readers were there sharing the stage with Derrida; his banter cajoles and admonishes readers for the supplements they bring to the text. This exotic and flouting conversational stylistic (exponentially more demanding than what occurs in conversation) is intimately connected to how language works (implicitly, ostensively, rhetorically, performatively) in conversation. It is a written stylistic emulating central conversational characteristics. Conversational is also a very apt way of characterizing the Platonic stylistic and that of other writers, like (the later) Wittgenstein. Though all three use radically different written genres, the term conversational indexes an important similarity between them. Part 5 takes a look at how Derrida exploits the opportunity presented by the Platonic text’s rich over-layering of other people’s voices in order to thematize his own critique. What I call the Derridian critique of writing highlights the epistemological problems stemming from the operations of the linguistic sign, its “decenteredness” and compulsive supplementarity.155 His critical matrix brings the qualities of language which have been discussed throughout this dissertation – language’s context dependency, extreme malleability, and lack of determinacy – onto centre stage. The voice-overs of western metaphysics are an effect of reader supplements – the losses and the gains entailed in each new reading – and the Platonic text an opportunity for Derridian display. Part 5’s analysis reveals the surprising proximity of Plato’s and Derrida’s views of language (a proximity that can be gauged by the contemporary relevance of the ancient question of linguistic mediation); and it shows how Plato’s views on language are relentlessly thematized in the dialogues (although the stylistic question about what it means to thematize will remain temporarily suspended). 163 See for example Christopher Johnson, Barbara Johnson (translator’s introduction Dissemination), Walter Brogan.156 I believe that the prevalence of this systematic view of Plato under the name of “Platonism” has resulted in reading157 failures of “Plato’s Pharmacy.” Presuppositions about Plato seem to have prevented some scholars from recognizing what Derrida is doing and saying and the kind of interest he has in Plato (see, for instance, Chris Mortensen, Yoav Rinon, Katherine Zuckert). Gonzalez’s own interpretation of the dialogues is mediated by an interest in coming up midway between systemic-158 doctrinal and non-systemic-non-doctrinal positions because he sees the absurd implications stemming from the systemic- doctrinal approach when compared to what Plato actually wrote in the dialogues; but he is (I think needlessly) worried about loss of meaning that might result from its rejection. See Gonzalez’s footnote on the various recent “skeptical” interpretations which his project “seeks to transcend” (Third 9). I see the systematizing trajectory as what might distinguish Baconian-type responses to Plato (together with the understanding of language informing such responses) from subsequent positivist developments in both Plato scholarship and language theory. Critics have observed the way that Derrida’s text enacts his critique. Sallis says that156 Platonic dialogue “not only speaks but also does something, accomplishes something in deed” – resulting in a mediation between word and deed (Being 17-18). In his commentary on the Phaedrus, Linck observes that “once again Socrates performs rather than explicates that about which he speaks” (275). It is this performative dimension of language use that characterizes the writings of both Plato and Derrida. Part 5 demonstrates how the conversational styles of the Derridian and Platonic texts perform and reveal their theories of language; and how their styles are inseparable from whatever “argument” they might be making: condensing and recording their sceptical-rhetorical views of language. Neither garment nor ornament, these stylistically inscribed meanings cannot be overlooked; they are not some outside form that can be removed from an inside content. This particular brand of form-content fusion is a fact that recent Plato scholarship has (sometimes woefully) had to acknowledge when recognizing that the dramatic and literary elements (not least the dialogue genre itself) present a picture (of both language and philosophy more broadly) which completely undermines still influential Enlightenment convictions about Plato’s systematicity and dogma. Sallis and Gonzalez both provide insights157 into how translation and conceptions of philosophical proposition – both factors of style and genre that have a large impact on meaning – have figured in “Platonism.” 5.2 genre and the style of “propositional knowledge” ~ Gonzalez disavows “Platonism’s” doctrinal-systematic readings, arguing against the view of the dialogues as containing what he calls “propositional knowledge” and logical argument on which so-called Platonic dogma depends (Dialogue 7-8). Gonzalez defines158 164 At the root of this nonpropositional knowledge, Gonzalez seems to imagine a stable nature (something like forms)159 that words can make manifest in a more determinate way than I believe is possible. I don’t share his confidence that the nature that is revealed in this way has any certainty. propositional knowledge as follows: The characterization of knowledge as propositional means not simply that we can speak about the objects of philosophy, but rather that we can do so without profound distortion. In other words, in order for philosophical knowledge to be propositional, it must be the case not only that we can form propositions about the fundamental principles of reality, but also that these propositions can express these principles as they truly are. . . . Knowledge that a triangle has three sides can be seen as knowledge of the proposition that a triangle has three sides. However knowledge is nonpropositional when it cannot be identified with knowing what can truly be asserted of its object. . . . The distinction between nonpropositional and propositional knowledge parallels a distinction . . . between the “manifest” and the “describable.” Nonpropositional knowledge is of what is manifest without being describable. (Dialogue 7-8) However problematic the attempt to define knowledge in this way is, the distinction between descriptive propositions and what becomes “manifest” in Plato’s conversational stylistic is of interest. I believe that the propositional/non-propositional distinction Gonzalez is making has159 to do with what the text shows us and points to, rather than what it “explicitly” states. This is a distinction between the precision required for logical proposition that Wittgenstein talks about and the more general kinds of things philosophy might be interested in. We shall see that the Platonic text is all about ostension and implicitness, entirely context dependent dimensions of language use. Context, for example, will determine whether the banning-of-poetry episode in the Republic is censorship of tragic poetry from ideal political states or whether it has something to do with human conduct and how language works. Sallis speaks at length about the focus in the dialogues on demonstration, on making things manifest: the way that the dialogues mediate the opposition between “logos and ergon” (word and deed), the way that the dialogues are themselves deeds or activities. He says that “the dimensions of ergon, no less than that of mythos, belongs together with the dimension of logos” in the Platonic text (Being 18). This “making-manifest” activity, entailed in the dialogues, is 165 As per Wittgenstein, Sperber and Wilson and Van Dijk.160 Recall that logos means all of the following: word, myth, account, argument, discourse. In the Platonic text Sallis’161 preferred translation is account or discourse, though it sometimes means word, and sometimes something like argument. Derrida, famously takes up the translation of pharmakon, observing that “blocking the passage of opposing values [evident in the assumption that Plato’s text would not intend the two sided pun entailed in the term pharamakon] is itself already an effect of ‘Platonism’” (D 98). See also Gill on translation. part of the above-noted performative dimension of language use at work in them, but not necessarily very exact. Remember that precision is required for logic, not language use, the effects of communication are often vague, and generalization and abstraction have important cognitive functions. The conversational stylistic is not analytic philosophy’s signature160 stylistic, not a language of logical proposition and argument (which is a rhetorical effect, in any case). Gonzalez’s conception of “propositional knowledge” is a response to misguided desires for propositional logic that have clouded perceptions of the Platonic text: desires reflecting a descriptive-referential view of language that no longer holds. Addressing the question of logical argument as it is related to the history of the interpretation of the Platonic text, Sallis states that we must resist the tendency to interpret post-Platonic determinations back into the work of Plato. When we regard it as self-evidently correct to allow logos to be taken as ratio, hence, as reason, we re- affirm without even really considering the matter, one of the most overwhelmingly decisive transitions in the movement away from the Greeks that constituted the course of Western thought. To assume, in advance, a specific, well-defined determination of logos in this direction – to take for granted in an interpretation of Platonic writings subsequent notions of “rationality” and “demonstrative argument” – is to be misled by the tradition . . . (Being 14-15). Sallis reminds us here of the problem of translation that Derrida also addresses: in particular, readers should be wary of the term argument being used to translate logos and its relatives in the text of the dialogues. Post-Platonic interpretations of ancient Greek language have distorted161 and obscured Plato’s writings in various ways. As Derrida observes, “all translations into languages that are the heirs and depositories of Western metaphysics thus produce on the pharmakon an effect of analysis that violently destroys it” (D 99). Post-Platonic determinations have also reduced Plato’s minutely orchestrated dialogues with their remarkable images, ironies and myths into “propositional knowledge” of the sort 166 See Sallis’ footnote on the form/content fusion and artfulness in Platonic dialogue: he cites Schleiermacher’s view162 that “form and subject are inseparable, and no proposition is to be rightly understood except in its own place and with the combinations and limitations which Plato has assigned to it.” He notes also Strauss’ statement that “nothing is accidental in a Platonic dialogue; everything is necessary at the place where it occurs,” and Bloom’s comments that “the dialogues are so constructed that each part is integrally connected with every other part; there are no meaningless accidents,” and “every word has its place and meaning” (Being 17). She is specifically sent out of the city “ejk th:V povlewV ajpestevllomen.”163 expected in the genre of Aristotelian-type treatise – as though Aristotle’s preferred genre were the standard by which Plato’s thought should be measured. Readers’ expectations of162 “propositional knowledge” from the dialogues indexes some kind of genre confusion. It is not a question of what “proposition” Socrates’ response at any moment represents, any more than a line that Lear or Hamlet speaks represents Shakespeare’s opinion or the play’s meaning. The dialogues’ (fictional) conversations are a kind of literature, a serio-comic genre according to Bakhtin. The conversational stylistic cannot be reduced to pronouncements of “exactly what Plato really meant”; it does not provide the kind of “explicitness” propositional utterances convey; rather it develops a network or maze of (often ironic) interconnections that all depend on one another for meaning, always shifting back and forth, inter-illuminating one another. The form/content, style/argument fusion makes the Platonic text’s consistent focus on how things are said a subject of great interest to this dissertation’s stylistic analysis. Consider, for example, what kind of “explicit” propositional statement appears (if any) in the infamous Book-10-banning-of-Poetry episode. Here Socrates appears to speak out of both sides of his mouth, as both poet and philosopher, when he proposes to invent a mythical war: “let’s also tell poetry that there is an ancient quarrel between [her] and philosophy” (Grube modified 607b). Examination of the whole passage shows that Socrates oscillates between and conflates “the topic of poetry” and its personification as the “pleasure giving Muse” who is banished or “ajpestevllomen” (sent off). But is it poetry – the style of writing – or the Muse who is163 banished? . . . you should also know that hymns to the gods and eulogies to good people are the only poetry we can admit into our city. If you admit the pleasure-giving Muse, whether in lyric or epic poetry, pleasure and pain will be kings in your city instead of law or the thing that everyone has always believed to be best, namely reason. 167 The proposition is ironic whether or not it is sacrilegious: “to betray what one believes to be the truth would be164 impious” (Grube 607c). Perhaps Socrates is suggesting a banning of deities. It seems impossible to distinguish assertions with metaphorical status from those without metaphorical status. Explicature is nowhere to be found. That’s absolutely true. Then let this be our defence – now that we’ve returned to the topic of poetry – that, in our view of its nature, we had reason to banish it from the city earlier, for our argument compelled us to do so. But in case we are charged with a certain harshness and lack of sophistication, let’s also tell poetry that there is an ancient quarrel between [her] and philosophy (Grube 607a-b). Socrates treats Poetry here as a Muse in the tradition of ancient poets and tells Glaucon that they are to address Poetry, to tell her there is an old quarrel, and she is not welcome. The quasi-imperative voice of the verb orders Glaucon (proseivpwmen de; aujth:/) – literally “let us address her.” Socrates’ injunction inverts the traditional invocation-of-the-muse device – dismissing the goddess rather than beseeching her assistance, as if enacting a divorce between poet and muse – but ironically from the stance of a poet who speaks to a muse. The force of the imperative poetically mimes the authoritative approach Socrates is taking to the Muse of Poetry: to both address her and dismiss her. In a typically poetic gesture, Socrates invents a mu:qoV (a ‘myth’) in order to tell Poetry to leave – an odd combination of poetic ethos and disrespect. Whether Socrates is metaphorically discussing the concept of poetry as a style of writing by ironically adopting the posture of a poet, and like a poet personifying the subject, or simply offending the deity in a poetic fashion is unclear; what is clear is that he is doing it poetically; he is either imitating a poet or asserting his own poetic ethos (implicitly and ironically). This164 stability-defying passage brims with irony. Socrates’ poetic lovgoi in which he imitates the stance of a poet would be dismissing his Muse by beguiling her with myth or proposing to reinvent a history that surely the deity would be better apprised of than he. The determinate reference and descriptive language use on which “propositional knowledge” depends seems to evaporate at every turn in this passage. Pronouncements on “exactly what Plato meant” might have missed something here. We will return to Poetry’s banning and how it works in the context of the Republic. But we can see in this brief glimpse the way that Plato intertwines the treatment of poetry as a style 168 Although I consider Sperber and Wilson’s use of the term “explicature” somewhat problematic, they do indicate165 that “in many cases – perhaps most – the propositional form of an utterance is not an explicature at all. This is true of tropes on the one hand and of non assertive speech acts on this other” (224). Tropes and nonassertive speech acts might be the norm in the conversational stylistic In Sperber and Wilson’s terms, “the aim of communication in general is to increase the mutuality of cognitive166 environments rather than guarantee an impossible duplication of thoughts” (200). of writing and its personification as the Muse. Completely apart from the small matter of the amount of attention Plato gives to a style of writing (which should tell us something to begin with) we can see the way that the conversational stylistic (which easily assumes an ironic voice) oscillates with implicitness. Even the assertion that the poetry making “hymns to gods and eulogies of good men” is permissible, is undermined by the irony of Socrates’ sacrilege towards the Muse (Grube 607a). If we define explicature as reader inferences that have been made so strongly manifest by the writer that the reader cannot easily avoid them, what possible explicature could be obtained from this irony-laden passage that seems to never sit still? If it165 is permissible for Socrates to be so poetic, what propositional “knowledge” might this passage convey about poetry as a style of writing? Only the much larger context of the Republic as a whole offers any stability to this passage, as we shall see. We shall also see that the ostensive implicitness of the conversational stylistic shares the qualities of Bakhtin’s novelistic image and the motion entailed in “becoming” (which is related to Plato’s educational ethos). Plato’s language scepticism lacks faith in the possibility and utility of “explicitly”-stated-determinate- propositional knowledge, whereas the serio-comic genre of Socratic dialogue exploits the performative resources of conversation’s implicitness and context sensitivity. What becomes manifest in Platonic dialogue is a myriad of interconnections, entangled threads, webs of relations: minutiae whose aim might well be shared evaluation.166 Derrida’s overwriting thematic takes aim at the systematicity and dogma of definitive- proposition-loving readers who decide “exactly what Plato really meant”: for instance, those who have determined that poetry is banned from Plato’s ideal republic. In the same vein, Gonzalez takes issue with a cornerstone of doctrinal interpretation: “Platonism’s” claim that the dialogues have a “theory of forms.” He says that “to the unprejudiced reader, it is evident that there is no ‘theory of forms’ in the dialogues” (Third 12). They are metaphors not some kind of 169 Note that Kahn considers the forms to be a theory despite their different incarnations. Gonzalez bolsters Plato’s167 theory of language (which he would say was not a theory per se) with some conception of stable nature that seems to be related to forms (as noted above). Though he modifies the Cornford view, his confidence in the ability of words to make things manifest is interesting. Kahn documents the diminished role of forms in most of the so-called “later” dialogues. Ackrill observes “I have gradually passed from talking about Forms to talking about concepts . . . I have implied that the task assigned in Plato’s later dialogues to the dialectician or philosopher is the investigation and plotting of the relations among concepts, a task to be pursued through a patient study of language” Essays 78). Perhaps Plato’s “forms” change in each different context because they belong to a discursive rather than an ontological field. “Platonism’s” reification of forms into ontological realities might be just the kind of “propositional knowledge” Plato doesn’t provide. Perhaps within the Platonic text they are metaphors treating the discursive nature of human thought, that change in different dialogues to suit different contexts. Although, they may be a failed metaphor, the failure for Socrates’ elaborate images to hang together might be a cue of its own; the disjunctions in Plato’s myths and images could be a reminder of their metaphoricity. “propositional knowledge,” and for Gonzalez, not a “theory” as such. Sallis suggests that to ask about “an eidos (or idea)” is to ask about the whatness of that thing – that is, to ask what it is, full stop. Socrates’ use of this metaphor varies. However enticing “Platonism’s” glimmering forms may be, Kahn admits that the presentation of the forms “is formulated differently each time it appears” (xvi). Gonzalez laments that “Plato shares very little with this ‘Platonism,’”167 which he feels stands in the way of a “real confrontation with philosophy as Plato understood and depicted it in his singularly elusive dialogues” (Dialectic 16). The conversational stylistic of the dialogues has quite different aims than say the Aristotelian treatise; and it is not amenable to “explicit” propositional statement. 5.3 genre and social relations ~ In this analysis of how language works in the Platonic and Derridian texts, questions of style and genre, and what motivates them, remain constantly on the horizon. Just as in conversation the punch line is often implicit, so too are the themes of the Platonic and Derridian conversational stylistic. Implicatures (inferences made by readers and hearers) accumulate in the background like loose ends, threads left dangling; and then they get twisted together, one thread leads to another, threads get tangled up and criss-cross back and forth, becoming a web. Implicitness invites reader inference, enabling the supplementary threads of “Platonism’s” voice-overs. Implicitness compacts utterance, making Derrida’s (and Plato’s) ironic, conversational stylistic do more than one thing at the same time. Derrida’s weaving and spinning metaphors in “Plato’s Pharmacy” reflect Plato’s celebrated use of them to characterize the linguistic operations of texts. Unravelling one strand produces a mass of tangled threads. Part 5 examines a few of the threads that come into contact with what I call the overwriting-of- 170 The place given to Lautréamont (pen name for Ducasse) in Derrida’s discussion of the preface suggests that uptake168 of Lautréamont in Derrida’s intellectual community gave some currency to this reader-writer thematic. Lautréamont’s translator, the poet Lykiard, observes that “his work more than most demands explication, dealing as it does with so many crucial ways of writing and reading a book” (12). Lykiard suggests this reader-writer theme is built into the text when he asks (about the Chants) “Who really is the author, what is his identity and relationship to the reader? Is the author himself a fiction, an idea (idée and I.D. have the same French pronunciation) in the reader’s mind? ‘Si j’existe,’ he wrote, ‘je ne suis pas un autre,’ yet Ducasses as author is elusive . . .” (8). the-text-by-readers thematic at certain critical moments in order to investigate the stylistic operations of the Derridian text and how they are interwoven with his commentary on Plato. But as Derrida warns, “the threads of the complicities are almost impossible to disentangle” (D 117). Language’s ability to do more than one thing at the same time corresponds to the multiple motivations for speaking and hearing, writing and reading. What motivates Derrida’s interest in Plato is particularly multifaceted. His often overlooked reverence for Plato has its counterpart in Derrida’s opportunistic use of the Platonic text in “Plato’s Pharmacy.” The reader-writer focus is especially apt for Plato’s use of the Socratic dialogue genre (the role- model of the conversational stylistic); and, as we have noted, the disjunct between multiple “Platonisms” and the Platonic text itself makes the dialogues prime material for the reader- overwriting thematic. Plato’s selectively over-interpreted writings are an opportunity to bring the Derridian critical matrix to bear on philosophy’s founding text. For example, the Platonic text is a particularly enticing site for developing the way that “reading is writing”(D 63): the way that reading involves writing, involves laying down one’s own counter-text, and the way that speaking entails hearing, both one’s own words as they are uttered and how the addressee interprets them (“reading” their response), etc. This reader-writer focus also had some currency in Derrida’s intellectual circles of the time: thus adding another layer of motivation not only168 for Derrida’s interest in the Platonic text, but also in the development of both his “critical 171 Derrida’s role in the interdisciplinary literary, theoretical, political thinking of the Tel Quel writers in the 1960s169 doubtless accounts, at least in part, for his stylistic development. Given fellow participants and thus an audience of the ilk of Roland Barthes, Maurice Blanchot, Julia Kristeva, Philippe Sollers, Tzvetan Todorov, Umberto Eco, Gérard Genette and other well known European intellectuals – together with their theoretical interests and the challenges they were making to distinctions between the literary, critical and political – gives some historical context to Derrida’s stylistic motives. Johnson’s “Introduction” states: “Both Numbers and “Dissemination” are attempts to enact rather than simply state the theoretical upheavals produced in the course of a radical reevaluation of the nature and function of writing undertaken by Derrida, Sollers, Roland Barthes, Julia Kristeva and other contributors to the journal Tel Quel in the late 1960s. Ideological and political as well as literary and critical, the Tel Quel program attempted to push to their utmost limits the theoretical revolutions wrought by Marx, Freud, Nietzsche, Mallarmé, Lévi-Struass, Saussure, and Heidegger. It is not surprising, therefore, to find that “Dissemination” operates at the very limits of intelligibility” (xxxi). I underline in Derrida’s text to add emphasis because of his frequent use of italics. All italics are Derrida’s. 170 matrix” and his style.”169 The playful opening lines of Dissemination begin an ironic discourse on genre and context vis-à-vis reading/writing practices at the same time as disseminating various features of Derrida’ critique of writing. This (therefore) will not have been a book. Still less, despite appearances, will it have been a collection of three “essays” whose itinerary it would be time, after the fact to recognize; whose continuity and underlying laws could not be pointed out; indeed, whose overall concept or meaning could at least, with all the insistence required on such occasions, be squarely set forth. I will not feign, according to the code, either premeditation or improvisation. These texts are assembled otherwise; it is not my intention here to present them. The question astir here, precisely, is that of presentation. While the form of the “book” is now going through a period of general upheaval, and while that form now appears less natural, and its history less transparent, than ever, and while one cannot tamper with it without disturbing everything else, the book form alone can no longer settle – here for example – the case of those writing processes which, in practically questioning that form, must also dismantle it. Hence the necessity, today, of working out at every turn, with redoubled effort, the question of the preservation of names: of paleonymy. Why should an old name, for a determinate time, be retained? Why should the effects of a new meaning, concept, or object be damped by memory? Posed in these terms, the question would already be caught up in a whole system of presuppositions that have now been elucidated: for example here, that of the signifier’s simple exteriority to “its” concept. One must therefore proceed otherwise. Let us begin again. [underline added] (D 3)170 Whatever genre expectations a reader might reasonably have brought to the text of a prestigious twentieth-century philosopher, Derrida promptly staunches them. As with Plato’s Poetry- 172 It would appear that the thematic focus on readers, texts and genre conventions quite marked in the “Hors livres”171 pertain to all three essays. For instance, in the “Double Session” Derrida writes: “One could indeed believe that although he is passive in reading, he at least has the active freedom to choose to begin to read, and that the same is true of Mallarmé; or even that you, dear everyreader, retain the initiative of reading all these texts, including Mallarmé’s, and hence, to the extent, in that place, you are indeed attending it, deciding on it, mastering it. . . .The question of the text is – (for whom are)/(for whoever reads) these sheets. (D 223-24). In “Dissemination” he writes: “The text is remarkable in that the reader (here in exemplary fashion) can never choose his own place in it . . . The moment ‘therefore’ is written, the spectator is less capable than ever of choosing his place. This impossibility – and this potency, too, of the reader writing himself – has from time immemorial been at work in the text in general” (D 290). See also Matthew S. Linck’s repeated reference to remarks “that can be read in at least two [or more] ways” (267)172 and Halperin’s reference to features that “can be accounted for in at least two ways that do not refer to the philosophical doctrines enunciated in the Dialogue” (100). banning passage, the sheen of irony marks Derrida’s opening paragraphs precluding any possibility of simple explicature; most strikingly ironic are the claims that the book is not a book, that it is not a collection of three essays, and that he will not present them – which according to the code of scholarly writing is required of an introduction or a preface. Of course he does introduce them, it is a collection of three essays (a fact which is carefully indicated by171 their original publication particulars appearing on the separate title page at the beginning of each essay) and it is a book. Its title is Dissemination; it appears in italics in this dissertation and in its “Works Cited.” It is a library book; it has a call number; and as we shall see it is a stage prop in the opening scene. At first glance readers might be unsettled by these genre ironies; the writer might not seem very serious; one might feel that one is missing something. Riffaterre claims that “unpredictability will compel attention”: