This essay deals with a Sophistic approach to argumentation known to ancient Greeks as antilogic and to Romans as controversia. I will use the terms interchangeably, along with other cognates like controversial reasoning and “in utramque partem,” or reasoning on both sides of a case. I will claim that controversia represents a major alternative to the Aristotelian tradition of argument. Broadly speaking, Aristotelian argument assumes an individual thinker who follows the dictates of deductive logic and who works to develop a sound proposition subsequently defended against all opposition. Controversia proceeds by placing multiple claims in juxtaposition and then negotiating the conflicts among them. It fully embraces the contingency of its setting, emphasizing dialogical interaction between specific parties, on a unique occasion, with a particular purpose. If Aristotelian argument is predicated on the drive towards formal validity and epistemological certainty, antilogic is based on the inevitable contention between probable opinions and the possibility of consensus among interlocutors. If Aristotelian argument proceeds in a linear, monological fashion, controversia approaches knowledge indirectly, tacking back-and-forth among opposing positions and assuming that “truth” is provisional and will reveal itself in mixed, ambigous form. Antilogic is thus dialogical, sceptical, contextual, and ultimately practical, all of which I will try to clarify as we proceed.
In previous work, I have traced the philosophic foundations of antilogic in the sceptical pragmatism of Protagoras and pursued the basic features of antilogical practice in a number of post-Periclean sources (Mendelson 1998). I have also explored Cicero’s De Oratore as an exemplary model of controversia (Mendelson 1997). As many of you know, the De Oratore displays considerable interest in an appropriate pedagogy for rhetoric, operating often as a master-class in the protocols of “in utramque partem.”[i] With the transition from Cicero to Quintilian, pedagogy takes center stage. The presence of controversial reasoning in Quintilian has, of course, been noted before (Bonner 1969, 1977; Clark 1957; Kennedy 1969; Marrou 1956; Murphy 1990). In the present essay, I will argue, however, that controversial reasoning is not just an incidental element, one techne “inter pares” (among equals); it is, instead, the very heart of Quintilian’s approach to rhetorical education. In other words, the Institutio Oratoria is principally involved in developing the concept of an “ideal orator;” and, as was the case with Cicero a century before, Quintilian is firmly committed to the notion that the “one and only true and perfect orator” is he who is able “to speak on both sides about every subject” (De Oratore 3.80). More specifically, I claim here that the pedagogy of controversia is ascendant in Quintilian because it fosters a sense of decorum (the ability to negotiate disagreement in ways appropriate to particular circumstances), while decorum, in turn, is essentially coordinate with prudence (the general ability to respond to controversy with dignity and common sense). Seen in this way, Quintilian articulates a syncretic vsion of argument, education, and culture, a vision of what Richard Lanham aptly describes as “the rhetorical paideia” (1993: 158; cf. 161).
In pursuit of this agenda, I will
1. briefly review the history of the controversial tradition,
2. explore Quintilian’s own method of argumentation and inquiry,
3. focus on the role of the progymnasmata exercises and declamation in the “Institutio,” and
4. extrapolate some general principles of controversial education from Quintilian and speculate on their potential contribution to a reconception of argument pedagogy today.

1. The History of Controversial Pedagogy
Quintilian is a neo-Sophist in the sense that his approach to education is pragmatic in focus and argumentative in nature (see Marrou 1956, Colson 1924, and Greer 1925). The first and, arguably, the most influential representative of Sophistic education was Protagoras, who declared himself “a Sophist and educator” and whose subject was the “proper care of [his students’] personal and public affairs,” so as to help them succeed as speakers and citizens (“Protagoras” 317b-318e). Among Protagoras’s many works, one book, the Antilogiae, appears to have been a textbook, and begins with the famous dictum that “on every issue there are two arguments (logoi) opposed to each other on everything” (Sprague 1972: 4). Marrou cites this concept as the core of Sophistic pedagogy and notes that Protagoras’s own educational program was “astonishing in its practical effectiveness” (1956: 51). Naturally, antilogical practice and pedagogy undergo significant transformation over time, most notably in the hands of the Academic sceptics.
In Book XII, Quintilian notes that the critical practices of the New Academy are particularly “useful” because their “habit of disputing both sides of the question approaches most nearly the actual practice of the courts” (12.2.25).[ii] In his commitment to Academic controversia, Quintilian is clearly following the lead of Cicero, who summarizes the Academic method this way: “. . . the only object of the Academics’ discussions is by arguing both sides of a question to draw out and fashion something which is either true or which comes as close as possible to the truth” (Academica 2.8). Such a position is founded on the antithetical scepticism of Pyrrho of Elias (4C BCE) who advocated a suspension of judgment during the assessment of alternative arguments in any particular case. Sextus Empiricus describes Pyrrhonistic scepticism as the ability to set up antitheses which account for the “equal weight of opposing states of affairs and arguments” (1.8). The sceptical tradition – as A. A. Long makes clear – is given institutional status in the New Academy first by Arcesilaus, who denies the existence of universal criteria adequate to warrant any claims to absolute truth. Instead, he transfers his attention from universals to the discovery of probable explanations arrived at through arguments between pro and contra positions (Long 1974: 91). Carneades continues the tradition by rejecting any dogmatic claims to certain knowledge, honing the practice of “in utramque partem” as a tool of critical scepticism, and insisting that prudential judgment is always contingent, never necessary. Judgment, in other words, cannot be dictated by criteria laid down in advance (see 2.13.2-5).
The principles of the New Academy pass into the Roman tradition though Cicero, who is unquestionably the major source for Quintilian’s own philosophical perspective. So while Quintilian may claim that it is unnecessary to “swear allegiance to any philosophical code” (12.2.26) and while his own philosophical interests tend in the direction of moral philosophy rather than epistemological speculation, his practice as a critic and educational theorist clearly reflect the traditions of the New Academy. In particular, he ascribes to the assumptions that all claims must be argued because more than one probable position exits, that judgment is best deferred as alternative logoi are weighed, and that criteria for judgment are developed out of the circumstances of the case. The pragmatism of his pedagogy is consequently grounded in a substantial philosophical tradition, a tradition that elevates the methods of argument themselves to the status of philosophical praxis.
I skip over here the interesting historical events that condition the adaptation of controversia in the Late Republic and Early Empire. I point out only that, as Chester Starr notes, “when one man became sovereign in Rome . . . the significance of political debate waned swiftly” (1965: 51). Indeed, the inevitable decline in oratory became a favorite subject for such first-century writers as Seneca the Elder, Petronius, and Tacitus. In this period of decline, says Grube, “rhetoric took refuge in the schools” (1965: 257), while much public oratory was given over to sententiousness and declamatory display. In such a climate, Quintilian is distinctly neo-Sophistic in his insistence on practical argument. Nowhere is this emphasis more emphatic than in his own methods of inquiry.

2. Quintilian’s Critical Method
Quintilian opens the Institutio Oratoria this way: “I was asked by certain of my friends to write something on the art of speaking . . . [because] they urged that previous writers on the subject had expressed different and at times contradictory opinions, between which it was very difficult to choose” (1. Pr. 1-2). Several books later, in his discussion of rhetorical invention, he notes that his first task is to canvas “the infinite diversity of opinions among writers on this subject” (3.1.7; cf. 3.1.1). The initial step for Quintilian, then, is to survey the “multiplex ratio disputandi” (the multiple ideas in dispute) that make-up the landscape of opinion on any point at issue. In the process, he is distinctly non-dogmatic, remaining independent of the various schools that dominated the educational theory of his day and allowing his readers to exercise their own judgment in reviewing a controversy.
Instances of Quintilian’s critical method are available at every turn in the Institutio. For example, after the reference to “contradictory opinions” that opens Bk. I, Quintilian immediately engages the question of whether or not it is better to educate a child at home or at public school (see 1.2.2-17). “Contradictory opinions,” he repeats, fully condition this topic and must be acknowledged, for while “eminent authorities” favor the public schools, “(i)t would . . . be folly to shut our eyes to the fact that there are some who disagree” (2.2.2). These critics, he goes on, are “guided in the main by two principles,” and he lays out each of these contra-arguments in significant detail. What is particularly interesting about this exercise in argumentation is that Quintilian begins with prolepsis, the anticipation of opposition, and in dealing with differences he avoids a simple claim/rebuttal structure, choosing instead to oscillate back and forth between contesting positions. The procedure as a whole operates, as Colson noted, more like a “discussion” than a treatise, and this dialogic approach becomes standard practice throughout the Institutio (1924: xxxix).

Similar examples of controversia are everywhere. In Bk. II alone, Quintilian takes up such issues as the choice of an appropriate teacher, memorizing commonplaces, the controversy over declamation, and the place of rules in oratorical training. The protocol of inquiry, analysis, and invention in all cases is controversia: the author first surveys the diversity of opinion on the topic in order to weigh the probabilities on each side. In his discussion of declamation, he writes that “I now come to another point in which the practice of teachers has differed. Some have not been content with . . . . Others have merely suggested [that] . . . . Both practices have their advantages . . . . But if we must choose one . . . ” (2.6.2). The dominant tropes of these supremely non-dogmatic inquiries are “on the contrary” and “on the other hand,” as the rhetor works his way through the various nuances of an argument and models for his readers the actual practice of controversial reasoning.
In sum, controversial methodology is ubiquitous in the Institutio because for Quintilian every question involves an “infinite diversity of opinions” (cf. 3.11.2). In confronting this multiplicity, Quintilian would himself reflect the breadth of interest advocated by Cicero’s Crassus and sample “all the available” arguments as a prelude to judgment. And while the argumentative exercises that fill out the Institutio may not always rise to the level of theoretical insight imagined by Crassus, there remains an admirable congruence between Quintilian’s own critical method and the practice of argumentation that he would advocate for his students (see 2.2.8).

3. The Progymnasmata, Declamation, and the Protocols of Argument
Roman students began composition study with a grammaticus, a teacher responsible for both grammar and an introduction to literature and literary style. The grammaticus would initiate composition training with the progymnasmata, a series of increasingly complex exercises fully involved with argument from the outset (Marrou 1956: 274ff, Bonner 1977: 213-49). At about sixteen, the student progressed to the tutelage of the rhetor, moving to the more difficult exercises in which the protocols of argument become the explicit focus of study. The exercises begin with a retelling of fables in which students “feign” the speech of given characters addressing contentious topics, such as monkeys deliberating on the founding of a city (in Clark 1957: 182). Composition, therefore, begins with imitation and impersonation, and in the context of mock-debate. Students pass next to “fictitious narratives” from literary sources and imitate the conversation of the people involved, like Medea justifying the theft of the Golden Fleece. These stories (called “argumentum”) were followed by chreia, exercises based on well- known maxims, like “money is the root of all evil.” In this case, the student was asked to provide the argumentative reasoning that supports the claim inherent in the maxim itself (see Hock and O’Neill). It is notable that even in these early exercises, the young rhetor is routinely given a specific character along with some situational data, so that invention always proceeds in relation to the requirements of a particular argumentative context. Moreover, rhetorical invention is, from the beginning, dialogical (always in response to previous speech) and practical (always generated with a particular occasion in mind).
While the early progymnasmata are often argumentative, argument itself comes to the foreground in the exercise of “refutation and confirmation.” Quintilian suggests that in response to a literary episode, students “annex” a number of claims on both sides of the case, thereby establishing dialogue between competing logoi rather than propositional reasoning as the framework for argument (2.4.18-19). And because the students would recite their compositions aloud to the class, all were exposed not simply to binary oppositions but to highly varied perspectives on such subjects as whether or not Romulus could actually have been suckled by a she-wolf (2.4.18). In these exercises, says Quintilian, “the mind is exercised” by the variety and multiplicity inherent in the topics, as the rhetor must deal not simply with abstract conceptions of pro and con but with “degrees” of vice and virtue (2.4.21).
The increasingly subtle challenges in argumentation progress to “comparisons” between characters and to “impersonations,” such as Priam pleading with Achilles for the return of Hector’s body. Finally, the progymnasmata culminate in philosophical “theses” and in debates on the law, both of which tend to complicate a priori assumptions, subvert simple binaries, and remind students that controversiality suffuses philosophical as well as literary composition. Throughout the exercises, the pedagogical focus remains essentially the same: the rhetor, unlike the philosopher and dialectician, is operating in response to specific contingencies by calculating the relative merits of opposing positions and developing the skills of sceptical inquiry, rhetorical invention, and pragmatic judgment. At all points in the process, the student-orator is guided by the principles of “in utramque partem” and contemplates not simply what can be said in behalf of a proposition but also what can be said in favor of the other side. Because each student must routinely compose orations that contradict each other, it is not so much the truth of one’s claim as it is the process of argumentation that is the ultimate subject of the progymnasmata and its elegant continuum of exercises.

Two additional ideas deserve mention here. First, Quintilian allocates a pivotal role to stasis theory (3.6); and, as Michael Carter points out, stasis – the effort to define the specific point at issue in argument – originates in the contention of opposing forces (1988: 98-99). The very act of arriving at a stasis is an act of controversiality, a conversation among contrasting opinions in a shared conflict. Second, in Bk. X Quintilian digresses to emphasize the role of “facilitas,” the resourcefulness and spontaneity acquired from continual interaction with other discourse. Such facility leads not only to a storehouse of materials appropriate to any argument, but also to the habit of easy exchange that allows orators to respond in accordance with all situations (10.1-2). Like sprezzatura, its Renaissance counterpart, “facilitas” is an element of character or ethos, a habit of mind to be nurtured by exposure to both opposition and variety. Both stasis theory and rhetorical “facilitas,” therefore, assume the importance of opposing positions in argument.
We pass now to declamation, which Quintilian calls “the most useful of rhetorical exercises” (2.10.2). The exercises themselves are mock forensic or judicial debates on specific points of law or history in which the student orator takes on a persona and works within the confines of a situational narrative. Indeed, if declamation is presented effectively, it should mimic, says Quintilian, the “real contests” and messy debates the student will encounter in public life (10.1.4). By the first-century CE, declamation had been divided into two kinds: the suasoriae or deliberative speeches on questions of history or politics, and the controversiae or forensic speeches on specific legal cases. As for suasoria, Philostratus lists these examples: the Spartans debate whether or not to build a wall and fortify themselves from attack, and Isocrates attempts to dissuade Athenians from their dependence on the sea (1965: 514 & 584). In most cases, student-orators were asked not only to respond to historical circumstances they had studied in literary sources, but to impersonate a specific character and address a particular audience. Matters of ethos, audience, and decorum were therefore paramount. Before I take up these matters, it may help if we have some idea of the classroom practices that characterized instruction in declamation.
The procedure was as follows: the teacher would present a declamatory problem and provide some introductory analysis (“divisio”) of the case, addressing opposing perspectives and how these might be arranged and presented. The students were then assigned the same or a similar case and allowed to select a stand.
They would then write out and read their initial draft to the teacher, who would question all pupils carefully in order to “test their critical powers” (2.5.13). It was assumed that the young orator would deal with pertinent aspects on both sides of the case, not just those in favor of the chosen position. The student would next prepare a more polished composition for memorization and delivery before the class as a whole, and sometimes before the public at-large. A distinctive feature of the declamatory process, then, was that the speeches were constructed with a responsive audience in mind. Typically, all students would declaim either for or against the same case, so that each speech was subject to peer review and examined in the context of diverse opinion. Further, the public nature of individual performance tended, says Quintilian, to give these speeches the feel of “mimic combats” similar to “the actual strife and pitched battles of the law courts” (2.10.8). At the very least, students subjected to the arduous, confrontational, semi-public performance of declamation would quickly become aware that rhetorical argument is addressed to a critical audience, that argument itself was always at least dyadic, and that, under these circumstances, “the all-important gift of the orator” was a “wise adaptability” to “the most varied emergencies” (2.23.1).
Fanciful as they often were, the suasoriae (the declamatory impersonations of historical figures) nonetheless function as instruction in the principles of ethos and audience. The Roman student had been prepared for role-playing by earlier exercises, but suasoria evoke much greater depth of detail and a more specific question to be addressed. So, when faced with the case of Alexander debating with his generals over whether to ignore the oracles and enter Babylon (Seneca 1974: suasoria 4), the student was not simply acting “ex persona” (in the character of) and delivering a dramatic monologue like Browning’s Andrea del Sarto; he was arguing in a specific historical context, with well-defined positions on either side, to an audience fully alert to the circumstantial data of the case.
Quintilian’s refers to these exercises as “absolutely necessary” to the expansion of the pupil’s understanding of human motive and response and notes that his own students assume as many different roles in their declamations as comic actors on stage (3.18.51). When we recall that students often declaim on both sides of a case and must regularly defend a position contrary to their initial inclinations, it is easy to see how this variety of impersonation serves to break down one’s natural egocentrism and open the mind to claims that might well have seemed alien. Impersonation, in other words, tends to liberalize one’s allegiances and breed tolerance. In brief, declamation is a dramatic experience in occupying the space of the other, of giving voice to a person who speaks in a different key, of “identifying” to the point of consubstantiation. To act the part of someone else is to bring the theoretical concept of “in utramque partem” to life.
And then there is the matter of audience. At its best, suasoria goes beyond the notion of recognizing what is unique in an audience as a technique to effect persuasion. Such an effort remains monological to the extent that it does not admit the potential for difference that the audience always represents. When combined with the lessons of impersonation, the invocation of and address to the audience as persons in their own right serves to multiply the voices one responds to in controversy. If impersonation invites the dialogical extension of the argument beyond the orator’s initial presumptions, the presence of an audience (which is seldom uniform) expands the conversation into “multiplex ratio disputandi” and invites a more comprehensive vision of the topic. In the process, the opposing parties in dialogue generate new possibilities for invention, as ideas shift, oscillate, and transform in the give-and-take of alternatives. Invention takes place, as Montaigne says, by “polishing our brains through contact with others” (1948: 112). As we turn from suasoria to controversia (the declamatory exercise devoted to forensic rather than deliberative cases), we turn also from the theatrical to the dialectical, for the controversiae represent a substantial increase in logical rigor. Seneca the Elder records this popular topic of school debate: “A young man captured by pirates writes his father for ransom. He is not ransomed. The daughter of the pirate chief urges him to swear that he will marry her if he escapes. He swears. Leaving her father, she follows the young man, who, upon his return to his home takes her to wife. A well-to-do orphan appears on the scene. The father orders his son to divorce the daughter of the pirate chief and marry the orphan. When the son refuses to obey, the father disowns him” (in Clark 1956: 231). Obviously, any defense of realism in the practice of such controversia could not be based on the events of the case itself. It was the verisimilitude of the argument rather than the case itself that Quintilian saw as essential to controversial reasoning.

