For some time, coteries of philosophers, rhetoricians, social theorists, and various other students of public reasoning have thought and written about the possibility of resurrecting the presumably dead practice of rational public argumentation. They have sought, in the words of James Crosswhite (1996: 70), “not to expose [public arguments] for the wretched things they are, but to reveal the intrinsic hopes carried by the practice of argument.” They have pursued optimistic answers to questions that Michel Foucault (1993:18-19, qted. in Crosswaite 1996:13) asserts have been central to philosophy and critical thought since the eighteenth century: “What is this Reason that we use? What are its historical effects?
What are its limits, and what are its dangers? How can we exist as rational beings, fortunately committed to practicing a rationality that is unfortunately crisscrossed by intrinsic dangers?” More specifically, I have argued (1998) that most western general theories of argumentation have been grounded in understandings of specific relationships between knowers and the known:
* ideas as the contents of minds
* evidence as external to minds
* inference as grounded in both mental and linguistic operations, that is, as reflective of mental activity yet materialized in particular kinds of language use.
Those three epistemological assumptions were the foundations of the philosophy of science and then public argument theory that grew up in the nineteenth century (see Fuller 1993: esp. ch. 1), making argumentative discourse – a kind of logical talk – the link between the knower and the known, and hence the mechanism for reasoned decision making as it ought to occur in the worlds of all knowers from all eras of human existence.
Such assumptions have been under attack at least since Kant sought to collapse the Cartesian dichtomy between knowing and being and since Nietzsche declared the end of philosophy. Following World War I, there were concerted drives to save public, rational argumentation by eliminating fallacious reasoning (Lasswell 1928), by neutralizing and concretizing interpersonal talk[i], and by making the verifiability principle a weapon for distinguishing between sense and non-sense in all arenas of human affairs (e.g., Ayer 1936/1962).

The attacks on general theories of public argumentation took on radically different forms following World War II. If the post-World War I queries tended to emphasize cognitive deficiencies in rational thought, the post-World War II assaults featured social and political, which is to say contextualized, analyses of flawed rationality and discursivity. At the Alta conference last year (Gronbeck 1998), I reviewed five lines of interrogation of general theories:
1. the anti-foundational attack, which asserts that totalized thought systems of any kind have lost their power because human conflicts are localized and grounded in idiosyncratic cultural practices;
2.  the moves to segment knowledge and hence reasoning demographically, as Carol Gilligan (1982) did when demonstrating fundamental differences between male and female moral reasoning;
3. the idea that some kinds of knowledge are not accessible to everyone because of variability in life experiences especially by people differently raced, gendered, experienced, and acculturated;
4. arguments concerning all knowledge practices being ultimately political because differentiations in knowledge produce differentiations in power; and
5.  the (especially French) postmodern assault on the idea of evidence (from the Latin e-videre) as grounded in The Seen in an electronic era when The Seen is unremittingly manipulable and hence falsifiable.

I do not wish to review those arguments, but, rather, examine three sorts of responses that can be given to them, to test the attacks and, then, to see if it still is possible to posit a general theory of argumentation that is adaptable to both public and more technical (academic) sorts of argumentation. I will take us back to three presumably general theories of argumentation – those offered by Stephen Toulmin, Chaïm Perelman and the Belgian School, and Frans van Eemeren and the Amsterdam School – in order to examine these classic visions for possible counter-attacks to those who find no hope for a unitary conception of argumentation. I have selected these three because they represent not only tested visions but, more important, because when taken together they provide us with epistemic, rhetorical , and linguistic responses to assaults, which is to say, responses featuring the very dimensions of the Enlightenment theory that made a serviceable general theory of argumentation possible in the first place. After I’ve reviewed the three positions, I’ll offer a few observations about the adequacy of such thinking as counter-attacks to public argumentation’s current detractors.

1. Toulmin: Field-Variant Modes of Rationality
Stephen Toulmin’s 1958 publication of The Uses of Argument (1958/1964: 1-6) forced students of logic – and hence both technical and everyday argument – to confront explicitly the sources of power underlying inferential thinking: is logic to be understood apodeictically, as a series of inviolable laws controlling assertions about conceptual relationships, or psychologically, as a reflection of the most correct sort of cognitive activity, or sociologically, as a technology groups of people have stipulated as recipes for special kinds of communication? He then neatly sidestepped that overwhelming question by saying that it was too reductionist. We must not ask what logic is in an essentialist way, but, rather, what is the process by which human beings reason together, that is, “the procedures and categories by using which claims-in-general can be argued for and settled” (p. 7)? It was such a focus on process, Toulmin argued, that would allow him to bring together the apodeixis (or force-of-proof) with the episteme (or science of knowing) in a unitary conception of arguing. The process of arguing, to Toulmin, comprehends both mechanisms of logical force with conceptualizations of substantive arenas for decision making.
Such ideas led Toulmin away from thought about logical theory as such and toward a focus on logical or argumentative practice. And the heart of argumentative practice, he said, was the claim: that which is to be not simply asserted but a series of “grounds (backing, data, facts, evidence, considerations, features) on which the merits of the assertion are to depend” (p. 11). Logical theory or form cannot dictate how we argue; rather, what we are trying to argue dictates the sorts of forms required for sound arguments.

This is not to say, however, that the what-we-are-arguing makes argumentation wholly relative to situation or what Toulmin calls the field of dispute. Rather, he says, the question is, “What things about the form and merits of our arguments are field-invariant and what things are field-dependent” (15)?
To answer that question, he first examined the little words – modals and adverbs of qualification – that appear often in arguments: words such “can” and “cannot,” “may” and “may not,” “probably,” “unless,” and the like. Such words, which express possibilities and impossibilities, proprieties and improprieties, are central to the argumentative process. One of Toulmin’s primary claims was that the force of such words is more or less invariable; whenever someone says that something is “impossible,” the word connotes a strong denial of possibility. Yet, he continued, the criteria one uses to define impossibility varies from field to field; impossibility is measured one way in mechanics (e.g., “A human being cannot lift a ton singlehandedly”), another way in parenting (e.g., “You cannot go visit your friend today”), a third way in linguistics (e.g., “You cannot have a male sister”). It is, then, the search for criteria-of-assessment that marks a key variation in argumentative strategies from field to field.

His second principal approach to the matter of field-variance is captured in his so-called layout of arguments. While the central terms of that layout – claim, data, warrant, backing, qualifications, and reservations – have been taught in basic communication textbooks since the early 1960s[ii], Toulmin’s rationle for constructing the layout has received little attention. His move to the six-part layout of arguments is predicated on the assumption that jurisprudence rather than traditional logic should be our primary model for argumentation if only because
1. it is most awkward, even extraordinarily difficult, to add qualifications to the universal premises of the syllogism and
2. the great variety of inferences that are needed for multiple kinds of arguments simply cannot be accommodated by traditional deductive machinery. Better still, the six-term layout allows critics to examine far more dimensions of rational decision making than do syllogisms with their narrow grounds for assessing validity. And finally, what the layout can do ultimately is feature backings for warrants – the deep assumptions or rules that govern correct reasoning in various fields. Backings for theological arguments are radically different from backings for statistical or jurisprudential arguments, for example. Field variance is most clearly visible in backing for warrants.
Once again, the notion of backing for warrants, that is, for inferential processes, takes Toulmin beyond the analyticity of formal logical inferences and into inferential processes arising out of particular substances or fields of human operation. He asks that we abandon the “Principle of the Syllogism” (128ff.) because arguments usually should be tested in three ways: not only with the tautology test of traditional logic but also the verification test of its statements and self-evidence test of commonsensical relations between and among its statements (esp. 131). Often in argumentation, for that matter, “proper” warrants are not only used in traditional ways but sometimes even must be established; in some fields, that is, audiences must be taught new ways of inferring conclusions appropriate to the subject matter.
The bottom line in Stephen Toulmin’s The Uses of Argument is captured in a single exhortation: “the need for a rapprochement between logic and epistemology, which will become not two subjects but one only” (254), which for him involves not only philosophical inquiry but also “the reintroduction of historical, empirical, and even – in a sense – anthropological considerations into the subject which philosophers have prided themselves on purifying, more than all other branches of philosophy, of any but a priori arguments” (ibid.). I will comment on such a goal at the end of this paper.

2. The Belgian School: The New Rhetorical Model
Writing about the same time as Stephen Toulmin were the Belgians Chaïm Perelman and Lucie Olbrechts-Tyteca. They were struggling to delegitimate the traditional logicians’ incarceration of rationality deduction as tools for reasoning in the post-Freudian age, yet they wished to accomplish these tasks without destroying the rigor and control of those artis cognescendi. Thus they launched their attacks on logical formalism, mere facticity as compelling of human action, the belief-opinion gap, and psychologism, all the while avering that “we have no wish to limit the study of argumentation to one adapted to a public of ignoramuses” (1958/1969: 7).
More specifically, Perelman and Olbrechts-Tyteca announce that “The domain of argumentation is that of the credible, the plausible, and the probable” (1). By “the credible” they refer to that which flows from acceptable human sources to perceiving minds; by the plausible, that which conforms to the experience of audiences; and by the probable, that which is likely true. The domain of argumentation thus is depicted as that of human conceptions of the world rather than features of the external world itself.
Yet, they nonetheless warn against treating argumentation “as a branch of psychology” (9). That is, an audience’s adherence to knowledge claims is not to be understood as attitudinal alignment or valuative correspondence, in part because the study of such effects cannot explain how or why arguments work, in part because argumentative force varies contextually, i.e., from science to law to philosophy. Adherence is to be grounded, not in psychological surrender, but in understandings of language formations. Language formations govern Perelman and Olbrechts-Tyteca’s effort to sort argumentative techniques into classes: quasi-logical arguments rely on linguistically manifest reasoning mechanisms; reality-based arguments, on the correspondence between language use and the outside world; associative arguments, on cognitive mappings of that world via relational concepts reflecting organizational relationships between aspects of interior or exterior life; and dissociative arguments, representing the disjunctive or dialectical separations of ideas as they are articulated in language. Perelman and Olbrechts-Tyteca argue that thinking occurs through, and not just in, argumentative-linguistic structures. Hence, a mentalist concept such as “idea” is collapsed into or made visible within the languages used to make it capable of being shared with others. Notice the moves that they have made. In seeking to avoid both logical formalisms and psychological subjectivities and in preferring talk about the construction of rather than the rehearsal of facticity in arguments, Perelman and Olbrects-Tyteca are closing the gap that John Locke (1695/1975) built into relationships between the knower and the known as well as between language and ideas. As well, like Toulmin, they recognize that the lifeworld is divisible into such varied arenas as the scientific, the legal, and the philosophical, and that human beings inhabit varied psychological and sociological realms. Because audiences can be mixed, arguers must learn to employ “a multiplicity of arguments” (22), except when dealing with “an incompetent mob” (25).