the unpredictable or surprising elements of a text that produce stylistic effects depend on genre expectations (Criteria 158). Derrida’s enigmatic opening line shatters all genre expectations. And his playful ironies entice readers to take a second glance, to reread, and investigate his text more thoroughly. The reader-writer focus, as we have noted, bears on “Plato’s Pharmacy” for at least two distinct reasons (so, like other features of conversational language use more than one thing is relevant at the same time). On the one hand, there are the voice-overs of readers writing172 western metaphysics, drowning out the voices of the dialogues’ characters; and on the other is genre – the dialogue structure itself involves readers in an ongoing transaction with the lines of the dialogues’ dramatis personae. Readers must negotiate the question of who speaks for Plato, 173 Sallis argues that “it is not a matter of seeking Plato’s opinions at all, for philosophy is what is fundamentally at173 issue in the dialogues” (Being 3). Gonzalez remarks “I find it plausible that the dialogues contain some of Plato’s own views, just as I find it plausible that King Lear and Hamlet contain some of Shakespeare’s own views” (Third 9). Also see Gerald Press’ “Introduction” in Who Speaks for Plato?. I write “NO!” in the margin when Socrates tricks Protagoras into agreeing that “for each thing that can have an174 opposite, there is only one opposite, not many” (Cooper ed. 332d). Halperin comments that “Plato’s texts read us, evidently, as much as we read them, even if they also seem to write,175 to prescribe, our own response to them” (125). if anyone at all, or if there is even any point to figuring out what Plato’s opinion is. Readers173 are always having to interpret the responses of the participants, always having to pay close attention to how things are said – not just questions of precisely what is said, but also the way it is said. Derrida’s conversational style puts the reader in a position of wanting to answer back (“it is too a book!”), not unlike the reader of the Platonic text who is always implicitly invited to find a better response than the interlocutor of the moment. Unlike Wittgenstein, in Plato and174 Derrida such reading responses often involve irony. This taxing written stylistic – far too demanding ever to be imported into everyday conversation and media – structurally elicits reader response. We might find too that it anticipates responses and already responds to those anticipations.175 Derrida’s ostentatious opening line – “This (therefore) will not have been a book” (D 3) – launches the overwriting-of-the-text-by-readers thematic interwoven throughout “Plato’s Pharmacy.” The rhetorical antics of Derrida’s opening deictic “this” provokes social relations, welcoming his readers into the scene of writing as though both he and the reader were performers in a play, with the book (a stage prop) open before them. The opening “this” is a deictic pronoun which presupposes a referent that the reader knows, that the reader will recognize. In a written text, like conventional academic prose, this refers to a referent in prior (or antecedent) text that the reader infers (that is, the reader has to infer what the referent is and in conventional academic prose the referent must be “clear” and obvious). At a very-fine grain level deictic determiners are presuppositions which are, by definition, invitations to make 174 Deictic pro-forms are “cohesive” in Halliday and Hasan’s cohesion analysis. Cohesion occurs when one element of176 the discourse is dependent on another for interpretation; cohesive elements thus have a presupposing nature make texts hang together for the reader. Normally “this” in a written text refers to a previous textual referent. The reader effort to link the one to the other coheres the text. However, the context of this instance forces the reader to pause and try to figure out what the referent is. Its cohesion is contrived; the reader really has to work for it. We can infer from the subsequent paragraph’s reference to “a collection of three ‘essays’” that the opening line’s177 “this” refers to the book. Had Derrida used any typical term to qualify the referent – i.e., writing, text, analysis, preface, introduction – the effect of physical presence would not have been created. Had he not set the opening line off in its own paragraph, it might not have created this non-textual effect. Textually, in its French edition “ceci” (this) could refer to the preceding page’s PRÉFACES or Hors Livres, or the178 text’s title, La dissémination. Arguably none of these are intuitive sites for the deictic in the body of a text, and all come with ironic overtones pertaining to the critique of genre underway. inferences; and Tannen considers such invitations to make inferences involvement strategies,176 involving the reader in the interpretation of the text. It is an opportunity for reader input, a kind of sharing of the meaning-making, as Tannen puts it. “This” is a good way to get someone’s attention, it signals something that the reader must do: i.e., fill in the referent usually from the prior text, but sometimes after it. But “this” is the opening word of the “book” which has no prior textual referent (except titles and subtitles), nor indeed an antecedent in the opening line (which is also a separate paragraph). It sits there in the opening line (and paragraph) forcing readers (trying to determine its referent) out of the textual and into an imaginary physical realm where the referent seems to be the material book itself (Dissemination). Contrary to expectation, “this” in the opening line177 of Derrida’s book is a physical deictic creating an imaginary physical “co-presence,” pretending to refer to something both parties can see or sense viscerally – “this” book.178 Accordingly, “this” alerts us that we are entering an imaginary zone to start with. As we read “this” we are to imagine that we see the book, as though reader and writer were in the same room together, as though the reader were there looking at the book that Derrida is pointing at. We are placed in the scene of writing as a scene of reading: as though the reader were an active participant in this imaginary physical scene. “This” is not just an invitation to make an inference, but a challenge – a demand for imaginative reader involvement. On second glance, one might observe that Derrida’s opening lines sound as if they are part of a conversation, something someone says out loud; it sounds as if Derrida were continuing an argument with his addressee: “still less, despite appearances . . .” 175 5.4 “Always with irony” (D 67) ~ In order to analyze how “literary” and “poetic effects” (particularly its layers of irony) produce implicatures and become relevant in this opening passage, we turn to Sperber and Wilson’s account of the operations of relevance and the way a conversation gives rise to a background against which relevance is judged. Some information is old: it is already present in the individual’s representation of the world. . . . Other information is not only new but entirely unconnected with anything in the individual’s representation of the world. It can only be added to this representation as isolated bits and pieces . . . . Still other information is new but connected with old information. When these interconnected new and old items of information are used together as premises in an inference process, further new information can be derived: information which could not have been inferred without this combination of old and new premises. When the processing of new information gives rise to such a multiplication effect, we call it relevant. The greater the multiplication effect, the greater the relevance. (48) As discourse proceeds, the hearer retrieves or constructs and then processes a number of assumptions. These form a gradually changing background against which new information is processed. Interpreting an utterance involves more that merely identifying the assumptions explicitly expressed: it crucially involves working out the consequences of adding this assumption to a set of assumptions that have themselves already be processed. In other words, it involves seeing the contextual effects of this assumption in a context determined, at least in part, by earlier acts of comprehension. (118) Hearers are constantly interpreting and enriching utterances into propositions; they then make inferences from them, determining what about that implicature fits or is useful, what about it is worth pointing out in the context, whether the implicature is strong (relevant in its own right) or part of the background against which other implicatures are assessed. Sperber and Wilson argue that “metaphor and irony are not essentially different from other types of ‘non-figurative’ utterances” in terms of the way hearers produce implicatures and gauge their relevance (243). They “give the name poetic effect to the peculiar effect of an utterance which achieves most of its relevance through a wide array of weak implicatures” (222). A weak implicature generally contributes to background information defined as implicatures which are not immediately relevant in their own right (209). “Some implicatures,” 176 they say, “are made so strongly manifest that the hearer can scarcely avoid recovering them. Others are made less strongly manifest. It is enough that the hearer should pay attention to some of these weaker implicatures for the relevance of the intended interpretation to become manifest” (197). But background information not only provides context for other implicatures, it can also be strengthened by a variety of emphasis-producing devices. A wide array of weak implicatures produces poetic effects which encourage the hearer to take a large share of responsibility in imagining possible implicatures – they invite substantial reader participation (221). Most importantly, weak implicatures can be strengthened by a variety of emphasis producing mechanisms like repetition; repeated items become stressed and then produce relevant implicatures of their own. Such strengthening operations are extremely useful for understanding some of the more oblique and poetic operations of both the Derridian and Platonic texts. Previously stated information, whether stressed or unstressed, forms part of the contextual scene (or background information) which informs the inferencing process. Later unstressed items repeated enough eventually become strengthened and produce their own relevant implicatures. Consider again then the imaginary atmosphere created by Derrida’s pretended conversation: by his demanding “this” (an involvement strategy) drawing readers into an imaginary scene, and how this scene’s fictional dimension is enhanced by Derrida’s playful (and equally attention-grabbing) future anterior verb form. The text seems to say that “this (book) is not a book,” or more accurately, “(at some future time) this (book) will be deemed (in some past time) not to have been a book.” The fortune-telling pretense of this verb form adds irony to an already ironic line; the Cassandra-like future-predicting voice combines with the imaginary of Derrida’s physical co-presence with the reader (as though they were in the same room together having a conversation). The fictional-storytelling voice creates implicatures about the playfulness of the genre and the “literary” register of Derrida’s text. So that when, referring to his three essays, Derrida says “mon intention n’est pas ici de les présenter” (it is not my intention here to present or introduce them), one might get the sense that he is refusing to 177 Derrida links his critique of the preface to the temporal movement of the future anterior (or perfect) verb form. “As179 though – right here – the preface could be calmly installed in the ample presence of its future perfect [futur antérieur], in the mode of that attending discourse whose definition you will have read later on” (D 56). Johnson too observes that “this book begins with a denial both of the book and of the beginning. The opening sentence, ‘This (therefore) will not have been a book,’ written in the future perfect tense, marks itself as presentation (‘this’), anticipation (‘will’), negation (‘not’), recapitulation (‘have been’), and conclusion (‘therefore’). The juxtaposition of the title (Hors livres, lit. ‘outside the book’) and the opening sentence is thus designed to map out the play of anticipatory retrospection and internalized exteriority involved in that metalinguistic moment of self-reflection traditionally known as the Preface” (xxxii). introduce his friends to his reader. Then when, still on the first page, Derrida playfully says “Recommençons” translated as “let us begin again,” the formal, French, first-person-plural, essay-style, nous convention gets combined with an imperative to start over again: inviting the reader to do something with the writer, reinforcing the imaginary reader-writer co-presence, creating implicatures. Come with me, he seems to say. Perhaps Derrida’s opening line – “this (therefore) will not have been a book” – will be deemed to have been a meeting between a reader and a writer, a conversation, a writing collaboration with readers who, we shall see, get their fingers caught in the threads. Or perhaps it will be deemed to have not been a meeting at all, but a pretense, an imitation or playful demonstration of a meeting. The fairy-tale-like opening at once invoking imaginaries of scene and time raises the question of whether even Derrida can see “the book” for which he is writing the preface, or whether he can see the rest of his preface as he composes this opening line. This question directs attention to the peculiar temporal posture of scholarly writing generally and of the preface-genre in particular. The temporal theme, raised by the future anterior verb form,179 seems to say, “let’s pretend,” directing attention to the present-tense pretence of scholarly writing. Derrida’s critique of the preface (or introduction) genre is not just the idea that the preface’s summary cannot master the already written text – with all the threads it disseminates in the supplementary operations of each new iteration – but the whole present-tense temporal posture of conventional scholarly writing. In particular, he is highlighting the pretense of an introduction or preface written after the fact: a completely false present tense, showing that writing conventions, like direct reports in conversation, are not what they seem. The present tense (which Derrida also, of course, uses) invokes the illusory sense of presence on which the 178 Supplementarity predicts that signs accrete and lose meaning with each new iteration: thus every different temporal180 moment brings novel nuances to words and parts of words, each new context, a different set of nuances, which are also always temporally shifting. It is necessary to pretend that time stops so that such constantly shifting nuances might have clearly determined limits, and so that complex articulations can take place in a time-suspended presence, so the meaning might be (momentarily) stable. Our ability to read/comprehend and write/articulate depends on the pretense of stable meaning. Derrida’s neologism “logocentrism” (which involves faith in the metaphysics of presence, that essence is absolutely present, and that a dualistic word/reality relationship is also rational or logical) is the conviction that a sensible signifier points to a determined- intelligible signified, and the sign itself to a determined referent. However, because they point to something else – signifier and signified, sign and referent are necessarily apart from that to which they point. Thus the operation of the sign itself is caught up in the movement of différance. Linguistic differentiation depends on temporal deferral, just as stable meaning depends on presence. operations of the linguistic sign depend. 180 The preface would announce in the future tense (“this is what you are going to read”) the conceptual content of significance . . . of what will already have been written. And thus sufficiently read to be gathered up in its semantic tenor and proposed in advance. From the viewpoint of the fore-word, which recreates an intention-to-say [un vouloir-dire] after the fact, the text exists as something written – a past – which, under the false appearance of a present, a hidden omnipotent author (in full mastery of his product) is presenting to the reader as his future. Here is what I wrote, then read, and what I am writing that you are going to read. After which you will again be able to take possession of this preface which in sum you have not yet begun to read, even though once having read it, you will already have anticipated everything that follows and thus you might just as well dispense with the rest (D 7). Amidst this discussion of the preface’s temporality problematic which is deeply intertwined with the critique of writing, Derrida’s text erupts into addressing the reader in the second person – “Here is what I wrote, then read, and what I am writing that you are going to read” – the deictic here reverberates with both physical and textual significance as readers are drawn into the conversation. Inferencing opportunities for readers abound in the Derridian text; but many of the weak implicatures resulting from the poetic effects of these opening lines become clear only after the fact, after beginning again, after rereading. 5.5 themes, theses, and “the ability to follow the given thread” (D 63) ~ The opening passage of Dissemination toys with style practices that are conventions of scholarly writing, practices which Derrida importantly also uses; this focus on writing practices begins with the opening line’s parentheses around “donc,” (‘therefore’) the subsequent 179 The Derridian critique of writing is not an argument against the use of language or its conventions. This critique181 (perhaps initially and primarily in La Voix and le Phénomène, Of Grammatology and Writing and Difference, but apparent in all of his writings) like his critique of Husserls geological and weaving metaphors (in “la forme et le vouloir-dire”) do not imply a “negative” or dismissive judgement about these features – Derrida is analyzing them, revealing them, bringing them into view. My phrase critique of writing reflects Derrida’s (opportunistic) choice to use the term writing writ large to displace language more generally. His critique of implications and features of language is not an ethical injunction not to use them, but an examination of their operations and significance – an exposure or outing, at another time one might have said “a linguistic- consciousness raising experience.” Each of the “three” essays was originally published in two parts, one could claim that there are seven essays182 counting the “Hors livres”; they are all unique in genre and therefore it is questionable whether or not they should be called “essays” at all. Elsewhere, for example, we will be enjoined to “pretend”: “Feignons de croire à cette opposition” (G 145). This183 strange imperative may echo Lautréamont’s unusual engagement with his reader (“pay heed,” “beware,” “believe,” “wave aside” 37). italicization of “three,” and the quotation marks around “essay.” Irony reverberates181 throughout the passage: there is the opening line’s claim not to be a book (and a “therefore” that sounds as though it might have been the conclusion of another conversation, the end of a text rather than the beginning). There is an ironic disclaimer that he will not “present” (or introduce) his three essays (a claim undermined by the “Hors livres” clear development of themes whose182 threads are caught up in each of the three texts presented in Disseminations); and there is the opening passage’s abrupt ending with the interesting imperative “Recommençons” to start183 over again. This bit of discourse is an echo of the Platonic text: and not just the most celebrated iterations that might first come to mind from the Timaeus, which is all about beginnings, and famously includes the lines “I must go back once again to the beginning” (Zeyl 48b) and “So let me begin again” (Zeyl 48e), but also references in Charmides (as translated by Jowett as) “let us begin again” (167b) and Laches “I must beg of you Nicias to begin again” (198a). It is like a prior text that a writer like Derrida (and Plato’s translators) might reach for; and as an echo of another text it brings with it a host of associations, not the least of which is irony itself. The implicatures of Derrida’s opening ironies (like the “Recommençons” reminder of the Timaeus) are all tied to various points of interest that provide linkages between a variety of themes, as we shall see. Sperber and Wilson define irony as an echoic text: a type of text for which the paradigm case is reported speech. The disjunct between a speaker’s meaning and someone else’s (evaluation, attitude or interpretation) expressed by the repetition of the original speaker’s 180 Quotation marks being themselves the scholarly short form for “so-called.”184 Sallis states that “a dialogue always makes manifest more than is explicitly said, and what is explicitly said takes185 on its full sense only as a moment within the whole of the mirror play. It is imperative that in seeking to interpret the dialogues we regard them in their character as dramas – that is, that we attend to their multiple dimensions and to the mirror-play which unfolds between and within these dimensions” (19). utterance is the norm in reported speech; similarly irony’s echoes consist of repetition, mimesis or duplication. There are two layers – a doubling, a tension between two things, an appearance and its other. Echo types range from positive (a repetition that reinforces or affirms a prior utterance) to irony, which Sperber and Wilson imagine as an echo which is designed to ridicule the opinion echoed (238-41). However, it is clearly not the case that ridicule as such is necessarily involved in Derridian and Platonic irony. Their ironic deployments set up oppositional tensions, although as Griswold points out irony in Plato is not simply “the opposite of what one is saying” (88). Griswold outlines the extreme complexity of irony in the Platonic text: pointing out that “irony is as much a way of revealing as it is of concealing, a way of teaching and motivating the reader/auditor to look further” (88). In terms of motivation Derrida’s attention grabbing opening line is a way of drawing readers into closer examination of what he is saying. He draws them into not only the question of how the texts collected together in Dissemination were originally published, under what conditions, and where (namely amongst the avant garde Tel-Quel group of writers who were challenging the limits of genre conventions in France), but also into the question of the genre of his texts (hence the italicization of “three” and the quotation marks around “essays”).184 Linck’s article “Unmastering Speech: Irony in Plato’s Phaedrus,” embraces the bewildering and breathtaking complexity of Platonic irony. He observes that “the art of Plato is such that the intricate cohesion of word and action reveals itself through many layers,” and discusses the way “the written text reveals itself as finely crafted in the subtlety of its inner connections” (264). Linck claims that “there is no simple opposition of dissembling veneer and authentic interior in Platonic writing” and that the irony is deployed to make the dialogues “mean more than they say, and yet also seem to mean what they say nevertheless” (275). Like185 a palimpsest, the image of Platonic irony seeps through Derrida’s ironically reverberating 181 Statkiewicz believes that the dialogues are rhapsodic, that is a poetic recitation, and “only when read rhapsodically186 can the dialogue be genuinely protreptic,” that is exhortation or instruction (12). In Statkiewicz’s Rhapsody of Philosophy: Dialogues with Plato in Contemporary Thought, rhapsody is a “mode of187 thinking” which he says is “Plato’s mode, replayed in the texts of Nietzsche, Heidegger, Jacques Derrida, Luce Irigaray, Deleuze, Nancy, Philippe Lacoue-Labarthe”; and he suggests that Plato’s “contemporary ‘continental’ readers” might have plumbed the poetic implications of the dialogues in a way that their Anglo-American colleagues have not yet done (4). opening lines introducing “Plato’s Pharmacy.” In the Platonic text, ironic tensions link multiple points, indirectly connecting and refracting diverse instances – creating a text which seems constantly to shift, to move – not just between two points but a myriad of interrelated points. Linck explains Plato’s use of irony in terms of both motivating readers to think, as well as the inaccessibility of linguistic mastery and certainty. Resolving the irony of a discussion that criticizes writing in written form, and which, in its artfulness, could perhaps only have been written, cannot be accomplished by attempting to figure out what Plato “really means.” In fact, the desire that stirs one to resolve such a paradox is perhaps more important than the destination. . . . Platonic writing exploits our desire for absolute mastery and yet simultaneously denies it in word and deed. We want the analogies to fit perfectly, we want Plato to explain why he both writes and condemns writing. We struggle to make Plato account for himself. In the process, hopefully, we find that he has anticipated our questions, but provides no easy answers. (275-76) Linck’s comments on Platonic irony reflect on Plato’s educational motivation; he remarks that we find in Platonic dialogues a distinction between “a teaching that attempts to impart information and a teaching that struggles to turn around a soul. The concealments of irony may in fact be necessary for this latter kind of teaching” (271).186 Although few seem to have gauged the scope of Plato’s stylistic motives as Linck has,187 there is widespread agreement that the dialogues have educational aims; these inferences are made from the focus on the question-and-answer method, use of the dialogue genre, and the participants espoused values. Sparshott takes this a step further (displacing “Platonism’s” doctrinal-metaphysical focus) when he claims that Plato is a philosopher of education and that his “fundamental concern is with education (paideia) – how knowledge and understanding are to be acquired, developed and transmitted in societies, how teaching is possible” (153). Questions about what is teachable, what is learnable, and how we learn are defining concerns of 182 Plato’s educational motives can be gauged by Socrates’ report (in the Phaedrus) of Thamus’ words: “by telling188 [students] of many things without teaching them, you will make them seem to know much, while for the most part they know nothing” (Hackforth 275a-b). Socrates later underscores that (whether read or recited) texts not engaging the reader or hearer in questioning or exploring do no more than remind those who already know (Hackforth 277e-78a). And implicitly if the writer or reciter is not there to answer questions (in order to contextualize the utterance) the text could do much less than remind the reader who knows. Bakhtin says “The word is born in dialogue as a living rejoinder within it; the word is shaped in dialogic interaction189 with an alien word that is already in the object. A word forms a concept of its own object in a dialogic way” (Dialogic 279). the dialogues as well as being complex matters at the root of Plato’s conversational stylistic. Derrida’s and Plato’s performing or enacting their theories of language in their texts indexes their interest in demonstration, in showing what is being said about language. Plato’s188 understanding of language as always in flux – a view also reflected in Voloshinov’s account of the social conditioning of the sign, his discussion of the grammaticalization of styles, and in Halliday’s grammar – is evident in his ironic-conversational stylistic, which never seems to sit still. This linguistic mobility or animation figures language’s changeability and promiscuity. Plato’s educational motives induce him to show, point to or demonstrate the malleability of the linguistic medium through which we understand the world; and at the same time this animation seems to emblematize the ground-shifting activity of learning. Bakhtin’s novelistic image provides a figuration of this active-semiotic motility. As previously noted, Bakhtin celebrates the dialogic notion of truth that he claims is at the root of the Socratic-dialogue genre. The “internal dialogism of the word” – its retention of different accents, its inter-subjective oscillations – guarantees its indeterminacy and changeability (Dialogic 279). The novelistic image (at the base of parody and close relative to the irony189 witnessed above) is a back-and-forth oscillation between an original thing and its distorted image (like the echo); it is an image that parodically imitates and subverts, thereby creating at least two voices, a dialogic gap between the two, and a movement or activity entailed in negotiating that gap; they are paired elements that stand in relation to each other and inter- illuminate each other (DI 43-46). The metaphors for this semiotic vitality are multiple and diverse. They collocate the motility of the trace, its receptivity and responsiveness (that will engender Derrida’s living tissue “histos” metaphors in “Plato’s Pharmacy”) – on one hand; and 183 Linck observes that “the type of irony Socrates employs in his first speech is not one that is highly visible when190 viewed prospectively. It only reveals itself through retrospection” (266). Halperin too refers to referring to Plato’s “retrospective irony” (100). they indicate the sense of change involved in learning, ground-shifting from one position to a new understanding (that heralds Bakhtin’s focus on “becoming,” a thematic from Plato with pre-Socratic roots) on the other. This movement, which appears as a rhetorical effect in the conversational stylistics of the Platonic and Derridian texts, both emblematizes and involves readers in linguistically mediated intellectual activity (i.e., thinking). It pertains fundamentally to linguistic undecideability and prompts the exigency to reread when what is evident at the first glance changes on review. “Lisons.” Derrida’s first person plural imperative will implore that we “read” (implicitly again) the dialogue in question (D 84). Only repeated rereadings of the dialogues can unravel the twisted strands of stylistically produced meaning in the Platonic text because “the threads of the complicities are almost impossible to disentangle” (D 117). But the exigency-to-reread device, characteristic of the190 conversational stylistic, is not welcome in many genres. Compare, for instance, the reader- writer motivations in two non-ironic genres: Aristotle’s treatise-type discourse versus Wittgenstein’s aphoristic-conversational writing. Completely apart from the question of whether there could ever be a reader who did not have to reread Wittgenstein, because of the thinking his text demands, he frequently directs his reader to reread a prior section (so that something new is revealed in an already-read text, so that the original oscillates in counterpoint to the revision like a novelistic image); and he frames matters consistently as questions, thereby avoiding possible “explicitness” altogether: “But what, for example, is the word ‘this’ the name of in language- game (8) or the word ‘that’ in the ostensive definition ‘that is called . . . ?’” (§38). Readers must make imaginative inferences from such questions; rereading often involves re-tracing previous steps, stumbling over new connections that have been made on old locations. Similarly snippets of dialogue are inference demanding; involvement and active thinking are unavoidable. Whereas Aristotle’s text provides the kind of “propositional knowledge” and supposedly “explicit” information (whether scholarly summary, example, qualification and clarification, 184 Aristotle’s readers must gather together and assimilate the information he provides: the thinking they do is191 independent of his stylistic. Aristotle’s texts are like the outcome of his engagement with the Platonic text – the dialectical process that Cherniss has described; rather than initiating or producing such a process in others, Aristotle’s text records his results of the process that the Platonic text elicits. Plato scholar A.E. Taylor directs attention to the dialogues’ absence of “theses,” “theorems,” and “positive results”192 (kindred of Gonzalez’s “propositional” knowledge) that stem from expectations of the scholarly thesis-and-argument stylistic: “Our very knowledge of the gravity of our spiritual malady will make us all the more unremitting in our determination to make the attempt to escape from our ignorance the great business of life. This, rather than anything more specific in the way of ‘positive results,’ is the conclusion Plato means us to draw from these ‘dialogues of search.’ It has been objected to Plato by unsympathetic critics, as he makes some of his characters object it to Socrates, that such a conclusion is not satisfactory. Socrates, Grote thinks, should have exchanged the easier part of critic for that of defender of theses of his own. He would have found that they could be subjected to a dialectic like his own with effects as damaging as those produced on his rivals’ theories by himself. The objection misses the mark. Plato’s object is not to propound theorems in moral science for our instruction, but to rouse us to give our own personal care to the conduct of our moral life by convincing us of the ignorance we usually disguise from ourselves. . .” (Taylor 29). existential statement, etc.) that we associate with the genres of scholarly writing (the species of pisteis are ethos, pathos and logos), Wittgenstein’s readers must be prompted by implicitness191 and attracted to a reading experience that makes active imaginative thinking demands. Remember Wittgenstein’s statement: “I should not like my writing to spare other people the trouble of thinking. But, if possible, to stimulate someone to thoughts of his own” (x). The understanding or thinking that is produced by the Platonic text does not depend on thesis and argument in Aristotelian terms. Brogan says the same sort of thing about Derrida192 when he states that: “Plato’s Pharmacy” does not pretend to “present a ‘thesis’ about Plato. Derrida, like Socrates, risks the accusation that he turns everything upside down and makes the lower appear to be the higher. Like Socrates, he risks the charge that he never says anything and refuses to be pinned down” (11). However apt Brogan’s commentary is with respect to the similarities between the two texts with their undecidability thematic, the “thesis” claim raises the whole question of the style of thematization that pertains to both the Derridian and Platonic texts. Derrida’s uniquely inflected, parodic-scholarly writing claims multiple “theses” for the “Hors livres” and his point that readers have overwritten the Platonic text – that the history of western metaphysics has translated the Platonic text according to the reductive regime of “Platonism” – qualifies as a thesis. Derrida thematizes the concept of thesis itself when he refers to the difficulty of “reducing a text as such to its effects of meaning, content, thesis or theme” and to the fact that dissemination results in more than one thesis: 185 Explicitly enacting is a kind of oxymoron. [La question de l’ici se trouve explicitement mise en scène dans la193 dissémination (la D 14).] in pointing out a single thematic nucleus or a single guiding thesis, it would cancel out the textual displacement that is a work “here.” (Here? Where? The question of the here and now is explicitly enacted [mise en scène] in dissemination.) Indeed, if such a thing were justifiable, we would have to assert right now that one of the theses – there is more than one – inscribed within dissemination is precisely the impossibility of reducing a text as such to its effects of meaning, content, thesis or theme. Not the impossibility, perhaps, since it is commonly done, but the resistance – we shall call it the restance – of a sort of writing that can neither adapt nor adopt such a reduction. Hence this is not a preface . . . (D. 7-8) Derrida’s attitude towards “the impossibility” of reducing his text to “its effects of meaning, content, thesis or theme” jives with Linck’s observations about ironic speech being “speech that can master itself but not its effects” and his comments about the way “Platonic writing exploits our desire for absolute mastery and yet simultaneously denies it” (269,75). In both Plato and Derrida we find images of language that undermine the possibility of linguistic mastery. The question of theme and thesis is linked (in Derrida’s “thesis” passage) to the process of dissemination (the collection’s title Dissemnation). Like the seed thrown by the farmer’s hand on a windy day, dissemination is the process whereby meanings, arising from factors outside of a writer’s mastery, attach to texts and then grow their own meanings (both the writers own unmastered intentions, and the myriad of contingent historical factors and reader supplements bearing on interpretation). The multiple threads of Derrida’s text form themes and theses, but they are presented according to the requirements of his performative stylistic, often demanding (rather than inviting) reader participation with challenging inferencing obligations. Derrida’s text disseminates. His flamboyant opening line “this (therefore) will not have been a book” both invites readers into the scene of writing and launches the critique of genre conventions of the scholarly preface (or introduction) that continues throughout the “Hors livres.” The “question of the here and now [that] is explicitly enacted” is a temporal thread of the critique of preface that at the same time indicates Derrida’s own performative intentions, and his intention to put into play and act out his critique. Note the repetition of the deictic193 186 Johnson’s excellent stylistic overview in her translator’s introduction addresses Derrida’s summary resistance when194 she observes that “because Derrida’s text is constructed as a moving chain or network, it constantly frustrates the desire to ‘get to the point’. . . In accordance with its deconstruction of summary meaning, Derrida’s writing mimes the movement of desire rather than its fulfilment, refusing to stop and totalize itself, or doing so only by feint. (emphasis added xvi). Sperber and Wilson treat this question in their discussion of explicitness relative nature. The greater the degree of195 “explicitness,” the more strongly manifest the assumption is and the less room there is for hearer inference. “here”: “‘here.’ (Here? Where? The question of the here and now is explicitly enacted.” The repeated attention-grabbing deixis (wobbling in this instance between textual and physical) will have achieved relevance for some readers by this stage: the vacillating physical and textual deictics rally with the opening deictic “this” reinforcing the conversation between reader and writer, reinforcing the implicature that the reader is an actor in an imaginary scene of writing. Derrida’s verb “mise en scène” reminds us that to perform as an actor in French is “jouer,” usually translated as “to play.” Derrida’s famous “play” seems to collocate with act, acting, actor, performance. Sceptical-language theory’s summary resistance is exemplified by the difficulty of summarizing the line: “(Here? Where? The question of the here and now is explicitly enacted in dissemination.)” Note too that “dissemination” in the “thesis” passage is194 not enclosed in quotation marks as one might expect were it referring to the last of the three essays collected together in Dissemination. Readers are invited to consider all the possibilities: to play, to act, and to engage with this text and its multiple ironies. Poetic effects invite (or even demand) substantial reader participation and involvement. Derrida’s ostentatious literary shenanigans perform his themes and theses. When he says “if reading and writing are one, as is easily thought these days, if reading is writing, this oneness designates neither undifferentiated (con)fusion nor identity at perfect rest; the is that couples reading with writing must rip apart” (D 63-64), this is as close as he comes to thematizing the reading-writing focus in his text. The conversational style of thematizing (which is also a style of contextualizing) in the Derridian and Platonic texts often works by the accumulation of weak implicatures: a series of loose ends whose repeat appearance or ironic entanglement brings them into view as they gain emphasis. The weaker the implicatures, the more dependent the text is on context, including what the reader brings to the discussion in background knowledge and effort. As a written genre the conversational stylistic’s dependence on weak implicature195 187 Derrida’s attention seems cued to what linguists call the pragmatic operations of texts which I subsume along with196 other qualities under the term performative. The oblique nature of Derrida’s opening and closings is quite typical. One of the stylistic mechanisms Johnson197 describes as frustrating the desire to get to the point noted above includes “fading in and out. The beginnings and endings of these essays are often the most mystifying parts. Sometimes, as in the description of Plato working after hours in his pharmacy, they are also cryptically literary almost lyrical. It is as though the borderlines of the text had to be made to bear the mark of the silence – and the pathos – that lies beyond its fringes, as if the text had first and last to more actively disconnect itself from the logos toward which it still aspires” (xvii). Living tissues present the contributions of live readers as compared with the textiles of dead writers in the first and198 second opening passages of “Plato’s Pharmacy.” makes it a style of implicitness that accommodates a socially gestated linguistic sign to the pragmatics of a certain kind of reader-provoking writing. It is an extremely taxing stylistic196 (for readers) rebelling, as it does, against the mythical-style-of-science ideals of clarity and transparency; it wants to puzzle its readers, wants to make them stop and to reread. Derrida’s style of thematizing in “Plato’s Pharmacy” includes traditional weaving and spinning metaphors (intertwined with non-traditional filial metaphors of webs and tissues) in a complete co-mingling of style and argument. If reading is writing, it is not surprising that the conventional weaving metaphor – representing both writing and storytelling – also represents the operations of reading, the textual supplements provided by readers. In a typically oblique opening gesture, the unnamed preamble to “Plato’s Pharmacy” offers an image in which the woven threads of a text are both the fabric produced by a writer and the embroidery of a reader following “the given thread” (63). Note below the underlined living tissue metaphors197 198 intertwined with weaving and spinning metaphors: [unnamed preamble:] The dissimulation of the woven texture can in any case take centuries to undo its web: a web that envelops a web, undoing the web for centuries; reconstituting it too as an organism, indefinitely regenerating its own tissue behind the cutting trace, the decision of each reading. There is always a surprise in store for the anatomy of physiology of any criticism that might think it had mastered the game, surveyed all the threads at once, deluding itself, too, in wanting to look at the text without touching it, without laying a hand on the “object,” without risking – which is the only chance of entering into the game, by getting a few fingers caught – the addition of some new thread. Adding here, is nothing other than giving to read. One must manage to think this out: that it is not a question of embroidering upon a text, unless one considers that to know how to embroider still means to have the ability to follow the given thread. If reading and writing are one, as is easily thought these days, if reading is writing, 188 Derrida says that “he who through ‘methodological prudence,’ ‘norms of objectivity,’or ‘safeguards of knowledge’199 would refrain from committing anything of himself, would not read at all” (D 64). this oneness designates neither undifferentiated (con)fusion nor identity at perfect rest: the is that couples reading with writing must rip apart. [underline added] (63-64) Derrida makes use of these metaphors to depict the way that readings create their own textured tapestry hiding and shaping what comes into view. The reader-overwriting thematic appears almost “explicitly” when the critic cannot help “getting a few fingers caught,” cannot help adding “some new thread.”199 5.6 weaving, meaning, and le vouloir-dire ~ Derrida’s deployment of Plato’s favoured weaving and spinning metaphors presumably reflects the Platonic text’s appropriation of these Homeric tropes; but these textile metaphors also characterize western metaphysic’s spin – the readings or voice-overs known as “Platonism.” Repeat performances of weaving and spinning allusions disseminate countless threads – stories about stories, writing about writing, talk about talk – that Derrida is keen to capitalize on in “Plato’s Pharmacy.” But what some critics seem to have missed in their readings of “Plato’s Pharmacy” is Derrida’s extreme deference to Plato. Listen to his tone of voice talking about the Phaedrus, where he begins with barely controlled sarcasm (about those voice-overs): “We are speaking of the Phaedrus that was obliged to wait almost twenty-five centuries before anyone gave up the idea that it was a badly composed dialogue” (66). Indeed as a counterpoint to western metaphysics’ “ludicrous” arguments about the Phaedrus, Derrida describes this “magisterial” dialogue in reverential tones: . . . The hypothesis of a rigorous, sure, and subtle form is naturally more fertile. It discovers new chords, new concordances; it surprises them in minutely fashioned counterpoint, within a more secret organization of themes, of names, of words. It unties a whole sumplokhv [intertwining, complication, interweaving] patiently interlacing the arguments. What is magisterial about the demonstration affirms itself and effaces itself at once, with suppleness, irony and discretion. This is, in particular, the case – and this will be our supplementary thread – with the whole last section (274b ff.), devoted, as everyone knows, to the origin, history, and value of writing. That entire hearing of the trial of writing should some day cease to appear as an extraneous mythological fantasy, an 189 For example, variations on the attention-grabbing phrases “there is nothing outside the text” and “there is no200 outside the text” that herald the Derridian critique of writing are interwoven into the “Hors livres.” Derrida writes about Hegel’s preface conundrums: “the preface is ruled out but it is still being written. . . . To allege that there is no absolute outside of the text [qu’il n’y a pas de hors-texte], is not to postulate some ideal immanence . . . .The text affirms the outside, marks the limits of this speculative operation, deconstructs and reduces to the status of ‘effects’ all the predicates through which speculation appropriates the outside. If there is nothing outside the text, [s’il n’y a rien hors du texte] this implies, with the transformation of the concept of text in general, that the text is no longer the snug airtight inside of an interiority or an identity-to-itself . . . In either case the preface is a fiction . . . ” (D 35-36) (la D 42). About Lautréamont he says: “there is nothing but text, there is nothing but extratext, in sum and ‘unceasing preface’” (D 43). Derrida’s ostentatious phrases “s’il n’y a rien hors du texte” and “qu’il n’y a pas de hors-texte” from “Hors livres” are reminders of Of Grammatology’s “il n’y a pas de hors-texte” and “qu'il n'y a rien hors du texte” and Derrida’s critique of writing (G 227, 233). appendix the organism could easily, with no loss, have done without. In truth, it is rigorously called for from one end of the Phaedrus to the other. Always with irony. But what can be said of irony here? What is its major sign? (67) Perhaps it is not surprising that Derrida might wish to re-perform Plato’s magisterial yet discreet irony – that “affirms itself and effaces itself at once.” As Walter Brogan notes: One of Derrida’s favourite “metaphors” for deconstruction is weaving, a metaphor that is also prominent in Plato’s dialogues. The web of writing is not constructed along the lines of simple hierarchies that interrelate fixed points that we call concepts and words. The woven text has a texture that stretches and shrinks, can expand, can be grafted onto, can fold, warp, and unravel. To follow the patterns and interlacing of the composition requires the weaver’s art of looping and knotting. The surface of the woven cloth dissimulates its complex and intricate networking. (12) Weaving a tapestry or spinning a yarn might be a better image for the kind of thematization that takes place in the Platonic and Derridian texts: these metaphors are both form and content conjoined, functioning at multiple layers at the same time, achieving relevance by repeat appearances, by attention-grabbing performances that depend on audience participation. Such a style of thematization revels in linguistic implicitness and ostension, calling into question a theory of language focussed on conceptions of “explicitness” and descriptive reference. Like the seamless surface of a moebius strip, the ligature of form and content, of style and argument is undetectable. The interweaving, the intertextuality – that tells us about the supplementary operations of reading and writing, about how language works – is presented by Derrida’s metaphors and his own pronounced interweavings of other texts. As we shall200 shortly see, what is significant here is not the (typical) scholarly practise of bringing together the voices of Hegel and Lautréamont (in the “Hors livres”), Plato, Mallarmé and Sollers (in the 190 collection) and also his own texts. What is decisive is that Derrida’s style of presenting other texts puts newcomers at a distinct disadvantage. We will follow one such thread in order to reveal the way that Derrida’s interweaving works, the way his own neologism can remain hidden from the first glance, the way that his style of interweaving is a warrant for rereading as a practice. When Derrida says that “the is that couples reading with writing must rip apart,” and when he highlights our “ability to follow the given thread,” he seems to suggest that finding and severing the original text and its voice-overs, including one’s own voice, is possible. But the Derridian critique of intentionality makes such an implication ironic. If reading is writing, it is not clear how can we distinguish between them because only the intentions of the writer and reader can distinguish the reading from the writing. Of Grammatology’s discussion of le propre and proper names, the play on le vouloir-dire vis-à-vis Rousseau, and Derrida’s critiques of Husserl’s notion of intentional meaning, call into question the concept of writer intentions and the question (considering Freud) of mastering even one’s own. What is of concern about this question of intention for “Plato’s Pharmacy” is not just whether or not the writing can be severed from the reading (which it can’t, despite such severing being an important scholarly ideal that Derrida aspires to practice however much he here toys with it). The impossibility is made evident in his text’s following line, as we shall see. What concerns us is the way that this suggestion of distinguishing the reading from writing invites review of Derrida’s critique of intentionality and is tied up with the expression “le vouloir-dire,” woven into “Plato’s Pharmacy.” This affiliation reveals the intertwining of lexico-stylistic choices and thematic argument. Derrida’s analysis of Husserl’s use of metaphors, especially the weaving metaphor, in his critique of Husserl’s faith in intentional-meaning in “La Forme et le vouloir-dire: Note sur la phénoménologie du language,” and in La Voix et le Phénomène (where he uses the expression le vouloir-dire, instead of la signification, to translate Bedeutung), discredits Husserl’s distinction between what is “meant” and what is “indicated.” In Husserl’s distinction, meaning is the interior and ideal essence of an outer shell that indicates. Derrida argues that 191 See especially “Le vouloir-dire come Soliloque” in La Voix et le Phénomène .201 Allison translates “La Forme et le vouloir-dire: Note sur la phénoménologie du langage.” as “Form and Meaning:202 A Note on the Phenomenology of Language.” Throughout the essay he translates “le vouloir-dire” as “meaning.” Replacing la signification with le vouloir-dire seems to focus attention on an individual’s intentions, what individuals want, wish, mean or intend to say. Bass has followed Allison in the translation of the title, but draws attention to the issue emphasizing “the relation of meaning to speech, to saying (dire)” but diffusing the attention to speaker intentions – the intended meaning (MP 159). A better translation might be intended-meaning which is what guides Johnson’s translation of “un vouloir-dire” (la D 13) as “intention-to-mean” (D 7). Both Spivak and Johnson note the importance of the phrase in their translations by frequently including the French phrase in the text, also see OG (200). See also Derrida’s toying with the idea that he can determine Plato’s intention in “Plato’s Pharmacy”: “Let us203 return to the text of Plato, assuming we have ever really left it. The word pharmakon is caught in a chain of significations. The play of that chain seems systematic. But the system here is not, simply, that of the intentions of an author who goes by the name of Plato. The system is not primarily that of what someone meant-to-say [un vouloir-dire]. Finely regulated communications are established, through the play of language, among diverse functions of the word and, within it, among diverse strata or regions of culture. These communications or corridors of meaning can sometimes be declared or clarified by Plato when he plays upon them ‘voluntarily,’ a word we put in quotation marks because what it designates, to content ourselves with remaining within the closure of these oppositions, is only a mode of ‘submission’ to the necessities of a given ‘language.’ None of these concepts can translate the relation we are aiming at here. Then again, in other cases, Plato can not see the links, can leave them in the shadow or break them up. And yet these links go on working of themselves. In spite of him? thanks to him? in his text? outside his text? but then where? between his text and the language? for what reader? at what moment? To answer such questions in principle and in general will seem impossible; and that will give us the suspicion that there is some malformation in the question itself, in each of its concepts, in each of the oppositions it thus accredits” (95-96). form (i.e., indicator-signifier) and content (i.e., intended-meaning-signified) are inseparable (or interwoven) even within Husserl’s own account. As Husserl’s distinction (meaning-201 indication) is ultimately based on speaker-intentions, Derrida mocks the distinction by using the ironic phrase “le vouloir-dire,” instead of la signification, to represent this intention-laden meaning. The substantive produced by a French infinitive verb is an emphatic and contrived202 form that calls attention to the activities it describes: here, the double activities of saying [‘dire’] and wanting (or wishing, intending, meaning) [‘vouloir’]. Both verbs stubbornly implicate a speaker in the noun – an agent of the uttering [‘dire’] and the intentions implicit in the (wanting, wishing, intending, meaning) valences of vouloir. By using the phrase, le vouloir-dire, Derrida playfully mocks Husserl’s desire for intentional-meaning, a desire for essential, contained and controlled meaning, external to the vicissitudes of linguistic signification and human (un)consciousness. 203 Note that in “La Forme et le vouloir-dire” Derrida not only critiques Husserl’s use of the weaving metaphor but also playfully and persistently uses it himself. At the beginning of the essay, for example, he states that the “the concept of form could serve as a thread to be followed in phenomenology’s elaboration of a purifying critique” [underline added] (157). Since it is precisely the examination of metaphors, and in particular the weaving metaphor, that Derrida 192 Derrida’s typically rhetorical critique highlights the way Husserl’s weaving metaphors are tangled up with204 geological metaphors (i.e. “substratum”). Derrida states that “the discourse on the logic of the discourse is entangled in a play of metaphors. The metaphor of the stratum, as we shall see, is far from being the only one” (MP 160). Drawing attention to the implicitly strange “logic” of imbricating discursive indication with non-discursive meaning, he says: “the interweaving (Verwebung) of language, the interweaving of that which is purely language in language with the other threads of experience constitutes a cloth. The word Verwebung refers to this metaphorical zone. The ‘strata’ are “woven,” their intercomplication is such that the warp cannot be distinguished from the woof . . . for cloth means text. Verweben here means texere. The discursive is related to the nondiscursive, the linguistic ‘stratum’ is intermixed with the prelinguistic ‘stratum’ according to the regulated system of a kind of text” (MP 160). Derrida plays on the ambivalence and complete instability in the meaning of form and its various oppositions in205 “La Forme et le vouloir-dire: Note sur la phénoménologie du langage.” It is not clear what “la forme” signifies: whether an outer shell, indicator or signifier through which an inner essence or content is conveyed, or is an essence, ideal, intended, signified content. He highlights a form/content binary which deconstructs “Platonism’s” term form (as essence) for åÉäïò. The binaries caught up in his play include: form/content, form/meaning, Platonic essence/secondary representation, form/sign. here undertakes, such overt use of the same weaving metaphor (that is the subject of his critique of Husserl’s account of language) creates an ironic tension. This practice of employing terms204 he critiques is so consistent in the Derridian text that it is relevant in its own right and has a counterpart in some of Plato’s apparently contradictory formulations. Derrida also ironically notes that “even if the word ‘form’ translates several Greek words in a highly equivocal fashion, nevertheless one may rest assured that these words all refer to fundamental concepts of metaphysics” [underline added] (MP 157). Not only are the foundations of metaphysics (and foundation itself) thus compromised by equivocation, but because he plays on the two opposing meanings of the term form rather than stabilizing it, Derrida’s deployment of the ambiguous term “form” in the title “La Forme et le vouloir-dire” also oscillates in Socratic (and Bakhtinian fashion).205 The phrase le vouloir-dire, which appears not only in the title of the essay, but also in the titles of several chapters of La Voix et le Phénomène , is thus closely affiliated with not only Derrida’s critique of Husserl’s intentional-meaning, but also the weaving-metaphor paradigm of how language works. It is particularly significant to this interrogation of Dissemination that the ironic and playful le vouloir-dire had already achieved celebrity status (outside of the essays on Husserl) in Of Grammatology where Derrida makes quite a joke of “intention” in connection with Rousseau. Note, for example, the following instances of what Rousseau meant, wanted or wished to say (as though Derrida, or even Rousseau himself has a clear understanding of these intentions): “Rousseau says it without wishing to say it [sans vouloir le dire]. What he wishes to 193 While Spivak’s translation accentuates Derrida’s playful implications of wanting in “vouloir dire,” an appropriate206 gesture for highlighting his irony, these instances could all be translated using the verb intended or meant. Later in the same paragraph there is continued playful focus not only on what Rousseau would wish to say but also to think and to efface: “Rousseau would wish to think,” [“voudrait penser”], “He would wish to pretend” [“voudrait faire comme si”] and “Rousseau would wish to efface” [“voudrait effacer”](OG 201, G 286-87). Note that in the original Tel Quel publication the separation of what I have called the “unnamed preamble” and this207 section labelled “I” did not exist. Thus the sense of the referent to the line “to a considerable degree we have already said all we meant to say” may have been construed as the oblique poetic offering preceding in the text that preceded it. One might speculate that separating these sections and creating a second “opening line” was rhetorically attractive to Derrida. say [ce qu’il veut dire] is: accessory accessory, accidental accident, the exterior outside, the supplementary evil or the accessory supplement” (OG 200-201, G 286). Observe the unlikelihood that Rousseau ever imagined, wished, meant, intended or wanted to say anything of the sort. It is worth callibrating the interweaving of le vouloir-dire with critiques of intention206 and the weaving metaphor along side Halliday’s insistence that grammar is emphatically “lexicogrammar”: bits of discourse with specific lexical affiliations. Derrida’s unusual and never “explicitly” defined neologisms bring their own histories and disseminate weak implicatures that gather force for those who recognize them while remaining hidden from newcomers. 