Students would begin their analysis of the controversia by first identifying the stasis and the likely arguments in opposition (10.5.20). Quintilian notes that it is simply not adequate in forensic argument to take up only accusation or defense, because “sufficient acquaintance with the other side of the case” is a prerequisite for effective persuasion (10.5.21). In the case of the pirate’s daughter, the controversy was likely to turn on a question of law vs. equity: is this law universally binding, or is equity a higher virtue than the written statute? Strong cases could be made on either side, and careful reasoning would be required. In another case entitled “The Poor Man’s Bees,” there is a controversy between the rich owner of a flower garden and a poor neighbor whose bees invade that garden (Quintilian 1987: #13). The rich man spreads insecticide on his flowers, kills the bees, and the poor man brings suit. In his sample declamation, Quintilian fills out the poor man’s speech in considerable detail, especially his refutation, which provides a comprehensive recapitulation of each point in the rich man’s case before the poor man’s detailed rebuttals (see Clark 1956: 247-50). What is interesting here is that the dialogue between opposing parties is incorporated into a single speech. As a result, declamatory orators become practiced not only in thinking “in utramque partem” as preparation for their own claims but also in providing what Bakhtin would call a “double-voice” within the boundaries of one’s own utterance. When Quintilian treats “altercatio” or debate proper (6.4), he reiterates the point that careful consideration should always be paid to “the arguments of the opponent” (6.4.14). Even when students find themselves in agreement, he says, it is best for them to practice their skills in “altercatio” by taking different sides and testing their ideas through “mimic battle” (6.4.21). And because students are regularly arguing both sides, their classroom experience may well serve, says Quintilian, to reduce the eristic ill-will often directed “at those who hold opposite opinions” (3.8.69).
There are, admittedly, problems with declamation, especially as the genre came to dominate Silver Age Roman letters and gave way to theatrical excess. Professors of rhetoric began to invite the public more and more often to open recitations, first to impress the parents of their students and to attract additional clients, later to display their own brilliance before ever-expanding crowds. Quintilian is himself candid in noting that declamation became “so degenerated that the license and ignorance of declaimers may be numbered among the chief causes of the decline of eloquence in Rome” (2.10.3). Marrou complains that declamatory narratives became much too fantastic; but he points out that declamation can be defended as an isolated opportunity for the practice of public eloquence during a period of decline in political freedom (1956: 288). It is Quintilian’s defense, however, that remains the strongest: for it is always possible, he claims, “to make sound use of anything that it naturally sound” (2.10.3). His method for insuring the soundness of declamation was to insist that they remained “modeled on the forensic and deliberative oratory” for which they were intended as training (2.10.8). Seen from this perspective – as “ foil(s) wherein to practice for the duels of the forum” – the progymnasmata and declamation represent a rite of passage, a transition from theory and exercise to a mature recognition of the requirements for successful advocacy in an environment conditioned by difference, disagreement, and change (5.12.17).

4. A Contemporary Role for Controversial Pedagogy
I would like to think that the presence and import of controversial reasoning in the “Institutio” has been sufficiently established to substantiate my principal claim that argument “in utramque partem” resides at the heart of Quintilian’s pedagogy. I have also tried to indicate that Quintilian’s pedagogy takes on its full resonance only when it is reassociated with its philosophical base, which is Sophistic in origin and sceptical in nature, which is firmly anchored in contingency and the unavoidable multiplicity that conditions all “res humana,” which casts a wide net in its search for knowledge and accepts a vision of truth that compounds opposing views, and which finally is thoroughly practical in its drive towards application in the world at-large. Only when Quintilian’s classroom protocols are placed in relation to their philosophical context can we begin to realize the rich possibilities that flow from the confluence of rhetorical theory and the pedagogical tradition.
The question before us now, however, is more pragmatic: i.e. what specific practices might be adapted from Quintilian’s pedagogy that, “mutatis mutandi,” can contribute to our rhetorical paideia? Thomas Sloane has recently noted that despite the revival of rhetorical studies, our conception of “inventio” remains “impoverished” and that, in general, rhetorical pedagogy has not kept pace with critical theory (1997: 127-28).[iii] To my mind, the study of Quintilian and the legacy of controversia puts us in a position to rectify this imbalance and reassert the connection between the rhetorical tradition and the classroom. The scope of the present essay, however, allows for only modest and provisional suggestions.
I begin with what Perelman might call “starting points,” preconditions for argumentation extrapolated from the practice of controversial reasoning and intended for discussion by students, provocative ideas antithetical to the traditional assumptions of what Deborah Tannen calls “the Argument Culture” (1998). Starting point #1: Argument deals with probabilities but does not preclude our ability to defend one position as stronger than others. On the contrary, controversia assumes (somewhat optimistically) that when “multiplex ratio” are weighed effectively, the preponderance of probability will favor one side over others. #2: All opening positions are partial in the dual sense that they are biased in favor of their own presumptions and they do not represent all that may be said about the subject. #3: If we accept our partiality, we must also accept the possibility that exchange with others could prompt us to change our minds. #4: If we accept our partiality, we should be inclined to suspend judgment until all positions have be addressed. And #5: the ground rules for judgment in the context of scepticism and probability cannot depend upon standards of certainty but will grow out of the exchange between parties engaged in conflict, what Blair and Johnson call the “epistemic standards of the audience” (1987: 49). Such are the preconditions for controversial argument that students might consider.
But what of practical methods, concrete extensions of Quintilian’s own practice that could contribute to our teaching? I will mention two possibilities, both of which fall under the heading of invention. In the first place, Quintilian’s curriculum identifies invention with dialogue and the process of symbolic exchange. Perelman and Olbrechts-Tyteca, of course, maintain that argument always develops “in terms of the audience” (1969: 5). The progymnasmata embraces and pragmatizes this essentially dialogical view by asking students to first imitate, then refute, then both agree and disagree with the claims of a text. Once this procedure of alternating support and critique has been established in the preliminary exercises, dialogical exchange is dramatized, as students first imagine, then (in declamation) actually confront other parties in controversy. Two implications follow from the primacy of dialogue: first, contact with other students in response to controversy should begin early and be repeated often. In other words, students need to come out from behind the keyboard and take their place in front of and face-to-face with other students and perspectives (cf. 1.11.9). For teachers of composition, this means an increased oral component in argument training. The second implication of dialogue’s primacy is that we must work harder to stimulate the continuous give-and-take that constitutes real-world argument. Argument “in utramque partem” implies repeated reversals: first one side speaks, then the other, then the other again, and so on. Instead of single-exchanges or the statement/rebuttal procedures of forensic debate, argument pedagogy must seek to simulate the ongoing conversation of actual controversy (see Leff 1987: 3).

The second potential candidate for pedagogical adaptation falls under the heading of “imitatio.” This subject is so vast and so diffused throughout Quintilian’s curriculum that I can scarcely do more than add my voice to those of James Murphy and Dale Sullivan in calling for a reassessment of its once-esteemed pedagogical role (1990: 44-53; 1989, resp.; cf. 10.2.1-28). Suffice it to say that our neo-Romantic tendency to equate imitation with the surrender of identity runs counter to the classical tradition. “Mimesis,” says Aristotle, is a natural part of the learning process (“Poetics” 1447a-b), but the degree of adhesion to the original source varies considerably. There is no reason to assume that imitation, as it “supplements, improves, and illustrates its ostensible models” is not a creative act (Russell 1981: 108). Within the general category of pedagogical “imitatio,” I would identify two specific options for adaptation to our classrooms.  The first is impersonation or role-playing. To impersonate is enter into dialogue with another perspective, to integrate into one’s self what had been unfamiliar (cf. 6.2.26). Conversely, impersonation allows students to distance themselves from their own presumptions and explore unexamined partialities. Furthermore, role-playing is fun; it evokes the ludic impulse in the service of instruction. It can transcend the appeal to reason alone and motivate the student in special ways. My own efforts to encourage role-playing in class have done more than any other technique to loosen the grip of dogmatic assumptions and to prompt an appreciation for the many-sidedness of argument. The second possible adaptation comes with declamation and the promise of case-study as a vehicle for experiencing the full complexity of circumstantial argument. Case-study exercises have been popular for some time in professional writing and legal studies, but they run counter to the emphasis of most argumentation texts on propositional structure and the demands of logos over audience, ethos, and situation (Mendelson 1989). What declamatory exercises can provide is a dramatic evocation of the multiplicity, ambiguity, and contingency that characterize actual controversy. Michael Billig points out that the nuance of human affairs can never be reduced to method, so “finite laws [or rhetorical precepts] are likely to be embarrassed . . . by novel particulars” (1987: 62 and 68). As Quintilian recognized, the well-conceived declamatory exercise is the capstone of rhetorical training because it exposes the rhetor to the complexity of novel particulars and requires a full measure of “facilitas” and decorum in return.
Of course, any pedagogical theory or method only has value to the extent that it serves a larger purpose. For Quintilian, that purpose was the cultivation of oratorical excellence in the service of moral dignity and public virtue (12.1-2; see Lanham). I would myself offer a variant rationale for the pedagogy of “in utramque partem.” A controversial pedagogy seeks at all points to generate two or more positions in conflict and to stimulate a productive dialogue among these sides as the appropriate means for understanding and perhaps even resolving the problem at hand. Because of the contingent nature of the problems that rhetoric is designed to address, problems about which there are always multiple points of view, judgment cannot proceed along abstract, technical lines (cf. Kahn 1985: 30-36). According to Cicero, decorum is that facility (“facilitas”) that allows one to comprehend what is appropriate in complex issues and to work expediently towards a viable resolution (“Orator” 71; Leff 1990). Decorum, therefore, is a “two-fold wisdom” which accommodates not only eloquence in an effort to articulate the issues but also persuasion in order to have an effect on the world. As such, decorum is ultimately cognate with prudence, the knowledge of appropriate action in response to specific situations (“De Oratore” 3.55 & 3.212). Classroom exercise in argument “in utramque partem” was, for Quintilian, the principle means of preparing students not only to respond to arguments with decorum but also to play their part in the public sphere with prudence.
In Aristophanes’ “The Clouds,” students go to the “thinking school” to learn to bicker with their parents and import corruption into the body politic. Quintilian reverses the moral orientation of advanced education, of “thinking schools,” but he continues to place argument at the heart of the curriculum. Only through the prudent management of controversy can the student become what Quintilian terms a truly Roman “wise man;” i.e. one who reveals his virtue “in the actual practice and experience of life” (12.2.7). The methods of controversial reasoning, of “in utramque partem” at work throughout Quintilian’s pedagogy are the tools that allow for the realization of this goal. For contemporary teachers, they are also the means by which we can invite the wisest of Roman teachers back into the classroom. I encourage you to welcome him.

NOTES
i. For a discussion of controversial reasoning in the De Oratore, see Thomas O. Sloane (1997: 28-53). The present paper was essentially completed before I could read Prof. Sloane’s distinguished new book (On the Contrary), which deals with many of the same ideas as this paper. I would, however, acknowledge, the influence on my own thinking of Prof. Sloane’s work and especially his earlier book (1985).
ii. All references to the Institutio Oratoria are to the Bulter edition and will include passage references in parenthesis. Unless otherwise indicated, all numerical references are to Quintilian.
iii. For two modern adaptations of the progymnasmata, see Comprone (1985) and Hagaman (1986).

REFERENCES
Aristotle. (1984). The Complete Works. 2 vols. Ed. Jonathan Barnes. Princeton: Bollingen Series.
Billig, M. (1987). Arguing and Thinking: A Rhetorical Approach to Social Psychology. Cambridge: Cambridge University Press.
Bonner, S.F. (1977). Education in Ancient Rome: From the Elder Cato to the Younger Pliny. London: Metheun and Company.
Bonner, S.F. (1969). Roman Declamation in the Late Republic and Early Empire. Liverpool: Liverpool University Press.
Carter, M. (1988). Stasis and Kairos: Principles of Social Construction in Classical Rhetoric. Rhetoric Review 7.1, 97-112.
Cicero. (1933). Academica. Trans. H. Rackham. Loeb Classical Library. Cambridge, MA: Harvard University Press.
Cicero. (1913). De Officiis. Trans. Walter Miller. Loeb Classical Library. Cambridge, MA: Harvard University Press.
Cicero (1929). De Oratore. Trans. H. Rackham. Loeb Classical Library. Cambridge, MA: Harvard University Press.
Cicero (1939). Orator. Trans. H.M. Hubbell. Loeb Classical Library. Cambridge, MA: Harvard University Press.
Clark, D.L. (1957). Rhetoric in Greco-Roman Education. Morningside Heights, NY: Columbia University Press.
Colson, F.H. (1924). Introduction. In: M. Fabii Quintilianii: Institutionis Oratoriae, Liber I. (xxxix), Cambridge: Cambridge University Press.
Comprone, J. (1985). Expressive Writing: Exercises in a New Progymnasmata. Journal of Technical Writing 4.1, 31-53.
Greer, W.J. (1925). Quintilian and the Declamation. The Classical Weekly 19.4, 27-31.
Grube, G.M.A. (1965). The Greek and Roman Critics. Toronto: University of Toronto Press.
Hagaman, L. (1986). The Modern Use of the Progymnasmata in Teaching Rhetorical Invention. Rhetoric Review 5.1, 22-29.
Hock, R. and E.N. O’Neill. (1986). The Chreia in Ancient Rhetoric. Athens, Georgia: Scholars Press.
Kahn, Victoria. (1985). Rhetoric, Prudence, and Skepticism in the Renaissance. Ithaca: Cornell University Press.
Kennedy, G. (1969). Quintilian. New York: Twayne Publishers.
Lanham, R. (1993). The Electronic Word: Democracy, Technology, and the Arts. Chicago: Chicago University Press.
Leff, M. (1990). Decorum and Rhetorical Interpretation: The Latin Humanistic Tradition and Contemporary Critical Theory. Vichiana 3, 109-26.
Leff, M. (1987). The Habitation of Rhetoric. In: Argument and Critical Practices: Proceedings of the 5th SCA/AFA Conference on Argumentation (pp. 1-8), Annandale, VA: Speech Communication Association.
Long, A.A. (1974). Hellenistic Philosophy. London: Duckworth.
Marrou, H.I. (1956). The History of Education in Antiquity. Trans. George Lamb. New York: Sheed and Ward.
Mendelson, M. (1998). The Antilogical Alternative. In: The Proceedings of the Tenth SCA/AFA Conference in Argumentation. Speech Communication Association (forthcoming).
Mendelson, M. (1997). Everything Must Be Argued: Rhetorical Theory and Pedagogical Practice in Cicero’s De Oratore. The Journal of Education 179.1, 15-47.
Mendelson, M. (1989). The Rhetorical Case: Its Roman Precedent and the Contemporary Debate. The Journal of Technical Writing and Communication 19.3, 203-26.
Montaigne. (1948). The Complete Essays. Trans. D. Frame. Stanford: Stanford University Press.
Murphy, J.J. (1987). Introduction. In: Quintilian on the Art of Speaking and Writing (pp. ix-xlxiii), Carbondale: Southern Illinois University Press.
Murphy, J.J. (1990). Roman Writing Instruction as Described by Quintilian. In: A Short History of Writing Instruction (pp. 19-76), Davis: Hermagoras Press.
Perelman, Ch. and L. Olbrechts-Tyteca. (1969). The New Rhetoric: A Treatise in Argumentation. Trans. J. Wilkinson and Purcell Weaver. Notre Dame: University of Notre Dame Press.
Philostratus. (1965). Lives of the Sophists. Trans. W.C. Wright. Loeb Classical Library. Cambridge, MA: Harvard University Press.
Protagoras. (1963). In: Edith Hamilton and Huntington Cairns (Eds), The Complete Dialogues of Plato (pp. 308-51), New York: Bollingen.
Quintilian. (1920). Institutio Oratoria. Trans. H.E. Bulter. 4 vols. Loeb Classical Library. Cambridge, MA: Harvard University Press.
Quintilian. (1987). Major Declamations. Trans. L.W. Sussman. Frankfurt: Verlag/Peter Lang.
Russell, D.A. (1981). Criticism in Antiquity. London: Duckworth.
Russell, D.A. (1983). Greek Declamation. Cambridge: Cambridge University Press.
Seneca the Elder. (1974). Controversiae and Suasoriae. Trans. M. Winterbottom. Loeb Classical Library. Cambridge MA: Harvard University Press.
Sextus Empiricus. (1994). Outline of Scepticism. Trans. Julia Annas and Jonathan Barnes. Cambridge: Cambridge University Press.
Sloane, T.O. (1985). Donne, Milton, and the End of Humanist Rhetoric. Berkeley: University of California Press.
Sloane, T.O. (1997). On the Contrary: The Protocol of Traditional Rhetoric. Washington D.C.: Catholic University of America Press.
Sprague, R.K. (1972). The Older Sophists: A Translation of Diels-Kranz. Columbia: University of South Carolina Press.
Starr, C.G. (1965). Civilization and the Caesars: The Intellectual Revolution of the Roman Empire. New York: Norton and Company.
Sullivan, D. (1989). Attitudes toward Imitation: Classical Culture and the Modern Temper. Rhetoric Review 8.1, 5-21.
Tannen, Deborah. (1998). The Argument Culture: Moving from Debate to Dialogue. New York: Random House.

1. The Semiotic Ornatus Perspective on Visual Rhetoric
In his article “The rhetoric of the image” Roland Barthes assumes that if classical rhetoric were to be rethought in structural terms it would “perhaps be possible to establish a general rhetoric of the signifiers of connotation, valid for articulated sound, image, gesture” (1977: 50):
“This rhetoric could only be established on the basis of a quite considerable inventory, but it is possible now to foresee that one will find in it some of the figures formerly identified by the Ancients and the Classics; the tomato, for example, signifies Italianicity by the metonymy and in an other advertisement the sequence of three scenes (Coffee in beans, coffee in powder, coffee sipped in the cup) releases a certain logical relationship in the same way as an asyndeton” (: 49f).
This ‘figurative’ approach to visual rhetoric is pursued more fully in the text “Rhétorique et image publicitaire”. Here Jacques Durand defines rhetoric as the art of fake speaking (“l’art de la parole feinte”) (1970: 70), and describes its task as transforming or converting the proper expression (“le language propre”) into a figurative or rhetorical expression (“language figuré”). What is said by using a rhetorical figure or trope could also have been said in a different, or normal, manner. Durand sought to “find a visual transposition of the rhetorical figures in the advertising image” (1987: 295) by examining more than one thousand magazine advertisements. This was done by considering “a rhetorical figure as a transformation from a ‘simple proposition, to a ‘figurative proposition’” (: 295). In these cases Barthes and Durand are exponents for what I will call a semiotic ornatus perspective on visual communication and argumentation, i.e. a search for meaning through a search for metaphors, metonymies, repetitions, inversions, and the like in visual communication.
My point here is not to dismiss or reject the great importance and semiotic value of a text such as “The Rhetoric of the Image”. Indeed, in this paper I use the concepts of anchorage and relay taken from Barthes’ influential article. However, as the major point of departure for both theoretical and analytical texts dealing with visual rhetoric, such a semiotic perspective is problematic in several ways. In this working paper I will briefly touch upon four arguments where this is the case. I will then try to sketch an alternative approach to visual rhetoric by taking the point of departure in the rhetorical art of inventio, rather than in the art of elocutio.

2. Four Arguments for the Lack of Usefulness of the Semiotic Ornatus Perspective
Argument 1: The ‘transformation theory’ is problematic.
The ornatus perspective on visual rhetoric is based on what we could call the ‘transformation theory’, i.e. the presumption that expressions (either verbal or visual) are transformations from a ‘natural’ or ‘normal’ way of expressing the same thing. A point can be expressed in ordo naturalis, the natural or ordinary way. However, if we want to add more emotional power and better adherence, the same point can also be expressed in ordo artificialis, the artful or artificial way. So, we have a distinction between the proper way of saying something (langage propre), and the rhetorical or figurative way of saying something (langage figuré).
The theoretical problem with this theory of transformation from the natural to the figurative expression – which is a traditional rhetorical view – is, of course, that it is difficult, if at all possible, to distinguish between the two ways of expression, and to define what the so-called natural expression is. It is easy to presuppose a ‘natural order’, but rather difficult to say what this natural order of a figurative expression might be. The transparent or ‘sober’ expression is itself a rhetorical choice and strategy. What then, is this kind of expression a transformation from?
This presumption of a ‘natural’ or ‘normal’ expression is equally problematic when dealing with visual representations. A distinctive feature of an iconic representation is that it has a ‘natural presence’ in its own right. In other words, it is what it shows. When dealing with images one can choose between countless expressions created by techniques of editing, framing, duration, mise-en-scène, and so on. Often, it is rather difficult to judge one expression as more ‘natural’ than another. Of course, we tend to notice when the regular conventions of a particular genre of images are changed: If the commentator in a news programme is seen in extreme close-up or from a bird’s-eye perspective, or if the characters in a movie suddenly face the camera and start talking directly to the audience. In rhetoric, however, the main purpose of figurative language is to stir the emotions unnoticed, without drawing attention to the language style itself. In fact, a general rule of rhetoric is that the language and the language form must be transparent – as an unnoticed window through which we see the message.