Yet, not every discourse-of-influencing has equal value to all others. They stipulate the following: “We are going to apply the term persuasive to argumentation that only claims validity for a particular audience, and the term convincing to argumentation that presumes to gain the adherence of every rational being” (28). After making this distinction – a part of rhetorical theory in the form they are offering it at least since Richard Whately (1828/1963) – Perelman and Olbrechts-Tyteca use it to define three audiences for whom conviction based on argumentation is possible: their famed universal audience, an interlocuter or single hearer, and oneself (30, 31-45). They generally accept the views
1. that arguers deliberating with themselves are able to guarantee their own “value and sincerity” (41);
2. that arguers with opponents, insofar as they engage in discussion rather than self-interested debate, will meet the “duty of dialogue” that is our Platonic inheritance (56); and
3. that the universal audience is the ultimate totalization of reason or argumentative rationality, an article of faith in actual audiences’ abilities to elevate their rational and moral sensibilities upon occasion so as to demand the best from a disputant and their commitment to not let argumentation be destroyed by skeptics and fanatics (31-35, 62). All the distinctions that result from separating persuading and convincing, finally, are understood by Perelman and Olbrechts-Tyteca to be “unprecise and in practice must remain so” (29).
Here, then, is a general theory of argumentation focused on both the rhetoric- and language-using habits of peoples as well as their abilities to make reasoned judgments in particular circumstances. Inferences themselves are to be found in language use, though the actual achievement of audience depends upon rhetorically sound selections of particular arguments as laid out in the classification scheme. The bulk of Perelman and Olbrechts-Tyteca’s book is description of the various arguments with advice on when and how to use them strategically.

3. The Amsterdam School: The New Dialectical Model
Setting themselves against “The New Rhetoric” model of the Belgians are members of the University of Amsterdam’s program in Discourse and Argumentation Studies (DASA). I will work in this paper primarily from a single source, Frans van Eemeren’s keynote address given to the 1987 Conference on Argumentation at Alta, Utah, entitled “Argumentation Studies’ Five Estates” (1987) because its purview is the broadest I can find from a member of the Amsterdam School and because van Eemeren talked in the first-person plural “we” in the address, giving it the feeling of a collective manifesto.
His central task in this work is to distinguish the so-called “New Dialectic” from Perelman’s rhetorical projects. He identifies the general project as one grounded in “Normal Pragmatics” or “the required integration of normal idealizaion and empirical description perspectives on discourse” (p. 9). He pursues the distinction between the Belgian and the Amsterdam projects through five so-called estates, with each estate “a subject of research in its own right which is a necessary component of a full argumentation school” (10). The five estates include
1. the philosophical estate, where “the fundamentals” of “any scholarly dispute” are subject to “reflection” (11);
2. the theoretical estate, where ideal models of the argumentative process are constructed (13-14);
3. the reconstruction estate, wherein particular viewpoints toward those ideal models are fashioned (14-16);
4. the empirical estate, wherein particular argumentative practices in various arenas are examined (16-17); and
5. the practical estate, or the place where pedagogies for teaching well-grounded argumentative practice are constructed (17-20).

Working with a pair of concepts they loosed from Toulmin, the idea of anthropological and critical approaches to the study of argumentation, the Amsterdamers construct two versions of the five estates comprising the realm of argumentation theory: the Belgian school is depicted as developing from an “anthropological-relativist perspective on reasonableness,” and the Amsterdam school, from a “critical-rationalist perspective” (11, 12). The Belgians’ effort to ground their theory of argumentation in audience assent or adherence is sharply contrasted in van Eeemeren’s keynote address to the Amsterdamers’ desire to ground their theory, not in geometrically inspired logics, but “modern logic,” which they find flexible enough to “be made dialogical (which syllogistic logic was explicitly not intended to be)” (12, 13). More explicitly still, he proceeds to center their version of modern logic in speech acts wherein language uses are to be understood in their “contexts, terms and expressions” that are meaningful in “their social function, and at the same time [in] their specific meaning” (13).

Such a distinction, in turn, leads them to contrast the Belgians’ ultimate court of argumentative appeal, “agreement with the standards prevailing among the people in whose cultural community the argumentation takes place,” with theirs, “agreement with discussion rules which are instrumental in resolving a dispute and which are acceptable to the parties concerned” (ibid.). Such rules, of course, are to be found articulated in speech acts committed by particular disputants in particular situations. From there, the fundamental philosophical differences they see between themselves and followers of Perelman, they move through the other estates making similar contrasts that I can only suggest in a short paper:
1. Arguers working within the rhetorical school must understand particular audience’s “stock of knowledge about [its] systems of beliefs” vis-à-vis those working within the dialectical school, who work from “an ideal model of a critical discussion and a code of conduct for the performance of speech acts” (19).
2. So far as modeling particular arguments is concerned, the rhetorical school examines past performances to discover the habitual rhetorical patterns used in a society, while the dialectical school examines the dialectical tensions existing in disagreements, looking for reconstructions of discourses that will resolve the dispute.
3. Empirically, rhetorical disputants examine past disputes to see which sorts of rhetorical patterns worked with particular sorts of audiences, i.e., persuasiveness, while dialectical disputants try to understand language usage well enough to realistically assess what reflective interlocuters will demand in argumentative exchanges, i.e., convincingness.[iii]
4. Practically, rhetorically oriented arguers work from models of previous argumentative successes to see what has worked in a society, while dialectically oriented arguers are more reflective, studying “the dispute-resolving capacity of argumentation and the need for dialogue in order to be really convincing” (18).
‘
DASA’s attitudes toward the rhetorical school are laid out clearly under this heading: The discussion rules, however, do not provide a simple trick that merely has to be learned by heart to be applied successfully in practice…. Argumentation is not an abstract nor a mechanical process, but a verbal and social activity aimed at convincing another person of one’s points of view by systematically conquering his doubts (19).
The “discussion rules” and “a code of conduct” are the foundational commitments of the pragma-dialecticians of Amsterdam. Similar to the various so-called felicity conditions discussed by American students of speech acts, the ideas of discussion rules and codes of conduct construct a sociolinguistic basis for argumentation, that is, a series of socially sanctioned rules for interpersonal language use. The rules are not to be found in language per se, as in, for example, Toulmin’s examination of the force of certain modal verbs and qualifying adverbs. Rather, they are to be found in social agreements about how members of some collectivity wish to conduct their business. And thus, van Eemeren argues, “argumentation should be studied and taught as a specimen of normal communication and interaction between language users” (19).

4. The Rebirth of Public Argumentation?
It now is time to return to the post-World War II assaults upon general theories of especially public argumentation. Recall the Enlightenment’s projects that urge us to understand that external facts are used to validate the internal lifeworld, that truth is to be understood empirically but in terms of universal generalizations, that language is capable of being studied independent of the human ideas to which it supposedly gives expression and can ground generally valid forms of reasoning, and that the visual is the dominant sense by which human beings access the external world and hence truth. These tenets of modernism in general produced a theory of public argumentation whereby validity was assessed formally and truth was determined with empirically testable relationships between premises and conclusions. The post-World War II assault upon such a model of argumentation, in contrast, challenged the idea of formal validity, replacing it with an experientially based concept of reasoning, and the idea of a reality separable from our experience of the lifeworld, replacing the notion of so-called brute reality with an emphasis upon so-called social reality. Thus, the anti-modernist attacks came as a firestorm, grounding validity and truth in life experience – but life experience understood within dominating social categories, that is, life as constructed on the basis of gender, race, social economic status, age, disability, and any other category by which groups of a society’s citizens might classify themselves and find significance in their experiences. Can a general theory of public argumentation be saved after such a conflagaration?
Toulmin, the Belgians, and the DASA scholars share some tactics when attempting to defend a positive answer to that question. All three become situationists to one degree or another; Toulmin talks of field variance, Perelman and Olbrects-Tyteca (1958/1969: 19), of “the ensemble of those whom the speaker wishes to influence by his argumentation,” and van Eemeren, of contexts. To be sure, they see situations in quite different ways: Toulmin, as arenas-of-talk governed by expectations, Perelman, as places where particular sets of auditors reside, and van Eemeren, as social agreements wherein discussion rules and codes of conduct are specified.
Yet, all three recognize that the universalism that seemed to echo through so many discussions of rhetoric and logic in the Enlightenment must be sacrificed if a general theory of public argumentation is to stand.

But, one might ask, if a sense of universalism is gone, what’s “general” or generalizable about the resulting theory of argumentation? To this question we get three quite different answers from the three schools. Toulmin provides two answers: a general language of reasoning that suggests the rational force of certain kinds of discourses and a series of required types of discourse captured in his layout of argument (even if the content of those discourses vary from field to field). Thus, to Toulmin, thinking itself and the language of thought are generalizable within a collectivity; thus, a general theory of argumentation can be found in epistemic and linguistic principles. Perelman provides a patently rhetorical answer to the question, suggesting that general principles for testing argumentative reasoning lie within the idea of the universal audience. In fact what is constant in all argumentative practice is the need to create adherence in situated human beings; again, adherence is not simply a matter of psychological assent, but, rather, a matter of constructing discourses within the domain of the credible, the plausible, and the probable – but as those notions are understood by particular audiences. Perelman’s generalizable theory is tied, as he says by the end of The New Rhetoric (1958/1969: 513), to “the language of a community, be this a community bound by biological ties, or by the practice of a common discipline or technique.” A version of communitarianism is what Perelman offers as a base for a general theory of argumentation. To van Eemeren, such a ground moves too close to a kind of relativism, so the DASA scholars expand their understanding of the social to includes rules for both language use and social expectations. That is, they find adherence to sociolinguistic concepts of speech acts the true foundation for a general theory of dialogic conversational engagement of others. Pragma-dialectics or normative pragmatics is grounded a generalizable theory of “the institutions of social life” (1987: 20).
In eschewing universalism yet seeking a version of generalizability, therefore, all three schools reviewed in this paper return us to the fundamental requirements of any large-scale theory of human community: the epistemic-linguistic study one finds in Toulmin, Perelman’s emphasis upon shared, intersubjectively validated relationships between individuals, and van Eemeren’s dialogical-linguistic study of speech acts. In spite of their differences, all three schools reaffirm the centrality of epistemic, social, and linguistic dimensions to argumentation.
No one of the post-World War II schools of argumentation reviewed here, however, is adequate to the challenge of post-positivist, anti-totalizing, culturally radical, postmodernist, ideological thought. The Amsterdam school’s adherence to the five estates as realms or areas within which any argument theory must have commitments to be complete is innovative and potentially powerful, yet the relatively little time and space devoted to serious epistemological justification means that the assaults upon totalizing  concepts, masculinist understandings of reasoning, and apolitical conceptions of convincingness can stand. What is needed within the school is the strong attention to epistemology that came through in Toulmin’s early (1958/1964) and later (1972) work on epistemic communities, as well as Perelman and Olbrechts-Tyteca’s work on understanding the power of socially situated experience as conditioning of argumentation’s force.
Additionally, argumentation needs to be become sensitive to what is now called medium theory (e.g., Deibert 1977) to pursue the varied forms in which argumentas can be presented, not only linguistically but also acoustically and visually; only then will the postmodern assault on visuality and evidence be met head-on. Work on a general theory of public argumentation in the face of post-World War II attacks on it must continue, though it assuredly can build upon the strong bases provided by Toulmin, Perelman and Olbrechts-Tyteca, and the Amsterdam school, whose pioneering work deserves our continued attention.