5.7 reading Derrida writing about Plato ~ In order to understand what Derrida is saying about Plato, it is necessary to embrace the relevance of the repetitions, the implications of his conversational stylistic: it is not an outer garment which can be removed so that the inner content can be laid bare. The verb form of “le vouloir-dire,” laden with playful irony, makes a repeat performance in the second opening line of “Plato’s Pharmacy” (in a section labelled “I” in the (book) publication of Dissemination.)207 I To a considerable degree, we have already said all [“nous avons déjà tout dit”] we meant to say. Our lexicon at any rate is not far from being exhausted. With the exception of this or that supplement, our questions will have nothing more to name but the texture of the text, reading and writing, mastery and play, the paradoxes of supplementarity, and the graphic relations between the living and the dead: within the textual, the textile, and the histological. We will keep within the limits of this tissue: between the metaphor of the histos and the question of the histos metaphor [“iJstvoV” web, tissue, occurring with sense ‘tissue’ in various biological terms]. Since we have already said everything [“nous avons déjà tout dit”], the 194 Disseminations’ clearly presented reminders of the original publication contexts also highlights Derrida’s weaving208 activity, emblematizing the operations of language. Johnson notes that the epigraph of “Plato’s Pharmacy” (which is an entry from the Greek to French Dictionary Bailey) is linked to the final discussion of the “Hors livres”: “it should be noted that the Greek word kovlafoV, which here begins the essay on Plato, is the last word printed in Littré’s long definition of the French word coup, with which the Hors-livre has just playfully left off” (63). reader must bear with us if we continue on a while. If we extend ourselves by force of play. If we then write a bit: on Plato, who already said in the Phaedrus that writing can only repeat (itself), that it “always signifies (semainei) the same” and that it is a “game” (paidia). (D 65, la D 73) The marked (in italics) reappearance of “vouloir-dire” in this high-profile line emblematizes the weaving metaphor and the interweaving of Derrida’s other texts that has been at play from the opening lines of Dissemination. When Derrida says “if reading and writing are one, as is208 easily thought these days” it seems to echo Of Grammatology’s “everything that for a least some twenty years that tended toward and finally succeeded in being gathered under the name of language is beginning to let itself be transferred to, or at least summarized under, the name of writing” (6). He again brings into play writings where he has made similar claims, especially Of Grammatology, where the interdependency of reading and writing are treated programmatically. The second and third openings to “Plato’s Pharmacy” illustrate the way these openings were very intentionally heralded by the “Hors livres” and interwoven with it. Derrida’s interweavings spawn endless disseminations, implicatures that gather strength: for instance, the interweavings of his text exemplifies the Platonic text’s stylistic virtuosity and salient weaving metaphors, and they model not only how language works – as Bakhtin says our words are always overflowing with other peoples words – but specifically the supplementary operations of reading. The “vouloir-dire’s” italicized appearance in this opening line – “To a considerable degree, we have already said all we meant to say” – acts as a little irony flag signalling intentional-mastered meaning (poking fun at attempts to control meanings). The playful suggestion at the beginning of a text that the writer has already said all that he meant to say, as if to suggest, at the outset, that he doesn’t have anything else to say (why say more?) is underscored by the repetition (the hyperbolic claim of having already said everything) which echoes in the second paragraph “since we have already said everything” [“nous avons déjà tout 195 Derrida notes: “our questions will have nothing more to name but the texture of the text.” However a considerable209 list of items his questions will “name” follows: “la lecture et l’écriture, la maîtrise et le jeu, les pardoxes de la suuplémentarité aussi et les rapports graphiques du vivant et du mort: dans le textuel, le textile et l’histologique” (la D 73). Certainly, Derrida seems to have quite a lot to say, since a very long two part essay follows these opening comments. Derrida’s written performance – like that of Plato’s Phaedrus – seems to contradict what the second opening line seems to say. Also see Protarchus’ suggestion that they must “repeat the [lovgon] all over from the beginning” [ejx ajrcV210 ejpanalabei:n dei:n lovgon] (D. Frede Philebus 66d). dit”]. The possible referents for what Derrida had already meant to say include the unnamed preamble, to “Plato’s Pharmacy,” the preface or “Hors livres,” and previous texts which as noted are frequently interwoven into Dissemination. (Or perhaps “we” philosophers, have already said everything; perhaps Plato already said everything.) The weak implicatures of these ironies and repetitions ripple through this text with continually wavering meanings. After the209 (second) opening section is a new section titled “1. Pharmacia,” and a “new” but conspicuously familiar third opening line which referentially calls attention to repeated openings: “let us begin again” [Recommençons.]. Therefore the dissimulation of the woven texture can in any case take centuries to undo its web” (D 65, la D 74). “Recommençons” draws attention not only to repeated beginnings, but its repetition here both recalls the first page of Dissemination and the enigmatic opening line which invited readers into the scene of writing, and it reminds us of Appollodorus’ line “I’d better begin again at the beginning” in Symposium (clearly a phrase Plato translators reach for) (Joyce 174a). The third opening’s “the dissimulation of the woven210 texture can in any case take centuries to undo its web” exactly reiterates a line in the unnamed preamble two pages earlier: this is repetition drawing attention to itself in a loud voice, repetition screaming, vying for the spotlight. The weak implicatures arising from these echoes – the centuries it takes to unravel the Platonic text, for example – gather together and accumulate focal stress. The prominent role that repetition plays in Derrida’s conversational stylistic dovetails with the rereading exigence. Derrida begins “Plato’s Pharmacy” with the abstruse line “a text is not a text unless it hides from the first comer, from the first glance, the law of its composition and the rules of its game” (D 63). This first opening line of “Plato’s Pharmacy,” like its counterpart in the “Hors livres” (this therefore will not have been a book), might have needed 196 Linck’s comments that the ironies were only apparent after the fact in the Phaedrus parallel Halperin’s remarks211 that “the erotics of oral narrativity is accessible only to a careful reader of the Symposium’s text” (109). For Halperin erotics have a structure of desire that “springs from a sense of lack” and “pursues a fullness of being212 that forever eludes it” (a dialectic of presence and absence) (101). several readings. Derrida’s stylistic directly imitates the Platonic text when it sometimes invites and sometimes demands several readings. The syntax of this opening line invites interpretation of the first clause as though it were a line of poetry: “a text is not a text unless it hides from the first comer,” a sentiment that keeps re-surfacing. Looking back we see that “the is that211 couples reading with writing” will only “rip apart” if one has the “ability to follow the given thread. That is,” he says playfully, “if you follow me, the hidden thread” (D 63). The thread will be hidden because, as he tells us in the opening line, a text hides the laws of its composition and the rules of its game from the first glance. This quality of a text (being hidden from the first glance) nicely characterizes the interfused gestures of the Platonic text and what Halperin calls “the erotics of rhetoric” – a quality that we have seen demonstrated or performed in the opening lines of both “Plato Pharmacy” and Dissemination (124). However, we must already have212 read the Platonic text to recognize such an allusion. This reading-recognition caveat is “thematized” almost immediately. In the meantime it appears that Derrida was only fooling around with the concept of intentionality when he suggested that the reading and the writing will rip apart. He tells us immediately afterwards that “one must then, in a single gesture, but doubled, read and write” at the same time: a statement that tells us about the nature of writing and reading more generally. We will always be rereading just as we are always compelled to repeat (D 64). This aspect of language use, the exigency to both read and write in a single gesture, is also signalled by Bakhtin’s notion of dialogism (the multiply-subjective, heteroglossic disposition of language use) as Tannen observes when she reports on the consensual and jointly produced nature of conversation (27). She highlights the title of Duranti’s article “The audience as co-author” (a title that Derrida himself could have used) to argue that “both speaking and listening include elements and traces of the other” (27-28). This reading-writing necessity (entailed in Derrida’s critique of writing) is here intertwined with the reader-writer focus 197 Statkiewicz complains that modern readers of Plato often read collections or anthologies which excerpt the213 dialogues, arguably a complete bastardization of their minutely orchestrated textual operations (6). It is interesting the Booth fails to recognize such elements of irony in A Rhetoric of Irony.214 (occasioned by the voice-overs), the weaving metaphor, and Derrida’s sustained focus on metaphors in “Plato’s Pharmacy” which he says “must be tirelessly questioned” (78). The repetition of “recommençons” makes clear that the “presentation” of the “three ‘essays’” in Dissemination is a carefully contrived and interwoven text, despite ironic disclaimers to the contrary. The three emphasized opening lines in “Plato’s Pharmacy” call attention to what constitutes a text, an “essay” or a “book,” and where a text begins. Derrida’s conversational213 stylistic displays or illustrates his views about language. The threads of Dissemination create a web in a stylistic of implicitness which highlights the role of repetition in language use (letters, words, phrases, sentences, stories, histories, whole texts), especially evident in conversation as the latching, repetition, and chiming of participants’ voices in Tannen’s transcripts show. These multiple repetitions and interweavings perform familiar features of Derrida’s critique. The imperative “recommençons” invites the reader, not to observe, consider, or note, (as one might expect of an imperative in formal academic prose), but to begin again, to do something with the writer, to participate in the process of writing itself, as a kind of co-author. 5.8 Derrida’s reader and the addressee of “Plato’s Pharmacy” ~ The problem of not being able to recognize what one does not know is particularly vexed in the presence of irony: the feature of the Derridian and Platonic stylistics that so complicate readings. While from a sceptical perspective irony is enormously rich – at the same time symbolizing play of meaning, producing ironic tensions, forming webs of relations, and214 refracting multiple interconnected meanings – it is also challenging; it imposes both the inferential task of how to read it, where readers are being directed to look, and the larger difficulty of recognizing it at all. Irony dangles various possible threads of meaning, diffusing clarity, permitting multiple potential implicatures, and leaving a great deal of work for readers. Carefully orchestrated ironic tensions spin arrays of weak implicatures: many weakly manifest assumptions, rather than obviously relevant ones. But in order to draw appropriate inferences, 198 This said, the specific addressee Derridia’s text identifies might not be the only target audience; bystanders and215 overhearers might also be included but positioned differently than the addressee. one requires certain background knowledge; and while weak ironic implicatures build on background information, the ironic tension must be recognized to begin with. The ability to recognize Derrida’s ironies somewhat defines the addressee of his text. For instance, one must know what a book is and that Dissemination is a collection of “three ‘essays’” in order to recognize some of the irony in the opening line of “Hors livres.” Similarly, one must be able to recognize le vouloir-dire as an ironic cue in the second opening of “Plato’s Pharmacy” in order to understand the playful context that it establishes. Arguably, readers who do not recognize such cues are not the addressees of these texts: a fact which positions the first-time reader in an odd spot, almost like a bystander or overhearer to a conversation. The Derridian and Platonic215 texts are designed for rereading: “a text is not a text unless it hides from the first comer, from the first glance.” The third opening to “Plato’s Pharmacy” provides some interesting cues about the reader’s knowledge of the Platonic text: Therefore the dissimulation of the woven texture can in any case take centuries to undo its web. The example we shall propose of this will not, seeing that we are dealing with Plato, be the Statesman, which will have come to mind first, no doubt because of the paradigm of the weaver, and especially because of the paradigm of the paradigm, the example of the example – writing – which immediately precedes it. We will come back to that only after a long detour. (underline added, D 65) Derrida here addresses a very specific and tiny audience – “the example . . . seeing we are dealing with Plato . . . which will have come to mind first” would be the Statesman (emphasis added). This passage presupposes that the audience addressed will recognize the weaving metaphor’s role in the Statesman: an esoteric topic for the 1968 Plato-studies community, the only audience in which the Statesman would likely have had such currency. Derrida presupposes not only that his reader is familiar with the Statesman (not a commonly read dialogue except by Plato scholars), but also that the reader would have so much remarked the weaving metaphor that it would come to mind first (a distinctly literary focus); and he 199 A Parisian colleague has suggested to me that the audience familiar with the Statesman might be larger in Europe216 where it appears that classical texts were displaced more slowly on university curricula than in North America. Addressee, bystander and overhearer are positions in Herbert Clark’s discussion of “audience design” in Arenas of217 Language Use.. Arguably Derrida displaces academic writing’s stance of neutrality with a stance of treating his reader as an equal,218 a friend, someone he is having a chat with, or a serious conversation punctuated with congenial humour. He assumes that his reader is eager, diligent, playful, and curious, and ready for a challenge or duel. presupposes that the reader would be familiar with “the paradigm of the paradigm” (writing). While the audience who might have recognized these allusions (who might have thought of the Statesman as exemplary of the weaving metaphor and who might have wondered why he wasn’t using it in preference to the Phaedrus) doubtless amounted to more than twenty souls (but not many more), it is doubtful how many of them would have also been readers of Derrida (in 1968 when this text was published). Derrida’s move here amounts to pragmatically situating most216 readers of this text in bystander or overhearing roles, rather than as addressees. The fact that217 Derrida cites the entire “paradigm” passage he is referring to in a footnote, and then provides an explanation of the relationship of this passage to the weaving metaphor indexes, Derrida’s awareness that his audience might not know these things since he is not always so helpful to his readers.218 In Sperber and Wilson’s relevance model, presuppositions cue background information. User-friendly presuppositions like this one – which tells us what we need to know – is a way of politely informing readers of the kinds of things they need to already know. Presuppositions require reader-hearer inferences. Here Derrida’s presuppositions alert readers that the addressee of “Plato’s Pharmacy” is intimately acquainted with the Platonic text: in other words, as bystanders or overhearers they might not recognize things without this background. The footnote kindly provides an indication of where to start looking for the material and even why it might be interesting, why Plato might be worth reading. These are relevant implicatures; the inferences we make about the meaning of a text depend on what we understand as relevant information. This audience positioning evinces another aspect of Derrida’s “playful” relationship with his reader, which is quite distinct from the stance of neutrality of scholarly writing (e.g., “this text assumes a knowledge of . . .”). He invites his reader to play his game. It 200 Derrida’s citation here is a performative-rhetorical effect, that typifies the Derridian style. Derrida’s use of citation219 in Dissemination warrants its own study – a matter too large to undertake here. Since his translation is ripped out of its context and unreferenced, and since Plato doesn’t “speak” anywhere we don’t even know how to find out what was said to be a paidia. In any case, the context established by this line is one of playful irony, not least because, as we have noted, Derrida’s writing “a bit on Plato” here results in a very long two-part article. While logography seems to mean speech writing generally, the passage actually contains a reference to a speech220 writer (logographon) and the written speech of Lysias the speech writer is thematic from the outset of the dialogue. The passage also refers to writing (gravfein) however; and perhaps this is what logography is supposed to mean here. is important to keep in mind the playful and ironic tone of the short second opening (cited above), which immediately precedes this third-opening (Statesman) passage. It refers to play: “if we then write a bit: on Plato, who already said in the Phaedrus that writing can only repeat (itself), that it “always signifies (semainei) the same” and that it is a “game” (paidia).” The sense of being involved in a “game” permeates this (subsequent) third opening.219 The previously discussed rereading exigence which results from a variety of rhetorical gestures and which may be a defining feature of the conversational stylistic (in Plato, Derrida and Wittgenstein) sometimes acts as a written counterpart to oral conversation’s repairs, efforts to correct misunderstandings where it is better to repair an error than to over-inform. A few pages after the three openings of “Plato’s Pharmacy” we find the (previously noted) imperative to read “Lisons,” translated as “let us read this more closely.” Here Derrida continues “at the precisely calculated centre of the dialogue – the reader can count the lines – the question of logography [‘logographie’] is raised (257c)” (D 68, la D 84). This very notable moment in the dialogue is the pivot point when Socrates turns from making speeches to analyzing them. But220 the strange invitation to count the lines of the dialogue should give readers pause. And it is not the comical image of Derrida bent over his text counting the lines of the dialogue under his breath that is conversation stopping. But the reference to counting the lines in order to determine “the precisely calculated centre of the dialogue” should attract attention. Since the number of lines in a dialogue are not uniform, it is not a consistent measure. Scholars familiar with translations of the Platonic text – especially those, like Derrida, who work with the Greek on facing pages like the French “Robin” and the English “Loeb” editions – would be aware that the number of Greek “lines” is different in each publication with different page sizes (so much 201 It depends on how long the lines are to begin with, which depends on such mundane material matters as page and221 font size. If readers are reading translations, the number of lines will differ quite markedly (because of the Greek syntax and punctuation). The precisely calculated middle of the text in terms of the lines would be different in every different publication and translation and especially different in French, English and Greek texts. In Derrida’s Léon Robin version of the text, the page numbers are 1-96 and the passage referred to appears on page 56 – clearly not the middle which would have been 48. the more so the translated lines). The mid-point in the Stephanus numbers – the standard221 referencing practice for the Platonic text which Derrida uses and which is itself variable – is 253d (279c-227=253d) not as Derrida here claims 257c. While this may be the centre of the dialogue in some sense, such a centre has nothing to do with the number of lines. Derrida is involved in a very playful, challenging, trying-to-trip-you-up game with his readers. His line-counting joke is followed guilelessly by a remark Phaedrus makes about the derogatory connotations of the epithet “speech writer” which might erroneously invite us to think that speech writing (logography) itself were considered shameful. Derrida’s text acts as if he were implying that the reputation of speech writers in the dialogues were an indictment of writing – as if Derrida believed that Plato was condemning writing (as “Platonism” has determined). But irony is a Janus-faced device: on the one hand creating the impression that the writer shares a view with the reader, something they have in common (Sperber and Wilson 242); on the other hand, failure to recognize the ironies results in misreading the text. Derrida appears to be willing to mislead his reader here (or allow his text to be misread) – which is just what he (quite justifiably) accuses Plato of doing. In order to understand the relevance of the shamefulness of speech writing, one would have to know the context in which Phaedrus makes this statement, what was said before and afterward, how Socrates responded, and whether or not the dialogue concluded that speech-writing was indeed shameful (which of course, it didn’t). In fact, Derrida had, at this point, (on the preceding page) just finished saying that “only a blind or grossly insensitive reading could indeed have spread the rumour that Plato was simply condemning the writer’s activity” (D 67). Thus we can see in this Derridian writing – where readers end up getting their fingers caught in the threads of the text – not only a reflection of the Platonic stylistic but the palimpsest of a mutually participative image of communication, where writers point and readers 202 probe, and an image of the ostensive-performative operations of language use both at work and on display. We also catch a glimpse of Derrida’s indirect method of thematizing the effect of the voice-overs of western metaphysics on the Platonic text: the idea that Plato was simply condemning a writer’s activity, for example. The opening clause of the “Plato’s Pharmacy” (“a text is not a text unless it hides from the first comer”) is less a general comment about language or writing, than it is a summary of a very specific kind of Platonic writing that Derrida emulates and re-performs. This Plato-Derrida pairing lines up with Halperin’s reading of Plato “not as a metaphysical dogmatist but as a kind of deconstructionist avant la lettre, a cunning writer fully alive to the doubleness of his rhetoric who embraces différance and who actively courts in his writing an effect of undecidability” (118). We have already witnessed the extreme irony in the banishment of Poetry episode from Book 10 – irony vacillating so much that we might have not known what to make of it. Statkiewicz posits the possibility that Plato expects his ironies to be recognized; he writes: “when Plato introduced – in a poetic form – his quarrel with the poets, when he rejected – in mimetic mode – mimesis, he certainly did not think that generations of readers would take his play all too seriously (or perhaps not seriously enough!)” (6). Derrida’s playful imperative to (re)read and to count the lines is a corrective gesture: a sharp reminder of the knowledgeable addressee to whom “Plato’s Pharmacy” speaks, a rather serious way of being playful, or playful way of being serious. Derrida’s imperative to read more closely doesn’t just serve as a caveat to know the text he is analyzing, but also indicates that the Platonic text is worth reading and rereading, that it is worth listening to. One might also wish to reread the Republic more carefully before running around announcing that Plato banned poetry from his ideal political state. Statkiewicz argues that Plato’s writing “should be understood as a dialogue/confrontation between poetry and philosophy rather than as a condemnation of the former by the later” (8). 