Argument 2: Ornatus is a very limited part of rhetoric, and the semiotic ornatus approach therefore contains a limited understanding of rhetorical persuasion.
Ornatus is but one of four elements of elocutio, in addition to perspecuitas, puritas and aptum. Furthermore, elocutio is but one of the five stages of composition. To make tropes and figures the starting point of a discussion of visual rhetoric is therefore a violent limitation of the art of rhetoric, because it only entails a fourth of a fifth of the art. Consequently, we no longer talk about rhetoric but rather of stylistics.
Tropes and figures are primarily means of expressing arguments – found in the stage of inventio – as evidently as possible. They are means for catching audience attention, making the audience remember the arguments in the speech, and, most importantly, stirring the emotions of the audience. Of course tropes and figures can have a persuasive effect, and they can show or illustrate important arguments or lines of reasoning. But they do not constitute the argument or the reasoning itself.
From an argumentative point of view, tropes and figures constitute the micro perspective whose main task is limited to creating rhetorical pathos. In this sense, ornatus performs a rhetorical and a persuasive appeal. But the emotional appeals of ethos and pathos do not give a comprehensive and understanding view of rhetoric unless they are connected to the most important rhetorical appeal, – logos. A unity of ethos, logos and pathos is thus a prerequisite in the search for a theory of visual rhetoric.

Argument 3: Ornatus is embedded in verbal language.
Because of the strong connection between ornatus and the verbal language – where the first in a sense is embedded in the second – the ornatus perspective gives us a very unhelpful and unmanageable starting point for critical and theoretical treatment of visual rhetoric.
Whereas the general and universally valid thoughts of argumentation and topoi in inventio are more or less free from the constraints of verbal expression, the tropes and figures of ornatus often are their verbal form or shape. The meaning of tropes and figures such as prosopopoeia (confirmatio), anaphora, and alliteration are embedded in the expressions themselves. Expressions and meanings such as these are either impossible to find in visual representations or can only be located with an unreasonable constraining of both the figurative expression and the visual representation.

Argument 4: The semiotic ornatus approach can say nothing about hierarchies of values, or of the importance of the rhetorical situation.
Because the semiotic ornatus approach neither deals with hierarchies of values nor with the rhetorical situation, it provides only a limited contribution to knowledge about the structures, elements and effects of visual argumentation. The fundamental structuralist view of pictures and visual argumentation in this approach also tends to concentrate primarily on relations inside the picture frames, and therefore tends to overlook the rhetorically very important aspects of the rhetorical situation: For instance the classic concepts of the right moment of speaking, kairos, and of proper adaptation of the speech to the occasion, aptum (decorum). These are necessary and important rhetorical considerations concerning the relations between the five constants in the rhetorical situation. Cicero puts it this way: “no single kind of oratory suits every cause or audience or speaker or occasion” (De Oratore III.liv.210).

Along with the importance of the rhetorical situation itself, also the concepts of topoi and hierarchies of values are important for understanding argumentation. In The New Rhetoric, Chaim Perelman & Olbrecths-Tyteca says that “all argumentation aims at the adherence of minds” (1971: 14). Adherence of minds requires that the rhetor finds a common ground of values or attitudes both for himself and the audience. A common ground – or warrant – is required in order to persuade. Basing the argumentation on the common ground that “democracy is good”, a politician opposed to membership of the EU can try and persuade an audience that the EU is an undemocratic institution. If members of the audience accept that the EU is undemocratic, they will be influenced (or even persuaded) into casting a “no” vote to membership of the EU, on the basis of their adherence to the warrant that democracy is good. We cannot make considerations like these through the semiotic ornatus approach. This is because it is not a theory about argumentation, merely one about semiotic signification. Of course, semiotic theories are significant. But it is important to remember that analysis of semiotic signification does not automatically include analysis of argumentation. The attempt to understand persuasive signs and discourses through tropes and figures, or through concepts such as denotation, connotation, paradigm, and syntagm, does not entail thoughts or concepts that in a reasonable way can account for situational constraints or for the elements, structures, and hierarchies of argumentative topoi and values. Neither can the semiotic ornatus perspective in a practical analytical way distinguish between a statement and an argument, or distinguish between a good and a bad argument.

3. A Rhetorical Conception of Argumentation – Inventio as the Point of Departure
As already indicated, the project of Roland Barthes – and of his followers – is more semiotic than it is rhetorical. “The Rhetoric of the Image” is more about semiotic signification than it is about rhetorical argumentation. It is furthermore doubtful that we can find one general or universal rhetorical form independent of medium or substance, and if possible, it is certainly doubtful both that such a form represents a truly persuasive rhetorical operation, and that such an operation has its ontological foundation in ornatus. We are more justified in claiming that such universal ways of argumentation and appeals are to be found in the rhetorical art of inventio, which is not in the same way tied up in and embedded in verbal language. I believe that two assumptions are important with inventio as the point of departure for a theory or an analytical view of visual argumentation:
(A) Rhetorical argumentation is an attempt to gain adherence to a claim or an attitude among an audience. This is done by strengthening and changing relevant hierarchies of lines of reasoning, values or viewpoints (common topics), by appealing through the three rhetorical proofs: ethos, logos and pathos.
(B). Practical rhetoric can be characterised as situational intentionality. Rhetoric rests on the orator who tries to promote his intention and gain adherence to his points in a particular situation through the use of language.

Let’s take a closer look at these two points:
(A) The Understanding of Argumentation as Creating or Changing Persuasive Hierarchies
According to Aristotle (A.I.3; 1354a), we can distinguish between proofs that belong to the art of rhetoric, ‘intrinsic proofs’ (entechnoi) and proofs or things that do not, ‘external proofs’ (atechnoi).[i] The “intrinsic proofs” are proofs that are furnished through the speech and which may reside in the character of the speaker (ethos), in a certain disposition in the audience (pathos) or in the speech itself (logos). Only these proofs – or ways of appeal – Aristotle says, are intrinsic to the art of rhetoric. He considers the rational logos appeal as the most constitutive point of departure for rhetorical argumentation, while the emotional appeals of pathos and ethos are necessary supports for logos. They are supports or pillars that indicate the degree of credibility, importance and value in the argument.
Aristotle then ascribes two modes of argument to rhetoric: the enthymeme, which is a rhetorical syllogism, and the example which is considered a rhetorical induction. The enthymeme is viewed as the most important kind of deductive demonstration and proof. This significant rhetorical way of providing proof is characteristic in its dealing with topical reasoning and thought patterns which arrange information and unite it in a coherent and persuasive form of argumentation. By topical reasoning I mean topics in Aristotle’s sense of the word: structural argumentative forms without content in their own right (B.XVII; 1391b). These are structures of rational argumentation that are manifest as common topics, or common structural forms of argumentation.
Aristotle points to “the possible and the impossible” as an example of a common topic. For instance: “[I]f one like thing is possible, so is the other” (B.XVIII.5; 1392a). This latent persuasive structure can be found in practical everyday argument such as: “When countries similar to ours can do without the EU, Norway too can do without the EU”.
In other words, our use of specific arguments is based on a variety of common topics in which the arguments and their premisses are embedded. The rhetorical appeal of a specific argument is placed on this foundation of common topics, and is furthermore based on common social, cultural and universal human values and premisses.
In their treatment of such common topics – or loci according to their terminology – Chaim Perelman and L. Olbrehts-Tyteca in The New Rhetoric talk about the quantity locus and the quality locus (1971: 85-92). The first term implies that something is better than something else for quantitative reasons, such as the superiority of that which is accepted by the majority. Thus, the quantity locus is the foundation of the democracy warrant mentioned above.
Opposed to this, there is the quality locus which emphasizes superiority of the unique, and it therefore implies that one bright person may be more right than several who are not so bright. Common topics such as these can be found both in verbal and in visual argumentation. For instance, in advertising it is possible to argue both by means of images and in words that a product is a good one because may people use it.
If we accept this line of reasoning, that some topical arguments can be manifested both in verbal and in visual communication, we can also assume that although visual and verbal argumentation are different forms or substances of communication, they do at least share some kind of common argumentative ontology. If this is the case, we may use the art of rhetoric to say something about visual argumentation. Contrary to what is the case with the semiotic ornatus approach, this kind of general perspective may run into fewer problems in the inter-semiotic translation of rhetorical appeal from one substance or medium to another.

In Aristotle’s Rhetoric, logos is described as the primary and only independent rhetorical proof (A.I.3, A.II). The proofs of ethos and pathos are always secondary, and they are always dependent on logos. The rhetorical enthymeme is, as he describes it, “the flesh and blood of proof” (A.I.3, 1354a; p. 66). By looking at Aristotle’s rhetorical enthymeme we can locate its persuasiveness in two assumptions:
1. The existence of common and interconnected topoi in the form of human values, attitudes and convictions, that tie social and cultural groups together and create the foundation upon which the persuasive appeal can be built.
2. The assumption that a person will accept the conclusion in the rhetorical enthymeme, if he or she accepts the premisses in the same enthymeme.

This Aristotelian conception of enthymemic argumentation presupposes that a strong stirring of emotions will follow from the acceptance of an attitude or an assessment. As pointed out by for instance Edwin Black (1978: Chapter IV & V), the emotional effect is, in a way, a consequence of the attitude or assessment that the argumentation creates.
If the rhetorical proofs and the use of topoi/loci are to function in a persuasively controlling way, they need to function in a structured hierarchy of values. Hierarchies such as these arrange our conception of the world, and hence our attitudes and actions. Broadly speaking, we induce change in actions and attitudes by introducing different structures or compositions of these hierarchies, or by exchanging the values or common topics upon which they are  based. To label the EU as an undemocratic institution is to categorize EU into a persuasive hierarchy of values based on the locus of quantity, or more specifically on the grounding value or warrant: “Democracy is good”.
An understanding of verbal as well as visual rhetoric requires an understanding of rhetorical operations such as the cognitive structuring of topoi, values and attitudes. We cannot find any good explanations or accounts of conditions and circumstances such as these by using the semiotic ornatus approach. Instead, we may use for instance Stephen Toulmin’s model of argumentation (1958, Toulmin et al. 1978), which contains the possibility of placing argumentative elements in a structured hierarchy.
Toulmin’s model takes a pragmatic and analytical approach to argumentation by focussing on the process of argumentation and on the structuring of elements. Hence, we may learn something about the function of the various elements in a persuasive discourse by using the model.
It is of course not possible to unfold neither the argumentation theory of Toulmin nor its implications here. But I believe that a model of argumentation such as the one from Toulmin can give us not only the possibility of seeing the structures both of a single argument (the micro level) and of a more elaborate string of reasoning (the macro level). It can also provide us with a view of the hierarchical layout of arguments. By determining which elements function as claim, datum and warrant, it can illustrate the connection between the elements, and indicate which elements that are based on one another.
Let us now go to the second assumption for inventio as a starting point for a theory or analytical view of visual argumentation. My argument so far presupposes that rhetorical discourse is always driven by intention in a particular situation, and that is has the persuasiveness as its most important constitutive feature. I have chosen to term this conception of rhetoric as situational intentionality.

(B) Rhetoric as Situational Intentionality – The Persuasive Continuum
With very few exceptions, rhetorical theorists generally agree that rhetoric has to do with persuasive discourse. Rhetoric is not constitutively about style, form or genre, but rather about intentionality. Placing intentionality at the core of rhetoric gives us an useful limitation and distinction. Consequently, a discourse is not rhetorical if it is not consciously intentional. I do not behave rhetorically when screaming “ouch!, that hurts!” when I accidentally hit myself with a hammer and thereby unintendedly “persuade” my wife to come to my rescue.
Even if we limit rhetoric to intentionality, we are still left with a tremendously broad topic which is hard to get into proper theoretical perspective. One may say that I behave intentionally when asking for the salt, or when I slam the door during a quarrel. But is it rhetoric?
As I indicated above, it may be hard to distinguish between what is rhetoric and what is not. With the limited propositional syntax of images (Messaris 1997:x), this distinction turns out to be even more problematic in visual argumentation. Maybe such a distinction is not very practical. Maybe we should rather distinguish between different forms or degrees of rhetoric or intentionality, depending on how “much” rhetoric is needed to get the adherence of minds in the audience.
In this manner, we can distinguish between different forms of rhetoric according to the relationship between the orator and the audience, and according to the degree of their disagreement, divergence or opposition. In a rhetorical perspective it is the positions in the communicative situations that are interesting, as different positions lead to, or at least demand, different forms of rhetoric.
When a teacher explains how the EU is functioning, the teacher is using rhetoric in a broad persuasio sense. Here, the teacher’s intention is to create an understanding of the EU, and in so doing, language is mainly used referentially. If a student objects to the truthfulness and relevance of the account, the teacher’s subsequent attempt at persuading or convincing the student of the accurateness and the relevance of the argument would maybe still be dominated by referential language. What is important here, however, is that it is also likely that the teacher’s discourse would now contain a higher degree of persuasiveness because of the student’s opposition. The teacher would arrange or manage his discourse according  to the objections of the student, and he would try to put forward the best reasons and arguments for his own view. He would thus exercise rhetoric in a restricted persuasio sense.
We can thus place the different rhetorical appeals and addresses on a continuum between a slightly opposed audience and a strongly opposed audience. This is what I will term the persuasive continuum. It is common and classical rhetorical knowledge that an orator cannot successfully speak in the same way to audiences that are either negative or positive to the message. We can find it in the already mentioned remark of Cicero that an orator should not always speak in the same way to everybody, against everybody, for everybody or with everybody, and we can also find it in Socrates’ remark that it is not difficult to praise Athenians in Athens.

4. Can This Understanding of Argumentation Contribute to an Illumination of Visual Rhetoric?
Towards what kind of analytical approach to visual rhetoric do  these considerations about rhetorical argumentation point? Of course, this is neither the time nor the place to unfold a full theory of visual rhetoric. Still, it is clear, I think, that at least three elements must be more central to such a work:
1. The rhetorical proofs (ethos, logos and pathos)
2. The argumentative hierarchies of values and topoi
3. The situational intentionality of rhetoric

A few remarks are needed to point the direction of such a rhetorical inventio approach to argumentation in visual argumentation. First of all, the difference to the semiotic ornatus approach lies in the possibility and choice of questions one is directed to, and may ask, in connection with a treatise of visual argumentation.
While the semiotic ornatus approach will lead the examiner of visual rhetoric to ask questions of how to find visual elements which somehow fit the rhetorical figures of ornatus, the approach lacks the possibility of asking questions about the kinds of proof, the argumentative hierarchies, and the situational intentionality. These kinds of questions, I believe, may not only be asked, but will also be satisfyingly answered through the approach such as the one I indicate here.
Before continuing with the remarks about which questions and possible answers the inventio approach might direct us towards, it is necessary to provide a more precise indication of what I mean with the term visual rhetoric, and what the particular visual contribution in a piece of visual rhetoric might be. This we will do with a short – and by no means complete – listing of different kinds of visual techniques and manifestations that can perform visual rhetoric. This overview covers visual rhetoric in moving images, although it also includes the rhetoric of non-moving images. We can distinguish at least three basic kinds of visual rhetoric, or main areas where the visual plays an important role in the argumentation.

1. The Rhetoric of Mise-en-Scène
The term rhetoric of mise-en-scène includes the visual aspects within a single shot (or picture or photograph) that are used to support or co-create the rhetorical intention of the message. This may for instance be setting, colours, shapes, symbols, and cameramovement, -angle, -perspective, and -distance.
The rhetorical function of such visual techniques, or visual rhetorics, is to induce general moods and feelings in the viewer, and to create associations. Primarily, they are emotional appeals (ethos and pathos) and particularly dependent on anchoring in order to create a complete rhetorical argument, including the appeal of logos. The concept of actio, as it is treated by traditional rhetoric, can be seen as a special and significant part of the mise-en-scène.

2. The Rhetoric of Editing
The rhetoric of editing includes the creation of meaning and argumentation through the connecting of different images; The use of fades, dissolves, cuts, following or breaking the rules of continuity to support the rhetorical message; The use of editing pace, for instance rapid editing as a way of signifying energy and youth, and thereby performing a certain ethos appeal.

3. The Rhetoric of Dispositio
The rhetoric of dispositio concerns the global form of and organising of either a single image or a longer construction of moving images. In a treatise of images in advertising, Scott (1994: 266) talks about “the arrangement of visual argument”, and how the order of argumentation may be guided by the layout of an advertisement. The film theorists Bordwell & Thompson discuss the rhetorical form (1990: 99ff.) of a film, and illustrate with a film that begins with “an introduction of the situation, goes on to a discussion of the relevant facts, then presents proofs that a given solution fits those facts, and ends with an epilogue that summarizes what has come before”. This thus follows the traditional rhetorical dispositio. However, we should not necessarily think of the traditional rhetorical dispositio when we are talking about the rhetoric of dispositio. By rhetorical dispositio, we here mean a global arrangement of the visual elements which convincingly supports – or even creates – the intentional message.
We have to remember, however, that these kinds of visual rhetorics are not rhetorical in their own right. Yellow colour, fast editing, round or square shapes or lines, the global form or dispositio of a film, are all elements that acquire their rhetorical significance from the rhetorical discourse which they are a part of.
The viewers’ determination of the rhetorical significance of or meaning of a particular discourse does partly take place through what we may term the rhetoric of anchoring and relaying. The rhetorical meaning is in part created horizontally or diachronically, when we as readers of a text or viewers of a television programme are continuously evaluating and perceiving the elements and events in a discourse. We do this while keeping in mind our expectations for the future of the discourse and our experience with the discourse so far (Holub 1984: 90). Within reception theory (see for instance Iser 1978) this particular creation of meaning is described by the terms ‘wandering viewpoint’, ‘protension’, and ‘retention’.
But the rhetoric of anchoring and relaying is also partly a vertical or synchronous creation of rhetorical meaning. The reader or viewer create meaning of the rhetorical discourse through a continuous hermeneutic movement between the visual expression and for instance a written text, spoken words, sounds or music.
Not even the rhetorical discourse itself is rhetorical entirely in its own right. Rather, the discourse gains its rhetorical significance from a rhetorical situation (As pointed out by for instance Bitzer 1968). The viewer thus performs several intermingling rhetorical hermeneutic movements when trying to recreate a mediated argument: A horizontal and a vertical hermeneutic movement between the different elements in the rhetorical discourse, a movement between the rhetorical discourse and the rhetorical situation, and a movement between the elements in the discourse and the rhetorical situation.
Keeping in mind that the rhetorical situation is created by, or even has its ontological foundation in, an instance of situational intentionality, we can now more clearly see the importance of the concept of situational intentionality. We may also understand why it is problematic that the semiotic ornatus approach, with its inherent structuralist view, overlook the significance of situational considerations.