NOTES
i. I am thinking here of the rise of the General Semantics movement, inspired by Alfred Korzybski’s Science and Sanity (1933) and I. A. Richards’ The Philosophy of Rhetoric (1936/1964, p. 7), where he defined rhetoric as “the study of misunderstanding and its remedies.” Training in the neutral, concrete, overwhelmingly descriptive use of language and reasoning spread across the U.S. in the form of General Semantics workshops, and scholarship proporting to validate that training appeared in two journals, General Semantics Bulletin and the more scholarly Etc.
ii. So far as I know, the first textbook to teach the Toulmin layout was a debate book, Douglas Ehninger and Wayne Brockriede, Decision by Debate (1963).
iii. Cf. Perelman and Olbrechts-Tyteca on the distinction between persuasion and conviction being suggested here. While van Eemeren does not evoke the universal audience in offering the distinction, his discussion of “reflection” comes close to suggesting it.

REFERENCES
Ayer, A. J. (1936). Language, Truth and Logic. London: V. Gollanz.
Crosswaite, J. (1996). The Rhetoric of Eeason: Writing and the Attractions of Argument. Madison: University of Wisconsin Press.
Deibert, R. J. (1997). Parchment, Printing, and Hypermedia: Communication in World Order Transformation. New York: Columbia University Press.
Eemeren, F. H. van (1987). Argumentation Studies’ Five Estates. In J. W. Wenzel (Ed.), Argument and Critical Practice: Proceedings of the Fifth SCA/AFA Conference on Argumentation, 9-24. Annandale, VA: SCA.
Ehninger, D., and Brockriede, W. (1963). Decision by Debate. New York: Dodd, Mead.
Foucault, M. (1982, March). Space, Knowledge, and Power. Skyline, 18-19.
Fuller, S. (1993). Philosophy, Rhetoric, & the End of Knowledge: The Coming of Science & Technology Studies. Madison: University of Wisconsin Press.
Gilligan, C. (1982). In a Different Voice: Psychological Development and Women’s Development. Cambridge, MA: Harvard University Press.
Gronbeck, B. E. (1998). The Crises in Scholarly Argument. In J. F. Klumpp (Ed.), Argument in a Time of Change: Definitions, Frameworks, and Critiques, 110-115. Proceedings of the tenth NCA/AFA conference on argumentation. Annandale, VA: National Communication Association.
Korzybski, A. (1933). Science and Sanity: An Introduction to Non-Aristotelian Systems and General Semantics. New York: Internatinal Non-Aristotelian Library Publications.
Lasswell, H. (1927). Propaganda Technique in the World War. New York: A. A. Knopf.
Locke, J. (1695/1975). An Essay Concerning Human Understanding. P. H. Nidditch (Ed.). Oxford, Eng.: Clarendon Press.
Perelman, C., and Olbrects-Tyteca, L. (1958/1969). The New Rhetoric: A Treatise on Argumentation. J. Wilkinson and P. Weaver (Trans.). Notre Dame, IN: University of Notre Dame.
Richards, I. A. (1936/1964). The Philosophy of Rhetoric. New York: Oxford University Press.
Toulmin, S. E. (1958/1964). The Uses of Argument. Cambridge, ENG: Cambridge UP.
Toulmin, S. E. (1972). Human Understanding. Princeton, kNJ: Princeton University Press.
Whately, R. (1828/1963). Elements of Rhetoric, Comprising an Analysis of the Laws of Moral Evidence and of Persuasion, with Rules for Argumentative Composition and Elocution. D. Ehninger (Ed.), D. Potter (Fore.). Carbondale: Southern Illinois University Press.

1. Introduction
The Vatican document ‘We remember: A reflection on the Shoah’, (issued on March 16, 1998) has led to many critical reactions throughout the world. The main reason for this is that it did not contain the generally expected apology to the Jewish people for the Roman Catholic Church’s complicity in the Holocaust but, instead, turned out to be an apologia in which the Church pleads not guilty. The apologia is based on a twofold distinction:
(1) between the Church as an institution and its individual members, and
(2) between anti-Semitism and anti-Judaism.

In this paper, I argue that these distinctions both constitute the fallacy of incorrect dissociation. The concept of dissociation was introduced by Perelman and Olbrechts-Tyteca. It is one of the two main principles of argumentation they discuss – the other one being association. In pragma-dialectical terms, dissociation aims at changing one of the the protagonist and the antagonist’s common starting points. If this is not done properly, the dissociation constitutes a violation of one of the rules for critical discussion. In this case, I contend that Rule 6 has been broken because the document presents the distinctions as self-evident and is therefore guilty of begging the question.
In Section 2, I describe the historical background of the document and sketch its outlines. In Section 3, I summarize the main reactions to it. In Section 4, I explain why the two distinctions made in the document can be analysed as dissociations in the Perelmanian sense. In Section 5, I argue that these dissociations violate Rule 6 of pragma-dialectics and constitute the fallacy of ‘innocence by dissociation’, being a special case of the fallacy of incorrect dissociation. Finally, in Section 6, I conclude that this fallacy is the terminological counterpart of the well-known fallacy of ‘guilt by association’.

2. Background and outline of ‘We remember: A reflection on the Shoah’
‘We Remember: A reflection on the Shoah’ is a 14-page document issued by the Vatican Commission for Religious Relations with the Jews on March 16, 1998.[i] It is a long-awaited document because it addresses the role of the Roman Catholic Church in the Holocaust during the Second World War. The document acknowledges that individual Catholics did things that were wrong or even sinful in their support of anti-Semitism and of Nazi persecution of Jews, and it repents for this – using the Hebrew word teshuvah. But it also absolves the Church as such from complicity in the Holocaust. It even warmly praises the controversial wartime Pope Pius XII (who has long been accused of remaining silent in the face of Nazi genocide and even of pro-German tendencies) for saving hundreds of thousands of Jewish lives ‘personally or through his representatives’.
The Vatican commission took up the task of creating this document at Pope John Paul II’s request eleven years ago, in 1987 – a year after the pope had a historic meeting with Rome Chief Rabbi Elio Toaff in Rome’s central synagogue. It is the third formal document prepared by the commission, following the landmark Nostra Aetate declaration of 1965, which marked the first official gesture of reconciliation by the Church to the Jews by repudiating the concept of Jewish guilt for Jesus’ death and by calling for mutual respect and dialogue between Catholics and Jews. By the way, it was not until 1965 that the Vatican eliminated the phrase ‘perfidious Jews’ from the liturgy of the Holy Week service.
The Vatican statement takes pains to distinguish anti-Judaism from anti-Semitism, suggesting that only the Nazis were guilty of anti-Semitism. It also stops far short of taking responsiblity as a religious institution from promulgating the tenets of anti-Judaism, in particular the teaching that the Jews killed Jesus. The widely accepted view is that this central Christian teaching provided the theological foundation for the anti-Semitism of the Nazi years that culminated in the murder of six million Jews by the Nazis. Instead, the Vatican’s document distances Christianity from the Holocaust. ‘The Shoah was the work of a thoroughly modern neo-pagan regime,’ it says.

3. Reactions to the Vatican document
It is an understatement to say that the document did not meet with general approval. Though Cardinal Edward Cassidy, the Head of the Vatican Commission, said that the Vatican’s statement amounts to an act of repentance as well as an apology, most of the reactions to the document clearly indicate that it does not live up to its expectations. The Vatican’s apology to the Jewish people still refuses, it is said, to accept full responsibility for the Catholic Church’s failure to take action to stop or slow the Holocaust. In this respect, many see the document as a step backwards compared to recent statements by Catholic Bishops in France, Germany and Poland, who admitted that the Church was at fault for its failure to react to Jewish persecution half a century ago. Oddly, they add, the Vatican document fails to do what the current pope, John Paul II, himself has done in less formal documents and speeches – that is, take direct responsibility for the Church’s failure to try to ameliorate the attempted genocide of the Jewish people.
Many representatives of Jewish groups voiced their disappointment about the document and declared that it ‘did not go far enough’. Some news agencies even claimed that the document has been ‘greeted with nearly universal dismay and anger by Jewish experts’. Perhaps this is a exaggeration but only a slight one because it cannot be denied that many expressed their dissatisfaction.[ii]
Among the dissatisfied critics were Rabbi Leon Klenicki, director of the Department of Interfaith Affairs of the Anti-Defamation League, who called the paper ‘a real insult’ and ‘a pretext for an apology for Pius XII,’ and Goldie Hershon, President of the Canadian Jewish Congress, who criticized the Vatican as follows: ‘It is inconsistent to admit the failures of ordinary Christians to speak out against the Holocaust, but to ignore the deafening silence of the Pope.’[iii]
Others were even more outspoken in their criticism. For example, Yitzhak Minervi, a former Israeli envoy to the Vatican, said: ‘All the responsibility is rolled onto the church’s flock […] while the church and its institutions emerges spotless.’ And Abraham Foxman, national director of the Anti-Defamation League in the United States said: ‘The document rings hollow. It is an apologia full of rationalization for Pope Pius XII and the Church. It takes very little moral and historical responsibility for the Church’s historic teaching for the contempt of Jews.’[iv]
Rabbi Mark Winer, a White Plains, N.Y. rabbi who is president of the National Council of Synagogues, finally, said that ‘the “remembrance” is incomplete, the “repentance” is lacking and the “resolve” for the future is pretty weak-kneed.’[v]

4. The distinctions in the Vatican document as dissociations
The first distinction in the Vatican document is that between ‘anti-Semitism’ and ‘anti-Judaism’:
[…] we cannot ignore the difference […] between anti-Semitism based on theories contrary to the constant teaching of the Church […] and the long-standing sentiments of mistrust and hostility that we call anti-Judaism, of which unfortunately, Christans also have been guilty.
Here, we see the first move towards the conclusion that the Roman Catholic Church is not guilty and never has been guilty of anti-Semitism. This point is repeated even more explicitly a little but further in the text:
The Shoah was the work of a thoroughly modern neo-pagan regime. Its anti-Semitism had its roots outside of Christianity […]’.