5.9 Poetry’s banishment re-visited ~ In order to track its ironic tensions, to see the way that the Platonic text guides, beckons and sometimes misleads its readers, we revisit Plato’s infamous banning of Poetry here: a 203 shining example of a text that is not what it seems at first glance, a text whose shimmering irony thwarts efforts to pin it down. When one seeks evidence for “Platonism’s” pronouncements about “Plato’s thought,” we find instead endless dialogue: suggestive questions, slippery reasoning, ironically poetic utterances, and apparent paradoxes. When one looks more closely, one finds finely nuanced self-referential utterances that always seem to be aware of how things are said. The banishment-of-Poetry episode offers the opportunity to show how the received view that Plato banned poetry from his ideal republic is a (gross) misstatement and misunderstanding of what is written in that dialogue. It shows in broad strokes how the Platonic text works by weak implicature, how its metaphors and analogies might not be what they seem (and need, as Derrida says, to be carefully scrutinized); it shows how its ironies are interconnected threads to be followed, and how its self-referentiality directs readers to read more closely. While Derrida is understood to be discussing language, Plato is not. Thus the focus of the following analysis is to show how the Platonic stylistic works, the way that it talks about language, and what is says about it. I want to direct attention to the linguistic-mediation thematic in the dialogues. Because it is what everyone already knows about Plato’s “ideal state,” I examine various contextual implications of the Poetry-banning episode that I think are relevant; I then consider the way poetry is one of several linguistic binaries with celebrity status, that were invented, like many of Derrida’s neologisms, to seed a discourse about language. I consider some of the self-referential ironies in the second and third books of the Republic, before moving on to the larger discussion of how things are said in other dialogues. While the discussion of how things are said dominates the first book of the Republic, and it is in the second and third book that Homer’s discreditation appears to take place (when Socrates turns to the question of the kinds of poetry that will be permitted and the kinds of representations poets should make), it is not until Book 10 that Poetry is finally “sent off” or banished. This rhetorical arrangement (the crucial ordering of the text’s events) offers an excellent opportunity to expose how the Platonic text develops its complex ironic network – 204 This is what seems proposed in Laws: “we’re tragedians ourselves, and our tragedy is the finest we can create. At222 any rate, our entire state has been constructed so as to be a ‘representation’ of the finest and noblest life – the very thing we maintain is most genuinely a tragedy. So we are poets like yourselves, composing the same genre, and your competitors as artists and actors in the finest drama, which true law alone has the natural powers to ‘produce’ to perfection” (Saunders 817b). Clay reminds us of what many have observed: that the life of the guardians (with which the dialogue is223 preoccupied) is hardly ideal (21-22). whose interconnected threads invite us to re-evaluate prior impressions. As Griswold says, “irony is as much a way of revealing as it is of concealing, a way of teaching and motivating the reader/auditor to look further” (88). Consider to begin with the placement, in the middle of Book 10, of Socrates’ purported dismissal of the Muse of Poetry. This mid-Book 10 event takes place long after Socrates has established himself as “greedy for images” (as though he were himself a poet) and just before launching into the (creative-imaginative) myth that closes the dialogue (Reeve 488a) . When we add these ironies together – Socrates’ professed greed for images, his own myth-making activities, and the stylistically poetic register and mythical invention to attack Poetry – we could, as some have done, interpret Socrates’ poetic voice as indicating that in the ideal polis (city-state) “instead of poetry we will have Platonic dialogues” (Murray 22). In this vein we witness the way that irony produces the possible implicature that the image-and-myth-making literary skills of the philosopher might displace those of the poet quite naturally.222 While as far as I know this implicature is never negated, the ironic tensions also lead elsewhere. For example, the banning (in Book 10) supposedly takes place in an ideal polis, but the ideal polis of the Republic falls apart in Books 8 and 9, well before this poetic line appears, begging the question about what Poetry is banned from. Furthermore, Socrates’s explanation of the deconstruction of the supposedly ideal polis – that, as has long been noted, no one would have wanted to live in – is, like the banning episode itself, made in a confusing and self- referentially poetic text. And there are other factors that should attract reader attention. The223 whole demise of the polis is contrived: the destruction is initiated by presupposing an emergence theory of polis evolution (as though cities were a form of biological life). It represents a radical shift from the narrative quest for implicitly ideal justice: they weren’t looking for something that was corruptible, but rather something that was, by definition, 205 At the beginning of Book 8, they “have already described the man who is like aristocracy, whom we correctly state224 to be the good and just man” (Grube 545a); what should follow is a description of the unjust man or city, or as Socrates has now set it up, a number of different unjust men or cities. Instead, at the beginning of Book 8, Socrates simply pre-supposes the emergence theory: “let us try to explain how timocracy emerges from aristocracy” (545c). And then in a typically sophistic move, he obtains agreement for his presupposition by asking a question unrelated to the presupposition, about “the cause of change in government” being found in “the ruling group” (545d). Note that whether or not change in government comes about from the ruling group, it is not clear that cities degenerate in this fashion from aristocracy into timocracy and then oligarchy and so on. This emergence theory is never established in a convincing way and seems unlikely. He says “Do you want us to do like Homer and to pray to the Muses that they may tell us ‘how discord first broke225 out,’ and shall we say that they express themselves to us, as to children in tragic language, playfully and in banter, though their language is lofty, as if in earnest” (Grube 345e) uncorruptible. Socrates obtains agreement for his (presupposed) emergence theory using224 suggestive (but actually non-sequitur) questions in order to affirm (in a typically circular fashion) his uncalled-for emergence theory. This move seems to demonstrate the way that, as he says in the Phaedrus, “when you shift your ground little by little you are more likely to pass undetected” (Hackforth 262a). It is just such self-referential comments that invite readers to look more closely at the text. And arguably when Socrates produces a poetically framed degeneration myth in order to explain this degradation of the polis, the self-referential foregrounding of his decision to use Homer’s discredited tragic language for his account is just such an invitation. There is no225 stable ground here – only a network of interconnected ironies. Socrates asks whether they should imitate Homer in order to say how discord first broke out, and then in lofty tones explains that “‘It is hard for a city composed in this way to change. But everything that is born must perish’” (545e-546a). We observe intertwined a shift of focus (from ideal polis to degenerating polis), typically sophistical “reasoning,” and self-referentially poetic language; this net of ironies invites us to query a host of statements the text has seemed to make (about tragedy and Homer, for example) and seems to be making (about the aptness of the emergence theory, for example – the text seems to invite us to question it, to question why we are suddenly discussing the degeneration of the polis that was supposed to be ideal). It invites us in a myriad of ways to read more carefully, motivating the reader to look more closely, as Griswold says. The ironies lead back to prior utterances. When the implicature that Homer has been discredited is undermined, the relevance of Book 3’s critique of Homer needs re-evaluation. It is especially when one looks more closely that the contours of these interwoven and indirect rhetorical 206 devices surface. Sperber and Wilson say that “whereas a direct answer leaves the hearer free to process the information offered in whatever way he likes, an indirect answer suggests a particular line of processing in the computation of contextual effects” (219). Indirectness is in a high-tax bracket when it comes to hearer/reader inference; it is demanding and depends on multiple contextual effects. Weak implicatures are often already available in the background or context (including what the reader already knows) but their relevance needs to be recognized; although reader involvement and effort are required, Sperber and Wilson here suggest that, despite its demands, such indirectness actually provides more detailed and nuanced direction on interpretation. Socrates’ account of the degeneration of the polis is launched quite typically from the weak implicatures arising both from questions that leave a great deal for reader inference and from a slippery use of analogies that move back and forth from cities to men, making it easy to lose track of the focus of discussion. It is easy to see why Derrida uses web metaphors for the networks of interwoven ironic tensions to characterize the Platonic text; and failure to calibrate the implicatures of the demise of this singularly unattractive “ideal” polis could lead to all kinds of mis-takes about what is important in the Republic. In fact, the irony is so thick towards the end of Book 8, that Socrates resorts to open sarcasm when he states that “the finest [most beautiful, fairest] government and the finest kind of man remain for us to discuss, dictatorship and the dictatorial man” (562a). This of course is the most unjust man. Remember that, according to Sperber and Wilson, “metaphor and irony are not essentially different from other types of ‘non-figurative’ utterances” in terms of the way hearers produce implicatures and gauge their relevance (243). Poetic effects achieve most of their relevance through a wide array of weak implicatures” which generally contribute to background information; and repetition strengthens background information so that a weak implicature can become relevant in its own right (222, 209). When the ideal polis begets various, discordant, increasingly dystopian types of government, the cyclically deteriorating condition of the Republic’s polis obviates the very 207 possibility of a single unified ideal political state. The question about why Socrates deconstructs his own polis, and why he has constructed something necessarily faulty while pretending that it was ideal, and then why he tells us about its demise, might warrant re-evaluation of “Platonism’s” account of the Republic. Even though some pretense of discussing a polis is maintained up to the very end of Book 9, the tone of the discourse ripples with irony when Glaucon and Socrates finally observe that a just person would not “be willing to go into politics at all” (Grube 592a). This should be the end of any illusion that the aim of the text were a utopian political state. It is clear that discord doesn’t break out in an ideal polis; if it does, it isn’t ideal. We are left to consider what Poetry is banished from (in Book 10), or what the term poetry and what the Muse of Poetry might represent in Socrates’ poetic imagination. If we estimate the relevance of the polis’ failure as a stable and desirable ideal, together with Socrates’ penchant for waxing poetic at crucial moments (like the poetic banning of Poetry) and the innumerable citations of Homer in the text (which continue to proliferate after his apparent discreditation), we might be prompted to revisit the larger context of the discussion and to interrogate the original analogy between the individual and the polis on which the text is initially premised. (“Lisons” as Derrida has said.) Recall, then, that the initial aim and context of the Republic – contrary to what many have been taught and what many continue to teach – is not to construct a utopian political state; rather it is to discover what justice is and how an individual can maintain it. Book 1 begins with Cephalus’ definition of what justice is for him – “speaking the truth and paying whatever debts one has incurred” (331c). In Book 2 the difficulty of examining small things (i.e., justice in the individual) is compared to reading fine print and warrants the analogy with the polis – a decision to seek a larger-print example (i.e., to find justice in a polis resembling justice in individuals): “So, if you’re willing let’s first find out what sort of thing justice is in a city and afterwards look for it in the individual, observing the ways in which the smaller is similar to the larger” (369a). There is no assumption of identity here; Socrates does not argue that a city is the same as a person, nor that justice in the polis and 208 See Clay’s documentation of the scholarship on the problematic nature of this analogy and his conclusion that226 “personal justice” is the “end” of the Republic (note 16 271). It is worth remembering in this context that Nietzsche’s concerns with the way that every word instantly becomes a227 concept have precursors in ancient philosophy’s preoccupation with linguistic mediation. Clay is not the first to observe that, in the end, the Republic seems not to be a city at all, but “a projected image of228 the human soul” (19). the individual would be the same; he suggests only that there would be a similarity. It is the226 easier-to-see image, large-text version of justice in the individual that invites the analogy, not the construction of an ideal polis. We shall shortly see the way that discussing the analogy and its purpose is part of the text’s self-referentiality that highlights the way things are said. In addition, the multiplying ironies direct us to reread more carefully exactly how things are said (which we do usually through layers of translation, already shaped by “Platonism’s” voice-overs, as Derrida has noted). And things are said, as Socrates frequently reminds us, by means of likeness (image or analogy). French’s mistaken identification of the bidet as a toilet reminds us of the227 problematic nature of identification by analogy: since things might be similar in a variety of different ways, different aims produce different analogies for the same thing and vise versa. It depends on what motivates the analogy to begin with. Glaucon and Socrates’ “city of words” at the end of Book 9 turns out to be “a model laid up in heaven” for the government of the soul and not an actual city that would exist anywhere on earth (592a-b). This model government of the soul appears to be the aim of the analogy. Since the writer’s motives determine what features are foregrounded, Plato’s polis needs to look like a person with a tripartite psyche (or soul). There is no reason for the metaphor to be reversible, however – a person designed to look like a polis might not have a soul at all. Analogies are always error prone as the Platonic text228 is constantly demonstrating. Perhaps the joke that Glaucon and Socrates share in Books 8 and 9 as they continue to talk about a polis (that no one would want to live in) entails a critique of analogical reasoning – showing us in deed, as Sallis says, the problem of inapt (or “false”) analogy (the absurd notion that a city is like a man with a soul, for example). Perhaps Plato is showing us the careful distinctions we need to make between the meaning of terms (metaphors and analogies) we use in different contexts. In the end the phases of the polis’ destruction (in 209 See, for example, Clay’s analysis of the injustice on which Kallipolis is founded: “He forces upon us the reflection229 that all civilization rests on injustice and the uncontrolled need of the lowest and most passionate element of the soul to violate the frontiers of what is necessary to life for what is thought to be necessary to the good life” (28). Clay also observes that though “the theme of the individual” is central, “it cannot be understood in isolation from other themes,” especially the relationship between philosophy and society (32). See also Freydberg (101). See Tigerstedt’s analysis of nineteenth-century views (31-33). Note also Clay’s observation that at the end of Book230 9, when this conversation takes place, “our confidence in the possibility of Kallipolis is irrevocably shaken” (29). Books 8 and 9) seem to correspond to stages in the human struggle to maintain that harmony of the psyche: the struggle that constitutes justice in the individual. At the end of Book 9, when we are reminded about what the “city”in Plato’s poetic text represents, the discussion of poetry becomes more conducive to a language-theory than a political-theory interpretation. While the relations between polis and individual can be fruitfully explored in the Republic as many have observed, the ideal polis as such seems to vanish as we read. It is gone229 well before the banning of Poetry. Contrary to the nineteenth century face-value interpretations of it as a political utopia, it seems reasonable to assume (as many scholars have) that the Republic’s polis is a model for justice in the individual. Glaucon’s assumption that a just230 person would not want to get involved in politics at all and Socrates’ response that the polis being described could never actually exist on earth, but that it is an ideal pattern on which to form oneself as a citizen completely subvert utopian interpretations. These factors together with the markedly anthropomorphic image of the polis, its failure as an ideal, and Socrates’ ironically poetic register at crucial moments are all implicatures that might prepare us to think about the banishment of Poetry in terms of its relationship to justice in the individual. The Platonic concern with the effects of language on the soul’s harmony are related to (Nietzschean sounding) views about the relationship between how things are said and individual human behaviour. When poetry is discussed in Republic, there is nothing wrong with poetry per se; it is the powerful influence of the affective dimensions of language use on the soul that are at issue, the convincing effects it has on what people believe (like Simonides’ definition of justice), and the subversive effects that appealing representations of evil behaviour and attitudes have on the soul. In the Republic, poetry plays an ubiquitous role – as both subject of discussion and style 210 of writing; but it resurfaces regularly in both capacities elsewhere. The treatment and use of poetry, rhetoric and sophistry in the Platonic text signal Plato’s concern with the productive discursive effects of language. Plato seems to anticipate Tomkins’ scripts and the crucial role linguistic scripts have on human psychology and social development; he also seems concerned with the productive effects of discourse that Foucault identifies and the linguistic mediation evinced by Halliday’s semogenic conception of grammar. As with the Derridian text, the discussion of language itself is always in view in Plato’s dialogues. 5.10 some polemical mechanics of style ~ The triumvirate poetry, rhetoric and sophist performs (in partnership with philosophy) a variety of tasks which direct attention to linguistic mediation. But the binary distinctions are deployed in complex ways that do other work than simply privileging one term. So, for example, Freydberg and Statkiewicz see the poetry/philosophy binary as a kind of dance between equals rather than an overthrowing of the one by the other. Freydberg discusses the “conflict” and “kinship” between philosophy and poetry in the Republic and assesses the distinctive way that Plato’s use of these “two ways of logos [or styles of writing] perform the same function” (100). He notes, for example, that the first critique of poetry (in Book 3) places a specific line of Homeric verse at the top of a list of poetry to be “expunged” from the polis when Socrates says: We shall expunge all that sort of thing, I said, beginning with these lines: I would rather labour on earth in service to another, to a man who is landless, with little to live on, that be king over all the dead. (Grube 386c) Freyberg then observes that this same line is later reused with a radically different meaning by Socrates in Book 7. And not surprisingly, not only does Socrates reuse the same line but he highlights the reiteration (by referring to Homer) when he asks: “Instead wouldn’t he feel with Homer, that he’d much prefer to ‘work the earth as a serf to another, one without possessions,’ and go through any sufferings, rather than share [the] opinions [of the cave dwellers] and live as they do?” (Grube 516d). This re-accenting of the meaning should be no surprise after Part 3’s discussion of the way direct reports perform a host of functions that have little to do with their 211 descriptive-referential meaning. The line’s emphatic reuse suggests not only that the once banned line is now acceptable, but that it means something different in this different context. A line once placed at the top of a list of utterances not permitted and then later emphatically reused and approved in its changed meaning seems contrived to catch our attention. Derrida might ask pointedly “is it an accident?” (D 70). Plato’s reuse of this line, on the prominent theme of the censorship of writing, in these two contrasting interpretative contexts produces implicatures about context dependency. While it is easy to see how failure to recognize this repetition could mislead readers, Plato seems to have gone out of his way to make it recognizable, to point out that the same words have different meanings in different contexts. Freydberg observes that it is the framework of the line’s interpretation that is at issue here (102). The reuse of the line performs its duty rather than describing it; the reuse demonstrates how not only words, but whole lines change meaning in different contexts – showing us in deed the way that language works, emphatically illustrating the way that meaning is produced contextually. In an exemplary instance of Plato’s ironic conversational stylistic at work, the citation performs or shows us that words are context dependent: something other than what it seems to describe. Censorship and language use are both the subjects of a discussion; at the same time, discussion of them also demonstrates how language works; it is a dissimulating moebius-strip-like interfusion of word and deed. When the implications of Plato’s performative treatment of language are explored, as we shall shortly see, it is hard to avoid the inference that language’s context dependency was a main factor in Plato’s decision to write using the Socratic dialogue genre. In fact, Plato’s renowned language oppositions – philosophy/poetry, philosophy/rhetoric, philosopher/sophist – seem to have been created to orchestrate ironic tensions within the Platonic text. They are a device invented for discussing language. It is clear 212 Murray credits Plato for starting the war between philosophy and poetry (230-31), which Shorey observes “still231 goes on in modern times” (Loeb VI, 464). Freydberg comments that “contrary to received wisdom on Plato and Homer (a wisdom which, happily, is receding more and more), the Platonic dialogues do not advocate censorship, nor are they ‘against’ poetry” (99). Desjardins’ insightful explanation of Plato’s dialectical relationship to tradition, as an engagement between relative and absolute linguistically-mediated understandings, also foregrounds his positive relation to poetry. Not only was the historical Socrates himself known as a sophist, but the methods of Plato’s Socrates are232 increasingly seen as sophistic as discussions like George Klosko’s on Plato and fallacy show. Barrett imagines the-late-fifth- century sophists as like modern-day intellectuals and academics, as local celebrities subject to both admiring ridicule as well as conservative suspicion. Sallis is not alone in foregrounding the way that the opposition between sophist and philosopher in the Platonic text reveals Plato’s interest in vindicating his beloved mentor from the charges that resulted in his execution. Scholars have recently documented Plato’s positive views of rhetoric (see Al Miani’s notes for commentary dating233 back to 1974) and his rhetorical skills have long been recognized. Recall Benitez’s previously noted observation that “the greatest rhetorical display” of the Protagoras is “the dialogue as a whole” (222). Schiappa argues that no distinct disciplinary rJhtorikhv or art of rhetoric as such pre-existed Plato’s time. He finds that sophists often taught a predisciplinary lovgwn tevcnh (speech skill/language arts) related to their other subjects of instruction He claims they were not defenders of any theory known as rJhtorikhv. Naas observes that there is clearly a practice of oratorical speech arts in Homer and that codification of the practice seems to have followed the practice by several centuries (Turning 133-34). that Plato invented the now legendary war between philosophy and poetry, just as he invented231 the sophist/philosopher antinomy which is most emphatically deconstructed in the Sophist itself (by the account of the similarities between Socrates and the sophist). And it is Plato’s defamation of the sophists that is responsible for their extreme disrepute (and the modern-day meaning of “sophistry”), as both Grote and Sidgwick long since observed. Similarly and232 despite the earlier noted widespread recognition that “Plato’s dialogues themselves exhibit an unparalleled mastery of rhetoric” (Mitscherling Ancient 281), he seems to have invented the rhetoric-philosophy binary too as Schiappa’s evidence that he coined the term rJhtorikhv (the art of rhetoric) suggests. Keeping in mind that Plato is also responsible for the celebrity status of233 the term philosopher, it seems likely that the term rJhtorikhv was coined to invent a linguistic opposition with which he could promote a critique of language. Perhaps this does make him the founder of rhetorical theory. He seems to have launched his language-focussed binaries in order to differentiate and make fine distinctions between one type of language use and another. Naas asks how we should interpret “Plato’s condemnation of persuasion when the philosophical enterprise seems only able to come into its own by distinguishing itself from rhetoric and persuasion” (Turning 220). But (as Naas’ inquiry also seems to recognize) Plato does not simply condemn persuasion: an inference made from examples of people being wrongly persuaded. Plato’s frequent references to discovering how best to persuade others attests to a complex understanding of how language works, which fully embraces the role of persuasion, and his 213 Many neutral or even positive uses of peivqw (I persuade) occur in the Republic, see 357b, 388e, 399b, 414d, 589c. 234 Socrates does not suggest that he himself is already a dialectician. In the Phaedrus, drawing on the authority of235 Homeric verse, the dialectician is depicted as a god: “I follow ‘straight behind, in his tracks, as if he were a god’” (Hackforth 266b). See also Mitscherling’s “Plato’s Misquotation of the Poets.”236 positive use of this term, which foregrounds use of language to persuade, attests to his interest in directing reader attention to his sceptical (eyes-wide-open) image of language. 234 Socrates distinguishes and identifies philosophy by differentiating it from rhetoric: it is not only knowledge of how language works, but it is that together with other kinds of wisdom. But there is no evidence that Plato ever imagines an unrhetorical or unpersuasive use of language. His oppositions have dialectical (dialektikovV) purposes: the dialectician is he who can see a collective unity and plurality. Platonic dialectical engagement means perceiving fine distinctions between things and then gathering them up into a larger context – to see the way that specific parts fit into a more general whole. Despite Plato’s images of a godlike dialectician whose divine theorizing (qewrei:n) or pure perception might entail extra-linguistic visions of parts and whole, in the textual world of Platonic conversation (diavlektoV) these distinctions235 depend on the words and utterances of the phenomenal world. The dialectical engagement of the dialogues is an engagement through (diav) human speech (diavlexiV from dialevgomai): a back and forth movement between binaries that reveals their interstices, both similarities and differences. Clay sees “deficiencies, paradoxes, tensions and even fallacies . . . as intentionally designed invitations to the reader to sort through the topic at hand” (emphasis added, cited in Griswold 5). Plato seems to launch his illustrious linguistic binaries – between philosophy236 and sophistry, rhetoric, and poetry – as dialectical invitations to explore both the spinning out of specific terms and the fundamentally rhetorical way that language works more generally. They are a lexico-stylistic-thematic fusion that emblematizes the impossible to disentangle complicities of the Platonic text. 5.11 self-referential text as discourse about discourse ~ One of the most important aspects of Plato’s stylistic complexity – the feature of the Platonic text that makes it so elusive, as Gonzalez says – is the way that his subtle, measured 214 As Clay comments: “One problem with this model for inquiry is that the state might not prove to be an adequate237 representation of the human soul writ large”(24). Gallop reports what Robinson observes about the polis analogy: “To get ideas by analogy about the actual human soul it seems that he [Plato] ought to look at actual cities. In particular, to answer his main questions, ‘What constitutes a just man?’ and ‘Is a just man reputed just, and try to see how they differ from actual unjust cities and whether they are happier” (cited in Images 166). steps often produce implicatures in multiple contexts at the same time. Recall the paradigm of reported speech which can cue climactic emphasis, offer an evaluation, invite participation, and impart information at the same time. The temptation to find comfort in a single tenable interpretation of any instance can lead to misunderstanding as Linck’s above-noted references to utterances that “can be read in at least two [or more] ways” attests (267). Consider for example the double (triple or quadruple) duty that the failure of the polis performs in Books 8 and 9. The degeneration of the “ideal” republic both directs our attention to the text’s initial analogy and the focus on justice in the individual as noted above, and at the same time the degeneration portrays the vicissitudes of harmony in the psyche. Thus the polis’ demise both provides a cue to recall the original analogy and itself presents a new analogy illustrating the inevitability of the psyche falling out of tune. And, from slightly different vantage, we can see that since the analogy between polis and individual is so tenuous to start with and since there is no reason to assume that an ‘ideal’ polis, designed to look like an individual, could function as a polis, it237 might fall apart to remind us that it is an analogy (as a critique of writing). Perhaps Socrates’ images and myths that don’t quite hang together (regardless of how astonishingly compelling they are – winged horses, the cave, the divided line, Aristophanes’ lovers, etc.) are reminders of the analogous nature of discourse, of language’s metaphoricity. Like other structural features of the dialogues, the failure of the polis not only re-directs our attention back to the individual soul and illustrates the vicissitudes of striving for the just life, but at the same time reminds readers that the polis is an analogy and that we think using analogies and metaphors. Likewise, it is not just Socrates’ preference for metaphor and analogy that is signalled by his statement “listen to my [image], and you’ll appreciate all the more how greedy for images I am” (Reeve 488a), but also his choice to remind us of it. It is Plato’s conversational way of including such meta-rhetoric in his own text. Gallop’s apology for the Republic’s poetic stylistic 215 See below for the way that this reference to “words” being “well spoken” ties in with Book 1's focus on how things238 are said. Later Socrates discusses his own speech: “You are making fun of me after driving me into such an impasse of (lovgoV) argument. But, all the same, hear my image so that you may still better see how I strain after imagery” (Loeb 487e-88a). infers – from Socrates’ repeated references to both his polis analogy and his images (that is, repeated references to his own representational practices) – implicatures about the superiority of Socrates’ poetic philosophy (Image 155-58). For example Socrates asks Glaucon: “Do you think, then, that he would be any the less a good painter, who, after portraying a pattern of the ideally beautiful man and omitting no touch required for the perfection of the picture, should not be able to prove that it is actually possible for such a man to exist?” “Not I, by Zeus,” he said. “Then were not we, as we say, trying to create in words the pattern of a good state?” “Certainly.” “Do you think, then, that our words are any the less well spoken if we find ourselves unable to prove that it is possible for a state to be governed in accordance with our words?” (Grube 472d-e) 238 While the Socrates-as-philosopher-poet implicature is one that can be inferred from these rhetorical gestures, they are also quite obviously (at the same time) meta-rhetorical comments (about representation, about analogy, and about language) which emphatically draw attention to both the stylistic-argumentative device of analogy and to Plato’s strikingly poetic stylistic. This discussion-of-representation passage evinces the frequently commented on “self-referentiality” of the Platonic text. It not only draws attention to its own rhetorical operations, but also makes implications about the beneficial effects of representation itself: the benefits stemming from language’s semogenic and productive powers. While we might not want to say exactly what Plato really meant at any one moment, it is certain that he often means more than one thing. As Linck observes the text means more than it seems to say. The nature of reported speech and the entirely reported nature of the Platonic text guarantees this multiplicity of meaning. Let us take a brief detour here back through Books 2 and 3 in order to gauge the way Plato’s discourse about discourse is interfused in the discussion about the education of the guardians: that is, the way that the self-referentiality is meta-rhetorical and at the same time interfused with philosophic considerations. Having arrived at the state’s concern with education, Socrates directs the conversation to the use of “untrue stories” that have some truth in them for the education of the young (like his own myth of the cave, for example), the fact that stories or 216 Shorey’s Loeb translation takes pains to point out that “we must not infer that Plato is trying to sophisticate away239 the moral virtue of truth-telling” (195). The strangeness of shift back to generalities about gods prompts Shorey to comment (in his translation) “generalizing after the exhaustive classification that precedes” (196) as if he had to justify or explain this radical shift from the complexities and productive effects of representation to banal-generalized assumptions about how gods behave. Such repetitions indicate the significance of the role of repetition in “Plato’s Pharmacy.” 