Some Questions and Considerations Concerning the Rhetorical Proofs:
When using the rhetorical appeals in criticism and analysis of visual discourse, we must first consider whether visual argumentation is actually able to persuade in a traditionally rhetorical sense. In Aristotle’s view, the emotional proofs of such pure verbal texts are thought to function as supporting pillars for logos, which is the primary proof and the most constitutive point of departure in rhetorical argumentation. Does visual argumentation function in the same way? Can visual expressions rather be expected to evoke emotional dispositions that in turn create an attitude that fits the emotional disposition? Does visual argumentation operate in a different order, where the emotional effect does not emanate the acceptance of an attitude, but rather produces it?

Is it typical for visual argumentation to evoke and stir emotions, and then (for instance through verbal support) to legitimate these emotions with fitting attitudes? We may ask whether the basic persuasive elements and structures are common to both visual and verbal argumentation, but that their place or order in the persuasive motion are different in the two instances. A discussion of questions such as these constitutes one of the many small steps towards a more comprehensive understanding of visual rhetoric.
A reasonable point of departure might be an investigation of the use of more particular analytical considerations about the rhetorical appeals in visual rhetoric. Possible questions might be: Which appeals are mainly made by the visual part and which are made in the verbal part of the expression? Which are present and which are absent?

Some Questions and Considerations Concerning the Argumentative Hierarchies of Values and Topoi:
The above reflections about argumentative hierarchies of values and topoi indicate another group of appropriate considerations and questions both in the theoretical uncovering of structures and elements in visual rhetoric, and in the practical critical analysis. These are considerations and questions such as: What is the topical foundation for the argumentation? Which topoi and values constitute the persuasive hierarchies, and how is the argumentation and its elements structured in these hierarchies? Which place and function does visual communication occupy in this structure of argumentation?
These circumstances can favourably be uncovered through argumentation analysis by using Toulmin’s model of argumentation. This is so first of all because this type of analysis can illuminate both the hierarchies and structures of the argumentation, and the foundational values and topoi in the appeal. Secondly, this type of analysis may place a single argument into a larger structured hierarchy of arguments, topoi and values.
In the illumination of the function and value of images and visual representation in rhetorical utterances, the advantage of the Toulmin model is that it can more clearly show the function of the visual expression in the arguments of a persuasive discourse. Does it function as claim, data or warrant? What is the relation between the visual expression and the degree of explicitness in the argumentation? What is the connection between the visual expression and the kinds of claims, data and warrants in the argumentation?

Some Questions and Considerations Concerning the Concept of Situational Intentionality:
We should consider and clarify the communicative situation both in the attempt to say something about how a rhetorical discourse works and how well we can expect it to work. As previously mentioned, there are two significant elements: the rhetor’s intention with the message and the discourse, and the audience’s opposition.

Generally speaking this perspective implies that the stronger the opposition, the greater the necessity of using verbal anchoring in the structuring of the desired hierarchy of topoi and values. The opposite also applies: the slighter, or weaker, the opposition, the less important the verbal anchoring will be. For instance:
The weaker the opposition in the audience
– the better is the possibility of succeeding rhetorically by visually confirming and supporting the present hierarchy of values and topoi in the audience,
– the greater is the possibility of succeeding rhetorically with hidden, indirect and vague argumentation through visual expressions.
– the more indirect and ambiguously advocating can the rhetor be,
– which is best done visually. And the lesser is the importance of giving clear and explicit guidance about what the audience is to do, or how or why, – which is very difficult to do visually.
– the more dominating can the aesthetic and emotional appeal through ethos and pathos be, – which is best done visually.
– the greater is the possibility succeeding rhetorically by mere creation of associative effects, – which is best done visually.
– the greater the value of what in advertising is known as product knowledge and product memory, – which is easily performed visually.
And the less the necessity of attitude – and action-changing rhetoric, which is difficult to perform visually. That is, the more functional will what we could call affirmative rhetoric be.

The stronger the opposition in the audience
– the greater is the demand for rhetor to create changes in the topical hierarchy of values in the audience, – which is rather difficult to do visually.
– the greater is the demand for explicit, direct and specific argumentation, – which is best performed verbally.
– the greater is the demand for discursive or analytical argumentation. That is a more “rational” line of reasoning, where the appeal of logos is central. This does, of course, not mean that emotional appeals are out of the question.
– the less is the value of product knowledge rhetoric and product memory rhetoric, and the greater the demand for rhetoric designed to change attitudes and action. In other words, the less effective affirmative rhetoric is.

5. A Few Concluding Remarks
This has been a very short and tentative account of some problems in the use of the semiotic ornatus approach to visual rhetoric, and a very limited indication of an alternative possibility. Even though this is truly work in progress, hopefully these considerations have made it somewhat clearer that a turn from elocutio to inventio is required in the quest for a more comprehensive theory of visual rhetoric.
Compared with the semiotic ornatus perspective, such a turn improves the possibility of understanding visual rhetoric on its own terms without a distorting reliance on the formal structures of the verbal language. It can also better take the more general considerations about the rhetorical proofs, the argumentative hierarchies, and the situational intentionality into account.
Furthermore, an approach of this kind can more fully and precisely make explicit and explain the invisible and implicit macro level, supporting – and to a certain degree creating – an instance of visual argumentation. It is an approach that has the potential of uncovering the connections between such a macro level and the micro level of a particular piece of argumentation.
Of course, this rhetorical inventio approach is also problematic in several ways. For instance, in its present form there is a tendency to rely on a purely rational, Aristotelian understanding of rhetoric and argumentation, with the risk of neglecting some of the more irrational elements in visual argumentation. However, even though both the rhetorical art of inventio and the Toulmin model of argumentation are in many ways attached to rational – and in some degree verbal – argumentation, it still seems to entail the most comprehensive and illustrating approach. Although the semiotic ornatus approach leaves no room for the inventio approach, the latter can actually embody the first.
Here we have only briefly looked at a small part of what a rhetorical inventio dominated theory of visual argumentation would consist of and implicate. Naturally, adjustments will be necessary in the further search for a truly visual, comprehensive and illustrating theory of visual rhetoric.

NOTES
i. We here use Lawson-Tancred’s translation of entechnoi and atechnoi, what Perelman & Olbrechts-Tyteca (1971: 9) term ‘technical’ and ‘extra-technical’ proofs, what L. F. Bitzer (1968: 8) terms ‘artistic’ and ‘in-artistic’ proofs, and what the Loeb translation terms ‘artificial’ and ‘inartificial’ proofs.

REFERENCES
Aristotle (1991). The Art of Rhetoric. Translated with an introduction and notes by Hugh Lawson- Tancred. Penguin Classics Middlesex.
Barthes, Roland (1977). Image, Music, Text. England: Fontana Press.
Black, Edwin (1978 [1965]). Rhetorical Criticism. A Study in Method. USA: The University of Wisconsin Press.
Bitzer, Lloyd F. (1968). “The Rhetorical Situation”. Philosophy & Rhetoric 1:1, pp. 1-14.
Bordwell, David & Kristin Thompson (1990 [1979]). Film Art. An Introduction. Third Edition. New York: McGraw-Hill Publishing Company.
Cicero. De Oratore. Translated by E. W. Sutton and H. Rackham. Loeb Classical Library. Cambridge and London: Harvard University Press.
Durand, Jacques (1970). Rhétorique et image publicitaire. Communications 15/1970, pp 70-95. Paris: Centre d’Etudes des Communications de masse. Ecole Pratique des Hautes Etudes.
Durand, Jacques (1987). Rhetorical Figures in the Advertising Image. In: Jean Umiker-Sebeok (Ed.) Marketing and Semiotics (pp 295-318), Berlin, New York, Amsterdam: Mouton de Gruyter.
Holub, Robert C. (1984). Reception Theory. A Critical Introduction. London: Methuen, Inc.
Iser, Wolfgang (1978). The Act of Reading. London: John Hopkins Univiversity Press.
Messaris, Paul (1997) Visual Persuasion. The Role of Images in Advertising. Sage Publications.
Perelman, Chaim & L. Olbrechts-Tyteca. (1971 [1969]). The New Rhetoric. A Treatise on Argumentation. London: University of Notre Dame Press (First paperback edition).
Scott, Linda M. (1994). Images in Advertising: The Need for a Theory of Visual Rhetoric. Journal of consumer research, 21, pp 252-273.
Toulmin, Stephen. (1994 [1958]). The Uses of Argument. Cambridge: Cambridge University Press.
Toulmin, Stephen, Richard Rieke & Allan Jannik 1984. An Introduction to Reasoning. Second Ed. Macmillan Publishing.

1. Introduction
One of the forms of rule skepticism, found both in legal practice and in legal theory, learns that the law is what the courts say it is and nothing more. In his study The Concept of Law (1961) Hart criticizes this form of rule skepticism. Decisions of a court he says, are statements with a certain authority making them final but not also infallible. To clarify this, Hart uses the example of an umpire in a game. In a game the judgements of an umpire – for instance about the scoring – have a certain authority. His judgements are given, by the secondary rules of the game, a status which renders them unchallengeable. In this sense it is true, says Hart, that for the purposes of the game ‘the score is what the scorer says it is’. But it is important to see that there is a scoring rule and it is the scorer’s duty to apply this rule as best he can.[i] It is this scoring rule which makes decisions of the umpire, though final, not infallible, for this scoring rule offers reasons for criticizing the decision.
According to Hart the same is true in the law. Like the umpire’s decision in a game, the decisions of a judge like ‘X is guilty’ or ‘X has a right’ are – up to a certain point – final. But, like the umpire in a game, the judge has an obligation to apply the rules correctly according to the secondary rules in a legal system.[ii] As a result judicial decisions are fallible.
Austin (1962) made similar observations about the nature of judicial decisions. He argues that if it is established that a performative utterance is performed happily and in all sincerity, that still does not suffice it beyond the reach of all criticism. It may always be criticized in a different dimension, a dimension comparable with the true/false criterium used to evaluate constative utterances: ‘Allowing that, in declaring the accused guilty, you have reached your verdict properly and in good faith, it still remains to ask whether the verdict was just, or fair’ (1962:21)
Since the publications of Austin en Hart, the observations about the character of judicial decisions give rise to the question what type of speech act is involved. Both in legal theory and in argumentation theory it is posed as a problem whether these speech acts are, or are to be reconstructed, as declarative, or as assertive speech acts. For on the one hand, the judge declares that somebody is guilty, but on the other the judge justifies that this decision is right according to the law. And this justification is a reason to reconstruct the decision as an assertive or, to be more precise, as a standpoint in a context of a discussion.
In this paper, I want to discuss the problem of the speech act character of a judicial decision within the framework of the pragma-dialectical argumentation theory. My basic starting point is that it is a misunderstanding to treat speech acts in judicial decisions as either assertive or declarative speech acts. I think that, for an adequate analysis of the speech act, one has to make a distinction between at least two discussions in a legal process and related to this distinction different functions of the speech act in a final judicial decision.
I will proceed as follows. First, I will discuss the merits and demerits of reconstructing a final judicial decision as the mixed speech act called assertive-declaration. Then, I will differentiate between two discussions and two types of speech acts in a legal process. Finally, I will discuss how these two different types of speech acts can be reconstructed as a standpoint.

2. Final judicial decisions as an assertive-declarative speech act
For those who are familiar with Speech Act Theory, it will be clear that it is possible to analyse a judicial decision as ‘X is guilty’ as the combination of an assertive and a declarative speech act. Searle (1979) contends that at least some members of declarative speech acts overlap with members of the class of assertive speech acts. These assertive-declarative speech acts have two illocutionary points. First, they have the assertive illocutionary point, according to which a speaker succeeds in achieving on a proposition P (X is guilty), if and only if he represents the state of affairs that P as actual. Second, they have the declarative illocutionary point, according to which a speaker succeeds in achieving on a proposition P if and only if he brings about the state of affairs that P.
Searle illustrates this double character of the assertive-declarative speech act with the example of the umpire who decides: ‘You are out’. In certain institutional situations, he explains, we not only ascertain facts but also need an authority to lay down a decision as to what the facts are:
Some institutions require assertive claims to be issued with the force of declarations in order that the argument over the truth of the claim can come to an end somewhere and the next institutional steps which wait on the settling of the factual issue can proceed (Searle 1979).
So, in Searle’s perspective an assertive declaration can be simultaneously conceived as a representation of a state of affairs (which is in keeping with the assertive point) and as the constitution of a state of affairs (which is in keeping with the declarative point). In Searle’s interpretation of the relationship between the two illocutionary points, the rule for assertive declarations would be as follows:

A speaker succeeds in achieving with respect to a proposition P the assertive illocutionary point if and only if he represents the state of affairs that P as actual, and, in addition, he succeeds in achieving with respect to P the declarative illocutionary point if and only if he brings about the state of affairs that P (Ruiter 1993: 61).
According to Ruiter (1993) this rule has paradoxal implications. Imagine, he says, that a judge decides ‘X is guilty’, with regard to a situation in which he is not guilty according to the law. On Searle’s account it must be accepted that the judge’s false decision is not only unchallengeable but actually true. For under the second part of the above rule, ‘X is guilty’ becomes a state of affairs owing to the judge’s decision, in consequence of which the assertion that ‘X is guilty’ is true under the first part.
Ruiter tries to solve this problem by making a distinction between the institutional world and the surrounding world of utterance. In the institutional world the judge’s decision ‘X is guilty’ constitutes the institutional fact that ‘X is guilty’. When this decision fails on the assertive point in the surrounding world of utterance, the fact still counts as an institutional fact.

3. Two discussions in a judicial decision
The main problem with Ruiters solution is that the difference between the ‘institutional world’ and the ‘surrounding world of utterance’ is rather general and not very clear. Another way to analyse the speech act (or speech acts) in a judicial decision – as I said in my introduction – is to make a distinction between different discussions in a legal process and related to these discussions different functions of a speech act like ‘X is guilty’.
Let me start with the analysis Feteris (1989) proposed. Following the pragma-dialectical model of a discussion, she gives a reconstruction of judicial decisions. In this model four stages are distinguished. At the confrontation stage, it is established that there is a dispute. A standpoint is advanced and questioned. At the opening stage, the decision is taken to attempt to resolve the dispute by means of a regulated argumentative discussion between a protagonist and an antagonist. At the argumentation stage, the protagonist defends his standpoint and the antagonist asks for more argumentation from him if he has further doubts. Finally at the concluding stage, it is established whether the dispute has been resolved because the standpoint or the doubt concerning the standpoint is being retracted.
Feteris locates the judicial decision in the concluding stage of the discussion between two parties in a process. If the facts stated can be considered as established facts and the judge has decided that there is a legal rule which connects the claim to these facts, the judge will grant the claim. But the secondary rules of a legal system do not only oblige a judge to give a decision in the dispute, but also to give a justification for this decision. The parties have a right to know which considerations underlie the decision. When a party does not agree with the decision, he can appeal the decision on the basis of the argumentation given in the justification.
On the basis of this analysis, Feteris concludes that a final decision of a judge can be seen as an assertive-declarative speech act. She proposes to reconstruct this speech act as an assertive speech act because the judge is bound to the acceptability of the propositional content of the speech act.
Relating the question of speech act character of a final decision to a stage in a legal discussion is an important step forward in solving the problem, but it leaves a few questions unanswered. The first question is: how can we conceive the final decision both as a standpoint of the judge – an assertive – and as a part of the concluding stage of the discussion between the parties? For in the concluding stage it is established whether the dispute has been resolved. Why should a standpoint and argumentation be part of a concluding stage? For an answer to this question, I think, we must make a distinction between at least two discussions in legal decision making. When we make this distinction there is another way to solve the paradoxes concerned to the assertive-declarations as observed by Ruiter.

The first discussion is the one between the parties. In this discussion the two parties defend and criticize a standpoint and the judge is a third party to the dispute. In this discussion the decision of the judge is part of the concluding stage where the discussion is brought to an end. The judge has the extra linguistic position to declare that somebody is guilty. This utterance of the judge must be reconstructed as a declarative speech act, for the fact that the judge says that ‘X is guilty’ brings about the state of affairs that ‘X is guilty’.
This reconstruction is in line with the pragma-dialectical theory about a critical discussion and the difference that is made between resolving a difference of opinion on the one hand and settling a dispute on the other. The declaration of the judge, seen from the perspective of the discussion between parties, is not a part of a critical discussion but a form of dispute settlement. So, the declarative is legal according to the rules.
As Feteris (1989) points out, a judge does not only declare that somebody is guilty, the judge also justifies why he is guilty according to established facts and legal rules. In other words, the judge defends the standpoint that the decision is acceptable. This standpoint is not a part of the discussion between the parties, but a part of the discussion between the judge and the parties, or between the judge and other explicit or implicit antagonists. In this discussion the standpoint of the judge is part of the confrontation stage. And, since the argumentation the judge gives is meant to convince the parties that his standpoint is right according to the law, his standpoint and argumentation are part of a critical discussion aimed at resolving a difference of opinion. When a party does not agree with the decision, he can appeal the decision on the basis of the argumentation given in the justification. And then, there is an explicit discussion between a party and the judge.
So, when we reconstruct the final decision of the judge both as a part of a concluding stage in the discussion between the parties and as a part of a confrontation stage in the discussion between the judge and one or more parties, the decision can be reconstructed both as a declarative and an assertive.
Does this resolve the paradox that Ruiter observes? I think it does. For the judge’s decision that somebody is guilty constitutes the institutional fact of his being guilty in the concluding stage of the discussion between parties. The decision has of course success of fit on the assertive illocutionary point only if he really is guilty according to the law. But if X is not guilty, he will nevertheless institutionally be guilty as long as the judge’s decision is not redressed. Though the falsity of assertion ‘X is guilty’ may offer a reason for invalidating it, it remains valid unless it is invalidated. In this way a false representation of a state of affairs counts legally as a state of affairs notwithstanding its lack of correspondence with reality.

In answering the question whether the decision is a declarative or an assertive, I have said that it is a declarative from one point of view and a standpoint from another point of view. Untill now it was understood that the speech act advancing a standpoint is an assertive speech act. But the next question is: what type of assertive is involved? I will now discuss some implications of the pragma-dialectical characterization of the assertive speech act ‘advancing a standpoint’ as given by Houtlosser (1994 and 1995), for the standpoint character of legal decisions. Houtlosser characterizes the speech act advancing a standpoint as a complex assertive that is at a higher textual level than the sentence connected to an expressed opinion that is confronted (or assumed to be confronted) with doubt or contradiction on the part of a critical listener. According to the essential condition, advancing a standpoint counts as taking the responsibility for a positive or negative position in respect of an expressed opinion, i.e. as assuming an obligation to defend this position in respect of the expressed opinion if called upon to do so. In principle, Houtlosser explains, the assertive speech act advancing a standpoint is related to assertive speech acts, but it can also be related to non-assertive speech acts. In the latter case, the expressed opinion consists of an assumption concerning the acceptability of a speech act that has become the object of contention in a debate or a text.
What are the consequences of this characterization of a standpoint for the discussion between the judge and a party in a legal discussion? Let us look at the example where the judge finds X guilty of murder and one of the arguments is that X had the intention to murder his wife. According to Houtlosser we can analyse this example as follows. The judge asserts that X is guilty of murder. This assertive presupposes its own acceptability. In his argument ‘X had the intention to murder his wife’ the judge reacts to or anticipates on the criticism of the accused (‘it was self-defense’) by supporting the disputed presupposition that his assertive is acceptable.  In doing so, he makes it function as a standpoint. He supports his standpoint with an argument supporting the propositional content of the assertive. According to Houtlosser we can reconstruct the standpoint as follows: ‘It is my standpoint that the assertion that X is guilty of murder is acceptable’
This example shows how we can reconstruct the assertive of the judge as a standpoint in the discussion between the judge and one of the parties in a process. What about a declarative of the judge in the concluding stage of the discussion between parties? Is it possible that this speech act – so to say – develops into a standpoint? I that is possible. As I have said, Houtlosser explains that the assertive speech act advancing a standpoint can be related to non-assertive speech acts. In these cases, the acceptability of the non-assertive speech act has become the object of discussion. Let us look at the example where the judge finds X guilty of murder and one of the arguments is that X murdered his wife in London. How can we reconstruct this argumentation? Let us start by analyzing the utterance ‘X is guilty of murder’ as a declaration in the concluding stage of the discussion between the parties. As we have seen, it is a necessary condition for a succesfull performance of this speech act that the judge has the extra linguistic position to declare something. Let us assume that it was this aspect of the speech act, that was (or was expected to be) criticized by the accused in saying that the judge has no jurisdiction in this case. By criticizing the acceptability of the judge’s (expected) declarative, the accused turns the presupposition that the declarative could be succesfully performed into an issue for discussion.[iii] The judge reacts to or anticipates on this criticism supporting the disputed presupposition that he could perform the declarative because he had jurisdiction. He supports his standpoint with an argument relating to the conditions for performing a declarative. The standpoint of the judge can be reconstructed as ‘It is my standpoint that the declaration that X is guilty of murder is acceptable’.