Anti-Judaism, on the other hand, does have Christian roots. According to the document, it can be traced back to ‘certain interpretations of the New Testament’, beit that these interpretations were totally mistaken:
In the Christian world […] erroneous and unjust interpretations of the New Testament regarding the Jewish people and their alleged culpability [for murdering Jesus Christ] have circulated for too long, engendering feelings of hostility towards this people.
The second distinction in the document is that between the Roman Catholic Church as an institution and its individual members. On the one hand, the document emphasizes that the Church, including its leader, Pope Pius XII, has done everything to resist and fight racism and Nazi anti-Semitism:
During and after the war, Jewish communities and Jewish leaders expressed their thanks for all that had been done for them, including what Pope Pius XII did personally or through his representatives to save hundreds of thousands of Jewish lives.
On the other hand, the document suggests that in ‘some Christians minds’, ‘anti-Jewish prejudices’ were ‘imbedded’ which made them ‘less sensitive, or even indifferent to the persecution launched against the Jews by National Socialism’ and observes that:
[…] the spiritual resistance and concrete action of other Christians was not that which might have been expected from Christ’s followers.
The ‘call to penitence’ is, therefore, only directed to the individual members of the Roman Catholic Church, not to itself or to its leaders, because, again, they are not guilty. When the document refers to the Catholic Church’s desire ‘to express her deep sorrow’, it is not because of the things the Church did wrong, but ‘for the failures of her sons and daughters’. The ‘act of repentance’ (teshuva) is carried out only indirectly, ‘since,’ – according to the document – ‘as members of the Church, we are linked to the sins as well as the merits of all her children’.
To sum up: the Roman Catholic Church pleads ‘not guilty’ with respect to the horrors of the Holocaust, first by distinguishing between pagan anti-Semitism and Christian anti-Judaism, and second by distancing itself from its individual members.
This twofold distinction amounts to what Chaim Perelman and Lucie Olbrechts-Tyteca, in Chapter 4 of their landmark study The New Rhetoric. A treatise on argumentation, call a ‘dissociation’ (1969: 411-459).[vi] They contrast dissociation with association:
By processes of association we understand schemes which bring separate elements together and allow us to establish a unity among them […]. By processes of dissociation, we mean techniques of separation, which have the purpose of dissociating, separating, disuniting elements which are regarded as forming a whole or at least a unified group within some system of thought […]. (1969: 190).

Perelman and Olbrechts-Tyteca mention several examples of dissociated concepts: ‘scientific truth’ and ‘religious truth’, ‘reality’ and ‘appearance’, ‘verbal’ and ‘real’, et cetera. They also point out that paradoxical expressions such as ‘learned ignorance’, ‘happy misfortune’, ‘bitter joy’, ‘thinking the unthinkable’, and ‘expressing the unexpressible’ ‘always call for an effort at dissociation’. Another example of a paradoxical expression is: ‘I do not mind dying. But it grieves me to depart from life.’ Here, the dissociation is ‘the result of opposition between a word and what is ordinarily regarded as a synonym for it’ (1969: 443).
If association unifies elements which were previously regarded by the audience as separate and dissociation separates elements which were previously regarded by the audience as a unit, it will be clear that the twofold distinction in the Vatican document is, in fact, a double dissociation. The document introduces a division into a concept the audience previously regarded as constituting ‘a single entity’, ‘a natural unity’ or ‘an indivisible whole’: first there was only ‘anti-Semitism’ and ‘the Roman Catholic Church’, now there is ‘pagan anti-Semitism’ versus ‘Christian inspired anti-Judaism’ on the one hand and ‘the Church as an institution’ versus ‘the individual members of the Church’ on the other.
Although Perelman and Olbrechts-Tyteca clearly believe that dissociation, just like association, is a general principle for defining argumentation schemes, the only loyal supporters of this idea I know of are Warnick and Kline (1992: 10). But then, they admire Perelman and Olbrechts-Tyteca’s work so much that they seem to accept almost everything they say without question.[vii] There are several other authors, however, who express their doubts about dissociation as an argumentation scheme. Schellens, for instance, observes that ‘it is unclear which argumentation forms or schemes make use of dissociation’ and concludes that ‘the dichotomy between association and dissociation is impractible’ (translated from 1985: 59). Kienpointner restricts his overview of argumentation schemes to those based on association because he finds the dissociative schemes ‘less systematic’ (translated from 1991: 189). Garssen even claims that ‘dissociation is neither a specific type of argumentation nor an argumentation scheme.’ His reason for this is that ‘dissociation provides no specific way to connect a starting point with a thesis in such a way that acceptance of the latter is increased’ (translated from 1997: 72).
In the Vatican document, the twofold dissociation aims at changing the audience’s beliefs about the Roman Catholic Church’s role in the Holocaust. The initial dialectical situation the document encounters is the general opinion which holds the Church jointly responsible for the terrors of the Holocaust inspired by Nazi anti-Semitism: ‘The Roman Catholic Church is accessory to the Holocaust because it has done too little to resist it and has always endorsed or even promoted anti-Semitism.’ The result of the twofold dissociation desired by the Vatican is that after reading the document the audience will believe, first, that only some individual members of the Church have done things to be blamed for and, second, the Church has never adopted an anti-Semitic attitude.
In order to succeed in the endeavour of changing the audience’s starting points, the document must convincingly show that the two distinctions (Church as an institution versus members of the Church and anti-Semitism versus anti-Judaism) are justified. The burden of proof is a heavy one. To what extent has the attempt been succesful? In my opinion, the attempt has failed totally.

5. Incorrect dissociations as pragma-dialectical fallacies
The twofold dissociation in the Vatican document would have been succesful only if it would have proved convincingly that there is no connection whatsoever between anti-Semitism and anti-Judaism on the one hand and between the Church as an institution and its members on the other. Judging by the reactions to the document the intended proof was not convincingly at all.
First, though the document admits that anti-Judaism has Christian roots which is based on ‘erroneous and unjust interpretations of the New Testament’, it ignores the fact – as is rightly observed in some of the reactions to the document – that the official Roman Catholic doctrine taught the ‘sons and daughters’ of the Church for centuries that the Jews murdered Jesus Christ. After all, it was only in 1965 that this doctrine was renounced by the Second Vatican Council.
Moreover, the document denies every relation between Christian anti-Judaism and pagan anti-Semitism, as if the second was not at all inspired and legitimized by the first. In this respect, the document is a step backwards compared to other statements, for example, by Dutch bishops who declared already in 1955 that ‘the tradition of theological anti-Judaism has contributed to a climate in which the Shoah could take place.’
Second, though the document states that ‘the Catholic Church expresses her deep sorrow for the failures of her sons and daughters in every age’, it maintains a sharp distinction between the Church as an institution on the one hand and its individual members on the other – as if the latter are not supposed to do what their religious leaders tell them to do.

The document’s failure in convincingly making the twofold dissociation is clearly illustrated by Rabbi Mark Winer: ‘In ascribing sinfulness to individual Catholics, it sidesteps responsibility on the part of the church […]. It never says that Catholic teaching was central to the teaching of contempt about the Jewish people.’ Dr. Geoffrey Wigodor, one of the two Israeli representatives on the International Jewish Committee for Interreligious Consultations with Christians, is even more outspoken: ‘In the document, the line is that it is not the Church that was to blame, but individuals who fell short of the Christian ideal. This flies in the face of history, noting it was the Church fathers themselves who interpreted the New Testament in an anti-Jewish manner; it was the Church councils which ruled against the Jews; and it was the popes themselves who drove the Jews out of civilized life, locking them up in gettos.’[viii]
One may add, as an aside, that if it would really be true that the Roman Catholic Church as an institution has done nothing to be blamed for, one may wonder whether the ‘call for penitence’ is, in fact, not totally out of order. Repentance always comes too late, the proverb tells us. But what is repentance without guilt?
To come back to my original question whether the twofold dissociation is justified, it is now possible to analyse the incorrectness of the dissociation in terms of the pragma-dialectical rules for critical discussion (van Eemeren et al. 1996: 298-306). Since the Vatican document presents the distinctions as self-evident and ignores the obvious relations between the two pairs of dissociated elements, the document’s arguments violate Rule 6 of pragma-dialectics: ‘A party may not falsely present a premise as an accepted starting point […]’ (van Eemeren et al. 1996: 284). The protagonist who violates this rule in this way (here: the authors of the Vatican document) is guilty of begging the question (van Eemeren et al. 1996: 305). One cannot resolve a dispute succesfully by presenting a dissociation as if it were already accepted by the antagonist (here: the readers of the Vatican document). This special case of begging the question may be christened (no pun intended) the fallacy of incorrect dissociation.