240 tales shape the souls of those who hear them at a young age and in a lasting way (warranting control of storytellers), the problem of representing the gods and heroes as doing bad things or being bad people (which is considered false representation and likened to inept representation), and the irrelevance of whether stories are allegorical or not. Above all, “the first stories they hear should be well told and dispose them to virtue” (Grube 376e-381e). Socrates thus creates room for allegories, images, and untrue representations in education, and it is no accident that these are all features that characterize the Platonic text. Then in an extraordinarily convoluted passage consisting largely of questions (thus avoiding explicature) Socrates argues for the beneficial use of lies as a means of education – “the lie then becomes like a drug [pharmakon]” – in which we can “make the fiction as like the truth as we can, and so make the lie or untruth useful” (Grube 382d). After implicitly justifying the use of the verbal lie, Socrates then239 radically shifts the conversation back to the not-obviously-related generalization about whether the divine and godlike would lie or deceive at all. The shift is another attention-grabbing contrivance; we seem to move jarringly from self-referential technical discussion back to allegory and myth (as if from science to fairy tale) in an instant. At the beginning of Book 3 after running rough-shod over various instances of Homeric verse to be expunged from the polis for fear of setting a bad example (but which at the same time articulate all the qualities that Socrates thinks are important in the guardian, like courage, moderation, self-control, truthfulness, etc.), Socrates comes to reiterate that “untruth is of no use to the gods, though useful to men as a kind of medicine [pharmakon]” (Grube 389b). It is240 important to recognize that in this context Socrates is talking about writing and poetry which is recorded in the soul of the rhapsode or inscribed in a manuscript (like the Platonic text itself). The self-reference is intertwined with the topic of discussion. After going over, in some detail, why false representations must be stopped “lest they produce in our youth a considerable 217 The guardians must imitate only “what is suitable to them, namely brave, self-controlled, pious, and free men and241 their actions. They must not be clever at doing or imitating mean actions or anything shameful, lest from enjoying the imitation they come to enjoy the reality. Have you not noticed that imitations, if they last from youth for some time, become part of one’s nature and settle into habits of gesture, voice and thought?” (Grube 395c-d). tolerance of evil,” he then makes it quite clear that it is not verse – i.e. not poetry itself – that is problematic. He says that: “both poets [poihtai;] and prose-writers [logopoioi;] tell bad tales about men in the most important matters” (Grube 392a-b). It is not just poetry, but writing more generally that is at issue. Socrates next turns “explicitly” to questions of style or levxiV (a way of speaking, diction, style) so that we will “have examined fully both what should be said and the manner of saying it” [“a” te lektevon kai; wJV lektevon”] (Grube 392c, repeated at 398b). Perhaps Aristotle overlooked this stutter or didn’t quite appreciate its implications for reading the Platonic text: that the manner of Socrates’ speech is important. Socrates elaborates the distinction (that Sternberg claims is misunderstood) between mimetic and diegetic discourse, explaining that: 241 when a moderate man in his narrative comes upon the words or actions of a good man he will be willing to expound it in character and will not be ashamed of that kind of imitation; he will impersonate this good man acting in a faultless and intelligent manner, but he will do so much less when the good man is overcome by disease or sexual passion, or by drunkenness or some other misfortune. When he comes upon a character unworthy of himself, he will be unwilling to make himself seriously like that worse character, except perhaps briefly when he is doing a good deed (Grube 396c-d). Although these multiply-qualified claims might produce ambiguous implicatures (which allow themselves to be misread), imitation itself has no moral disposition in this context. Imitations nevertheless have powerful effects on both the people who are doing the imitating as well as those who are hearing the imitation – the way that Judith Butler points out that the performance of gender has on producing gender. As it turns out the pure style that is selected for the guardians’ education has nothing to do with the selection of words or what we think of as style. The fact that Socrates calls it levxiV (‘style’ or ‘selection of words’) – in a context where he has repeatedly underscored attention to the manner of speech and fingered poets as culprits – indexes a willingness to let his text be misread by the first glance. What is actually at issue is 218 Either irony of poetic expression or full-blown-belly laughter about windbags and flatulence.242 the ethical posture of the imitated. The mixture of narrative and imitation, in which only the good is imitated, is opposed to the (also mixed) bad “style,” which is not a style at all but an indiscriminate willingness to imitate everything good, bad and in between. The sheer proportion of text that Socrates devotes to discussing all these aspects of writing (or language use) and style in particular should be enough to produce relevant implicatures about Plato’s interest in foregrounding how language works. That they are also at the same time a reflection of the Platonic text itself is also relevant. Ironically, in contrast to the impression the first glance at this text has frequently made, both allegorical and “untrue” stories are valorized, but neither mimetic nor diegetic discourse is. The discussion is very repetitive: Socrates repeatedly talks about the harmful effects of misrepresentations of gods and heroes as role models on children, as well as repeating that “untruth” is useful as a pharmakon. The repetition might strengthen impressions that rereading exposes as unwarranted. The discourse on style, of how things are said, is discourse about writing and self-referentially about the (educational) writing in the Platonic text. This thematic doubleness is completely interfused with stylistic devices and as always with the dramatic setting. It becomes increasingly ironic, when Socrates describes Adeimantus saying: “I divine,” he said, “that you are considering whether we shall admit tragedy and comedy into our city or not.” “Perhaps,” said I, “and perhaps even more than that. For I certainly do not yet know myself, but whithersoever the wind, as it were, of the argument [lovgoV] blows, there lies our course.” “Well said,” he replied [kai; kalw:V g=, e[fh, levgeiV] ( Loeb 394d-e). The irony – signalled by this use of poetic language about the (breezy) structure of the242 discourse in this supposedly anti-poet section of the text – undermines its apparently anti-poetic layer. The suggestive nature of Adeimantus’ comment about admitting poets to the polis fosters mistakes. When reread carefully, the dialogue advocates against neither poetry, nor mimesis. It is the “Muse inspired and harmonious man” who “achieves the most beautiful blend of physical and artistic culture” and who is opposed to the discourse-hating man who “never associates with 219 the Muse”; importantly the discourse-hater makes no “use of persuasion but bulls his way through every situation”; this condemnation highlights the importance of persuasion and discourse (411d-412a). It turns out that the guardian is a “Muse inspired and harmonious man” who “far more so than the musician . . . harmonizes the strings of his instrument.” So much for the discrediting of Homer; the guardians of Book 3 are not muse banishers. Socrates explains that “a god has given men these two means, artistic and physical education, to deal with these two parts of themselves, not the body and the soul except incidentally but the spirited and the wisdom-loving parts, in order that these be in harmony with each other, each being stretched and relaxed to the proper point.” This metaphorically presented, muse-inspired harmony is justice. Plato’s discussion about the education of the guardians is also a discussion about what is considered virtuous behaviour, and also about the productive effects of language, and also about writing styles at the same time. These themes are interwoven so tightly that one might be at any moment unsure of what layer of the discourse is the focal point: “the dissimulation of the woven texture can in any case take centuries to undo its web.” When thematic topics are discussed by questions and answers between different speakers – where the asker is presumed to be the voice guiding the discussion (i.e., Socrates’ suggestive questions) – the effect is a complete absence of explicature and a preponderance of arrays and layers of weak implicatures. This manner of writing, punctuated with a profusion of ironic tensions produces Plato’s shimmering, ironic- conversational stylistic which hides the laws of its composition and the rules of its game from not only the first glance, but often from the third and fourth careful examination as well. The context shifts with each new reading. Plato’s self-referential meta-rhetoric is interfused with other aspects of the discourse – and not just topics or layers of discussion but sometimes dramatic elements, sometimes literary rhetorical flourishes. Plato’s awareness of the metaphoricity of analogy itself – its foreign, potentially chimerical nature – seems embedded not only in his critiques of poetry, sophistry and rhetoric, but also in the frequent meta-rhetoric on method, which, as we shall see, inevitably foregrounds the problem of interpretation. 220 This fine and tightly interwoven tapestry of ironic and conversational threads produces an ubiquitous discourse on language – a logos about logos. It is such conversational interweaving of writing about writing and philosophy at the same time that Derrida re-performs. We turn here to the way that Plato’s attention to how things are said is part of a meta-rhetoric which reveals his sceptical understanding of how language works. This involves a more fine- grained analysis in which it is not just Socrates’ utterances and argumentative lead that count. It is not just in Socrates’ voice that we hear what Blondell calls Plato’s “dazzling interplay of unity and multiplicity”; no single voice performs this interplay; neither is a single voice meaningful if it is extracted from the specific situation created by other speakers in Plato’s dialogical texts (1). In order to track this meta-rhetoric we need to attend not only to Socrates’ remarkable manner of speech, but we also need to gauge the responses of other characters, their confusion, their simple-mindedness, their sarcasm. Even apparently innocuous expressions of bewilderment by Theaetetus and the Young Socrates, for example, produce contextual effects. Their banal remarks sometimes indicate a failure of discrimination and willingness to bend to authority (directing attention to what they miss), and at other times their expression comments on the obscurity of the Visitor’s or Socrates’ utterances, directing attention to how things are said, and to the meta-rhetorical discourse hovering just below and sometimes breaching the surface of the text. 5.12 linguistic cues, ‘how things are said’ & meta-rhetoric ~ Halperin’s analysis of the Symposium’s narrative structure, “Plato and the Erotics of Narrativity,” observes that “the formal, theoretical or conceptual, distinction between dramatic and narrative literature is not one that is likely to have been lost on Plato. For the very distinction originated with Plato” (94). He argues that the Symposium is the dialogue in which “Plato most fully thematizes” narrative structure and that this thematization exactly reverses what he does in the Theaetetus: making the Symposium and the Theaetetus companion pieces when it comes to the question of narrative structure (96, 99). What Halperin considers the “thematization” of narrative structure I consider self-referential meta-rhetoric that is designed to 221 attract attention to itself and to put linguistic mediation on the horizon. Halperin notices (as I do) the way that reporting draws attention to the act of speech. He comments that Plato constantly reminds the reader of the many narrators that intervene between the reader and the transmitted story – he emphasizes the reported character of the account – by sprinkling throughout Apollodorus’ narrative such phrases as ‘he said that he said’ (ephç phanai, or simply phanai), phrases often omitted in translation through a wish to avoid unnecessary awkwardness but so copious in the original as to make the text of the Symposium an ideal object lesson in the use of indirect discourse in Attic Greek. (97) Socrates’ above noted reporting of Adeimantus’ response to his breezy metaphors “‘Well said,’ he replied [kai; kalw:V g=, e[fh, levgeiV]” (Loeb 394d-e), illustrates just such emphasis on reporting. As Halperin observes, such reports are often translated out as awkwardness: for example, Grube’s translation reduces this line to “right” without any reporting verbs at all, without mention of how “well” it was said. Sallis stresses the way the term lovgoV (word, speech, discourse, account, explanation) figures in the Platonic text. We have witnessed the way that the discussion of poetry in Books 2 and 3 interweaves language use, self-referential commentary, style, educational ideals, and human virtues. Now we will focus on the way that lovgoV’ relative, levgw (‘I speak’) and various other reporting verbs and discourse markers foreground the way things are said in the dialogues; and we focus on the way this fine level of attention to how things are said is intertwined with topics of discussion and self-reflective commentary in a seamless fusion of form and content. After reviewing the way these linguistic cues perform the duty of foregrounding language use, we will very briefly reflect on the Protagoras’ meta-rhetoric on method, the Theaetetus’ expressed intention to omit verbs of reporting, and the role that referential promiscuity has in different ways of saying what knowledge is. We touch on the findings of the Cratylus and the way that Socrates plays with language in order to show that it has no essential meaning, and then we return to Theaetetus, with its determination that the interlocutors don’t know what they are saying, its reflections on linguistic instability, and its reflections on the problematic relation between linguistic semiotics and a world in flux; and 222 Desjardins argues that ths discussion of interpretation is pervasive in the dialogues.243 finally close with Theaetetus and Socrates’ joke about the nature of knowing. Just as what looks like a topic of discussion is also self-reflective commentary, what looks like a style of reporting, or storytelling is also a linguistic cue. A variety of devices – reporting verbs, references to speaking, and discourse markers – cue attention to the proto-post- structuralist discussions of language embedded in the Platonic meta-rhetoric. These self- referential references to speech mark language use. In Plato’s nimble choreography, they act as cues or indicators of linguistic performance implicitly highlighting the role of lovgoV in philosophic engagement. Plato frequently inscribes these speech references (or linguistic cues) across the voices of interlocutors – as though others than Socrates were in on the game. For example, the following playful exchange between Protarchus and Socrates in the Philebus, exhibits a focus on things being said, ways of things being said, and how well said they are (italicized here and in all subsequent citations): Prot: This much is clear, Socrates, that in such a case there would not be either any pleasure or pain at all. Soc: Very well put [kavllist= ei\\peV or beautifully said]. But I guess what you meant to say is that we necessarily are always experiencing one or the other, as the wise men say. For everything is in an eternal flux, upward and downward. Prot: They do say that, and what they say seems important. (D. Frede 42e-43a) In this dialogue – which is emphatically self-referential and preoccupied with discussing its own discussion – the speakers are depicted as stumbling over references to speech, as though it were a game to talk about talk. Quite typically the reporting verbs and discourse markers are linguistic cues interfused with a meta-rhetoric about the meaning of words or interpretation. It243 is this combination of attention to linguistic meaning, the pronounced foregrounding of speech, and the self-referential commentary about style and method that reveal Plato’s concern with linguistic mediation and his sceptical (eyes-wide-open) view of language. Such embedded speech references will already be familiar to readers of the Republic, where the words of both Socrates and Cephalus emphasise speech from the outset. Socrates 223 says, for example: I wondered at his saying this and I wanted him to say more, so I urged him on by saying: Cephalus when you say this, I don’t think most people would agree with you; they think you endure old age easily not because of your manner of life but because you are wealthy, for the wealthy, they say, have many things to encourage them. (329d-e Grube) Despite a tendency of reporting to become invisible as a matter of stylistic form (perhaps warranting translators’ tendency to edit them out), the first person reporting of a dialogue (such as takes place in the Republic) increases the instances of reporting verbs; and whether or not we notice them, the reporting verbs themselves constantly draw attention to the speakers speaking, their utterances, and their language use. Then when Socrates’ reporting lapses (he frequently forgets to report who is saying what, and one finds oneself having to reread in order to figure out who said what) their reappearance sets the reporting verbs in relief, bringing them into the limelight. In this passage, a superabundance of verbs about speaking, which highlight what people have said about various things, meshes with the thematic of finding an apt definition (or account in words, a lovgoV) of justice, and the question of whether Simonides’ poetic utterance is well said: “Simonides . . . says that to return what is owed is just” (332a Grube). Note that Socrates’ reporting of Cephalus’ reporting is tangled up with other linguistic cues in Cephalus’ response to Socrates: What you say is true, he said. They would not agree. And there is something in what they say, but not as much as they think. What Themistocles said is quite right: when a man from Seriphus was insulting him by saying that his high reputation was due to his city and not to himself, he replied that had he been a Seriphian, he would not be famous, but neither would the other had he been an Athenian. (329e-330a Grube). In this instance, Socrates reports Cephalus’ reporting of other’s speech in a way that seems to foreground linguistic cues that could have been eliminated. For example, Socrates might have reported Cephalus’ speech as follows: “Socrates, this is true, but others would not agree and they have a point. Themistocles remembers being insulted by someone who thought Themistocles’ reputation was a result of where he was from.” Such a formulation would have 224 (i.e., Simonides’ poetic definition versus Socrates’ subsequent poetic “muse-inspired harmony” of “the spirited and244 wisdom-loving parts” of the soul in Book 3) eliminated all the reporting and other speech verbs. Instead Plato has both characters stumbling over efforts to talk about talk: to say what others have said and what they themselves are saying, in a highly marked, contrived, and repetitive written style. Again the interlocutors seem to be imitating one another as though both were in on the game. In this opening of the Republic, the discourse on how things are said is interwoven with thematic issues which govern the entire text, beginning with Simonides’ poetic pronouncement on justice and including other matters like opinions and beliefs (about justice, for example). It weaves together discussions of reputation and what people believe with ways of saying things: poetic ways (lines of Sophocles and Pindar), as well as commonplace anecdotes, and beliefs attributed to other familiar figures (like the politician Themosticles). The frequency of these linguistic cues in less than two pages of Grube’s text (between 329d above to 331c) provides an indication of the emphasis on how things are said in the opening of Book 1. After the above- mentioned interchange representing Socrates and Cephalus stumbling over references to speech, a line from 329e is repeated: “what you say is true, he said” (330c) followed by, “many people in saying this, Socrates, he said” (330d); “it is then that the stories we are told about the underworld” (330e), “as Pindar puts it [levgei ‘says’]. The poet has expressed this charmingly [carievntwV ei\\pen, ‘beautifully said’]”(331a); “this is wonderfully well said. It is in this connection that I would say” (331a); “beautifully spoken, Cephalus, said I, but are we to say that justice or right is simply to speak the truth” (331c). This marked discourse, which is followed by the discussion of writing and style in Books 2 and 3 that we have just reviewed, sets up Book 1’s discussion of the interpretation of specific ways of saying what justice is as244 well as the text’s central concerns with poetry and opinion/knowledge. As noted earlier the artifice of reported speech – that is speech within speech and speech about speech – guarantees multiple operations being performed at the same time. The Platonic text is all reported, thus both highly contextualized and about much more than its descriptive referential meaning (descriptive reference having little to do with what reported speech 225 The methodological discussion is woven into the fabric of the dialogue from the outset. When he sets out to clarify245 how virtue is teachable, it is Protagoras who asks what form of explanation Socrates prefers: that of telling a story or of developing an argument (320c).This question is addressed to Socrates, and although only Socrates and Protagoras have been involved in the discussion and Socrates should answer the question himself, he evades the issue by offering an account of what the group answered. Methodological issues surface again when Socrates remarks that Protagoras possesses the rare accomplishment of being able to deliver long beautiful speeches and also of being “able to reply briefly when questioned, and after asking a question to await and accept the answer” (Lombardo and Bell 329b). The methodological focus culminates in the centre of the dialogue when Socrates insists that Protagoras provide brief answers to his questions (334d-339a). performs). What I call its meta-rhetoric is the logos about logos. The meta-rhetoric circulates between a topic of discussion and its stylistically inscribed imprint (like the reporting of conversation) and between a topic and its dramatic presentation (like Socrates petulance noted below). But it could be argued that there is no meta-rhetoric in the dialogues. As a part of the conversation between the interlocutors, the discussion of logos is part of a discussion; the linguistic cues are a manner of speech and a manner of reporting conversations. This logos about logos is in every way interfused: integral to the dialogue structure (its dramatic, literary, stylistic mechanics), integrated with the topics of the discussion, and integrated with the implications stemming from the interface between these diverse textual dimensions of the dialogue. The meta-rhetoric is embedded in a stylistic of implicitness, conversational through and through. In the Protagoras, the meta-rhetoric is typically thematized as methodological (in this case, the preference for short question and answer over long speeches). Midway through the245 dialogue Socrates threatens to go stomping off like an angry child if Protagoras does not converse using the question-and-(brief)-answer method, and then Socrates himself ironically holds forth at some length on the subject of poetry. (He discusses various lines from Simonides, in one of which Simonides is reporting what Pitticus said.) Immediately following this discourse on poetry, Socrates disparages discussing poetry: But if he does not mind, let us talk no more of poems and verses . . . For it seems to me that arguing about poetry is comparable to the wine-parties of common market-folk. . . . But where the party consists of thorough gentlemen. . . the company contenting themselves with their own conversation . . . each speaking and listening decently in his turn. . . . such men as most of us claim to be, requires no extraneous voices, not even of the poets, whom one cannot question on the sense of what they say; . . . we are generally told by some that the poet thought so and so, and by others, something different, and they go on arguing 226 about a matter which they are powerless to determine. (Loeb 347c-e). This disparagement of discussing poetry makes Socrates’ linguistic deeds (talking at length about poetry) act in complete defiance to what his words seem to say. It is utterly ironic. The Protagoras’ focus on language use – which draws attention to forms of discourse, conversation, poetry, ways of speaking, problems of interpretation, and multiple meanings of words – typifies the Platonic text’s ubiquitous discussion of language use (of poetic styles, of mimesis and diegesis, of images and likenesses, of the question-answer genre, of rhetoric, of sophistry). When Socrates demonstrates the question-answer method to Protagoras by performing a dialogue within a dialogue in which he plays all the parts (a burlesque of the concept of conversation), his own dramatic choices produce ironies with arrays of weak implicatures. Socrates frames one of his favourite topics (opinion or what most people think) in this long- winded display by ironically miming both sides of the conversation and answering his own questions (352b). Emphatically foregrounding the meta-rhetoric on method, this parody produces a distracting, doubled-up reporting (not unlike Adeimantus’ e[fh, levgeiV [he said, you say]). The focus of the conversation is quite typically on how things are said and what the citation “being overcome by pleasure” means. Socrates imitates both what “they” (the opinion holders) say and how he would reply: “If they should ask us: ‘we have been speaking of “being overcome by pleasure.” What do you say this is?’ I would reply to them this way . . .” (Lombardo and Bell 353c). Here how things are said, what they mean, and opinion are completely intertwined in this absurd and contrived tripping-over-itself style of reporting. The attention-grabbing rhetorical antic of answering one’s own questions, the meta-rhetoric on method, and the linguistic cues are all interfused in Plato’s logos about logos. Lest this reporting extravagance appear incidental, its significance is made clear in the Theaetetus. Here Plato highlights the reporting contrivance by beginning with Euclides’ singular and much remarked announcement that his account will omit verbs of reporting. We might then anticipate a reduction in linguistic cues. However, instead we hear a strange echo of the Protagoras’ doubled-up-reporting contrivances when Socrates ironically uses absent voices 227 i.e., “as the saying goes” (Levett and Burnyeat); “as the saying is” (Loeb),246 to create an artificial reporting exigency – a reporting drawn into relief by its anticipated absence. This reporting exigency results in such statements as Socrates’ imaginary account of how Protagoras might respond to their conversation: “He will say all the things that we are saying in our attempt to defend him; and then, I imagine . . . him saying: ‘This is good, Socrates . . .’” (Levett and Burnyeat 165e-166a). This imaginary reporting of other voices blends together, in Theaetetus, with discourse markers like as they say (e.g., 162c), and the other insistent246 linguistic cues: “Therefore you were quite right in saying that knowledge is nothing else than perception . . . since these things are true, perception is knowledge. Eh, Theaetetus? Shall we say that this is, so to speak, your new-born child and the result of my midwifery? Or what shall we say?” (160d-e, Loeb). Observe that these slippery assertions about what knowledge is are quickly qualified by questions, inviting readers to make their own inferences. The linguistic cues highlight “what people say”; but of course, Plato’s point here is partly that “what they say” is not necessarily (or even likely) true. This instance of Theaetetus’ docility illustrates how Plato intertwines thematics about knowledge and “how things are said” in an interfused meta- rhetorical-conversational stylistic of implicitness. There is a complete collapse of argument into style or style into argument, and no line in between the two. As we noted in Part 1, language’s freedom from referential constraint is an epistemological problem because linguistically mediated knowledge of the world depends on descriptive reference. Since the notion that we think in language appears to have been unexceptional in ancient Greece (see Theat 190a, Soph 263e, Phaed 266b, see also the proto- Lacanian account in Phil 38e-39a), we should not be surprised that the Theaetetus’ inquiry into what knowledge is addresses epistemological implications of language’s referential incompetence that sound Nietzschean and Derridian. The Theaetetus’ discussion of knowledge investigates different ways of saying what knowledge is: focussing on the ties between how things are said and the epistemological problems it explores, particularly that of linguistic 228 So, for example, Socrates or the Visitor will say things like “since there’s a single kind of expertise involved in all247 of them, then according to what we have said, we’ll expect it to have a single name” (White Soph 226c). This statement stands in contrast to the numerous names that appear for the sophist, as well as to other statements like “it doesn’t matter to our method which name would seem to be the most appropriate” – any name will do (Soph 227c). The earlier expectation of a single name is a lesson on names: the dialogue demonstrates that we should not expect a single expertise will have a single name. The Cratylus, Theaetetus, Statesman and Sophist all make numerous observations and assumptions about names that are either supported or undermined by what the main speaker shows in his analysis. These dialogues evince the fact “that views tried out at one place are [frequently] refuted at another” (Ackrill Comments 610). The Theaetetus repeatedly draws attention to different ways of saying things. For instance, when Theaetetus cannot248 follow Socrates slightly incomprehensible abstract account in 192a-d, Socrates tells Theaetetus to ajnapavlin a[koue [‘listen over again’] and then he provides a different way of saying the same thing. (Socrates’ imperative here might remind us of Derrida’s “recommençons.”) Regarding names specifically, observe that the Statesman starts by out pointing out that there are many names for one expert knowledge (259c) and later observes that “the five names of what are now called constitutions have become only one” (Rowe 301b). instability. In this discourse about knowledge, opinion and commonplace assumptions (or beliefs) often seem to be espoused, but turn out to be part of an account that undermines their validity, reminding us to heed the difference between word and deed, and then estimate its relevance. So while Socrates sometimes provisionally pretends a contrary view, his accounts frequently demonstrate that different names are used to describe the same things. He responds247 to Theaetetus’ claim, for instance, that “knowledge is simply perception” (151e) by suggesting that this is the same thing that Protagoras said: “Perception, you say, is knowledge . . . . If I may venture to say so, it is not a bad description [lovgon] of knowledge that you have given, but one which Protagoras used to give [e[lege said]. Only he has said the same thing in a different way” (151e-152a, Loeb). Here, according to Socrates, entirely different names seem to say the same thing: i.e., “knowledge” is the same as “perception,” and to say that “knowledge is perception” is the same as to say that “man is the measure.” Avid rereaders of the Platonic text will248 instantly recognize that – despite appearances to the contrary – Socrates subscribes to none of these formulations, these opinions, these different ways of saying of what knowledge is. However, what is of most interest is the piling up of linguistic cues in this dialogue that omits verbs of reporting. This discussion – about how we say what knowledge is and the different ways of saying the same thing – both underscores the way that language’s referential promiscuity precludes determinate distinctions and illustrates Plato’s ironic-conversational fusion of style and argument. This referential promiscuity thematic appears repeatedly. The use of one name for 229 Quite apart from the peculiar arguments at the end of the dialogue that are generally recognized to have249 undermined the determinations about essential meaning, the etymologies “explicitly” raise the spectre of names as a sound images. If names (as sound images) had essential meaning, as Cratylus believes, sounds and syllables would have consistent and essential meaning, two different things would not have the same name, a name would not have more than one referent, and things of similar essence would have similar names – a condition that would seem to obviate the need for interpretation, make definition effortless, and arguably render large portions of the Platonic text unnecessary. Of course it is on the basis of sound similarity that Socrates makes his first etymological analysis (392a-e). Ironically this plausible etymology is nevertheless radically undermined, when Socrates immediately afterwards offers the caveat that he himself does not understand what he has said; this admission prompts Hermogenes’ after-the-fact retraction of agreement; he announces “then I certainly don’t” (Reeve 392e). This opening sets up the ironic banter, parading Hermogenes’ willingness to go along with the joke, that runs through the entire etymological comedy from beginning to end. Interestingly, Cratylus’ faith in a divine essential language does not prevent him from acknowledging “incorrect” names are used and effective. While ejrwtavw (‘I ask’) is not the same as levgw (‘I speak’), ei[rein means both ‘to ask’ and ‘to speak’; and there is a250 complex mesh of etymological links between all three verbs. ei[rein is thus a choice exemplar of the way that meaning slides around since it not only means both speaking and asking, but also has a completely different meaning (to fasten together in rows, to string). Similarly, levgw means to select/choose and thus not to speak in some contexts. different things and more than one name for the same thing is thematized in the Cratylus, where the different names used by foreigners, Greeks from different locations, gods, poets, and so on are foregrounded. Here Socrates’ etymologies undermine any possibility that names have essential connections with referents (apart from some onomatopoeic words). In this context,249 not surprisingly, the image-greedy Socrates explores his own favoured metaphorical usage. He considers Homer’s reference to a “golden race” and explains that “I don’t think he’s saying that the golden race is by nature made of gold, but that it is good and fine” (Reeve 398a). He thus points out that the word golden was used metaphorically. By pointing out that Homer doesn’t mean that the race was actually a set of shiny metal figurines on someone’s mantle, Socrates – contrary to Cornford’s still influential pronouncement – denies that words have essential meaning: a fact which Plato’s footloose stylistic attitude should have informed us of in advance. Socrates nevertheless goes on ad nauseam in the Cratylus with his mostly ridiculous etymologies showing that words have overlapping meaning, that words that sound the same mean different things, that different words are used to say the same things, that the same word is used to mean different things, and finally that sounds and syllables have no essential meaning of their own. For example, he runs trippingly from Homer to heroes and then (absurdly) to eros, when he makes the stunningly sophistic argument that equates a form of ejrwtavw (‘I question, ask, inquire’) to the statement to; ga;r ei[rein levgein ejstivn (‘for to question is to speak’) (Crat 398d). Although the verbs mean different things, the statement that “to ask is to speak” is250 230 Robinson’s analytic-philosophy perspective that “names refer” speaks to their capacity to identify things, a251 capacity requisite for knowledge (334). But to what does the term knowledge ‘refer’? Is it “wisdom” (145e), or “perception” (151e), or “judgement”: “the name, Socrates, I suppose is judgement” (187a). entirely consistent with earlier determinations of speech as action (Crat 387b), and indeed from one perspective to ask a question is to utter or do something. Socrates is the master of asking questions (a master of implicitness); questions by definition do something, leading readers to consider a certain perspective or way of seeing something, inviting them to participate in the discourse by making their own inferences. We might here be reminded that Socrates looks a lot like a sophist: perhaps most like his buddy the word-master Prodicus, the emblem of making fine distinctions in language. When words or names of such unstable condition shape thought epistemological problems are implicated. The link between the knowledge of what a thing is and knowledge of its name is foregrounded early in the Theaetetus when Socrates asks whether or not “anyone can understand the name of a thing when he doesn’t know what the thing is” (emphasis added Levett and Burnyeat 147b). Socrates claims at the outset that he “can’t get a proper grasp of what” knowledge actually is (Levett and Burnyeat 146a) and then later comments that they are using names (like knowledge) whose meanings they do not understand: “You realize that our whole discussion from the beginning has been an inquiry about knowledge, on the assumption that we do not yet know what it is” (Levett and Burnyeat 196d). The dialogue explores the issues conversationally: leaving matters unanswered, circling back, starting again. The knowledge-naming-identification problem – at the root of Nietzsche’s critique of metaphysics – is partly that people don’t know what a thing is and accordingly use its name mindlessly or without understanding. The tension between knowledge of a thing and its name exposes the dependency of linguistic identification on understanding what the referent is. In the251 Theaetetus, Plato explores the epistemological problem of linguistic mediation that results from our dependence on names (or some kind of semiotic) for identification. He shows that one cannot recognize what one does not know and one cannot name what one cannot identify. As the Republic’s discussion about the divided line suggests, human knowledge depends on 231 Recall the perception-shadow problem for the cave dwellers and Socrates’ suggestion that “they would suppose252 that in naming the things they saw they were naming the passing objects” (Loeb 515b). See also the four repeated queries: “Socrates, what do you mean?” (152d, 154b, 172c, 174a Loeb); these echoes all253 use the verb levgeiV and translate more literally as what are you saying. analogies, images, and, as Nietzsche two millennia later reminds us, metaphors. 252 On one level the focus on exactly how things are said reveals the instability of names themselves, their context dependency, their various shades of meaning in different settings. So, for example, this saying “that knowledge is perception” has, in short succession, made its foray into the conversation (151e), been knocked down (157e-158b), and then (at 160d) dramatically (and ironically) recovered – an emphatic demonstration (in deed) of the way that our interpretation of words and utterances can change and how the specific context shapes our interpretation. From the outset of this dialogue, linguistic cues highlight queries into what an utterance means and how it should be interpreted. Theaetetus repeatedly asks Socrates “What do you mean by that” (146d, Loeb)? In addition to the linguistic highlighting, the implicatures253 about Socrates’ language use, about Theaetetus’ understanding, and about meaning more generally arise from the direct reporting. The weak implicatures gather steam from repeat appearances. Socrates’ answer playfully foregrounds the issue: “Nothing perhaps, but I will tell you what I think I mean” (146d Loeb). The play of meaning resulting from the irony imbricates knowing and language, suggesting 1) that Socrates doesn’t know what he means, 2) that he doesn’t know what his words mean, 3) that he can nevertheless say what he thinks he means. The language-knowledge problematic, the dramatic presentation’s playfulness and the conversational reporting stylistic are intertwined. The false determination – that knowledge is perception – reveals the way that how things are said can be misleading. Yet we still don’t know what knowledge is. And there is nothing that can be done about this, about words and utterances getting knocked about – no way of putting things can ever protect the saying from these aporias. The dialogue offers ongoing queries into meaning and context specificity: “I think he means what you say he does” (153d Loeb) and “what does this tale mean for us, Theaetetus, with reference to what was said before” (156c Loeb)? In other words, “what does this tale mean in this specific context?” 232 The understanding of language change evinced by the discussions of the Cratylus sets modern linguistics’ discovery of it in relief and exposes the naiveté of language correctionists frothing over “incorrect usages” that corrupt the Queen’s English. The Theaetetus seems to go a step further into proto-post-structuralist linguistic insights. Heraclitean theories of motion are imbricated with names, knowledge, and interpretation from the beginning of the dialogue (152e). After discussing “motion” from the outset, in the middle of the dialogue, when Socrates asks “what is this thing that they are talking about when they say that all things are in motion” (Levett and Burnyeat 181c)? The problem of using terms and utterances whose meaning one does not understand is interwoven with the interpretation thematic. These rhetorically orchestrated developments are intersected by an ironic discourse proscribing word use. Highlighting the disjunction between our perceptions of the world and language, Socrates makes an ironic prohibition of the verb “to be” – a proscription that might have staunched the entire Platonic enterprise, since, for example, this would mean that one could not even venture to imagine what knowledge might “be.” First Socrates claims “We wrongly say [that those things] ‘are’ – wrongly, because nothing ever is, but is always becoming” (152d-e Loeb). He later suggests altogether prohibiting the verb “to be” and adds ironically and in jest that “we ought not, the wise men say, to permit the use of ‘something’ or ‘somebody’s’ or ‘mine’ or ‘this’ or ‘that’ or any other word that implies making things stand still” and anyone who uses such terms “is easily refuted” (157b Loeb). This purported failure of reference (i.e., between ‘this’ and its shifting and thus unidentifiable referent, between ‘to be’ and a world in flux, a world of becoming) occasions Socrates’ ironic and comical prescriptions about the impropriety of using these commonplace terms: because they make things stand still. Commenting on both the thematic and imagery of motion Sallis observes that “the Theaetetus (e.g., 183b-c) and the later part of the Cratylus itself (439d) drive home the point that that which is perpetually moving cannot be named correctly” adding elsewhere that “perpetual movement” itself “would amount to a total lack of any distinctness” (239, 255). Names which are used to differentiate, to distinguish one thing from another, have yet another 233 We are, Derrida points out, dependent on our linguistic semiotic system (which include metaphors like forms): 254 “there is no sense in doing without the concepts of metaphysics in order to shake metaphysics. We have no language – no syntax and no lexicon – which is foreign to this history” (WD 280). problematic quality: that of falsely arresting motion. This feature of langauge was discussed in Part 3 when Halliday comments of the way that nominalized grammar problematizes “conceptualizing the fluid and indeterminate ‘reality’ required by quantum physics” (OL 258). Signs convey a pretence of stasis, an illusion of presence: as though “now” is not already past when it is written, as though the referent and sign exist eternally in some stable and unchanging universe. Language can differentiate (or identify) only by this pretence of arresting motion. Names change. But if the named, the referent itself, is in constant flux how can we tell what it is? If things are in constant motion, then the referent is never what it was at the moment of being named. How can we know anything if it is constantly changing, how can we recognize it? The Platonic insight that signs demand a pretence of presence is parallel with contemporary post-structuralist critiques, reminiscent not only of Derrida’s critique of temporal conventions of the preface genre, but more profoundly his observation that, the sign is itself always apart from the referent, both different from it and temporally deferred. Plato’s conversational stylistic of implicitness contains a meta-rhetorical critique of language that make post-structuralist linguistic insights seem like old news. By the end of the Theaetetus, Socrates brings the matter of using language to a humourous apotheosis by reminding Theaetetus of all the words that they have used in their discussion whose meaning they were attempting to discover – i.e., they did not understand what they were saying (196d- 197a). Again, both Theaetetus and Socrates seem to be sharing a joke when they acquiesce to their dependency on the promiscuous linguistic sign (since we have no other language, no254 other instrument for thinking). At the end of the dialogue, Socrates musingly asks “Shall I venture to say what the nature of knowing is?” Theaetetus responds playfully “you shall have full pardon for not refraining from those terms [that is, either ‘wrong’ ones, or ones whose meaning we do not know].” And then demonstrating that how things are said is inevitably problematic, Socrates replies with a tautological and would-be “empty” definition of 234 knowledge: would be if, that is, its linguistic cues and interconnected ironies did not produce relevant implicatures about opinion, the limitations of human understanding, and language’s referential incompetence. He says musingly to Theaetetus: “Have you heard what they say nowadays that knowing is? . . . They say it is having knowledge” (Loeb 197a). There is no natural tie between knowing and the word knowledge, no umbilical chord between words and things, as Cratylians with their code metaphors and rule-governed grammar dream. Knowledge is not one thing, but a multiplicity of things. Human language works because of the sign’s promiscuity. We can see in the variety of implicatures readers might infer from what “they say” knowing is (given its context at the end of Theaetetus) what Halperin might mean when he refers to Plato’s “self-cancelling textual practices” (120). Plato’s conversational stylistic of implicitness overflows with ironies and other involvement strategies, inviting (or forcing) readers to decide whether, for instance, it was Socrates’ obscure expression, Theaetetus’ obtuseness, or a thematic implication about the context dependency of meaning that was most relevant in the reporting of Theaetetus’ question “levgeiV”; (‘what do you mean?’). The Platonic text’s superabundance of implicitness demanding that readers decide makes it easy to see how Derrida comes to his reader-writer thematic – the reader’s role in the scene of writing – when he undertakes to talk a bit about Plato. Halperin formulates Plato’s textual practices as producing an “erotics of reading”; this interpretive activity and affective interest in texts stems from what he calls “Plato’s theory of desire” in which “the perpetual loss and renewal of understanding on the part of the interpreter . . . reflects a familiar erotic operation, namely the dialectic of presence and absence that structures the phenomenology of desire – in this case the phenomenology of hermeneutic desire” (120-26). The motion or movement entailed in the loss and renewal of understanding in hermeneutic desire conveys a sense of the activity of reader inference, deciphering the characters reported speech. Hermeneutic desire stimulates active thinking. Plato’s ironic-conversational-textual strategy entails perpetual movement in “the 235 dialectical alternation of comprehension and incomprehension” (Halperin 120). His strategy has educational aims – moving from one position to another, mental callisthenics, wrestling with ideas – demands created by a stylistic, that never seems to sit still. Plato produces a philosophic thinking activity: “texts that mimetically construct the desires of their readers, engaging them in a hermeneutic activity that imitates the philosophical activity” (Halperin 125). Naas observes that “there is a long tradition in philosophy of taking on the tradition, calling into question certain assumptions, canons, and institutional practices” and suggests that taking on the tradition is another way of glossing deconstruction (Taking xviii, xix). Since Plato founded the tradition by taking on what people say, Derrida might be less like an adversary and more like a advocate. Naas argues that Derrida is concerned with “the reception of Plato in the tradition” (Receiving 7). Derrida doesn’t seem so much to take Plato on as he does to re-perform him, to reproduce the thinking experience, the discovery of what we do not know, the fine distinctions that the Platonic text exacts from its readers. The Platonic text gives us all kinds of clues and leads, Plato’s indirection provides increasingly nuanced understandings of particular contexts, but we discover more about what we don’t know than what we do. The more we reread the more difficult it is to imagine the reader who would to be able to say “exactly what Plato really meant.” Although it might have been difficult to get Derrida to admit it, he seems to share Plato’s educational motives. Halperin’s comment – that “Plato’s texts read us . . . as much as we read them, even if they also seem to write, to prescribe, our own response to them” – might equally apply to Derrida’s (125). “Plato has devised for us, not to put an end to the dialectic by resolving it in favour of one alternative or another and thereby closing down the circuit through which our desire circulates but to keep it moving so as to prolong the erotic tension that animates our existence as readers and lovers of Plato’s texts” (Halperin 129). This erotic tension that makes us reread, that draws us into its grip inspires Plato’s celebrated analogy between Eros and philosophy and signals the kinds of motivations that might complicate language theory’s reader-writer relations, the way that writers welcome their readers to the scene of 236 writing. The currency of contemporary re-evaluations of Plato – the fact that over two millennia later his text warrants such interest – might give us a glimpse into some of the complexities of Derrida’s relation to Plato and what motivates this mimesis. But then, if one aspired to a 2500- year love affair with one’s readers, what better role model? 237 Post Script Language use is always multiply motivated; it always entails more than one thing going on at the same time. The conversational stylistic actively exploits language’s ability to do many things at the same time: this multi-functionality creates a degree of compression in the piling on of meanings, interconnections and rhetorical effects. Compression accomplishes a variety of ends, in addition to the avoidance of prolixity. The compression of the conversational stylistic presents the image of linguistic complexity that a sceptical-rhetorical theory of language embraces. The multiple purposes motivating this dissertation included pointing out the similarities, identities and congruities between twentieth century views on linguistic indeterminacy and the instability of reference, the role of language in knowledge formation ,and views of language from antiquity. Such a project also invited a degree of stylistic compression – a piling on of interconnections between theorists from diverse backgrounds and eras. Compression in this dissertation is a way of presenting the linguistic complexity I have sought to expose: despite the inimicability of such meta-discursive, condensing stylistic to scholarly writing with its ideal of artfully produced transparency. The goal of this dissertation is to promote consciousness of linguistic mediation and linguistic complexity. Its goal was not “a single guiding thesis,” but a serviceable, generalized understanding of language’s diverse nature: an understanding which might facilitate keeping linguistic mediation in mind, for keeping linguistic consciousness at the intellectual scene (D 7). In particular, a broad understanding of language’s performative propensities – an understanding of the interplay between language’s performative and referential capacities – seems called for. I have lumped diverse characteristics under the term ‘performative’ to describe the way language acts; but, since I have only scratched the surface of language’s performativity, this is a risky reduction . The potentially diverse, non-descriptive-referential dimensions of language have been lumped together in order to put descriptive reference into relief. I have described reported speech as paradigmatic of language’s performativity because speech reports do something: and they do something entirely other than their descriptive- 238 referential meanings. Speech reports have been described as duplicitous and contradictory – suggesting that performativity is similarly underhanded. Quite apart from the findings of scholars of reported speech about the artful nature of ordinary conversation (which calls into question the relationship between ordinariness and artfulness in language use), I want to problematize the polarity of the characterizations that have come up in this dissertation (ordinariness, naturalness, duplicity and artfulness). If language’s performative inclinations in direct discourse look like extreme duplicity, then descriptive reference is sitting in the “norm” position. But we need to query how descriptive reference came to be the benchmark. The binary attributing “naturalness” to descriptive reference and “artifice” to the performative might not take into account the nature of the symbolic order itself. Symbols are multi-purpose and perhaps fundamentally artificial by nature. The semiotic realm’s artificiality – its tropological reality – would be natural. Perhaps for such an artificial sphere the illusion of a simple, natural word- thing relation – the essence of descriptive reference – is the apex of artifice. It is the extreme apotheosis of artfully produced transparency, testifying not only to language’s excellence, but to the fundamentally performative nature of the linguistic semiotic realm. 239 Works Cited Ackrill, J. L. “Comments on Raphael Demos's Plato's Philosophy of Language.” Journal of Philosophy 61 (1964): 610-2. Print. ——. Essays on Plato and Aristotle. Oxford: Clarendon, 1997. Print. Adorno, Theodor W. Minima Moralia: Reflections of a Damaged Life. Tran. E. F. N. Jephcott. London: Verso, 2005. Print. Akehurst, Thomas L. “Writing History for the Ahistorical: Analytic Philosophy and its Past.” History of European Ideas 35.1 (2009): 116-21. Print. Al-Maini, Doug. “Technique and Teleology in Plato’s Rhetoric.” The European Legacy 10.4 (2005): 283-98. Print. Aristotle. Art of Rhetoric. Tran. John Henry Freese. Loeb Classical Library ed. XXII Vol. Harvard University Press, 1926. Print. ——. Categories, on Interpretation, an Prior Analytics. Trans. Hugh Tredennick and H. P. Cooke. Loeb Classical Library ed. I Vol. Harvard University Press, 1938. Print. ——. Introduction to Aristotle. Ed. Richard McKeon. New York: The Modern Library, 1947. Print. ——. Metaphysics. Stevenson, Daniel C.;. 2009.Web. The Internet Classic Archive. . ——. On Rhetoric: A Theory of Civic Discourse. Tran. George Alexander Kennedy. Ed. George Alexander Kennedy. New York: Oxford University Press, 1991. Print. ——. Posterior Analytics and Topica. Trans. Hugh Tredennick and E. S. Forster. Loeb Classical Library ed. II Vol. Harvard University Press, 1960. Print. Arthos, John. “Gadamer's Rhetorical Imaginary.” Rhetoric Society Quarterly 38.2 (2008): 171-97. Print. Austin, J. L. How to do Things with Words. 2d ed. 1955 Vol. Oxford Eng.: Clarendon Press, 1975. Print. ——. The Works of J.L. Austin Electronic Edition. InteLex Past Masters ed. Charlottesville, Virginia, USA: InteLex Corporation, 2000. Web. Bacon, Francis. Francis Bacon: free web books, online.Web. 4/10/2009 . Bacon, Sir Francis. The Works of Sir Francis Bacon. Ed. Anniina Jokinen. 24 May 2010 2010.Web. Luminarium. . Bakhtin, M. M. The Dialogic Imagination: Four Essays. Trans. J. Michael Holquist and Caryl Emerson. Ed. J. Michael Holquist. 1 Vol. Austin: University of Texas Press, 1981. Print. ——. Problems of Dostoevsky's Poetics. Tran. Caryl Emerson. Ed. Caryl Emerson. 8 Vol. Minneapolis: University of Minnesota Press, 1984. Print. ——. Speech Genres and Other Late Essays. Tran. Vern W. McGee. Eds. Michael Holquist and Caryl Emerson. 