4. Conclusion
In this paper I have discussed some problems of the speech act character of a judicial decision. I have tried to show that it is a misunderstanding to treat speech acts in judicial decisions as either an assertive or a declarative. Instead we have differentiate between at least two discussions in a legal process. A discussion between parties and a discussion between the judge and his real or anticipated opponents. In these two discussions the judicial decision plays a different role. In the first he declares something, in the second he asserts something. Finally I have tried to show that the declaration of the judge can be questioned and then be the object of the assertive advancing a standpoint.

NOTES
i. Cf. Hart (1961:142): ‘“The score is what the scorer says it is” would be false if it meant that there was no rule for scoring save what the scorer in his discretion chose to apply. There might indeed be a game with such a rule, and some amusement might be found in playing it if the scorer’s discretion were exercised with some regularity; but it would be a different game. We may call such a game the game of ‘scorer’s discretion’.’
ii. Hart (1961:94) differentiates between primary rules and secondary rules in a legal system. Primary rules are concerned with the actions that individuals must or must not do. Secondary rules are all about primary rules: they specify the ways in which the primary rules may conclusively ascertained, introduced, eliminated, varied and the fact of their violation conclusively determined.
iii. Cf. Austin (1974:14) ‘[…] Our performative, like any other ritual or ceremony, may be, as the lawyers say, ‘nul and void’. If for example, the speaker is not in a position to perform an act of that kind, or if the object with respect to which he purports to perform is not suitable for the purpose, then he doesn’t manage simply by issuing his utterance, to carry out the purported act.’

REFERENCES
Austin, J.L. (1974). Performative-Constative. In: J.R. Searle (Ed.), The Philosophy of Language (pp. 13-22). Oxford: Oxford University Press. [Transl. of Austin, J.L. (1962). ‘Performatif-Constatif’. Cahiers de Royaumont, Philosophie IV: La Philosophie analytique. Paris: Les Éditions de Minuit, 271-304.]
Eemeren, F.H. van & R. Grootendorst (1992). Argumentation, Communication and Fallacies. Hillsdale: Erlbaum.
Feteris, E.T. (1989). Discussieregels in het recht. Een pragma-dialectische analyse van het burgerlijk proces en het strafproces. Dordrecht: Foris.
Hart, H.L.A. (1961). The concept of law. Oxford: Oxford University Press.
Houtlosser, P. (1994). The Speech act ‘advancing a standpoint’. In: F.H. van Eemeren & R. Grootendorst (eds.), Studies in Pragma-Dialectics (pp. 166-171). Amsterdam: Sic Sat.
Houtlosser, P. (1995). Standpunten in een kritische discussie. Een pragma-dialectisch perspectief op de identificatie en reconstructie van standpunten. Amsterdam: IFOTT.
Ruiter, D.W.P. (1993). Institutional legal Facts: Legal Powers and their Effects. Dordrecht etc.: Kluwer.
Searle, J.R. (1979). A taxonomy of illocutionary acts. In: J.R. Searle, Expression and Meaning. Studies in the Theory of Speech Acts (p.1-29). Cambridge: Cambridge University Press.

1. Introduction
Policy decisions can give rise to lively public debates. Should we build a new airport, expand the old one, or try to cut down on travelling by airplanes? Should we build more motorways or make the public transport cheaper in order to solve the traffic congestion problem? When a debate arises, each option will have its own proponents. They will try to persuade others that their option is indeed in everyone’s best interests. To achieve that goal, they put forward pragmatic argumentation. That is, they claim that their option will probably or certainly result in desirable consequences. The strength of their argument depends on two aspects: The consequence’s desirability and the consequence’s probability. A strong argument in favor of the option would be that the option will certainly result in desirable consequences.
Previous research has shown that people have more trouble evaluating arguments supporting a probablity claim than evaluating arguments supporting a desirability claim (Areni & Lutz 1988). In other words: The argument quality of a desirability argument is more transparent than that of a probability argument. O’Keefe (1995) suggested that argumentation theory provides a framework to study the concept of argument quality. However, he also warned that what should be convincing from the point of view of an argumentation theorist, is not always convincing from a layperson’s point of view.
In this paper, I will first discuss the different types of argument that can be used to support a probability claim. Next, I will review empirical research in which the actual persuasiveness of these types of argument is studied. However, in none of the studies, the persuasiveness of the different argument types has been compared directly. Section 4 contains the description of an experiment in which the same claim is supported by different types of argument. The actual persuasiveness of these argument types is measured, as well as the extent to which the participants think that they are convincing.

2. Types of argument
In policy debates, probability claims typically refer to future events, for instance: building a new airport will boost the economy. To support such claims, one can use inductive reasoning. Usually, three types of argument are distinguished in inductive reasoning (see, e.g., Govier 1992). Following the terminology employed by Rieke and Sillars (1984), these three types are the argument by analogy, the argument by generalization, and the argument by cause.
Rieke and Sillars (1984: 76-77) define an argument by analogy as follows: “(…) you compare two situations which you believe to have the same essential characteristics, and reason that a specific characteric which exists in one situation can be reasoned to exist in the analogous situation”. For instance, to support a claim about the beneficial economic effect of building a second airport, proponents may give the example of another country in which the building of a second airport had a strong beneficial effect on that country’s economy. Essential for the quality of this argument, is the extent to which the two countries are similar. The more similar the countries, the more valid the argument by analogy.

The argument by generalization proposes that “you look at a series of instances and from them claim a general principle” (Rieke & Sillars 1984: 72). For instance, instead of giving just one example of a country profiting from building a second airport, one provides a number of such examples. As the number of examples grows larger, the argument by generalization may result in a argument using statistical evidence. Instead of discussing several examples, one presents a percentage or some other descriptive statistic representing the proportion of countries profiting from building a second airport. The quality of this type of argument depends on the number of observations and the representativeness of the observations. For instance, an argument by generalization based on one hundred examples is normatively better than an argument based on two examples.
However, when (most of) the hundred instances are very dissimilar from the issue at hand, the argument should not be convincing. For instance, the effects of building a second airport in developing countries may not be comparable to building a second airport in The Netherlands.
The argument by cause provides an explanation why a certain effect may arise (Rieke & Sillars 1984: 74). In the case of a second airport, one might argue that building it will improve a country’s economic position because (1) building and running such an airport will provide employment for thousands of people, and (2) it will improve the country’s position as a major distribution point in the world’s economy thereby attracting foreign companies to settle there. The quality of this argument depends on the presence or absence of other factors that might cause the second airport to become a failure or a success.
From a normative point of view, an argument by generalization that is based on a sufficiently large sample of representative instances, should be more convincing than an argument by analogy, especially if the latter uses an example that differs strongly from the issue at hand. Whether an argument by cause should be more convincing than an argument by generalization depends on the extent to which the argument by cause identifies the most important possible causes. The question that will be addressed in the next section is the extent to which what should be convincing, is convincing in actuality.

3. Empirical studies on the persuasiveness of different types of argument
A number of experiments have been conducted to assess whether some types of arguments are more convincing than others. Especially the distinction between the argument by analogy and the argument by generalization has received much attention by researchers. In several reviews, it is concluded that the argument by analogy is more persuasive than the argument by generalizability (see, e.g., O’Keefe 1990: 168-169; Taylor & Thompson 1982: 163-164). Baesler and Burgoon (1994) found 19 experiments in which the persuasiveness of the argument by analogy was directly compared to that of the argument by generalizability. In 13 experiments, the argument by analogy proved to be more convincing than the argument by generalizability; in only 2 experiments, the opposite effect was obtained (No differences between types of argument were found in the remaining 4 experiments).
Based upon such reviews, O’Keefe (1995: 15) noted that there is a distinction between what constitutes a strong argument from normative point of view (i.e., the argument by generalization), and from a descriptive point of view (i.e., the argument by analogy). However, Baesler and Burgoon (1994) claim that the manipulation of the two types of argument is (often) confounded with a second factor: the argument’s vividness. That is, an argument by analogy usually presents an anecdote to support the claim; in an argument by generalizability, the claim is usually supported by statistics. In general, an anecdote is easier to imagine than statistics. Nisbett and Ross (1980) dub this the vividness effect. A vivid argument would be more convincing than a more pallid one. Following this line of reasoning, an argument by analogy would be more convincing than an argument by generalizability, not because it is based on a single instance, but because of its higher imagineability.

To test this explanation, Baesler and Burgoon (1994) manipulated not only the type of argument (argument by analogy or argument by generalizability), but the vividness of these arguments as well. That is, they provided vivid statistical and anecdotial evidence as well as pallid statistical and anecdotial evidence. Controlling the evidence’s vividness led to a pattern of results different from the usually reported one: The argument by generalizibility (employing statistical evidence) proved to be more convincing than the argument by analogy (employing anecdotial evidence). Hoeken and Van Wijk (1997) obtained a similar pattern of results using a different message on a different topic. The vividness of the argument by analogy may therefore be the reason for the often reported finding that the normatively stronger, but less vivid argument is less convincing than the normatively weaker, but more vivid argument.
Compared to the argument by analogy and the argument by generalizability, the argument by cause has received far less attention by researchers. Slusher and Anderson (1996) compared the convincingness of an argument by cause to that of an argument by generalizability. They used a message stating that AIDS is not transmitted by casual contact (including nonsexual household contact or contact through mosquitos). Evidence substantiating this claim was either causal or statistical. The argument by cause, for instance, ran that “The Aids virus is not concentrated in saliva, is not present in sweat, and has to be present in high concentration to infect another person. The argument by generalizability stated that in ”a study of more than 100 people in families where there was a person with AIDS without the knowledge of the family and in which normal family interactions (…) took place revealed not a single case of AIDS transmission.”

The results showed that the argument by cause was more successful at changing faulty beliefs about the ways in which AIDS can be transmitted than the argument by generalization. Because it is much more difficult to change an existing belief than to form a new belief, these results suggest that the argument by cause is a powerful argument. The superior effect of the argument by cause may have two reasons. Slusher and Anderson (1996) state that using arguments by cause result in the availability of explanations why AIDS cannot be transmitted by casual contact. As the availability of explanations increases, people are more inclined to accept the claim. In contrast, the argument by generalization does not lead to an increase of available explanations. A second explanation for the superior effect of the argument by cause may be that it enables people to build a model of why and how an effect may or may not occur. The argument by generalizability does not enable one to construe such a model. Having such a model, regardless of how tentative it may be, strengthens the belief that a certain effect will occur (Tversky & Kahneman 1982).
The empirical studies on the convincingness of different types of argument enable the following, tentative conclusions. Although most studies show that the argument by analogy is more convincing than the argument by generalizability, this effect may be the result of an artefact. In an argument by analogy, usually more vivid, anecdotial evidence is employed, whereas the statistical evidence typically employed in an argument by generalizability is more pallid. When the vividness of evidence is controlled, however, the argument by generalizability is more convincing than the argument by analogy. In the only experiment in which the convincingness of an argument by cause is directly compared to an argument by generalizability, the former proved to be more convincing than the latter. A tentative ordering of the different types of argument would be that the argument by cause is more convincing than the argument by generalizability, which in turn is more convincing than the argument by analogy.

4. The experiment
An experiment was conducted to address two topics. First, I tried to replicate earlier findings that an argument by analogy is less persuasive than an argument by generalizability, which in turn is less persuasive than an argument by cause. Replicating such effects employing arguments on different topics is an important precondition before general conclusions about message and argumentation effects can be drawn (cf. O’Keefe 1990: 121-129). Apart from replication, the experiment extends previous empirical studies. For the first time, the three different types of argument were compared directly. That is, the same claim was supported either by an argument by analogy, an argument by generalizability, or an argument by cause.

The second topic concerns the relation between the perception of the argument’s quality and its actual persuasiveness. In the experiments discussed above, the extent to which participants accepted the claim was measured. They were not asked whether they regarded the argument as strong. In this experiment, participants not only rated the extent to which they accepted the claim, they also indicated their opinion about the argument’s strength. One would expect these scores to correlate. That is, the type of argument being rated as strongest, should be the most convincing one as well. In an experiment by Collins, Taylor, Wood and Thompson (1988), however, participants rated one message as more persuasive than another, whereas in actuality they were equally persuasive. To assess whether the perception of argument strength corresponds with the actual persuasiveness, both variables were measured.

The discussion above leads to the following two research questions:
1. Do different types of argument lead to differences in actual persuasiveness?
2. Do differences in persuasiveness correspond to differences in perceived argument quality?

To answer these questions, an experiment was conducted in which a claim about the future financial success of a cultural centre was backed up by either an argument by analogy, an argument by generalizability, or an argument by cause. The argument by analogy was deliberately weakened through choosing an example that differed on essential characteristics from the issue under consideration.

4.1 Method
Material
The material consisted of three versions of a (fictitious) newspaper article on a council meeting in the Dutch town of Doetinchem. The meeting was about the mayor’s proposal to build a multi-functional cultural centre. It was reported that some of the council members doubted that such a centre would be profitable. They feared that the citizens would have to pay for the losses. The mayor argued that the centre would attract sufficient visitors and make a profit within four years. The argument to support this claim could be either an argument by analogy, an argument by generalization, or an argument by cause. All arguments consisted of 6 sentences and 75 words.
The argument by analogy stated that a similar centre in the city of Groningen had been very successful. It had made a profit within four years. Groningen differed from Doetinchem on several important dimensions. Unlike Doetinchem, Groningen has a university and is much larger than Doetinchem. Furthermore, it is situated in a different part of The Netherlands. In a previous experiment, size of population, type of city, and location in the country, were identified as the most defining characteristics of a town (Hoeken & Van Wijk 1997).
The argument by generalization referred to a study by the Dutch Organization of Municipalities. In the study, the profitability of 27 cultural centres in different towns of varying size, dispersed over The Netherlands had been assessed. On average, the centres had made a profit within four years. Finally, the argument by cause provided three reasons why the cultural centre would be profitable. First, many citizens from nearby towns went to a faraway cultural centre to see movies and plays. Second, a popular movie theatre in a nearby town had burnt down. It was believed that the visitors would find their way to cultural centre in Doetinchem. Finally, Doetinchem’s demographics showed that the number of well-educated people who are well-off increased. Such people like to visit cultural centres.

Participants
A total of 324 participants took part in the experiment. There were slightly more men (51.2%) than women (48.8%). Their age ranged from 17 to 72 with an average of 29 years. Education ranged from primary education to a master’s degree. The majority (67.7%) had completed at least grammar school.

Questionnaire
The questionnaire contained questions on a number of variables such as the participants’ cognitive responses, their evaluation of the article, their own behavior with respect to cultural activities, and some general questions about their level of education, sexe, and age. In addition, to test whether the argument by analogy was perceived as more vivid than the other types of argument, the text’s vividness was measured. The most relevant variables with respect to the research questions were those operationalizing the argument’s actual and its perceived persuasiveness. The argument’s actual persuasiveness was operationalized as the extent to which participants accepted the claim that the centre would make a profit within four years. The argument’s perceived persuasiveness was operationalized by having participants rate the argument’s strength and its relevance.

The acceptance of the claim
The acceptance of the claim that the centre is capable of generating money was measured by the clause “The probability that the cultural centre will make a profit within four years, seems to me” followed by four seven-point semantic differentials. Two of the four semantic differentials had the positive antonym at the left pole of the scale (large, present), the other two had the positive antonym at the right pole (probable, realistic). The reliability of the scale was good (Cronbach’s alpha = .89).

Perception of argument quality
The perceived argument quality was measured using four seven-point semantic differentials and one seven-point Likert scale. The semantic differentials were preceded by the clause “I regard the argumentation supporting the claim that the centre will attract sufficient visitors as”. Two of the four semantic differentials had the positive antonym at the left pole of the scale (sound, relevant), the other two had the positive antonym at the right pole (strong, convincing). For the Likert-item, the argument was repeated. For instance, in the case of the analogy-argument: The mayor referred during the council meeting to the profit made by a cultural centre in Groningen. How relevant do you rate this example with respect to the decision to build a cultural centre in Doetinchem? The participants indicated their response on a seven-point Likert scale ranging from “very irrelevant” to “very relevant”. The five items formed a reliable scale (Cronbach’s alpha = .83).

Design
A factorial design was used, that is, each subject read only one of the text versions. This resulted in three experimental groups. After reading the text, they responded to the various items of the questionnaire.

Procedure
Each participant was run individually. Participants were told that the Linguistics department of Tilburg University was interested in the way in which people made up their mind in case of a referendum. After this introduction, the participant received the experimental booklet. After completing the experimental booklet, participants were informed about the true purpose of the experiment and thanked for their cooperation. An experimental session lasted about 14 minutes.

4.2 Results
First, it was tested whether the different types of argument were rated as equally vivid. In previous experiments, the argument by analogy was often more vivid than the argument by generalizability thereby influencing the argument’s persuasiveness. An analysis of variance revealed no differences between the three types of argument with respect to perceived vividness (F 1).

The first research question was: Do different types of argument lead to differences in actual persuasiveness? Table 1 contains the mean ratings of the acceptance of the claim that the cultural centre will make a profit within four years and the mean ratings of the perceived argument quality.

TABLE 1 – The mean ratings and standard deviations of acceptance of the claim and perceived argument quality as a function of argument-type (1 = very negative, 7 = very positive)

An analysis of variance revealed a main effect of Argument type on the acceptance of the claim that the centre would be profitable (F (2, 321) = 5.31, p .01; eta2 = .03). Post hoc comparisons using the Tukey’s HSD test showed that the argument by generalization led to higher scores than the argument by analogy and the argument by cause. The latter two did not differ from each other. The second research question was: Do differences in persuasiveness correspond to differences in perceived argument quality?
Analyses of variance revealed main effects of Argument type for the perceived argument quality (F (2, 320) = 19.61, p .001; eta2 = .11). Post hoc comparisons using the Tukey’s HSD test showed that the argument by analogy was perceived as weaker than the argument by generalization and the argument by cause. The latter two did not differ from each other on perceived strength.