6. Conclusion
Only by committing the fallacy of incorrect dissociation, the Vatican document is able to maintain the Roman Catholic Church’s claim to guiltlessness of the Holocaust – a claim to ‘innocence by dissociation’, so to speak. This phrase is the terminological counterpart of the well-known fallacy guilt by association: an attempt to ‘transfer some perceived discredit to an opponent, based on some association that person has with a supposedly discreditable individual or group’ (Johnson and Blair 1983: 82). According to Johnson and Blair, the fallacy of guilt by association is ‘a special case of ad hominem, for it is an attack on the person (instead of the argument), but an indirect one – via some (alleged) association of the person’ (1983: 90). As is clear from this definition, the parallel really is only terminological. For Perelman and Olbrechts-Tyteca, the term association as well as the term dissociation does not refer to relations among people but between things.
Having said that the Vatican document commits the fallacy of incorrect dissociation, I am tempted to conclude that, after all, the Roman Catholic Church is guilty of something: if not of failing in fighting the Holocaust, then of committing a fallacy – albeit that the former is, of course, to be taken much more seriously than the latter. But then, I am sure that the Vatican could easily manage to produce a document in which even this less serious accusation would be refuted.

NOTES
i. The Vatican document is published on The Holy See’s Internet site (www.vatican.va).
ii. The quotations in this paragraph are taken from the Jewish Telegraphic Agency Inc., virtualjerusalem.com (www.jta.org), March 16 and 29, 1998.
iii. The quotations in this paragrapgh are taken from the BBC News Online (news.bbc.co.uk), March 16, 1998; the Jewish Telegraphic Agency Inc. virtualjerusalem.com (www.jta.org), March 16, 1998; and the Canadian Jewish Congress (www.cjc.ca), March 16, 1998.
iv. The quotations in this paragrapgh are taken from The Miami Herald, Heraldlink (www.herald.com), March 17, 1998; and The Jerusalem Post, Internet Edition (www.jpost.com), March 17, 1998.
v. This quotation is taken from the Jewish Telegraphic Agency Inc. virtualjerusalem.com (www.jta.org), March 16, 1998.
vi. The book was originally published in French as La nouvelle rhétorique: traité de l’argumentation (1958).
vii. Cf. Van Eemeren, Grootendorst et al. (1996: 124-125).
viii. The quotations in this paragraph are taken from the Jewish Telegraphic Agency Inc. virtualjerusalem.com (www.jta.org), March 16, 1998; and The Jerusalem Post, Internet Edition (www.jpost.com) March 17, 1998.

REFERENCES
Eemeren, F.H. van, R. Grootendorst, F. Snoeck Henkemans et al. (1996). Fundamentals of Argumentation Theory. A Handbook of Historical Backgrounds and Contemporary Developments, Mahwah, N.J.: Lawrence Erlbaum.
Garssen, B.J. (1997). Argumentatieschema’s in pragma-dialectsich perspectief. Een theoretisch en empirisch onderzoek (Argumentation schemes in pragma-dialectical perspective. A theoretical and empirical research). PhD dissertation University of Amsterdam. Amsterdam: IFOTT. Studies in Language and Language Use, Vol. 32.
Johnson, R.H. & J. Anthony Blair (1983). Logical Self-Defense, 2nd ed. Toronto etc.: McGraw-Hill Ryerson Limited.
Kienpointner, M. (1992). Alltagslogik. Struktur und Funktion von Argumentationsmustern (Everyday logic. Structure and function of argumentation schemes). Stuttgart-Bad Cannstatt: Frommann-Holzboog. Series ‘Problemata’, Vol. 126.
Perelman, Ch. & L. Olbrechts-Tyteca (1969). The New Rhetoric. A Treatise on Argumentation, Notre Dame 1969: University of Notre dame Press. Originally published in French as La nouvelle rhétorique: traité de l’argumentation (1958).
Schellens, P.J.M.C. (1985). Redelijke argumenten. Een onderzoek naar normen voor kritische lezers (Reasonable arguments. A study in criteria for critical reading). PhD dissertation University of Utrecht.
Warnick, B. & S.L. Kline (1992). The New Rhetoric’s argument schemes: A rhetorical view of practical reasoning. Argumentation and Advocacy 29, 1-15.

In his provocative work, Protagoras and Logos, Edward Schiappa (1991) suggests that the Presocratics, the Sophists and Plato shared a different approach to language and communication. Still constrained to varying degrees by their primarily oral culture, they nevertheless offered prose as an alternative to poetry, and “treated language itself as an object of analysis for the first time in Greek history” (31). While Schiappa treats the definition and historical manifestations of logos with great care, he fails to do the same with mythos; presumably the Presocratics, the Sophists and Plato offered an alternative not only to “poetry as a vehicle of wisdom and entertainment,” but also to mythic accounts and conceptions of persuasion (31). Hence it is possible to better understand the contributions of early theorists of logos by better understanding the mythic understanding of persuasion that was available to the Greeks. In this essay I will explore the Greek mythic beliefs that persuasion took place through the action of the deities Hermes, Peitho, and the Charites (Barthell 1971: 152). After considering the range of meanings that each represents, I will consider the meanings represented by various combinations of them. In pursuing these meanings, I’m attempting to understand what a Greek, especially an Athenian, would gain by asking, “How can I persuade x?” and receiving the answer, “By considering Peitho, the Charites, and Hermes.” This question would have acquired more urgency around 500 BC, after Kleisthenes’ reforms, when the Athenian Pnyx was reinforced and dressed for the first time, hence dominating the approach to the marketplace (Kournouniotes and Thompson 1932: 216). After considering likely answers to the question, I will return to Schiappa’s argument, and maintain that Protagoras and later theorists where not as revolutionary as Schiappa portrays them, when one treats the mythic-poetic tradition as more than a preference for poetry.
In the discussions of deities that follows, it would be well to keep in mind the following chronology. The Iliad and the Odyssey date from the eighth century BC; Hesiod’s poems date from the seventh century BC; and the Homeric Hymns date from the sixth and fifth centuries BC. The Homeric Hymn to Hermes is from the sixth, probably late sixth century BC. Protagoras arrived in Athens around 450 BC.

1. Analysis of deities associated with persuasion
1.1 Hermes
Hermes probably originally arose as a god of the stone heaps that marked property boundaries (Farnell 1909: 7; Brown 1917/1990: 32). Hermes was the power found in the heap (Burkert 156). Because many tribal activities took place at the boundary between tribal territories, Hermes took on a range of associated meanings. Trading took place at the boundaries, so Hermes became a god of the marketplace, which later moved into the center of towns (Brown 1917/1990: 37). At first the stone heaps marked a neutral and sacred spot where trading could be safely conducted by traveling tradesmen and tribal groups with surplus goods (Farnell 1909: 26). Later, trading could safely be conducted in towns themselves. For example, in archaic Athens, the marketplace was on the northwest slope of the Acropolis, but was moved further north by Solon to a more central, level location (Travlos 1971: 2). At the symbolic center of the new agora was the altar of the Twelve Gods, and a Stoa of large herms (21). From that center, beginning around 520 BC, distances were measured and marked with herms at halfway points along all the major roads leading to the city (Brown 1917/1990: 107). The herms were inscribed with a statement of ownership by the tyrant Hipparchus, and a maxim such as “Think just thoughts as you journey” (Brown 1917/1990: 111; Parker 1996: 80). In this aspect Hermes implies that persuasion was a key to the success of the marketplace.

Farnell (1909) was worried over the dual associations of Hermes with the marketplace and thieving. He resolved this contradiction by arguing it does not mean that the state would pray to Hermes when it was about to represent itself dishonestly, nor that the state tolerated dishonest trading, but that “he stood to preserve the public peace of the place,” since early assemblies and deliberations were held there ( 24, 27). Especially the use of Herms, inscribed with the names of public benefactors, serving as mileage markers near the end of the sixth century BC, “may have spread the belief that the god was interested in the general welfare of the city” (26). Brown (1917/1990) sought to preserve the contradiction because he believed it indicated social tensions in Athens. In the Homeric Hymn, Hermes’ desires and characteristics are those of the “merchant and the craftsman” working in the Agora (81). Hence the hymn celebrates the increasing commercialization of the agora and ridicules aristocratic disdain for the nouveau riche (82).
Wife abductions and livestock raids took place at the boundary, so Hermes became a god of marriage, seduction and stealthy thievery (Brown 1917/1990: 42; Kerenyi 1944/1976): 60). As a god of seduction, he was mainly a god devoted to fertility and increase (Farnell V 25). The phallus on herms was probably originally a fertility symbol to those herding at the boundary. Yet, as Parker (1996) notes, “It then got stuck; for herms soon entered the city as images to which cult was paid, but retained the gross appendage” (82). Hermes’ ithyphallic nature remained important as it then began to symbolize the growth and continuation of the city. Hence even the later mutilation of the herms provoked widespread shock and panic.
Hermes’ connection to thievery was recognized in common outcry at the beginning of a new undertaking – “Koinos Hermes,” “a theft done together” (Kerenyi 1944/1976: 60). Hermes’ thefts, both in Homer and the Homeric Hymn to Hermes, are aided by trickery such as mists, sleep induced by his wand or staff, or invisibility (Brown 1917/1990: 11). This aspect of Hermes, when applied to persuasion, connotes the use of persuasion to mislead.

Negotiations between tribes occurred at the boundary, so Hermes became a god of intergroup negotiation and cunning speech (Brown 1917/1990: 8). By Homer’s time heralds had gained some of Hermes’ sanctity, and presently adopted Hermes’ shepherd’s staff, the kerykeion: “hence he became specially their tutelary divinity and the guardian of such morality as attached to Hellenic diplomacy” (Farnell 1909: 20). For example, records of some sacrifices note that the tongue of the victim was reserved for Hermes and the heralds (30n.). A further link between Hermes and heralds is found in the Greek term for interpreter, hermeneus, which derives from hermes (Burkert 1985: 158). Hence Hermes as a god of persuasion would suggest ambassadors’ speeches, which are portrayed as early as the Iliad (Wooten 1973: 209).
More broadly than these state functions, Hermes was associated with all cunning speech. In the Odyssey, Odysseus is related to Hermes on his mother’s side (Kerenyi 1944/1976: 48). Odysseus’ grandfather Autolycos, a son of Hermes, is highlighted for his prowess in thieving and manipulating oaths (Burkert 1985: 158). In the Homeric Hymn to Hermes, there is an example of such an oath when Apollo brings Hermes before Zeus after accusing Hermes of stealing Apollo’s cattle. Hermes swears not that he did not steal the cattle, but that he never drove the cattle to his house, and never stepped across his threshold. This sounds like a denial until one remembers that Hermes had driven the cattle to a cave and had entered his house through the keyhole. Like his connection to thievery, Hermes’ connection to cunning speech suggests the use of persuasion to mislead. The connection to cunning speech also suggests a certain indirect style of speaking that may be useful in persuasion.
Finally, people traveled at the boundaries, and often delighted in finding and consuming offerings to Hermes at the base of the stone heaps. So, Hermes became a companion god of travelers, and a god of sudden windfalls and the propitious discovery of good things (Brown 1917/1990: 20, 44; Kerenyi 1944/1976: 58-59). Yet Hermes was not credited with all lucky events. If a mentally deficient person had good luck, it was attributed to Herakles (Kerenyi 1944/1976: 60). Not only did he accompany traders on their travels of discovery, but Hermes also was credited mythologically with inventing several useful items. For example, the Homeric Hymn to Hermes credits him with inventing the tortoise shell lyre, reed pipes, and starting a blaze using fire sticks. Further, Hermes could lend grace and ingenuity to human artists’ crafts (Grantz 1993: 109). With regard to persuasion, these qualities suggest the process that would later be called invention.