1st ed. 8 Vol. Austin: University of Texas Press, 1986. Print. Baumann, Nicola, and Julius Kuhl. “Intuition, Affect, and Personality: Unconscious Coherence Judgments and Self-Regulation of Negative Affect.” Journal of personality and social psychology 83.5 (2002): 1213-23. Print. 240 Bazerman, Charles. Shaping Written Knowledge: The Genre and Activity of the Experimental Article in Science. Madison: University of Wisconsin Press, 1988. Print. Becker, A. L. Beyond Translation. Ann Arbor: University of Michigan Press, 2000. Print. Benitez, Eugenio. “Argument, Rhetoric, and Philosophic Method: Plato's Protagoras.” Philosophy and Rhetoric 25.3 (1992): 222-52. Print. Bennington, Geoffrey, and Jacques Derrida. Jacques Derrida. Tran. Geoffrey Bennington. Chicago: University of Chicago Press, 1993. Print. Berkenkotter, Carol, and Thomas N. Huckin. Genre Knowledge in Disciplinary Communication : Cognition/ culture/power. Northvale, NJ: L. Erlbaum Associates, 1995. Print. Bett, Richard. “Nietzsche on the Skeptics and Nietzsche as Skeptic.” Archiv fuer Geschichte der Philosophie 82.1 (2000): 62-86. Print. ——. “On the Pre-History of Pyrrhonism.” Proceedings of the Boston Area Colloquium in Ancient Philosophy 14 (1998): 137-66. Print. Biletzki, Anat, and Anat Matar, eds. The Story of Analytic Philosophy: Plot and Heros. New York: Routledge, 1998. Print. Bitzer, Lloyd F. “The Rhetorical Situation.” Philosophy and Rhetoric 1 (1968): 1-14. Print. Blondell, Ruby. The Play of Character in Plato's Dialogues. Cambridge: Cambridge Univ Pr, 2002. Print. Bolinger, Dwight L. “Syntactic Blends and Other Matters” Language. 37. 3 (1961): 366-381. Print. Bowers, Kenneth S., and And Others. “Intuition in the Context of Discovery.” Cognitive psychology 22.1 (1990): 72-110. Print. Brogan, Walter. “Plato's Pharmakon: Between Two Repetitions in Derrida and Deconstruction.” Derrida and Deconstruction. Ed. Hugh Silverman. New York: Routledge, 1989. 7-23. Print. Burke, Kenneth. A Grammar of Motives. Berkeley: University of California Press, 1969. Print. ——. Language as Symbolic Action: Essays on Life, Literature, and Method. Berkeley: University of California Press, 1966. Print. ——. A Rhetoric of Motives. Berkeley: University of California Press, 1969. Print. Buxton, R. G. A. Persuasion in Greek Tragedy: A Study of Peitho. Cambridge: Cambridge University Press, 1982. Print. Calsamiglia, Helena, and Carmen López Ferrero. “Role and Position of Scientific Voices: Reported Speech in the Media.” Discourse Studies 5.2 (2003): 147. Print. Cavell, Stanely. Philosophical Passages: Wittgenstein, Emerson, Austin, Derrida. Cambridge: Blackwell, 1995. Print. Cavell, Stanley. “The Availability of Wittgenstein's Later Philosophy.” Philosophical Review 71 (1962): 67-93. Print. ——. “Must we Mean what we Say?” Inquiry: An Interdisciplinary Journal of Philosophy 1 (1958): 172-212. Print. Cherniss, H. The Riddle of the Early Academy. BERKELEY: UNIV OF CALIF PR, 1945. Print. 241 Clark, Herbert H., and Richard J. Gerrig. “Quotations as Demonstrations.” Language 66.4 (1990): 764- 805. Print. Clay, Diskin. “Reading the Republic.” PLATONIC WRITINGS: PLATONIC READINGS. Ed. Charles L. Jr Griswold. NEW YORK: ROUTLEDGE19-32. Print. Coe, Richard M., Lorelei Lingard, and Tatiana Teslenko. Eds. The Rhetoric and Ideology of Genre: Strategies for Stability and Change. Cresskill, N.J.: Hampton Press, 2002. Print. Cornford, F. M. Plato's Theory of Knowledge. LONDON: ROUTLEDGE, 1960. Print. Cossutta, Frédéric. “Dialogic Characteristics of Philosophical Discourse: The Case of Plato's Dialogues.” Philosophy and Rhetoric 36.1 (2003): 48-76. Print. Covino, William A. The Art of Wondering: A Revisionist Return to the History of Rhetoric. Portsmouth, NH: Boynton/Cook Publishers, 1988. Print. Culler, Jonathan D., and Kevin Lamb, eds. Just being Difficult?: Academic Writing in the Public Arena. Stanford: Stanford University Press, 2003. Print. Culler, Jonathan. “Bad Writing and Good Philosophy.” Just being Difficult? Academic Writing in the Public Arena. Eds. Kevin Lamb and Jonathan Culler. Stanford, CA: Stanford UP, 2003. 43-57. Print. ——. “Philosophy and Literature: The Fortunes of the Performative.” Poetics Today 21.3 (2000): 503. Print. de Man, Paul. Blindness and Insight: Essays in the Rhetoric of Contemporary Criticism. 2nd , rev. -- ed. 7 Vol. Minneapolis: University of Minnesota Press, 1983. Print. ——. “Dialogue and Dialogism.” Poetics Today 4.1 : 99-107. Print. ——. “Genesis and Genealogy in Nietzsche's the Birth of Tragedy.” Diacritics: A Review of Contemporary Criticism 2.4 (1972): 44-53. Print. ——. “Nietzsche's Theory of Rhetoric.” Symposium 28 (1974): 33-51. Print. ——. “The Resistance to Theory.” Yale French Studies 63 (1982): 3-20. Web. de Man, Paul. “Nietzsche's Theory of Rhetoric.” Symposium 28 (1974): 33-51. Print. de Man. “The Epistemology of Metaphor.” Cricial Inquiry 5.1 (1978): 13-30. Print. de Rijk, Lambertus Marie. Aristotle: Semantics and Ontology. Leiden: Brill, 2002. Print. de Saussure, Ferdinand. Course in General Linguistics. Tran. Roy Harris. Eds. Charles Bally, Albert Sechehaye, and Albert Riedlinger. LaSalle, Ill.: Open Court, 1986. Print. de St. Aubin, Ed. “The Joy and Shame of Intellectual Probate.” PsycCRITIQUES 42.1 (1997): 76-7. Print. Demos, Raphael. “Plato's Philosophy of Language.” Journal of Philosophy 61 (1964): 595-609. Print. Derrida, Jacques. Dissemination. Tran. Barbara Johnson. Chicago: Chicago University Press, 1981. Print. ——. La Dissémination. 265 Vol. Paris: Editions du Seuil, 1972. Print. ——. L'Ecriture Et La Différence. Paris: Editions du Seuil, 1967. Print. ——. Edmund Husserl's Origin of Geometry: An Introduction. Tran. John P. Leavey Jr. Ed. John P. Leavey Jr. Lincoln: University of Nebraska Press, 1989. Print. 242 ——. De La Grammatologie. Paris: Editions de Minuit, 1967. Print. ——. “The Law of Genre.” Critical Inquiry 7.1 (1980): 55-81. Print. ——. Limited Inc. Trans. Alan Bass and Samuel Weber. Ed. Gerald Graff. Evanston: Northwestern University Press, 1988. Print. ——. Marges De La Philosophie. Paris: Éditions de Minuit, 1972. Print. ——. Margins of Philosophy. Tran. Alan Bass. Chicago: University of Chicago Press, 1982. Print. ——. Monolingualism of the Other, Or, the Prosthesis of Origin. Tran. Patrick Mensah. Stanford: Stanford University Press, 1998. Print. ——. Of Grammatology. Tran. Gayatri Chakravorty Spivak. Correct ed. Baltimore: Johns Hopkins University Press, 1998. Print. ——. “The Retrait of Metaphor.” enclitic 2.2 (1978): 5-33. Print. ——. Speech and Phenomena, and Other Essays on Husserl's Theory of Signs. Tran. David B. Allison. Evanston: Northwestern University Press, 1973. Print. ——. Spurs: Nietzsche's Styles. Tran. Barbara Harlow. Chicago: University of Chicago Press, 1979. Print. ——. La Voix Et Le Phénomène, Introduction Au Problème Du Signe Dans La Phénoménologie De Husserl. Paris: Presses universitaires de France, 1967. Print. ——. Writing and Difference. Tran. Alan Bass. Chicago: University of Chicago Press, 1978. Print. Desjardins, Rosemary. “Why Dialogues: Plato's Serious Play.” PLATONIC WRITINGS: PLATONIC READINGS. Ed. Charles L. Jr Griswold. New York: Routledge, 1988. 110-125. Print. Deslauriers, Marguerite. “Plato and Aristotle on Division and Definition.” Ancient Philosophy (1990): 203-19. Print. Doerge, Friedrich Christoph. Illocutionary Acts: Austin's Account and what Searle made Out of it. http://tobias-lib.uni-tuebingen.de/frontdoor.php?la=en: Tobias-lib, 2004. Ellis, John M. Language, Thought and Logic. Evanston: Northwestern University Press, 1993. Print. Fahnestock, Jeanne. “Figures of Argument.” Informal Logic 24.2 (2004): 115-35. Print. ——. “Rhetoric in the Age of Cognitive Science.” The Viability of the Rhetorical Tradition. Eds. R. Graff, Arthur E. Walzer, and J. M. Atwill. Albany: SUNY Press, 2005. 159-179-135. Print. ——. “Rhetoric of Science: Enriching the Discipline.” Technical Communication Quarterly 14.3 (2005): 277-86. Web. ——. “Rhetorical Stylistics.” Language and Literature 14.3 (2005): 215-30. Print. Fish, Stanley E. “Epilogue: The Plain Style Question.” Self-Consuming Artifacts: The Experience of Seventeenth-Century Literature. Berkeley: U. of Calif. P, 1973. Print. ——. “Normal Circumstances, Literal Language, Direct Speech Acts, the Ordinary, the Everyday, the Obvious, what Goes without Saying, and Other Special Cases.” Critical Inquiry 4.4 (1978): 625- 44. Print. Fludernik, Monika. The Fictions of Language and the Languages of Fiction: The Linguistic Representation of Speech and Consciousness. London: Routledge, 1993. Print. 243 Follesdal, Dagfinn. “Analytic Philosophy: What is it and Why should One Engage in it?” Ratio: An International Journal of Analytic Philosophy 9.3 (1996): 193-208. Print. Foucault, Michel. The Birth of the Clinic : An Archaeology of Medical Perception. New York: Vintage Books, 1994. Print. Fowler, Roger. “Commentary.” New Literary History 12.3 (1981): 547-57. Web. Frede, Dorothea and Brad Inwood Eds. Language and Learning: Philosophy of Language in the Hellenistic Age Proceedings of the Ninth Symposium Hellenisticum. Cambridge: Cambridge Univ Pr, 2005. Print. Freedman, Aviva, and Peter Medway. Eds. Genre and the New Rhetoric. London ; Bristol, PA: Taylor & Francis, 1994. Print. French, Robert M. “The Computational Modeling of Analogy-Making.” Trends in cognitive sciences 6.5 (2002): 200. Print. Freydberg, Bernard. “Retracing Homer and Aristophanes in the Platonic Text.” Retracing the Platonic Text. Eds. John Sallis and John Russon. Evanston: Northwestern University, 2000. 99-112. Print. Gallop, David. “Image and Reality in Plato's Republic.” Archiv fuer Geschichte der Philosophie 47 (1965): 113-31. Print. ——. “Plato and the Alphabet.” Philosophical Review 72 (1963): 364-76. Print. Garber, Marjorie. “` ' (Quotation Marks).” Critical Inquiry 25.4 (1999): 653. Print. Garver, Newton. “Preface.” Speech and Phenomena, and Other Essays on Husserl's Theory of Signs. Tran. David B. Allison. Evanston: Northwestern University Press, 1973. ix-xxix. Print. Gill, Christopher, and T. P. Wiseman, eds. Lies and Fiction in the Ancient World. 1st University of Texas Press ed. Devon: University of Exeter Press, 1993. Print. Gill, Christopher. “Translating Plato.” Phronesis 43.2 (1998): 197-206. Print. Giltrow, Janet:. “'Curious Gentlemen': The Hudson's Bay Company and the Royal Society, Business and Science in the Eighteenth Century.” unpublished (2008)Print. Giltrow, Janet. “Genre as Engine of Heteroglossia.” unpublished (2007)Print. Giltrow, Janet Lesley. “”Anything You Say . . .”: Reported Speech in Judicial Opinion”. International Conference on Language and Law. May 2006, Duesseldorf, Germany. Print. ——. “Genre as Difference: The Sociality of Linguistic Variation.” Syntactic Variation and Genre. Eds. Heidrun Dorgeloh and Anja Wanner. Berlin: Mouton de Gruyter, forthcoming. Print. ——. “Modern Conscience: Modalities of Obligation in Research Genres.” Text 25.2 (2005): 171-99. Print. Glendinning, Simon. “Inheriting 'Philosophy': The Case of Austin and Derrida Revisited.” Ratio: An International Journal of Analytic Philosophy 13.4 (2000): 307-31. Print. Gonzalez, Francisco J. Dialectic and Dialogue: Plato's Practice of Philosophical Inquiry. Evanston: Northwestern University Press, 1998. Print. ——. “Introduction: A Short History of Platonic Interpretation and the ‘Third Way.’” The Third Way: New Directions in Platonic Studies. Ed. Francisco J. Gonzalez. Lanham: Rowman & Littlefield, 1995. 1-22. Print. 244 ——. The Third Way: New Directions in Platonic Studies. Lanham: Rowman & Littlefield, 1995. Print. Graham, Kenneth John Emerson. The Performance of Conviction: Plainness and Rhetoric in Early English Renaissance. Ithaca: Cornell University Press, 1994. Print. Grene, Marjorie Glicksman. A Portrait of Aristotle. Chicago: University of Chicago Press, 1963. Print. Grice, H. P. Studies in the Way of Words. Cambridge, Mass.: Harvard University Press, 1989. Print. Griswold, Charles L. Jr, ed. Platonic Writings: Platonic Readings. New York: Routledge, 1988. Print. Griswold, Charles L. “Irony in the Platonic Dialogues.” Philosophy and Literature 26.1 (2002): 84-106. Print. Groarke, Leo,. Ancient Skepticism. Stanford University.Web. The Stanford Encyclopedia of Philosophy. <>. Grondin, Jean. “Continental Or Hermeneutical Philosophy.” Interrogating the Tradition: Hermeneutics and the History of Philosophy. Eds. Scott Charles E. and Sallis John., 2000. 75-83. Print. Grote, George. Plato and the Other Companions of Sokrates. 1 Vol. London: Elibron Classics, 2005 (RP 1865). Print. Hacker, P. M. S. “The Rise of Twentieth Century Analytic Philosophy.” Ratio: An International Journal of Analytic Philosophy 9.3 (1996): 243-68. Print. Halliday, M. A. K., and Ruqaiya Hasan. Cohesion in English. London: Longman, 1976. Print. Halliday, M. A. K. “On Matter and Meaning: The Two Realms of Human Experience.” Linguistics and the Human Sciences 1.1 (2005): 59-82. Web. ——. “Towards a Language-Based Theory of Learning.” LINGUISTICS AND EDUCATION 5 (1993): 93-116. Web. Halliday, M. A. K. An Introduction to Functional Grammar. London: Edward Arnold, 1985. Print. ——. On Language and Linguistics. Ed. Jonathan J. Webster. 3 in the Collected Works of M.A.K. Halliday Vol. London: Continuum, 1985. Print. Halliwel, Stephen. The Aesthetics of Mimesis. Princeton: Princeton University Press, 2002. Print. Halperin, David M. “Plato and the Erotics of Narrativity.” Oxford Studies in Ancient Philosophy supplementary volume (1992): 93-129. Print. Hevern, Vincent W. Narrative Psychology: Internet and Resource Guide. Le Moyne College, Psychology Department. 28 December 2007 2007.Web. . Hobson, Marian. Jacques Derrida: Opening Lines. London: Routledge, 1998. Print. Holt, Elizabeth. “Reporting and Reacting: Concurrent Responses to Reported Speech.” Research on Language & Social Interaction 33.4 (2000): 425-54. Print. Honderich, Ted, ed. The Oxford Companion to Philosophy. Oxford: Oxford University Press, 1995. Print. Hyland, Ken. “Academic Attribution: Citation and the Construction of Disciplinary Knowledge.” Applied Linguistics 20.3 (1999): 341-67. Print. Jameison, Kathleen M. “Antecedent Genre as Rhetorical Constraint.” Quarterly Journal of Speech 61 (1975): 406-15. Print. 245 James, William. “The Sentiment of Rationality.” The Will to Believe. Cambridge: Harvard University Press, 1979. 57-89. Print. Jardine, Lisa. Francis Bacon, Discovery and the Art of Discourse. LONDON,: CAMBRIDGE UNIV PR, 1974. Print. Jarratt, Susan Carole Funderburgh. Rereading the Sophists : Classical Rhetoric Refigured. Carbondale: Southern Illinois University Press, 1991. Print. Johnson, Christopher. System and Writing in the Philosophy of Jacques Derrida. Cambridge: Cambridge University Press, 1993. Print. Johnson, Nan. “Reader-Response and the Pathos Principle.” Rhetoric Review 6.2 (1988): 152-66. Print. Kahn, Charles H. Plato and the Socratic Dialogue. Cambridge, England ; New York: Cambridge University Press, 1996. Print. Kennedy, George Alexander. A New History of Classical Rhetoric. Princeton, N.J.: Princeton University Press, 1994. Print. Kerferd, G. B. The Sophistic Movement. Cambridge: Cambridge University Press, 1981. Print. Kirby, John T. “A Classicist's Approach to Rhetoric in Plato.” Philosophy and Rhetoric 30.2 (1997): 190-202. Print. Kofman, Sarah. “Explosion I: Of Nietzsche's Ecce Homo.” Diacritics: A Review of Contemporary Criticism 24 (1994): 50-70. Web. ——. Nietzsche and Metaphor. Tran. Duncan Large. London: Athlone Press, 1993. Print. Kohrs Campbell, Karlyn, Kathleen Hall Jamieson, and Edwin Black. Form and Genre: Shaping Rhetorical Action. Falls Church, VA: The Speech Communication Association, 1979. Print. Kuhn, Thomas S. The Structure of Scientific Revolutions. 3rd ed. Chicago, IL: University of Chicago Press, 1996. Print. Lautréamont, Comte de. Maldoror & the Complete Works of the Compte De Lautréamont. Tran. Alexis Lykiard. Cambridge: Exact Change, 1994. Print. Liddell, H.G.and Robert Scott. An Intermediate Greek-English Lexicon. 7 ed. 2003. Print. (Citedth parenthetically as LSM) Leech, Robert, Denis Mareschal, and Richard P. Cooper. “Analogy as Relational Priming: A Developmental and Computational Perspective on the Origins of a Complex Cognitive Skill.” Behavioral & Brain Sciences 31.4 (2008): 357-414. Print. Leff, Michael. “The Uses of Aristotle's Rhetoric in Contemorary American Scholarship.” Argumentation 7 (1993): 313-27. Print. Leff, Michael. In Search of Ariadne's Thread: A Review of the Recent Literature on Rhetorical Theory. Central States Speech Journal, 1978. Print. ——. “The Topics of Argumentative Invention in Latin Rhetorical Theory from Cicero to Boethius.” Rhetorica: A Journal of the History of Rhetoric 1.1 (1983): 23-44. Print. ——. “Up from Theory: Or I Fought the Topoi and the Topoi Won.” RSQ: Rhetoric Society Quarterly 36.2 (2006): 203-11. Print. Lewontin, R. C. Biology as Ideology. New York: Harper Perennial, 1992. Print. 246 Linck, Matthew S. “Unmastering Speech: Irony in Plato's Phaedrus.” Philosophy and Rhetoric 36.3 (2003): 264-76. Print. Linell, Per. The Written Language Bias In Linguistics. 1982.Web. . Linell, Per, and Ivana Markova. “Acts in Discourse: From Monological Speech Acts to Dialogical Inter- Acts.” Journal for the Theory of Social Behaviour 23.2 (1993): 174-95. Print. Lipson, Carol S. “Francis Bacon and Plain Scientific Prose: A Reexamination.” Journal of Technical Writing and Communication 15.2 (1985): 143-55. Print. Lykiard, Alexis. “Introduction: Preface without Memoirs.” Maldoror & the Complete Works of the Compte De Lautréamont. Cambridge: Exact Change, 1994. 1-24. Print. Marx, Karl, and Friedrich Engels. German Ideology. New York: International Publishers, 1972. Print. Mayes, Patricia. “Quotation in Spoken English.” Studies in language 14.2 (1990)Print. McAdon, Brad. “Probabilities, Signs, Necessary Signs, Idia, and Topoi: The Confusing Discussion of Materials for Enthymemes in the Rhetoric.” Philosophy & Rhetoric 36.3 (2003): 223-47. Print. ——. “Reconsidering the Intention Or Purpose of Aristotle's Rhetoric.” Rhetoric Review 23.3 (2004): 216-34. Print. McIlwain, Doris. “Pleasure in Mind: Silvan Tomkins and Affect in Aesthetics, Personality Theory and Culture.” Theory & Psychology 17.4 (2007): 499-504. Print. McKeon, Richard. “Literary Criticism and the Concept of Imitation in Antiquity.” Modern Philology 34.1 (1936): 1-35. Print. McKeon, Richard. Rhetoric: Essays in Invention and Discovery. Ed. Mark Backman. Woodbridge, CT: Ox Bow Press, 1987. Print. Miller, Carolyn R. “The Aristotelian Topos: Hunting for Novelty.” Rereading Aristotle's Rhetoric. Ed. Arthur E. Walzer. Carbondale, IL: Southern Illinois UP, 2000. 130-146. Print. ——. “Aristotle's 'Special Topics' in Rhetorical Practice and Pedagogy.” Rhetoric Society Quarterly 17.1 (1987): 61-70. Print. Miller, James. “Is Bad Writing Necessary? George Orwell, Theodor Adorno, and the Politics of Language.” Lingua Franca: The Review of Academic Life 9.9 (2000): 33-44. Print. Mitscherling, J. “The Ancient and Current Quarrels between Philosophy and Rhetoric.” European Legacy 10.4 (2005): 271-82. Print. ——. “Plato's Misquotation of the Poets.” Classical Quarterly 55.1 (2005): 295-8. Print. Miyake, Akira, et al. “Inner Speech as a Retrieval Aid for Task Goals: The Effects of Cue Type and Articulatory Suppression in the Random Task Cuing Paradigm.” Acta Psychologica 115.2 (2004): 123. Print. Modrak, Deborah K. W. Aristotle's Theory of Language and Meaning. Cambridge: Cambridge Univ Pr, 2001. Print. Mortensen, Chris. “Plato's Pharmacy and Derrida's Drugstore.” Language and Communication 20.4 (2000): 329-46. Print. Murray, Penelope. Plato on Poetry., 1998. Print. 247 Myers, G. “Functions of Reported Speech in Group Discussions.” Applied Linguistics 20.3 (1999): 376- 401. Print. Myers, Greg. “The Pragmatics of Politeness in Scientific Articles.” Applied Linguistics 10.1 (1989): 1- 35. Print. Naas, Michael. Taking on the Tradition. Stanford: Stanford University Press, 2003. Print. Naas, Michael B. “Receiving the Tradition.” Interrogating the Tradition: Hermeneutics and the History of Philosophy. Eds. Charles E. Scott and John Sallis. Albany: Suny Press, 2000. Print. Naas, Michael. Turning: From Persuasion to Philosophy. Atlantic Highlands, N.J.: Humanities Press, 1995. Print. Nietzsche, Friedrich Wilhelm. Frederich Nietzsche on Rhetoric and Language. Trans. Sander L. Gilman, Carole Blair, and David J. Parent. Eds. Sander L. Gilman, Carole Blair, and David J. Parent. Oxford: Oxford University Press, 1989. Print. Nietzsche, Friedrich Wilhelm. Beyond Good and Evil: Prelude to a Philosophy of the Future. Tran. Marion Faber. Oxford ; New York: Oxford University Press, 1998. Print. ——. The Gay Science: With a Prelude in German Rhymes an an Appendix of Songs. Tran. Walter Kaufmann. New York: Vintage, 1974. Print. ——. “On Truth and Lie in the Extra-Moral Sense.” Philosophy and Truth: Selections from Nietzsche’s Notebooks of the Early 1870's. Tran. Daniel Breazeale. Ed. Daniel Breazeale. New Jersey: Humanities Press, 1979. 79-91. Print. O'Leary-Hawthorne, Diane. Not-being and Linguistic Deception. Edmonton: Academic, 1996. Print. Olson, Gary A. “Jacques Derrida on Rhetoric and Composition: A Conversation.” JAC: A Journal of Composition Theory 10.1 (1990): 1-21. Print. Parmenides, and David Gallop. Parmenides of Elea : Fragments : A Text and Translation with an Introduction. 18 Vol. Toronto: University of Toronto Press, 1983. Print. Pinker, Steven. Words and Rules: The Ingredients of Language. New York: Basic Books, 1999. Print. Plato. Republic. Tran. G. M. A. Grube. Indianapolis: Hackett, 1974. Print. Plato. The Collected Dialogues of Plato, Including the Letters. Eds. Edith Hamilton and Huntington Cairns. 71 Vol. New York: Pantheon Books, 1961. Print. ——. Euthyphro, Apology, Crito, Phaedo, and Phaedrus. Tran. Harold North Fowler. Ed. Harold North Fowler. Loeb Classical Library ed. I Vol. Cambridge: Harvard University Press, 1914. Print. ——. Laches, Protagoras, Meno, and Euthydemus. Tran. W. R. M. Lamb. Ed. W. R. M. Lamb. Loeb Classical Library ed. II Vol. Cambridge: Harvard University Press, 1924. Print. ——. Plato: Complete Works. Eds. John M. Cooper and D. S. Hutchinson. Indianapolis: Hackett Publishing Company, 1997. Print. ——. Republic 1-5. Tran. Paul Shorey. Ed. Paul Shorey. Loeb Classical Library ed. V Vol. Cambridge: Harvard University Press, 1930. Print. ——. Republic 6-10. Tran. Paul Shorey. Ed. Paul Shorey. Loeb Classical Library ed. VI Vol. Cambridge: Harvard University Press, 1935. Print. ——. Theaetetus and Sophist. Tran. Harold North Fowler. Ed. Harold North Fowler. Loeb Classical Library ed. VII Vol. Cambridge: Harvard University Press, 1921. Print. 248 Press, Gerald A., ed. Who Speaks for Plato?: Studies in Platonic Anonymity. Lanham: Bowman & Littlefield Publishers, Inc., 2000. Print. Preston, Aaron. Analytic Philosophy: The History of an Illusion. London: Continuum, 2007. Print. Richards, Jennifer. Rhetoric. New York: Routledge, 2008. Print. Riffaterre, Michael. “Criteria for Style Analysis.” Word 25 (1959): 154-74. Print. ——. “Stylistic Context.” Word 16 (1960): 207-18. Print. Rinon, Yoav. “The Rhetoric of Jacques Derrida I: Plato's Pharmacy.” Review of Metaphysics 46.2 (1992): 369-86. Print. Robinson, John Mansley. An Introduction to Early Greek Philosophy: The Chief Fragments and Ancient Testimony, with Connecting Commentary. Boston: Houghton Mifflin, 1968. Print. Robinson, Richard. “A Criticism of Plato's Cratylus.” Philosophical Review 65 (1956): 324-41. Print. ——. “The Theory of Names in Plato's Cratylus.” Revue Internationale de Philosophie 9: 221-36. Print. Román Alcalá, Ramón. “The Skepticism of the New Academy: A Weak Form of Platonism?” Philosophical Inquiry: International Quarterly 25.3-4 (2003): 199-216. Print. Rowe, Christopher. “Plato.” Phronesis: A Journal of Ancient Philosophy 153.1 (1998): 84-92. Print. Russon, John Edward, and John Sallis. Retracing the Platonic Text. Evanston, Ill.: Northwestern University Press, 2000. Print. Ryle, Gilbert. “Letters and Syllables in Plato.” Philosophical Review 69 (1960): 431-51. Print. Sallis, John,1938-. Being and Logos : Reading the Platonic Dialogues. 3rd ed. Bloomington: Indiana University Press, 1996. Print. Sapir, Edward. “Language as a Form of Human Behavior.” The English Journal 16.6 (1927): 421-33. Web. Schiappa, E. “Did Plato Coin Rhetorike?” American journal of philology 111.4 (1990): 457. Print. ——. “Rhetorike: What's in a Name? Toward a Revised History of Early Greek Rhetorical Theory.” The Quarterly journal of speech 78.1 (1992): 1. Print. ——. “Sophistic Rhetoric: Oasis Or Mirage?” Rhetoric review 10.1 (1991): 5. Print. Schiappa, Edward. The Beginnings of Rhetorical Theory in Classical Greece. New Haven, CT: Yale University Press, 1999. Print. Scott, Charles E., and John Sallis. Interrogating the Tradition : Hermeneutics and the History of Philosophy. Albany, NY: State University of New York Press, 2000. Print. Searle, John R. “Reiterating the Differences: A Reply to Derrida.” Glyph: Johns Hopkins Textual Studies 1 (1977): 198-208. Print. Sedgwick, Eve Kosofsky, Adam Frank, and Irving E. Alexander. Shame and its Sisters : A Silvan Tomkins Reader. Durham: Duke University Press, 1995. Print. Short, Mick, Elena Semino, and Martin Wynne. “Revisiting the Notion of Faithfulness in Discourse Presentation using a Corpus Approach.” Language & Literature 11.4 (2002): 325. Print. Shrimpton, Gordon S. “Accuracy in Thucydides.” Ancient History Bulletin 12.3 (1998): 71-82. Print. 249 ——. “Beyond the Limits: New Thoughts on the Origin and Development of Ancient Historical Writing.” Ancient History Bulletin 15.2 (2001): 50-62. Print. ——. “Herodotus' Intellectual Context.” Ancient History Bulletin 17.3-4 (2003): 149-57. Print. Shukman, Ann. “Linguistics/Language/Literature: A Commentary.” New Literary History: A Journal of Theory and Interpretation 12.3 (1981): 563-9. Print. Sidgwick, H. “The Sophists.” Journal of Philology 4 (1872): 288-307. Print. Skorupski, John. “Why did Language Matter to Analytic Philosophy?” Ratio: An International Journal of Analytic Philosophy 9.3 (1996): 269-83. Print. Smith, P. Christopher. Apodeiknusthai, Dialegesthai, Peithein: A Reconstruction of Plato's Methods of Argument in the Phaedo., 1995. Print. Sorell, Tom, and G. A. J. Rogers, eds. Analytic Philosophy and History of Philosophy (Mind Association Occasional Series). Oxford: Clarendon Press, 2005. Print. Sparshott, Francis. “Plato's Ethics Dead Or Alive.” Apeiron: A Journal for Ancient Philosophy and Science 33.2 (2000): 145-61. Print. Sperber, Dan, and Deirdre Wilson. “Pragmatics, Modularity and Mind-Reading.” Mind & Language 17.1 (2002): 3. Print. ——. Relevance: Communication and Cognition. 2nd ed. Cambridge, MA: Blackwell Publishers, 1995. Print. Sprat, Thomas. The History of the Royal Society.Print. Statkiewicz, Max. Rhapsody of Philosophy: Dialogues with Plato in Contemporary Thought. University Park, Pennsylvania: The Pennsylvania State University Press, 2009. Print. Sternberg, Meir. “Proteus in Quotation-Land: Mimesis and the Forms of Reported Discourse.” Poetics Today 3.2 (1982): 107-56. Print. Strauss, Leo. “Persecution and the Art of Writing.” Social Research 8.1 (1941): 488-504. Print. Swales, John M., et al. “Consider this: The Role of Imperatives in Scholarly Writing.” Applied Linguistics 19.1 (1998): 97-121. Print. Tannen, Deborah. Talking Voices: Repetition, Dialogue, and Imagery in Conversational Discourse. Cambridge: Cambridge UP, 1989. Print. Taylor, A. E. Plato, the Man and His Work. 7th -- ed. London: Methuen, 1960. Print. Tejera, V. Plato's Dialogues One by One : A Dialogical Interpretation. Lanham, MD: University Press of America, 1999. Print. Thesleff, Holger. Studies in the Styles of Plato. HELSINKI,: SOC PHILOSO FENNICA, 1967. Print. Tigerstedt, Eugene Napoleon. Interpreting Plato. STOCKHOLM,: ALMQVIST & WIKSELL, 1977. Print. Tomkins, Silvan Solomon. Affect, Imagery, Consciousness, Vol. 4: Cognition: Duplication and Transformation of Information. New York, NY US: Springer Publishing Co, 1992. Print. Topolinski, Sascha, and Fritz Strack. “The Architecture of Intuition: Fluency and Affect Determine Intuitive Judgments of Semantic and Visual Coherence and Judgments of Grammaticality in 250 Artificial Grammar Learning.” Journal of Experimental Psychology: General 138.1 (2009): 39- 63. Print. van Dijk, Teun Adrianus. Macrostructures : An Interdisciplinary Study of Global Structures in Discourse, Interaction, and Cognition. Hillsdale, N.J.: L. Erlbaum Associates, 1980. Print. Voloshinov, Valentin Nikolaievich. Marxism and the Philosophy of Language. Trans. Ladislav Matejka and I. R. Titunik. Cambridge: Harvard University Press, 1986. Print. Welch, Kathleen E. The Contemporary Reception of Classical Rhetoric : Appropriations of Ancient Discourse. Hillsdale, N.J.: L. Erlbaum, 1990. Print. Whorf, Benjamin Lee. Language, Thought, and Reality: Selected Writings. Ed. John B. Carroll . Cambridge: MIT Press, 1965. Print. Wittgenstein, Ludwig. The Collected Works of Ludwig Wittgenstein. Electronic Edition. InteLex Past Masters ed. Charlottesville, Virginia, USA: InteLex Corporation, 1998. Web. ——. Tractatus Logico-Philosophicus. 1922. Tran. C.K. Ogden. London: Routledge, 1992. ——. Philosophical Investigations: The German Text, with a Revised English Translation. Tran. G. E. M. Anscombe. 3rd ed. Oxford; Malden, Mass.: Blackwell, 2001. Print. Woolf, Raphael. “The Written Word in Plato's Protagoras.” Ancient Philosophy 19.1 (1999): 21-30. Print. Zappen, James P. “Francis Bacon and the Historiography of Scientific Rhetoric.” Rhetoric Review 8.1 (1989): 74-88. Print. ——. “Francis Bacon and the Rhetoric of Science.” College Composition and Communication 26.3 (1975): 244-7. Print. Zuckert, Catherine H. Postmodern Platos. Chicago: The University of Chicago Press, 1996. Print."@en ; edm:hasType "Thesis/Dissertation"@en ; vivo:dateIssued "2011-05"@en ; edm:isShownAt "10.14288/1.0071580"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "English"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "Attribution-NonCommercial-NoDerivatives 4.0 International"@en ; ns0:rightsURI "http://creativecommons.org/licenses/by-nc-nd/4.0/"@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "Linguistic returns : the currency of sceptical-rhetorical theory and its stylistic inscription in the Platonic and Derridian text"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/30556"@en .