There appears to be a discrepancy between the argument by cause’s perceived persuasiveness and its actual persuasiveness: Whereas the argument by cause is perceived to be stronger than the argument by analogy, it led to similar scores with respect to the acceptance of the claim. This discrepancy is corroborated by the correlations between the perceived argument quality and the acceptance of the claim. The correlations and the percentage of explained variance are displayed in Table 2.

TABLE 2 – The correlations and percentages explained variance between the acceptance of the claim and the perceived argument quality as a function of argument type

Whereas the correlations between perceived quality and claim acceptance are high for the argument by analogy and for the argument by generalization, they are much lower for the argument by cause.

4.3 Conclusion
The first research question was: Do different types of argument lead to differences in actual persuasiveness? The answer is affirmative: The types of argument had a different effect on the acceptance of the claim. However, the differences do only partly replicate the pattern of results obtained in other studies. In this study, the argument by generalizability proved to be stronger than the argument by analogy. As such, it replicates the results of Baesler and Burgoon (1994) and Hoeken and Van Wijk (1997). The expected superior effect of the argument by cause did not arise. On the contrary, the argument by cause proved to be equally convincing as the argument by analogy and less convincing than the argument by generalizability. This result deviates from the results reported by Slusher and Anderson (1996), who found the argument by cause to be more convincing than the argument by generalizability.
The second question was: Do differences in persuasiveness correspond to differences in perceived argument quality? Again, the answer is partly affirmative. In correspondence with the actual persuasiveness, the argument by generalizability is rated as stronger than the argument by analogy. Ratings of the argument’s strength are in both cases strongly related to the actual persuasiveness. In contrast, the argument by cause received higher ratings compared to its actual persuasiveness. It was rated as stronger than the argument by analogy despite the fact that both types of argument yielded similar claim acceptance ratings. The correlation between the perceived argument strength and its actual persuasiveness is much lower for the argument by cause compared to the correlations for the other two types of argument. In the next section, an interpretation for these results will be put forward and the implications discussed.

5. General discussion
The first research question related to the persuasiveness of different types of arguments. In reviews of empirical research, it is often concluded that the argument by analogy is more persuasive than the argument by generalizability. However, as shown by Baesler and Burgoon (1994), this pattern may be the result of confounding argument type with vividness of evidence. When the vividness of the anecdotial evidence employed in the argument by analogy is equally vivid as the statistical evidence employed in the argument by generalizability, the latter is more convincing than the former. In the experiment reported above, there was no difference in perceived vividness, and the argument by generalizability was more persuasive than the argument by analogy. Therefore, the results replicate the finding that the argument by generalizability is more convincing than the argument by analogy if the vividness of the arguments is controlled.
The results on the acceptance of the claim did not replicate previous results obtained for the argument by cause. Instead of being more convincing, the argument by cause proved to be less convincing than the argument by generalizability. A possible explanation for this difference may be the confounding of an argument by cause with an argument by authority. Slusher and Anderson (1996) attacked the claim that AIDS can be transmitted through casual contact or mosquitos. They stated that the AIDS virus has to be present in a high concentration. Neither saliva nor sweat contains a sufficiently high concentration to contaminate another person. This explanation was suggested to be the result of scientific research. Scientists are commonly regarded as competent and reliable sources, thereby lending the argument extra credibility.

In the experiment described above, the explanation of why the cultural centre would be a success was given by the mayor. The mayor himself proposed to build such a centre. Therefore, people may question his impartiality in this matter. Furthermore, a mayor is usually not an expert on the factors that contribute to a cultural centre’s success. Therefore, participants in this experiment may have regarded the source of the explanation as less credible than the (scientific) source in the Slusher and Anderson experiment. This difference in source credibility may have been responsible for the different pattern of results. In order to test this explanation, the causal argument why the cultural centre will become a succes should be ascribed to an independent expert. In that case, the causal argument should be more convincing than the argument by generalizability.
The second research question addressed the relation between perceived argument quality and actual persuasiveness. For the argument by analogy and the argument by generalizability, this relation was straightforward. The higher the perceived argument quality, the more convinced people were, and vice versa. For the argument by cause, the relation proved to be more problematic. Although the argument was perceived as strong, it was not very convincing. The correlation between the perceived argument quality and the actual persuasiveness was markedly lower than the correlations for the other two types of argument.

In the experiment, the participants first indicated to what extent they agreed to the claim that the centre would make a profit. After that, they rated the argument’s quality. The results suggest that only when asked to reflect upon the argument’s quality, the participants who had read the argument by cause realized that the argument was pretty sound. Apparently, the argument by cause needed closer inspection in order to be convincing. This should not lead to the conclusion that only when asked to reflect upon the arguments, people distinguish between strong and weak arguments. If that were the case, no effects of argument type would have been obtained. However, the argument by generalizability lead to a stronger acceptance of the argument’s claim than the argument by analogy. That effect was obtained before participants were asked to reflect upon the argument’s quality. Therefore, even when not instructed to reflect upon argument quality, people are sensitive to differences in argument type.
The discrepancy between the perception of argument quality and the actual persuasiveness only arises for the argument by cause. It is possible that people believe that an argument by cause is convincing whereas in actuality they are not persuaded by it. Collins et al. (1988) report a similar pattern of results on the effect of colourful language. They showed that a message containing colourful language was rated as more persuasive without yielding any significant attitude change. Collins et al. conclude that there is a widespread belief that colourful language facilitates persuasion, thereby influencing people’s ratings of a message’s persuasiveness. In actuality, people would not be sensitive to this message variable.
Something similar may be the case for the argument by cause. Our understanding of the world is largely based on laws of cause and effect. An argument based on such a relation may therefore give the impression of being very convincing without having this effect. The results of the experiment underscore two points. First, the results once again stress the importance of replicating the effects of message and argument variables. Seemingly small differences in argument manipulation can lead to large differences in persuasiveness. Second, it is important to distinguish between what is perceived as convincing and what actually is convincing. Opinions about what constitutes a stronger argument do not necessarily guarantee a stronger persuasive effect. Finally, the results do clarify the need of further study of the conditions under which the argument by cause is persuasive.

REFERENCES
Areni, C. S. & R. J. Lutz (1988). The role of argument quality in the elaboration likelihood model. In: M. J. Houston (Ed.), Advances in consumer research (Vol. 15, pp. 197-203). Provo, UT: Association for Consumer Research.
Baesler, J. E. & J. K. Burgoon (1994). The temporal effects of story and statistical evidence on belief change. Communication Research 21, 582-602.
Collins, R. L., S. E. Taylor, J. V. Wood & S. C. Thompson (1988). The vividness effect: Elusive or illusory? Journal of Experimental Social Psychology 24, 1-18.
Govier, T. (1992). A practical study of argument (3rd Ed.). Belmont, CA: Wadsworth.
Hoeken, H. & C. van Wijk (1997). De overtuigingskracht van anekdotische en statistische evidentie [The convincingness of anecdotial and statistical evidence]. Taalbeheersing 19, 338-357.
Nisbett, R. & L. Ross (1980). Human inference: Strategies and shortcomings of social judgment. Englewood Cliffs, NJ: Prentice-Hall.
O’Keefe, D. J. (1990). Persuasion. Theory and research. Newbury Park, CA: Sage.
O’Keefe, D. J. (1995). Argumentation studies and dual-process models of persuasion. In: F. van Eemeren, R. Grootendorst, J. Blair & C. Willard (Eds.), Proceedings of the third ISSA conference on argumentation, Vol 1: Perspectives and approaches (pp. 3-17). Amsterdam: SIC SAT.
Rieke, R.D. & M. O. Sillars (1984). Argumentation and the decision making process (2nd Ed.). New York: Harpers.
Slusher, M. P. & C. A. Anderson (1996). Using causal persuasive arguments to change beliefs and teach new information: The mediating role of explanation availability and evaluation bias in the acceptance of knowledge. Journal of Educational Psychology 88, 110-122.
Taylor, S. E. & S. C. Thompson (1982). Stalking the elusive “vividness” effect. Psychological Review 89, 155-181.
Tversky, A., & D. Kahneman (1982). Causal schemas in judgments under uncertainty. In: D. Kahneman, P. Slovic & A. Tversky (Eds.), Judgment under uncertainty: Heuristics and biases (pp. 117-128). Cambridge: Cambridge University Press.

The history of applied logic in the English-speaking countries in the twentieth century can be discerned in the curriculum students have been exposed to in logic courses. That curriculum is manifested most explicitly in the text books that have been used, primarily in logic courses offered by philosophy departments. One of the more interesting aspects of the evolution of the applied logic curriculum is the gradual expansion of interest of logicians in creating techniques for more and more kinds of arguments.
The first half of the century reflected an interest in techniques that could establish whether or not an argument was deductively valid as a consequence of its logical form. Until the thirties, syllogistic dominated as the technique of choice, as it had for centuries before. But the creation of the propositional and predicate calculi around the turn of the century, followed by Gentzen’s development of “natural deduction” versions of these, led to these systems superceding the syllogistic as the preferred tools for inference evaluation. This is reflected in the introductory logic texts that appeared in the late forties and early fifties. Among them was Irving Copi’s Introduction To Logic, which appeared in 1951 and ultimately became the template for many such texts.

An examination of even the latest edition of Copi’s text will show the deductivist orientation of these texts. By their tests, only a small subset of everyday arguments could qualify as having logically good inferences. This fact should have bothered logic teachers, since it was recognized even then that people, including themselves, were often persuaded to believe the conclusions of arguments whose inferences were not formally valid. But the formal techniques continued to hold sway, partly because of a lingering Cartesianism. It was difficult to let go of formal validity as a logical paradigm of good inference. Some of this reluctance has been due to the dubious conviction that logicians ought to have better logical standards than anyone else.
Some people did shake off the spell of formalism, however. I am thinking here of Max Black and Monroe Beardsley, who produced texts around 1950 that look surprisingly contemporary in terms of curriculum. But it was not until around 1970 that texts of this kind began to become popular. Names such as Howard Kahane, Stephen Thomas, and Michael Scriven come to mind. These texts have come to be considered texts in Informal Logic, a “movement” that became visible as a result of the conference organized by Anthony Blair and Ralph Johnson in 1978 at the University of Windsor.

In its narrower version, Informal Logic has focused on the evaluation of inferences made in everyday argumentation, using whatever criteria seem to be appropriate. These could be deductive or inductive tests. Expressed one way, the goal could be seen as that of arriving at a probability value for a conclusion, given the truth of the premisses (Of course, this judgment was not expressed numerically. The preference has been to use evaluative terms found in language).
In a broader version, one that not all logicians are comfortable with, Informal Logic is about argument evaluation. This involves arriving at an evaluative judgment of how likely the conclusion is, given the argument per se, rather given than the truth of the premisses. This broader concept takes account of the logical fact that the probability of a conclusion depends on the probability of premisses as well as inference quality.

Traditionally, logicians have seen their field of interest to be only inference quality. This is partly explained by the historical preoccupation with formal logic. If applied logic is applied formal logic, then obviously premiss evaluation is an empirical matter, to be relegated to the appropriate discipline or subject. However, once we assign logic a broader scope that includes inductive argument, the issue of premiss truth value can be included in the subject, since the issue of premiss truth value is whether or not we can infer the premiss from the information we have.
With the foregoing stage setting, I come to the purpose of this paper, which is to propose a further increase in the scope of Informal Logic. The motive for this proposed extension arises from the recognition that people who have arguments directed to them are interested in more than just arriving at a judgment of conclusion probability given the argument (i.e., argument evaluation).
Typically, people direct arguments to others when they think the “arguee” does not, prior to the presentation of the argument, regard the argument’s conclusion as true. This is why we say that arguments are artifacts for persuasion. The most important question for the arguee, then, is: should I now accept the conclusion as true, after hearing the argument?

Clearly, this question is broader in scope than the earlier question about how likely the argument itself makes the conclusion.
One reason why is that the arguee normally already has information relevant to judging the truth value of the conclusion in question. In some cases, the reason(s) given by the arguer might tip the balance in the direction of belief. In others it won’t, because of some weakness in the argument.
But there is another kind of evidence that can, and should, be taken into account before we decide how likely the conclusion is after hearing the argument. This is arguer credibility. Quite often we are recipients of arguments from people and sources that we recognize as dependable sources for claims of this epistemic kind. Thus, the fact that this source affirms the truth of the claim is itself evidence for the claim. So obviously, this evidence must be factored into our evaluation of the claim.
How these two extra sources of evidence (our prior evidence for and against the conclusion, and arguer credibility) are to be fitted into the theory of claim evaluation is the subject of the remainder of this paper. The basis for the analysis will be a simple model of an argument as a propositional complex.

When an arguer (S) presents an arguee (H) with an argument of the form ‘P, so C.’, he/she is relying on two claims to get H to believe C: (1) P is true, and (2) P, if true, guarantees the truth of C. This latter claim I shall call the “inference claim”. It can be written more familiarly in the form ‘If P then C.’. The sophisticated arguee, in deciding whether or not to accept C as true after hearing the argument, can be thought of as concerned to establish two probability values:
p(P) and p(If P then C). The latter can be written more concisely in the form p(C/P).

Let’s deal with getting p(P) first. The evidence we can have consists of (1) any information we may have that would lead us to assign a probability to P prior to taking account of S’s credibility in affirming it. We can call this “p(P)i”. The issue then is how to factor in S’s credibility. One way of conceiving the situation is to regard the proposition ‘S affirms that P.’ as a premiss for the conclusion P. In judging the probability of P given this little argument we need to use this formula:

p(P) = p(S affirms that P) x p(P/S affirms that P)

We can assume that we know that S has affirmed P, so: p(S affirms that P) = 1. We now have:
p(P) = p(P/S affirms that P)

Using Bayes’ theorem we can write:
p(P) = p(P/S affirms that P) = [p(S affirms that P/P) x p(P)i]
/ [[p(S affirms that P) x p(P)i] +

[(1 – p(S affirms that P / P) x (1 – p(P)i)]]

This is simpler than it looks, once we notice that ‘p(S affirms that P/P)’ represents S’s reliability in judging P. That is, it represents the number of times S would judge P to be the case, when P actually is the case. Let’s label this “RP”. We can now rewrite the complex equation as:

p(P) = p(P/S affirms that P) = [RP x p(P)i] / [[RP x p(P)i] +

[(1 – RP) x (1 -p(P)i]]

This still looks pretty complex, not something we can use without pencil and paper or a calculator. However, for practical purposes we do not need an exact result. A result accurate to one decimal place would be sufficient. In what follows I offer a simplified way of applying the Bayes formula.

By “cut-and-try”, I have found that this formula gives fairly accurate results: p(P) = r / (1 + r). Here “r” is what I call the “Bayes ratio”:
p(P)i / EP. Here “EP” is simply 1 – RP. That is, instead of working with arguer reliability, we use arguer’s error rate.

How close to the Bayes Theorem results are the results using the simplified formula? If we calculate p(P) for any pair of values for p(P)i and EP using the two formulas and round off to one decimal place (0.9, 0.8, etc.), the simple formula will yield a value accurate within one decimal place almost always. (That is, the error is +/-0.1.) For everyday purposes this is pretty accurate.

We could use the simple formula to get an approximate value for p(P), but we can simplify even further if we regard our “bottom line” task as one in which we must decide whether to accept P as true or not. This requires a decision as to what value of p(P) is high enough to warrant regarding P as true. No precise answer can be defended, partly because it depends on what would be at stake in accepting P as true, and partly because some of us are more cautious than others. For purposes of discussion I shall adopt a probability of 80% as a threshold for acceptance. That is, when a claim is seen as at least 80% probable, I will regard this as an adequate basis for taking it to be true.

Looking at our formula, what value does “r” have to have for us to accept P as true? Looking at the formula we can see that when r is 4, p(P) = 4/(1 + 4), or 4/5, or 0.8. So we can adopt the policy of deciding that P is true when r is 4 or greater. That is, when we judge S’s error rate to be less than 1/4 of the initial probability of P. Now let’s see how Bayes applies to the inference claim ‘If P then C’, which I shall abbreviate as “I” when necessary. Recall that an arguer wants to persuade us to believe his conclusion (C) by getting us to accept two other claims: (1) P is true, and (2) ‘If P then C’ is true. We can use the same analysis for the latter as for the former. We can make a judgment of p(C/P) (“p(I)”) prior to taking into account the fact that the arguer is affirming it. Then we can use Bayes to arrive at the following simplified formula:

p(I) = p(I/S affirms that I) = rI / (rI + 1) (Where rI = p(I)I / EI)

We are now in a position to determine how probable C is for us, given what we knew prior to hearing the argument for it, the argument itself, and the epistemic credibility of the arguer. This is simply p(P) x p(I). But the fact that this is a product relationship raises a problem if we want to decide whether or not to accept C as true now.

We noted above that, using an 80% threshold, we would accept P as true if EP was less than 1/4 of p(P)i. We could use the same threshold for I, but if we do, we will be accepting C as true in cases when p(C) is only 0.64. This is when p(P) = 0.8 and p(I) = 0.8. This looks a bit inconsistent, since we would require p(C) to be at least 0.8 if it were asserted without grounds. It is desirable, then, when judging the epistemic impact of an argument, that we use 90% as thresholds for p(P) and p(I). This gives a value for p(C) of 0.81, consistent with the general standard of 0.8.

Now we must revise our threshold values for rP and rI. Remember that, in each case, they occur in the form ‘r /(r + 1)’, we can see that their value is minimally 9 to get a formula value of 0.9. It might be convenient in practice to adjust the value of r to 10. This yields a minimal product value of 0.8264. The standard for accepting C as true now is: accept C as true when both S’s error rate in judging the premiss is less than 10% of the prior probability of the premiss, plus S’s error rate in judging the inference claim is less than 10% of our prior assigned probability value.

These criteria need to be incorporated into a strategy. One of the characteristics or ideals of logicality is that a person ought to be logically autonomous. In dealing with other people’s attempts at persuading us to believe things, we should rely in the first instance on what we already take to be true. Thus, if our information itself leads us to assign values above 0.9 to both P and I, then we can accept the conclusion without relying on S’s reliability. This is preserving our logical autonomy. On the other hand, being logical about an argument also requires us to take account of S’s credibility, so that when either p(P)i or p(I)i is less than 0.9, we need to see if rP or rI is high enough to warrant accepting the claim as true.

Thus, in this scenario, we rely first on our own information, then if accepting the conclusion as true is not warranted by this, we bring S’s reliability into the picture. Being logical involves thinking for oneself, but it is illogical to fail to take all the evidence into account, and this includes arguer credibility.

Taking arguer credibility into account, however, is not easy to do accurately. Cognitive psychologists have found that people do badly173 when required to factor claimer reliability into their claim probability estimates. By training and experience we are able to make judgments about claim probability, but arguer reliability is quite different. The evidence for it is, of course, the person’s background and behavior, but our evaluations can be distorted in a variety of ways. In most cultures we are taught who the knowledgeable people are on the more important subjects, but we do not learn any habits or strategies of reliability evaluation. These difficulties in using the procedure I leave for another time, but their existence does not invalidate the procedure itself. It just means that we need to expand our efforts in teaching critical thinking into this area.