Through analogical extension, Hermes became associated with other types of boundaries. One was the boundary between sleeping and waking. Farnell (1909) notes that “From the Homeric period onwards we have evidence proving the custom of offering libations to Hermes after the evening banquet, before retiring to rest; and we may believe such offerings aimed at securing happy sleep and freedom from ghostly terrors” (14). During Protagoras’ time in Athens, near the end of the fifth century BC, Hermes became explicitly associated with the new cult of Asclepias, the oracular physician, as the “bringer of dreams” (Parker 1996: 182). Yet Hermes’ association with omens had begun earlier. In the Homeric Hymn, Zeus gives to Hermes power over birds of omen (Gantz 1993: 106). In addition, Hermes was a patron of “divination by counters” (Farnell 1909: 17). Hence Hermes was implicated with ideas about the future.
Another analogical boundary associated with Hermes was that between the living and the dead. Farnell (1909) related that “On the Acropolis in the temple of Athena Polias, stood a very ancient wooden agalma of Hermes, said to have been dedicated by Kekrops, and as its form was almost invisible beneath the myrtle boughs wrapped around it, we may regard it as descending from the semi-iconic period” (5). Harrison (1966) believed Hermes was represented there as a snake, an unmistakable Greek chthonic symbol (295). In the Odyssey, dating from the same period as the temple, Hermes first appears in myth as a conductor of souls, when he awakens the souls of the slain suitors and safely guides them to Hades (Gantz 1993: 108). In addition, Hermes was commonly addressed as the agent of “binding” in written curses buried with the hope that Hermes would conduct the person named to Hades (Guthrie 1950: 271). In connection with persuasion this function most strongly suggests the possibility of using persuasion to gain revenge, especially through what would later be called forensic speaking.

The last boundary associated with Hermes was the boundary between public and private. Frequently at Greek house gates and temple entrances, herms were placed both to guard the emerging members of the household and to help keep daimons from entering the house (Farnell 1909: 18). As in the marketplace, Hermes served a protective at the house. Hence he might travel with citizens as they hurried to the market or the assembly.
Inconsistencies in Hermes’ family history provide clues to additional meanings. According to the Odyssey, and Hesiod’s Theogony, Hermes was the daughter of Zeus and Maia, the daughter of Atlas (Gantz 1993: 105-06). While Hermes emerges in epic and lyric poetry as the father of children by Polymele, Philonis, Aphrodite, and an unnamed daughter of Dryops, in the later Homeric Hymn to Hermes, Hermes claims Apollo’s place as consort to Mnemosyne, Memory, mother of the Muses. If Brown (1917/1990) is correct that the hymn was composed in the court of Hipparchus (c. 514 BC), frequented by the poet Simonides, then Hermes as a god of persuasion suggested the art of memory as well (92, 124, 130). Simonides, both in legend and in a c.264 public inscription, is named as the inventor of an art of memorization (Yates 1966: 28 ).
Finally, worship of Hermes was widespread, but not institutionalized. Rather, references to poorly understood “Hermaia” festivals were found throughout Greece (Farnell 1909: 31). There is reliable evidence that the third day of the Anthesteria festival was dedicated to sacrifices to Hermes as the escorter of souls (Simon 1983: 93). There may have been an archaic altar to Hermes in the Akademia neighborhood Travlos 1971: 42). Sacrifice of goats, Hermes’ sacrificial creature, was the second most common type of sacrifice (Burkert 1985: 54).Yet, again, the vast majority of these sacrifices were part of private ceremonies rather than public events. In all, he was rarely mentioned in the family trees of prominent clans or towns, and artistic representations of Hermes used in public worship are rare (Farnell 1909: 1, 32).

However, in the first half of the fifth century, sculptors “idealized and enobled” full-figure representations of Hermes, and representations of all types became more widespread (Farnell 1909: 55). Specifically, Hermes began to get younger and more athletic. Previously, almost all representations of Hermes were of a bearded, older man. Also, after Hipparchus set up the first Athenian herms, “the city was soon flooded with them. By the late fifth century the doorstep herm, that cheerfully shameless figure, must have been the most familiar divine presence in the streets of Athens” (Parker 1996: 81). It is tempting to connect these developments with the increasing importance of persuasion in Greek life.

1.2 Peitho
Peitho literally means “persuasion.” Since there were no clear capitalization patterns in archaic or classical Greek, there was no clear distinction between Peitho the goddess and peitho the abstract concept (Buxton 1982: 30). In addition, she seems to have operated simultaneously as a goddess of private and public persuasion. According to Buxton (1982), “to Greeks all Peitho was ‘seductive’. Peitho is a continuum within which divine, secular, erotic and non-erotic come together” (31).
The earliest poetic sources stress her erotic nature. In Hesiod’s Works and Days, Peitho gives Pandora golden necklaces to wear (69). According to Buxton (1982), “These were traditional instruments of erotic enticement” (37). Sappho portrays Peitho as a handmaiden of Aphrodite who can convince the object of her desires to love, and who can also “cheat mortals” when love turns out to be impermanent (38, 65).
Later cults and visual portrayals linked Peitho with marriage, which had a dual nature as a sexual relationship and an institution important to the state (Shapiro 1993: 187-88). In representations on vases predating Protagoras’ arrival in Athens, Peitho is portrayed playing a part in events of the Trojan War. She is either rewarding Paris for choosing Aphrodite, or coaxing a reluctant Helen to marry (Shapiro 1993: 189-91). Again, these private events had tremendous political implications. Buxton (1982) has argued that Peitho was often linked with other personifications with such a dual role:
“the characteristics of Eunomia, Euklia and Harmonia, like the characteristics of Peitho, span the erotic and public spheres. Seductiveness resides not only in Persuasion, but in Good Order, Noble Reputation, and Harmony” (48). In association with these other deities, Peitho may have suggested ethos and what would later be called arrangement.

At Athens, Peitho was at first linked in cult with Aphrodite Pandemos. Both shared a temple on the south or southwest slope of the Acropolis. Although inscriptions linked with the cult are late, they state that the cult was ancient. Again, there is confusion over Peitho’s nature, since Plato defined “Pandemos” as the vulgar sphere of Aphrodite’s activity (Farnell 1896: 660). Yet there is no evidence that Plato’s definition was shared, and it certainly did not exist before the fifth century BC. Rather, Apollodoros’ interpretation that the temple was built close to the old marketplace and so named because “all the people” gathered and deliberated there seems more correct (Buxton 1982: 34; Parker 1996: 49). As Farnell (1896) put it, “What we know is that until the declining period of Greek history, the cult of Aphrodite, so far as it appears in written or monumental record, was as pure and austere as that of Zeus and Athena, purer than that of Artemis, in nearly all Greek communities rules of chastity being sometimes imposed upon her priestesses . . . “ ( 663). Both Aphrodite Pandemos and Peitho were honored in a festival called the Aphrodisia; after “the sanctuary was purified with the blood of a dove, the altars were anointed, and the cult images conducted in procession to the place where they were washed” (Simon 1983: 48-49). In connection with Aphrodite Pandemos, then, Peitho connoted ties binding the citizens of the polis together.

In poetry, tragedy, and sophistic speeches, the nature of peitho was continually elaborated beginning ten years before Protagoras arrived in Athens. Buxton (1982) summarized the evidence as a set of manipulable polarities (62):

For example, he argues that a key theme of Aeschylus’ Suppliants (c. 460 BC) is that peitho is preferable to bia in both private and public spheres (Buxton 1982: 90). Just as with her secular fate, Peitho gained religious attention after the arrival of Protagoras. Both Isocrates and Demosthenes mention sacrifices to Peitho that make it clear Peitho began to be worshiped independently of Aphrodite Pandemos (Parker 1996: 234; Shapiro 1993: 202).

1.3 The Charites
The Charites, better known as the Three Graces, probably originated as agricultural gods. Rose (1958/1972) speculates that they “were pretty certainly to begin with agricultural deities whose function it was to make tilled ground look ‘winsome’ or ‘delightful’ because bearing a good crop” (16). Later they became deities who were also responsible for the things that bring people together, enhance life, “and induce men to accept gifts, especially a woman” (Tyrell and Brown 1991: 185). With regard to their worship, there was a shrine of the Charites associated with the archaic altar of Athena Nike near the Mycenaean entrance to the Acropolis
(Travlos 1971: 148).
Many groups of Charites were mentioned in poetry, “but no real myths about them and very little indication of any concrete function” (Gantz 1993: 54). All groups include a member who connotes brightness or light (Ann and Imel 1993: 145, 204, 206). Yet the “most widely accepted tradition” named the Charites as Aglaia, Euphrosyne and Thalia (Guirard 1963: 70). Like similar members in all other groups of Charites, Aglaia connoted brilliance, brightness, and splendor. She was said to preside at banquets, dances and social occasions (Ann and Imel 1993: 145). Euphrosyne, “she who rejoices the heart,” connoted mirth and hospitality (Guirard 1963: 70: Ann and Imel 1993: 174). Thalia, “she who brought flowers,” connoted gift giving and prosperity. She also later functioned as the Muse of comedy (Guirard 1963: 70; Ann and Imel 1993: 218). In sum, together the Charites connoted brilliance of style, symposia, and speeches of praise.