1. Nature and Consequences of the ‘Brent Spar’ Crisis
In June 1995, the giant oil corporation Shell attempted to sink its obsolete oil platform, ‘Brent Spar’, in the North Sea, 190 kilometers north-east of the Shetland Islands. Their plans were approved by the British government and by the signatories of the Oslo Convention for the protection of the marine environment (Shell ‘Brent Spar’ calendar of events: 1). Shortly before the scheduled deepwater disposal, the environmental organization Greenpeace began a ”high-profile campaign” (Thompson 7.3.96) in opposition to Shell’s plan. The ‘Brent Spar’ crisis started on the 30th of April when Greenpeace activists occupied the platform and held it for three months.
The ‘Brent Spar’ crisis was extremely complex because what Shell had considered to be a British domestic issue actually turned out to be an international ”fracas” involving the countries surrounding the North Sea (Seaman 1996: 4). Greenpeace’s and Shell’s actions caused a three month long conflict over the seas, disagreement among the European governments, public demonstrations and boycotts, fifty fire-bombed fifty Shell service stations, and a war of words in the European media. On the 20th of July 1995, Shell aborted its operation and towed the oil platform to the Norwegian Erfjord, where it was and is still moored and decaying. Up to the present, no clear answer has emerged as to whether an offshore or onshore solution is best. That the platform’s fate is still uncertain reveals the complexity of the issue and further, proves little about who (Shell or Greenpeace) is right or wrong.
The ‘Brent Spar’ crisis has long lasting consequences for the financial situation and the reputation of both parties. Greenpeace has spent a total of $1.4 million on their campaign in opposition to sinking the oil platform. Although Greenpeace was forced to apologize to Shell in September 1995 and admitted that ”their sampling on board of the ‘Brent Spar’ was flawed” (Shell press release 9.5.95), Greenpeace’s enhanced reputation, a result of the ‘Brent Spar’ crisis, remains unchanged. Shell’s position on ‘Brent Spar’ has led to long-term financial consequences as well as damage to their public reputation. Shell gas stations have experienced losses due to a ‘Brent Spar’ boycott (European Energy Report 3.29.95). Further, Shell pays $54,000.00 a month to ‘park’ its obsolete platform in the Norwegian fjord (Thompson 8.14.96). Shell has also spent enormous amounts of money in responding to the crisis, and public trust building, not to mention the new form of disposal.

2. Purpose of the Study
One question that arises when reflecting on the ‘Brent Spar’ crisis is how the newspapers’ communication created symbolic realities that motivated masses of people in different European countries to take sides for or against Greenpeace and a giant like the Shell oil corporation. My study provides an answer to this question by analyzing all press articles that appeared from April 30 to July 20, 1995 in two major German newspapers, ‘Die Frankfurter Allgemeine Zeitung’ (FAZ) and ‘Die Süddeutsche Zeitung’ (SZ), and in three major French newspaper, ‘Le Figaro’ (LF), ‘Le Monde’ (LM), and ‘La Libération’ (LB). Germany and France, which represent the core power group of the European Union, border the Northsea. Furthermore, the two nations are the subjects of my study because they reflect different national reactions to the crisis. Ultimately, the text analysis explains the persuasive appeal of the press and provides an understanding of the development of the crisis.

3. Bormann’s Fantasy Theme Analysis
The text analysis of the press texts is based on Bormann’s fantasy theme analysis which he developed on the grounds of Bales’ (1970) small group communication research his own ‘Symbolic Convergence Communication Theory’. Bormann (1972) states: ”The explanatory power of the fantasy chain analysis lies in its ability to account for the development, evolution and decay of dramas that catch up groups and change their behavior” (399). I use Bormann’s notions of fantasy themes and rhetorical visions to look for themes in the press texts in order analyze how argumentative discourse operated in the crisis and to demonstrate how attention was drawn towards Shell’s actions in Europe. A fantasy theme is a ”dramatizing message or part of a message and includes characters (personae) in action within a given scene” (Bormann 1977: 130). The symbolic reality that can be constructed from an accumulation of fantasy themes over time forms composite dramas and chains out among a mass public. This reality is what constitutes a rhetorical vision (130). In the following analysis, I examine recurrent rhetorical patterns that led to the creation of fantasy themes and visions that were created during the ‘Brent Spar’ crisis in Germany and France.

4. ”David against Goliath”: Fantasy Themes in Germany
Recurrent communicative patterns in the German press included the choice of words in the press coverage, the use of quotations, and the structure of the texts. They helped to establish narratives in which ‘dramatis personae’ were created and situated in a dramatic war-like scenario. Fantasy themes were created in the German press that depicted Shell as the villain, as the insensitive, capitalist giant whose only interest was profit. Greenpeace was characterized as the hero, the small non-profit organization that was concerned with the well-being of the environment and thus also with the well-being of humanity. The German press formed a rhetorical vision of a ‘green war’ referred to as the ‘Brent Spar’.
David against Goliath was an apt metaphor for the rhetorical vision surrounding the confrontation between Greenpeace and Shell. The German press used words with a positive connotation and expressions to describe Greenpeace. The organization was referred to as ”environmental protectors”, (e.g. SZ 5.23.95: 12; FAZ 6.9.95: 6), an ”environmental protectionist organization” (e.g. SZ 6.16.95: 7; FAZ 6.9.95: 1) or ”activists” (e.g. SZ 5.24./25.95, 6.8.95: 12; FAZ 6.12.95: 27). These positive names characterized Greenpeace as an organization that pursues altruistic goals, such as the protection of nature. The fact that the organization was represented by its members, ”the protectors” and ”the activists”, aroused sympathy and allegiance by making the organization more human and tangible, easy for the readers to identify with. Greenpeace was depicted as the hero.
In contrast, Shell was depicted as a villain. Shell’s image suffered because the corporation was depicted as a group of greedy capitalists. The ‘Frankfurter Allgemeine Zeitung’ labeled Shell a ”cool calculating corporation” (6.19.95: 20) and the ‘Süddeutsche Zeitung’ reported that ”Shell is saving money…” (6.19.95: 3). Another article criticized Shell and the British government for placing cost over environmental concerns and noted that ”the ecological consequences of the disposal did not play a role in the decision” (FAZ 6.21.95: N1). The article also reproached Shell with ”a form of economizing which buys short term savings of expenses with long term risks that are not calculable and expensive to pay for”.
An author of an article of the ‘Frankfurter Allgemeine Zeitung’ remarked: A lot of people mistrust this global corporation merely because of its size. They associate the corporation with political and economic power, and further with behavior that does not regard the so called little man, the average person (6.20.95: 16).
The German press portrayed the oil corporation as only interested in containing costs. Shell was characterized as a greedy, capitalist-mongering entity, and a selfish villain. The press aroused fear that Shell would harm nature, and, because Germans link their well-being as humans to the well-being of nature, the fear touched their very own existence.
Apart from the more obvious choice of words, the press also employed text structure and quotations as the subtle rhetorical devices which supported the construction of the hero and the villain, thus generating a dramatic scenario. The articles extensively affirmed Greenpeace’s dramatic description of the battle on the water and mostly quoted Greenpeace members at the beginning of the text; Shell’s point of view was only briefly cited near the end. In general, the structure of press articles is based on a hierarchy of relevance (Van Dijk 1988: 41): The title mirrors the most important information of the text, followed by the subtitle, the lead, the beginning of an article, etc. The further the article proceeds, the more specific the information becomes and thus less important to the everyday reader. Newspaper readers usually pay the most attention to the beginning of articles and often do not continue reading to the end (Van Dijk 1988: 142).
Almost every single article in the German newspapers placed dramatic messages from Greenpeace in top positions. Titles of articles fostered a good impression of Greenpeace, and portrayed an evil Shell. These are some of the titles: ”Greenpeace activists rammed on the Northsea” (SZ 6.12.95: 6), ”Despite international criticism: ‘Brent Spar’ on its way to the sinking spot” (SZ 6.13.95: 6), ”Christian Democratic Party furious at Shell because of oil platform” (SZ 6.13.95: 5), ”Garbage, Shell, and the sea” (FAZ 6.14.95: 17), ”Protest wave due to the sinking of the oil platform” (SZ 6.14./15.95: 1), ”The Shell boycott shows effects” (FAZ 6.16.95: 1), ”Contradictory statements from Shell” (SZ 6.17./18.: 6), ”The garbage cannot be sunk in the sea: A study of British scientists/Poisoned mud inside the platform” (FAZ 6.21.95: 3). Such powerful assertions, placed on the top of the articles, aroused strong emotional reactions for the environmental organization and against the oil corporation. Clusters of meanings unified in the media’s war scenario and created a rhetorical community with a rhetorical vision of a green war named ‘Brent Spar’.
The platform ‘Brent Spar’ became a symbol of the Shell corporation and the danger that was connected with it. The name ‘Brent Spar’ was made the keyword of the crisis. Anger over and fear of Shell’s actions were aggravated by the press reports which made the oil platform a symbol of the threat posed by Shell. The newspapers’ emphasis on the platform’s hazardous contents, its immense size, and its heavy weight all contributed to its symbolic status. In almost every article, the content of the rig was mentioned. For instance: ”According to Greenpeace, there are at least 100 tons of poisoned mud, such as arson, cadmium, lead and slightly radioactive waste” (FAZ 5.15.97: 3); or ”…’Brent Spar’ with 130 tons of poisoned waste on board” (SZ 6.17/18.95: 6). The mention of toxic waste aboard the oil rig scared the hyper-sensitized public.
There were constant allusions in the newspaper coverage to the rig’s size and weight: ”About hundred tons of poison would thus sink into the sea with the platform,” (FAZ 6.14.95: 17); or ”the whole station is 140 meters high, 32 meters are above the sea level; it was kept in position by chains and heavy anchor blocks,” (SZ 6.17/18.95: 4). The rig was described as a gigantic monster that could break free of its chains and destroy the Northsea and thus threaten human existence. In contrast to the rig’s dangerous waste and its massive size and weight, it was frightening for readers to discover that the ”outer jacket of the ‘Brent Spar’ is only two centimeters thick” (FAZ 6.21.95: 3). The German media’s representation of the oil platform signaled danger and inflexibility, characteristics that the press also attached to the oil corporation. For Germans, the oil platform took on the symbolic meaning of a monster, the ‘Brent Spar’, which also represented Shell, a destroyer of nature.
According to the press, the invasion of the Northsea had to be repelled and the sea had to be saved. Calls for action, such as ”the sea must not be misused as the garbage can of an oil corporation,” by the president of the Churches’ Week were accompanied by applause from 80,000 participants (FAZ 6.19.95: 2). These statements sounded like war chants which promoted the battle on the sea. ”The sea must not be misused as a garbage can” was stated by politicians and civilians as a war slogan and was frequently repeated by the press (FAZ 6.14.95: 17; 6.16.95: 6; 6.17.95: 1). As masses of people, both civilians and politicians, embraced the war fantasies, the drama escalated.
War analogies repeatedly appeared in the newspaper coverage: ”The battle against the sinking of the British oil platform ‘Brent Spar’ near the Scottish coast becomes more and more bitter,” (SZ 6.12.95: 6). Dramatic messages were reminiscent of war-time reports, for example:
Despite constant bombardment with water cannons, Greenpeace managed by helicopter to supply its two members, who landed on the platform on Friday, with food, clothes, and blankets (FAZ 6.19.95: 2).
Unequal battle: According to Greenpeace, an accompanying ship of the 65,000 ton oil platform ‘Brent Spar’ deliberately tried to spray one of the two occupants of the platform with a water cannon. The man did not fall overboard only because he got stuck in a barbed wire fence (FAZ 6.20.95: 3).

This sample of the press coverage illustrates how Greenpeace was symbolically ”humanized” because it was represented by the five demonstrators whereas Shell was ”dehumanized” because it was represented by a ship and the violence of a water cannon.
During the course of events, the German press labeled British members of the ‘Northsea Protection Conference’ ”outsiders,” (FAZ 6.9.95: 1) ”brake pads,” and ”the black sheep of the European Northsea Protection Conference” (6). Another articles stated that ”the British government, which deflected the massive protest with stoic composure, is also on the losing side” (SZ 6.22.95: 4). The derogatory remarks in the press clearly mirrored Germany’s disapproval of the British government’s support of the oil corporation.
The British public was referred to in a similarly derogatory manner by the German press: ”The fact that the British tolerate the pollution of the sea with great composure is not explicable by the difference in mentality,” (SZ 6.22.95: 4) and ”In particular the British, who, as inhabitants of an island, consider the sea as a way of transport and as a dustheap, receive minus points in their environmental performance” (FAZ 6.20.95: 3). According to the new meaning inhabiting the German newspapers’ rhetoric, the British government and the public became accomplices of the oil corporation.
Now Greenpeace and Germany were fighting together against the evil Shell and its British accomplices. Another brick was laid in the building of the scenario. Antipathy and anxiety towards Shell and its allies were aroused. The ‘Brent Spar’ vision became a symbolic reality and constructed a meaning for the ‘Brent Spar’ issue that neither Shell, nor any of the European governments had anticipated. The war-like scenario became so intense that individuals felt compelled to unify and take action. The early war chant ”the sea must not be misused as a garbage can,” became the aggressive slogan ”Shell to Hell” (FAZ 6.17.95: 2; SZ 6.17/18.95: 6).
The rhetorical vision of the green war committed people à la Robin Hood, so that even illegal means were justified in the battle for the good of environmental protection. Behavior such as occupying the platform, flying helicopters in illegal areas, exaggerating the amount of poison on board the rig, doing financial harm to Shell’s franchisers by boycotting their gas stations, attacking the owners of Shell gas stations all became justifiable, as did shooting at Shell gas stations. These were all illegal or unethical acts justified under the banner of ecological protection. The ‘green war’ reality produced a crooked logic. The evil, the violence and other illegal actions, were tolerated and even supported so that the preservation of the environment, would triumph. This demonstrated how the rhetorical vision of the ‘Brent Spar’ war created a new reality in which ethics and legality were reversed.