1.4 Hermes and Peitho
The meanings of both Hermes and Peitho become clearer when considering the contrast between peitho and dolos. Dolos can be translated as a “cunning trick” best exemplified by Hermes’ theft of Apollo’s cattle. Such cunning was a cardinal trait of Hermes. In most cases dolos is subversive, and is used to defeat a more powerful opponent (Buxton 1982: 64). In addition, when contrasted with dolos, peitho becomes frankness that is used to strengthen legitimate unions (65). Hence Peitho was usually portrayed speaking directly to an individual about to be married, while Hermes’ cunning, indirect designs could be promoted at a distance using mist, sleep, etc. Considering Hermes and Peitho together leads to consideration of two styles of speaking: direct and indirect. Also, Peitho’s connection to the interests of the state and Hermes’ connection to divination suggest what would later be called deliberation.

1.5 Hermes and the Charites
Hermes was “expressly assigned” to the Charites as their escort, and he was portrayed in that role in several places (Kerenyi 1944/1976: 110). For example, an archaic votive relief of Hermes and the Charites was displayed near Plato’s Academy (Travlos 1971: 51). Farnell (1909) believed these common representations showed Hermes leading the goddesses to the sacrifices being prepared for them, and perhaps connoted a role for Hermes as administering sacrifices for other gods (36). With regard to persuasion, the arrangement suggests that brilliance should follow invention, that brilliant words are direction less without clever ideas.

1.6 Peitho and the Charites
Peitho and the Charites were often depicted as attendants of Aphrodite (Rose 1958/1972: 55). In the Iliad, the Charites make a robe for Aphrodite; in the Odyssey, they bathe and dress her (Gantz 1993: 54). A surviving vase from the period of Protagoras’ stay in Athens explicitly links Peitho and the Charites at the birth of Aphrodite. Peitho pours a libation in her honor, while one of the Charites drapes a garment over her (Shapiro 1993: 200). In knowing Aphrodite’s secrets of adornment, they suggest that knowledge of what men desire is important to persuasion.
In Hesiod, the Charites help Peitho place the golden necklaces over Pandora’s head (68). Again the erotic and the publicly significant are conjoined. The conjunction might extend to the Charites also on those vases where Peitho is portrayed holding a flower in the midst of events linked to the Trojan War (Shapiro 1993: 190-92). Given their original nature as agricultural deities, they might well have been portrayed as blooming plants.

1.7 Hermes, Peitho, and the Charites
The five deities were grouped together in a few places during the archaic period. However, the total groups adds no extra meaning. With Hermes as a god of seduction, and with Peitho and the Charites knowing all Aphrodite’s secrets, this grouping might connote persuasion as sexual union. However, several facts make this interpretation unlikely. First, Hermes as a god of seduction and increase was grouped with the Nymphs, forest spirits without clear associations with the Charites or Peitho. Second, Hermes was never associated with plant fertility, as the Charites were (Farnell 1909: 11).
In addition, though Hermes and Aphrodite (so, through association, her attendants Peitho and the Charites) were linked in cults in several places, their connection stemmed from Aphrodite’s original nature as a Asian chthonic goddess (Farnell 1909: 653). Hermaphrodite was a child of Hermes and Aphrodite, but the Hermaphrodite myth occurs no earlier than Diodorus Siculus, who wrote during the late first century BC (Gantz 1993: 104). Finally, the Charites were not, during the archaic period, portrayed nude. Indeed, like Peitho, they were celibate. So, all five deities suggested the ideas of increase and prosperity, and the institution of marriage, but did not maintain these conditions through sexual union, but through clever words, frankness and brilliance.
In Hesiod, Hermes joins Peitho and the Charites in forming Pandora; he provides her with cunning and a knowledge of trickery (68). In the earliest surviving painted representation of Peitho, c. 510 BC, Hermes and the Charites join Peitho as she is about to crown Paris for making the right choice (Shapiro 1993: 189). In these two places, all five deities were implicated in events with simultaneous private and public significance.
A generation or two later, all five deities were represented on the Parthenon frieze and the throne of the Olympic Zeus, where the birth of Aphrodite was also portrayed (Farnell 1909: 705). So, just as with the individual deities, the whole group gained public significance after the arrival of Protagoras in Athens.

2. Reanalysis of the contributions of the sophists
Schiappa (1991) claims that Protagoras was “revolutionary” in his methods of teaching, and that with earlier philosophers he began a move toward prose and abstract expression from mythic-poetic expression. Based on the analysis of the five deities, Protagoras was not revolutionary, but was professionally successful because he fit well into Athenian mythic-poetic beliefs. In addition, Protagoras, later sophists, and Plato were not revolutionary because much that was associated with later dialogues about and manuals of rhetoric was already present in the mythic-poetic tradition. Finally, some of the elements of the mythic-poetic tradition were never translated into abstract, theoretical terms, so the tradition retained great power even as prose literacy grew.
Protagoras entered Athens using roads prominently marked with herms, and proclaimed himself a teacher of logos. As an itinerant teacher selling his services, he would seem to be claiming the protection and favor of Hermes whether he wanted to or not. As a traveler, a merchant, and a self-proclaimed inventor of useful instructional methods, he was triply associated with Hermes. As a proponent of debate, and a proclaimant of provocative, controversial aphorisms, he exhibited skills lent him by Hermes, and enjoyed a long, prosperous career as a result. A true historical reconstruction of Protagoras’ significance must take Greek social factors such as religion into account. These factors tend to temper claims about Protagoras’ and later sophists’ significance.
Protagoras did not write a practical manual of logos, and, if he did, it would have probably have been a collection of sample speeches (Schiappa 1991: 158). Such collections were new only in the sense that they were collections, for examples of speeches can be found throughout the Greek mythic-poetic tradition (Buxton 1982: 6-8). Yet the five deities associated with persuasion connoted much that would be treated in abstract, theoretical prose only much later than Protagoras. First, they suggested four of the later five “canons” of rhetoric: invention, style, arrangement, and memory. Specific techniques – three styles and a lost archaic art of memory – were also suggested. Since Peitho’s connection to good order was at one remove, it might be tempting to attribute the canon of arrangement to growing literacy. Yet it is important to keep in mind that while none of the five deities directly suggest arrangement, speeches presented as part of the mythic-poetic tradition did exhibit regular divisions.

Not only did the five deities suggest “canons,” they also suggested types or functions of persuasive speaking: forensic, eristic, deliberative, ambassadorial, and speeches of praise at symposia. Finally, the deities suggest several qualities of persuasion: it binds citizens together, despite economic tensions; it can be used to overcome a more powerful opponent; it requires knowledge of what men desire; it provides gifts, and can induce citizens to accept gifts; it can be used to mislead; and, it can be used to gain revenge. These “canons,” functions and qualities would not be systematically examined in abstract prose until Plato, Aristotle, and the author of the Rhetorica ad Alexandrum (c. 387-330 BC). Hence later theories of rhetoric can be seen in part as a working out and partial endorsement of the implications of the mythic-poetic tradition. Finally, Hermes unequivocally suggests ambassadorial speaking, a type of speaking that was never treated in an abstract, theoretical manner (Wooten 1973: 109). This is true even though Wooten argues that this type of speaking became increasingly important during the Hellenistic age (212). Hence some aspects of persuasion remained too sacred to be treated impersonally. Additionally, the five deities were worshiped more fervently at the same time as the mythic-poetic tradition was translated into prose. Representations of Hermes grew more youthful and widespread, Peitho began to be worshiped independently, and the Charites began to be represented on great state monuments at the same time that the sophists gained prominence.
Hence Protagoras deserves credit as one of the first prose writers, and as the inventor of new verb forms amenable to abstract expression. However, his personal circumstances and the content of his aphorisms fit perfectly well within the Greek mythic-poetic tradition, and the success of his and later sophists’ students might even have stimulated increased devotion to Hermes, Peitho, and the Charites.
Since the mythic conceptions of persuasion express much that would appear later in abstract prose, the contributions of later sophists must be differently assessed against the backdrop of the mythic-poetic tradition. Here, following Schiappa, I have focused especially on Protagoras. However, later sophist also deserve attention to determine their theories’ relationships to the views of persuasion implied by the mythic-poetic tradition.

REFERENCES
Ann, M., & D. M. Imel. (1993). Goddesses in World Mythology. Santa Barbara: ABC-CLIO.
Barthell, E. E., Jr. (1971). Gods and Goddesses of Ancient Greece. Coral Gables: Miami University Press.
Brown, N. O. (1917/1990). Hermes the Thief: The Evolution of a Myth. Great Barrington, Massachusetts: Lindisfarne Press.
Burkert, W. (1985). Greek Religion. Cambridge, Massachusetts: Harvard University Press.
Buxton, R. G. A. (1982). Persuasion in Greek Tragedy: A Study of Peitho. New York: Cambridge University Press.
Farnell, L. R. (1896). The Cults of the Greek States. Vol. II. New York: Clarendon Press.
Farnell, L. R. (1909). The Cults of the Greek States. Vol. V. New York: Clarendon Press.
Gantz, T. (1993). Early Greek Myth: A Guide to Literary and Artistic Sources. Baltimore: Johns Hopkins University Press.
Guirard, F. (1963). Greek Mythology. London: Paul Hamlyn.
Guthrie, W. K. C. (1950). The Greeks and Their Gods. Boston: Beacon Press.
Harrison, J. E. (1966). Epilogomena to the Study of Greek Religion and Themis: A Study of the Social Origins of Religion, New York: University Books.
Hesiod. (1983) Theogyny, Works and Days, Shield. Tr. Apostolos N. Athanassakis. Baltimore: Johns Hopkins University Press.
Kerenyi, K. (1944/1976). Hermes, Guide of Souls. Tr. Murray Stein. Woodstock, Connecticut: Spring Publications.
Kournouniotes, K., & H. A. Thompson. (1932). The Pnyx at Athens. Reprinted from Hesperia, I.
Parker, R. (1996). Athenian Religion: A History. New York: Clarendon Press.
Rose, H. J. (1958/1972). Gods and Heroes of the Greeks: An Introduction to Mythology. New York: World Publishing.
Sappho. (1958). Sappho: A New Translation. Tr. Mary Barnard. Berkeley: University of California Press.
Schiappa, E. (1991). Protagoras and Logos: A Study in Greek Philosophyand Rhetoric. Columbia: University of South Carolina Press.
Shapiro, H. A. (1993). Personification in Greek Art: The Representation of Abstract Concepts, 600 – 400 B. C., Zurich: Akanthus.
Simon, E. (1983). Festivals of Attica: An Archaeological Commentary. Madison: University of Wisconsin Press.
Thompson, H. A. (1936). “Pnyx and Thesmophoria.” Hesperia 5, 151-200.
Travlos, J. (1971). Pictorial Dictionary of Ancient Athens. New York: Praeger Publishers.
Tyrell, W. B., & F. S. Brown. (1991). Athenian Myths and Institutions: Words in Action. New York: Oxford University Press.
Wooten, C. (1973). “The Ambassador’s Speech: A Particularly Hellenistic Genre of Oratory.” The Quarterly Journal of Speech 59, 209-12.
Yates, F. (1966). The Art of Memory. Chicago: University of Chicago Press.