5. ”The Green Guerrilla against Shell”: FantasyThemes in France”
Contrary to the German newspapers, recurrent rhetorical devices in the French coverage of the ‘Brent Spar’ crisis, such as metaphors and similes, certain types of quotations, and the structure of the articles, helped to create fantasies about Shell as the victim of the villains, the green terrorists led by Greenpeace and backed by Germany.
According to the press coverage, France did not have an active role in the ‘Brent Spar’ drama but instead played a neutral part. Fantasy themes conveyed through the French caused anxiety that green issues could take over French policy-making and gain control over decisions in industry.
The title in ‘Le Figaro’ ”The green Guerrilla against Shell” (6.21.95: 12) reflects the fantasy theme that was created by the French press with respect to the battle between Greenpeace and Shell. Greenpeace was characterized as the leader of a ”green Guerrilla” troop that used physical force, radical means, and illegal action in order to interfere in Shell’s plans. In contrast, Shell was characterized as a corporation that simply tried to do its business, namely the sinking of their oil platform according to their best knowledge, but became the victim of Greenpeace’s zealous campaign. Greenpeace was depicted as an egotistic and radical villain that interfered in domestic British business and policy. The positively connoted term Greenpeace was rarely used in the French press coverage but instead was replaced with metaphors and similes. These metaphors and similes subtly portrayed Greenpeace as irrational, dangerous, radical, and terrorist, evoking antagonistic feelings.
According to Johnson (1987), new metaphors ”can give new meaning … to what we know and believe” (139). Lakoff and Johnson (1980) point out that a metaphor ”has an explanatory power of the only sort that makes sense to most people” (34). Metaphors have an illustrative and an affective function. Johnson (1987) further remarks that a ”metaphor can acquire the status of truth” (142) and illustrates ”the power of metaphor to create a reality” (144). Metaphors are very powerful rhetorical devices that contribute significantly to the creation of fantasy themes and rhetorical visions.
The following example of the French coverage of the ‘Brent Spar’ crisis is loaded with metaphorical expressions. The press declared that the environmental organization changed from ”crusades for baby seals” to one that took advantage of ”the unexpected opportunity to gild their escutcheon,” at a point in time when Greenpeace was ”confronted with difficult structural and financial problems” (LF 6.21.95: 2). Herewith, the French press suggested that Greenpeace, a non-profit organization, became capitalist and economically competitive. The assertions in the newspapers implied that Greenpeace used the ‘Brent Spar’ issue not for the purpose of fighting for environmental protection but rather to brush up its reputation and to motivate monetary donations. The French press presented an organization that, in protest against the sinking of the ‘Brent Spar’, did not pursue the altruistic goal to save nature like it used to, but instead was selfishly interested in its own success.
The metaphoric label ”muscular ecology” (LF 6.21.95: 2) was a title in reference to Greenpeace to ridicule the organization. The metaphor depicted Greenpeace as foolish and irrational because it used physical strength to present a show and attract attention. However, the metaphorical term also produced anxiety because it implied that Greenpeace actually was strong, powerful, and misguided.
Further, the French press observed that the ecologists had changed and their control had become stronger: ”They gazed at each other as their hair was growing longer in the same time the wool of the lambs from Larzac [a remote French village] was growing. Forget this, they cut their hair short, sometimes under the force of order” (LF 6.21.95: 2). This was a reference to cutting your hair as being ”gung-ho military.” Although the comparison of the ecologists’ hair to the ”wool of the lambs” drew an odd picture, the statement clearly illustrated that the ecologists had become more active and strictly organized, almost like a military unit. The French press implied that the ecologists had to be taken more seriously than before, that they had gained control, and that they might be dangerous in the future.
This impression was fortified when the press accused Greenpeace of ”triggering the revolt” (LB 6.19.95: 26) and members of Greenpeace were called ”militant ecologists,” (LB 6.15.95: 20; 6.18.95: 18) ”militants,” (LB 6.21.95: 5; LF 6.21.95: 12), and ”two militants, ‘green berets’ of a new kind…” (LF 6.21.95: 5). These terms for Greenpeace, emphasized the organization’s new radicalization. As mentioned above, the environmental organization was also equated with a ”green Guerrilla,” (LF 6.21.95: 12) which alluded to both unconventional warfare, such as engaging the enemy behind its own lines and to highly motivated revolutionaries who are willing to die for their cause. The picture of a ”green Guerrilla” encouraged to fantasize about a violent Greenpeace which would strive for victory by any means.293 Furthermore, one editorial mentioned that ”it is, without any doubt, too excessive to talk about ecological terrorism, when wilder activists act in countries like Algeria” (LF 6.21.95: 5). Although the metaphorical term ”ecological terrorism” was considered an inappropriately extreme label for this situation, it was nevertheless still used, which meant that the allusion to terrorism was embedded into the mainstream consciousness.
In comparison to the slanderous representation of Greenpeace as the villain, Shell was depicted in a neutral way, as ”the oil group Shell,” (LM 6.10.95: 2) ”oil people,” (LF 6.21.95: 1) ”Shell,” (LB 6.18.95: 18; 6.21.95: 6; 6.22.95: 21) ”the oil corporation Shell,” (LM 6.16.95: 1; LB 6.21.95: 1) and ”the firm” (LM 6.21.95: 25). The French press gave a picture of Shell that detached the oil corporation from the whole scenario on the Northsea. The non-accusatory description of Shell fit well with the media’s depiction of Shell as the victim.
In the French coverage of the ‘Brent Spar’ crisis, Shell was characterized as a rational and responsible corporation that became the victim of Greenpeace’s extreme reaction. The titles, ”Shell whom no one likes” (LM 6.20.95: 16) and ”It is Shell whom no one likes anymore” (LB 6.21.95: 1) implied pity for Shell. The press portrayed Shell as the whipping boy. In addition, the passive voice in the title ”It is Shell whom…,” implied that Shell was a victim.
In the media’s drama, the protagonist was forced to defend itself from the antagonist’s attacks. War metaphors and the reports of war-like situations, always with Greenpeace as the main antagonist, dramatized the scenario. For example: ”its [Shell’s] project… triggered an anti-Shell front,” (LB 6.18.95: 18) ”the platform was conquered by a helicopter of the Greenpeace organization that successfully brought two militants to the platform,” (LM 6.18./19.95: 3) ”ecologist extremist commandos,” (LM 6.20.95: 16) ”the iron arm that the ecologists aimed at Shell…,” (LF 21.6.95: 1) ”the muscled action is part of a deterrent arsenal of the tough wing of the ‘Greens,’” (LF, 6.21.95, p. 2) ”due to the impressive wall of shields, Shell gave up the sinking,” (LF 6.21.95: 12) and ”four more activists succeeded in taking over the platform by helicopter despite the efforts of Shell’s protection ships” (LM 6.22.95: 2). The French press coverage focused on Greenpeace’s occupation of the platform. The use of war terminology and imagery reinforced the fantasy of the green villain who initiated the conflict.
Slowly, the war fantasy chained out. By declaring that ”Greenpeace is on its war foot,” (6.21.95: 12) ‘Le Figaro’ conveyed the idea that it was Greenpeace that declared war. This statement implied that Greenpeace started the war. ‘Le Figaro’ continued: ”On Monday, the association sent the Solo, its fleet’s most powerful ship, and dared to oppose the sinking” (6.21.95: 12). This narrative sounded like a war report that vividly described Greenpeace’s attack and aroused tension and anxiety. In contrast to the detailed description of Greenpeace’s attack, once again, Shell’s response was not mentioned. The war scenario aroused hostility towards the villain and parlayed pity for the victim.
During the war, the French press also constructed fantasy themes of Germans as being ”fanatically ecologically correct” (LM 7.2./3.95: 1). The French attitude towards the Germans during the ‘Brent Spar’ crisis was further influenced by phrases in the press such as: ” ‘Stop this madness,’ screamed the General Secretary of the Christian Socialist Union” (LM 6.16.95: 1). The idea of the stern General Secretary of the CSU ”screaming” to stop the sinking was ridiculous. The reaction of Germany’s politicians was presented by the French press as hysterical, emotion clearly ruling over rationality. This method of reporting led to French antipathy towards Germany.
The strong disapproval of Germany’s reaction was further reflected in remarks such as ”It is a sign of these times that the oil corporation Royal Dutch Shell’s project to sink the oil rig ‘Brent Spar’, that had come to the end of 30 years of good and loyal service in the North Atlantic, aroused a big fuss in Europe, and particularly in Germany” (LM 6.20.95: 16). The personification of the oil rig created the illusion that the ‘Brent Spar’ needed to be treated like a loyal employee that had done his/her service for the public and now deserved honorable retirement. The French press accused Germany of unnecessary intervention into the affair of Shell’s oil rig.
The press continually articulated its belief that the disposal of the ‘Brent Spar’ was not Greenpeace’s or Germany’s business but rather a British domestic issue. The press wrote that Germany’s mass protests were extraneous since ”this collective phenomena is even more surprising as the German coasts are absolutely not menaced by a possible black sea” (LM 6.16.95: 1). This attitude that a country should only interfere in another country’s decisions when that country is directly endangered was clearly espoused in the French press. The quoted statement also implied that France was wary of mass protests against French policy, for instance their nuclear testing.
One ‘Figaro’ article, typical of the French press coverage, quoted Shell’s president who explained that Greenpeace’s estimation of the amount of toxic waste on board the oil platform was ”exaggerated, irresponsible, and alarming” (6.21.95: 12), thereby reinforced the fantasy theme of an extremist Germany that interfered with an innocent Shell’s plans. The article further printed the president’s detailed explanation of the exact content of the oil rig which included the following imagery: ”The very weak rate of radioactivity, which is naturally formed in the inside of the platform, is not higher than the rate that emanates from a couple of houses built on Aberdeen’s granite”. With this vivid comparison, the president explained that the oil rig’s amount of toxic waste was harmless. He further claimed that the sinking option ”is what is best for the oil industry of today.” The quotation from Shell’s president was followed by a lengthy description of the emotional uproar and bombing attacks in Germany (LF 6.21.95: 12). Germany became a companion villain with Greenpeace in the ‘Brent Spar’ crisis.
The whole scenario was dramatized when the German environmental movement was placed in an aggressive, humorous light. The ‘Libération’ used ridicule exaggerations to the green movement, writing that ”in Germany, a sport sailor who sails on the North Sea sees himself getting a ticket if he throws nothing more than a tissue over board” (6.15.95: 20). This imagery of polluters as law offenders presented the Germans as uptight and rigid. The antipathy was aggravated when the press explained that ”nothing provokes as much indignation in Germany as contempt of the environment. Polluters are considered criminals, and their carelessness is considered supreme contempt of your neighbor” (LM 6.16.95: 1). These two press statements exaggerated their claims by suggesting that polluters are treated like criminals or even murderers in Germany. This encouraged the idea of Germany that overreacts and French dislike of Germany.
Illustrations of Germany’s attitude toward the sinking of the oil rig and in-depth description of the protests of various German groups furthered the dramatization. The press vividly described the situation in Germany: ”Deserted gas stations, angry franchisers and a ruined image: the project of the British group Shell… ignited a very spectacular boycott movement in Germany. …a gas station in the region of Frankfurt was shot at six times by a driver, without the incident hurting anyone.” (LB 6.15.95: 20). The dramatic messages about the situation in Germany inspired the readers to fantasize about the radical, terrorist-like Germans fighting for the environment. The antipathy that was initially aroused turned into hostility as Germany became Greenpeace’s accomplice and a danger to France.
Negative feelings in France were fortified by constant details of the events in Germany (e.g., LB 6.15.95: 20; 6.18.95: 18; 6.19.95: 26; 6.21.95: 6; LF 6.21.95: 12). A typical description that French readers were exposed to looked like this: The protests against Shell’s plans have been particularly lively in Germany, where from the churches to the unions, from Chancellor Kohl to the east German ice skater Katarina Witt, from the social-democratic party to the popular tabloid Bild, everyone raised in opposition against the project of sinking the ‘Brent Spar’ (LB 6.21.95: 6).
The long description with its parallel form ”from… to…” exemplified the German situation and dramatized it by emphasizing how strong and unified the protest was in Germany. The dramatic messages portrayed the Germans as fanatic in their protest caused by an emotional uproar. The fantasy theme of Germans who transformed into radicals aroused the anxiety that France, with its plans for nuclear tests in the Murorora Atoll, would become the next target.
The French coverage of Germany’s reactions to the ‘Brent Spar’ crisis took on a general anti-German attitude in environmental matters. Many articles dealt with the protests in Germany rather than with the reactions in France or with the ‘Brent Spar’ issue itself. Articles were titled ”Shell boycotted in Germany,” (LB 6.15.95: 20) ”Shell’s anti-ecological move scandalizes Germany,” (LM 6.16.95: 1) ”In Germany, the boycott keeled Shell over,” (LB 6.19.95: 26) and ”In Germany, Robin Hood effect” (LB 6.21.95: 6). Although the protests in the Netherlands were as passionate as those in Germany and Dutch bombed gas stations, the French press focused exclusively on Germany, conveying an anti-German attitude to the readers.[i] 94
Moreover, the Germans were reproached: ”there is some hypocrisy on the part of the Germans to make themselves the moral censors of the behavior of a multinational oil corporation from which they consumed products with an indifferent greediness” (LM 6.20.95: 16). This form of criticism fed the new reality that depicted Germany as a second villain in the ‘Brent Spar’ war. Finally, the war came to an end. Metaphors depicting a downward direction were used to emphasize Shell’s defeat. Lakoff and Johnson point out the existence of ”orientational metaphors,” (14) in which spatial orientations up and down correspond with happy/positive and sad/negative (15). They also explain that ”Having control or force is up; being subject to control or force is down” (15).
The press in France reported that the war was over because ”the ecologists made the oil people fold” (LF 6.21.95: 1). In French, to ”fold” literally means to fold something in half, like a piece of paper. The oil corporation could no longer resist Greenpeace’s and Germany’s attack and consequently ”put down their arms” (LF 6.21.95: 12). The war resulted in the ”capitulation  of one of the largest oil corporations to the ecologists,” (LM 6.22.95: 2) and was a ”triumph for Greenpeace” (LM 7.2./3.95: 1) and Germany.
To sum up, a rhetorical vision of ‘ecological fanaticism’ was built by the accumulation of fantasy themes that characterized Greenpeace as a ”dreadful watchdog” and a militant policeman of the ”good world market.” The fantasy themes also portrayed Germans as fanatic green ”moral censors” (LM 7.2./3.95: 1) with extreme ecological demands. The French press implied that Shell was the victim, and next time the victim could be France. The rhetorical vision aroused fear that in the future, France might be targeted and treated like a criminal by the ”watchdogs” of the environment. Imaginary headlines reading ”France accused of eco-negligence” and images of hysterical Germans floated into French minds. The rhetorical vision of ecological fanaticism evoked anxiety.

6. Conclusion and Future Implications
This study illustrated how the media’s argumentative discourse created fantasy themes and rhetorical visions based on the symbolic potential of environmental issues in the 20th century. The analysis of German and French newspaper articles illustrated that the press used fantasy themes and rhetorical visions, which impacted the development of the ‘Brent Spar’ crisis. In Germany, the fantasy themes involved simple images which depicted Greenpeace and Germany as the hero(ines) of nature and guardians of human existence while, in sharp contrast, Shell and Great Britain were depicted as the greedy, environmentally hostile villains. The German press interrelated the fantasy themes to form a rhetorical vision of a green war which was given the name of the obsolete oil rig ‘Brent Spar’. The ‘Brent Spar’ issue was assigned a new meaning.
In comparison to the German press, the French national press constructed fantasy themes concerning the ‘Brent Spar’ crisis in direct opposition to Germany. For French readers, Greenpeace was depicted as a war-engaging, militant ”guerrilla” organization, while Germany was characterized as a fanatic bully for green issues. Both villains were accused of meddling in another sovereign nation’s domestic affairs. Furthermore, the French press propelled Frenchmen to consider Shell a victim. The fantasies gave rise to the rhetorical vision of ecological fanaticism of Greenpeace and Germany. The French press conveyed its disregard for the German response to the ‘Brent Spar’ crisis and an anti-Greenpeace and anti-German attitude was proliferated by the French press.
This study exposed the details in which the ‘Brent Spar’ issue took on a bizarre development whose outcome – the renouncing of the offshore disposal – is still in doubt. It is still uncertain whether the offshore or onshore solution will prove be more environmentally friendly and feasible. The Shell corporation and the British government obviously underestimated Greenpeace and the public’s position on the oil platform’s disposal. The creation of various fantasy themes (partly based on previously existing clichés), the internationalization of the ‘Brent Spar’ issue, and the public’s drive for participatory democracy went far beyond the consequences that were anticipated by Shell and Great Britain. The strong opposition in Germany against the sinking of the oil rig caused an oppositional reaction in the French press’ coverage that resulted in a common consciousness that violated the post-war friendship between Germany and France and the German-French axis of the European Union (EU).
Although the background information was abundant, the data rich and valuable, and the analysis in-depth, I do not claim that the study was exhaustive. Data from the Netherlands, Belgium, Denmark, Sweden, Iceland, and Norway, countries that also dealt with the crisis, were omitted due to the restrictions of a Master’s Thesis. Further, television coverage, which also plays an integral role in the creation of fantasy themes, was not included in the analysis. Overall, this study has significant implications for future research.
It revealed the effectiveness of Bormann’s method in improving our understanding of peoples’ thoughts, emotions, and motivations. Further, the study showed that the concepts of fantasy themes and rhetorical visions are universal and that the method is applicable across cultural and language boundaries. Similar analyses of crises would bring about significant insight into the their nature and could help to improve crisis communication and management. Future studies of rhetorical discourse should be generated to explore phenomena such as racism and sexism and thus raise our awareness and knowledge of the power of rhetoric and the construction of symbolic realities. Moreover, Bormann’s fantasy theme analysis, in combination with cultural studies should be applied to current written or oral accounts of other incidents: Researchers could study events such as the mass suicide of members of Marshall Applewhite’s Heaven’s Gate sect in California, separatist wars such as in the former Yugoslavia and Chechenya, the rebel war in former Zaire, or the violent historical development of relations between Palestinians and Israelis. These analyses would provide a better understanding of international crises and, in the best case, would lead to an improvement of peace processes.

NOTES
i. The newspaper’s focus on German protests could be related to a historical antipathy between France and Germany that caused several wars and can still be observed today in the permanent political and economic competition.

REFERENCES
Bales, R. F. (1970). Personality and interpersonal behavior. New York, NY Holt, Rinehart, and Winston.
Bormann, E. G. (1972). Fantasy and rhetorical vision: The rhetorical criticism of social reality. The Quarterly Journal of Speech, 58(4), 396-407.
Bormann, E. G. (1977). Fetching the good out of evil: A rhetorical use of calamity. Quarterly Journal of Speech, 63, 130-139.
Bormann, E. G. (1982). Colloquy: I. Fantasy and rhetorical vision: Ten years later. Quarterly Journal of Speech, 68, 288-305.
European Energy Report (3.29.96). Shell shrugs off Brent Spar, sect. energy. Lexis Nexis.
Frankfurter Allgemeine Zeitung (5.15.95). Greenpeace setzt Protest auf Ölplattform fort, 3.
Frankfurter Allgemeine Zeitung (6.9.95). Nordseeschutz-Konferenz in Esbjerg begonnen, 1.
Frankfurter Allgemeine Zeitung (6.9.95). Vierte Internationale Nordseeschutzkonferenz eröffnet: Großbrittannien wieder in der Außenseiterrolle, 6.
Frankfurter Allgemeine Zeitung (6.12.95). Greenpeace: Mal konfrontativ, mal kooperativ: Spendenrekord in Deutschland/ International Hauptgeldgeber, 27.
Frankfurter Allgemeine Zeitung (6.14.95). Müll, Shell und das Meer, 17.
Frankfurter Allgemeine Zeitung (6.16.95). Der Shell-Boykott zeigt Wirkung, 1.
Frankfurter Allgemeine Zeitung (6.16.95). Zunehmender Boykott von Shell: Außenminister Kinkel gegen Versenkung der Ölplattform im Meer, 6.
Frankfurter Allgemeine Zeitung (6.17.95). Schlechte Argumente, 1.
Frankfurter Allgemeine Zeitung (6.17.95). Weiterer Anschlag auf Shell-Tankstelle, 2.
Frankfurter Allgemeine Zeitung (6.19.95). Shell prüft Entschädigung, 2.
Frankfurter Allgemeine Zeitung (6.19.95). Shell-Chef Herkströter auf Tauchstation, 20.
Frankfurter Allgemeine Zeitung (6.20.95). Der Müll, das Meer und die Moral, 3.
Frankfurter Allgemeine Zeitung (6.20.95). Klimaschutz und Umweltpolitik in Grossbritannien: Beim Wettlauf der guten Vorsätze sehen sich die Briten in bester Position: Die Ideologie ist der Technik voraus/Ein Minister für Kommunalverwaltung und Umwelt, 3.
Frankfurter Allgemeine Zeitung (6.20.95). In der Nordsee steht der Ruf der Shell auf dem Spiel: Eines der größten und rentabelsten Unternehmen der Welt/Große Selbständigkeit der Konzerngesellschaften, 16.
Frankfurter Allgemeine Zeitung (6.21.95). ”Der Abfall kann nicht auf See versenkt werden”: Ein Gutachten britischer Wissenschaftler/Giftiger Schlamm in der Plattform, 3.
Frankfurter Allgemeine Zeitung (6.21.95). ”Brent Spar”-nicht des Meeres Tod: Sicherheit und Kosten bestimmen über die Entsortgung/Ölplattformen im Betrieb schädlich, N1.
Johnson, M. (1987). The body in the mind: The bodily basis of meaning, imagination,and reason. Chicago, IL: The University of Chicago Press.
Lakoff, G., & Johnson, M. (1980). Metaphors we live by. Chicago, IL: The University of Chicago Press.
Le Figaro (6.21.95). Les écologistes font plier les pétroliers: Shell abandonne son projet de saborder une plate-forme en mer, 1.
Le Figaro (6.21.95). Écologie musclée, 2.
Le Figaro (6.21.95). La compagnie renonce à immerger une plate-forme pétrolière: Guérilla verte contre Shell: Les organisations écologistes, au premier rang desquelles Greenpeace, appelaient à un boycottage des stations de la marque a la coquille, qui a baissé les armes, 12.
Le Monde (6.10.95). Conflit sur les plates-formes pétrolières désaffectées, 2.
Le Monde (6.16.95). Le dérapage anti-écologique de Shell scandalise l’Allemagne, 1.
Le Monde (6.18./19.95). Europe: Brent Spar, 3.
Le Monde (6.20.95). Shell qu’on n’aime pas, 16.
Le Monde (6.21.95). Le projet de sabordage d’une plate-forme déclenche une tempête contre Shell: La societé pétrolière continue d’affronter les protestations de Greenpeace, 25.
Le Monde (6.22.95). Shell cède à la pression de l’opinion européenne: Soumise à une campagne de boycottage lancée par Greenpeace et relayée largement en Allemagne et dans les pays nordiques, la compagnie anglo-néerlandaise renonce à saborder sa plate-forme pétrolière Brent Spar dans l’Atlantique, 2.
Le Monde (7.2./3.95). Le triomphe de Greenpeace, 1.
Libération (6.15.95). Shell boycotté en Allemagne: Pour son projet de couler une plate-forme en mer du Nord, 20.
Libération (6.18.95). Shell insiste pour démonter sa plate-forme, 18.
Libération (6.19.95). En Allemagne, le boycott fait chavirer la Shell: Le projet d’immersion d’une plate-forme au large de l’Ecosse coûte cher à la compagnie, 26.
Libération (6.21.95). La compagnie pétrolière cède au boycott: Mer du Nord: C’est Shell qu’on n’aime plus, 1.
Libération (6.21.95). Le projet d’immersion de la plate-forme pétrolière a été abandonné, 5.
Libération (6.21.95). En Allemagne, l’effet Robin des Bois, 6.
Libération (6.22.95). Shell: La mobilisation a payé: Bruxelles s’oppose à toute destruction de plate-forme en mer, 21.
Seaman, P. (1996). The media: Solutions to an international challenge (Presentation at the British Nuclear Industry Forum).
Shell Brent Spar calendar of events (1995). London: Shell-Mex House.
Shell press release (9.5.95). Greenpeace letter to Shell UK. London: Shell-Mex House.
Süddeutsche Zeitung (5.23.95). Räumung der Ölplattform vorerst gescheitert, 12.
Süddeutsche Zeitung (5.24/25.95). Polizei räumt besetzte Ölplattform, 12.
Süddeutsche Zeitung (6.8.95). Greenpeace-Aktivisten verlassen ”Brent Spar”, 12.
Süddeutsche Zeitung (6.12.95). Greenpeace-Aktivisten in der Nordsee gerammt, 6.
Süddeutsche Zeitung (6.13.95). CSU wegen Bohrinsel empört über Shell, 5.
Süddeutsche Zeitung (6.14/15.95). Protestwelle gegen Versenkung der Ölplattform, 1.
Süddeutsche Zeitung (6.16.95). Proteste gegen Shell weiten sich aus: Auch niederländische Politiker rufen zu Benzin-Boykott auf, 7.
Süddeutsche Zeitung (6.17/18.95). Shell am Pranger, 4.