1. Introduction
The notion of a constitutive rule was placed on the philosophical agenda by John Searle who opposed constitutive rules to regulative rules. Where ‘regulative rules regulate antecedently or independently existing forms of behaviour ….. constitutive rules do not merely regulate, they create or define new forms of behaviour. The rules of football or chess, for example, do not merely regulate playing football or chess, but as it were they create the very possibility of playing such games.’ (Searle 1969: 33).
If we take the notion of behaviour rather broad, to make it include not only physical, but also mental behaviour such as believing and making inferences, rules of inference can be considered as a kind of regulative rules. Rules of inference indicate what we are allowed to infer, and, in an epistemological context, what we are justified to believe, given our other beliefs. On this view, the distinction between constitutive rules and rules of inference is a special case of the distinction between constitutive rules and regulative rules.
In this paper I want to explore the distinction and the relations between constitutive rules and rules of inference. In section 2 I elaborate on the distinction between these two. In section 3 the distinction is exploited to explain the defeasibility of reasoning with rules of inference. In section 4 I will argue for the surprising view that propositional logic is in the first place an ontological theory, and only in the second place a theory of valid reasoning. The argument of section 4 is supported in section 5 with a sketch of the outlines of a general theory of valid reasoning. The paper is summarised in section 6.

2. Constitutive rules and rules of inference
The distinction between constitutive rules and rules of inference is based on another distinction, that is the distinction between the world and our beliefs about it. Let us follow Wittgenstein (1922, 1.1) in defining the world as the set of all facts. Facts are what corresponds in the world to true sentences. Since sentences are language-dependent, facts are also language-dependent. And so is the world, because the world is the set of all facts.
This view does not imply that the world depends completely on human culture, but rather that the world is captured by means of concepts that depend on human culture. The conceptual framework in terms of which the world is captured is a cultural phenomenon. That does not preclude the possibility that this conceptual framework has been adapted in time, e.g. through both physical and cultural evolution, to capture the world as well as possible.
The facts in the world are not independent of each other. There are physical laws that create law-like connections between facts of certain types. Physical laws should be distinguished from human attempts to describe them. These attempts form hypotheses that may be true or false. The laws themselves are not true or false, but exist or not. An example of a physical law is the law of gravitation. This law creates a connection between the facts that body 1 has mass m1, body 2 has mass m2, and that the distance between them is d, and the fact that between the bodies 1 and 2 there exists a gravitational force that equals _m1m2/d2, where _ is the gravitational constant. That the law of gravitation holds seems to be a fact about the world. This fact brings with it that other facts in the world are related in the way indicated in the law.

Our world is not merely a physical one. By means of culture, humans have created lots of additional facts. The fact that the guilder is (still) the monetary unit of the Netherlands is an example of such a humanly conditioned fact.[i] The same counts for the fact that John is obligated to pay Mary ten guilders, because he promised to do that. Humanly conditioned facts are related to other facts by means of constitutive rules. One such a rule is that if one promises to do A, one thereby incurs the obligation to do A. Another constitutive rule is the rule that if something is a rectangle with equal sides, it is a square.
Constitutive rules are similar to physical laws in that their existence is a fact about the world that brings with it that other facts in the world are related in the way indicated in the rule. They differ from physical laws in that their existence depends on human culture, while physical laws are assumed to exist independent of human culture.
In my examples I have presented the linguistic convention that squares are rectangles with equal sides as a constitutive rule. This view seems reasonable to me, because linguistic conventions are a kind of rules, and the existence of these rules has effect on relations between other facts. Because of the ways in which we use the words ‘rectangle’, ‘square’, ‘equal’ and ‘side’, squares are rectangles with equal sides. In general, conceptual relations based on linguistic conventions have repercussions on the world, because the world is language-dependent.

Where constitutive rules create relations between facts in the world, rules of inference create relations between our beliefs about the world. Rules of inference permit certain inferences, and thereby allow us to believe something if we believe something else. For instance,  the rule of inference that smoke signal fire allows us to believe that there is a fire if we believe that there is smoke. To be sure, rules of inference are usually not formulated in terms of beliefs, but rather in terms of the belief-contents. The rule ‘smoke signals fire’ is a case in point. It refers to the phenomena smoke and fire, and not to believes about them. Nevertheless, as a rule of inference, it is relevant for our believes and not for the phenomena themselves. The rule of inference does not indicate a relation between facts, but allows us to conclude to the presence of one fact if we believe or hypothesise that one or more other facts are present.
Although the rule of inference ‘smoke signals fire’ does not indicate a relation between the fact types ‘presence of smoke’ and ‘presence of fire’, our use of this rule is justified by such a relation. If smoke tends to go together with fire, and we know that, we are justified in using the concerning rule of inference. This use in turns justifies our believing that there is a fire if we believe that there is smoke.
Another way to state that we are justified in believing that there is a fire if we believe that there is smoke, is to say that the conclusion that there is a fire can validly be drawn from the premise that there is smoke. Validity in this sense does not imply that the conclusion must be true if the premise is true. It only implies that the inference at issue is a ‘good’ one.
Using a rule of inference can be seen as another kind of mental behaviour, next to believing and drawing inferences. Such a use may be justified or not. I have already given the example that if one knows that smoke tends to go together with fire, one is justified in using the rule of inference that smoke signals fire. In general the use of a rule of inference is justified if one has reason to believe that there is a big chance that the conclusion of the rule of inference is true if the rule’s antecedent is true.[ii]
One way to obtain such a reason is when one believes that a constitutive rule exists. For instance, if the rule exists that thieves are punishable, this is a reason to use the rule of inference that if somebody is a thief, (s)he is punishable. This rule of inference has the same formulation as its underlying constitutive rule. Nevertheless their functions are quite different. The constitutive rule makes that somebody who is a thief is punishable, while the rule of inference permits one to believe that somebody is punishable if one believes that this person is a thief. In general, the existence of a constitutive rule is a reason to use the corresponding rule of inference.
This relation cannot be turned around, however. Not every justified use of a rule of inference indicates that there is an underlying constitutive rule. The rule of inference that smoke signals fire is a case in point. There is no constitutive rule that makes that there is a fire whenever there is smoke.

3. Reasoning by default
The difference between constitutive rules and rules of inference has profound implications for the ways these rules behave from an (onto)logical point of view. Constitutive rules attach consequences on the level of the world (in contrast to our beliefs about the world) to facts in the world. In this respect they resemble physical laws that also create correlations between types of facts.
Because constitutive rules operate on the ontological level, all facts are potentially relevant for the operation of these rules. Take again the rule that thieves are punishable. This rule makes that thieves are punishable, but it is subject to exceptions. For instance, if a thief acted under force majeure, she is not punishable. As a consequence, the operation of the rule that thieves are punishable depends not only on the fact that somebody is a thief, but also on the existence of rules and principles that identify which facts count as force majeure, and the actual presence of these facts. Moreover, the rules that define force majeure may also be subject to exceptions, defined by still other rules. The existence of those additional rules and the presence of facts that satisfy their conditions are also relevant for the effects of the rule that thieves are punishable, and so on …
Because of these possibilities of exceptions, and exceptions to exceptions, it may be hard to establish whether some concrete thief is punishable, but given the constitutive nature of the rules involved, it is assumed that the question whether she is punishable is provided with an objective answer by the existing rules and the facts that match their conditions. The rules so to speak automatically attach legal consequences to cases, and in doing this, they take all relevant facts into account, including the facts that are exceptions to exceptions, etc…
These ontological effects of constitutive rules must be distinguished from the epistemic role played by the rules of inference that are based on these constitutive rules. The judge who must decide whether some thief is punishable gives a decision on whether the person in question legally is punishable. The presumption is that the issue of punishability is decided by the law, and that the judge only tells us what the law is.
One might object that in hard cases the judge does not tell us what the law is, but rather makes the law herself. Even if this is correct, it does not subtract from it that in easy cases the judge describes the legal consequences of the case. The operation of legal rules is at least in easy cases assumed to be on the ontological level, independent of human activities such as the active application of the rules.

A parallel with physical laws may be useful here. What happens with a particular object in a field of forces depends on all physical laws and on all facts that are relevant for these laws. It may be very difficult, if not impossible, to predict what will happen, but that does not stand in the way of all laws and all facts contributing to the actual movement of this object. At least ideally the same counts for constitutive rules. All constitutive rules and all facts are presumed to contribute to the constitution of the world, even if it is hard to determine the outcome.
To find out whether some person is punishable, the judge draws a conclusion from the fact that this person is a thief. The validity of the conclusion is based on the rule of inference that thieves are punishable. Rules of inference are applied to draw conclusions about the effects of constitutive rules. These facts themselves are assumed to obtain independently of human reasoning. The effects that constitutive rules create in the world are mimicked by rules of inference in our beliefs about the world.
Because of the ontological level on which constitutive rules are presumed to operate, it is not possible that rules ‘overlook’ some facts that might be relevant for their outcome. For instance, it is not possible that some thieve is punishable, because the fact that she operated under force majeure was overlooked by the constitutive rules of law. Constituted facts are not revisable on the basis of additional information, because constitution by definition takes everything that is relevant into account. Constitutive rules may have exceptions, but the facts that form exceptions are automatically taken into account in the operation of the rule. It is not possible that an exception comes up later and causes the effects of the rule to disappear.
The punishability of a person as an ontological issue is established by taking all relevant factors in consideration.
With rules of inference, matters are very different. Essentially a rule of inference indicates what one is allowed to believe given one’s other beliefs. The beliefs from which one starts may be incomplete in the sense that relevant information may be lacking.[iii] Rules of inference may give a definitive answer to the question what other beliefs may be held given these beliefs, but they cannot give an answer to the question which beliefs may be held that is not relative to some set of beliefs.
But then conclusions drawn on basis of good rules of inference may be revisable in the light of new information. If one only knows that somebody is a thief, one may infer that this person is punishable. If later on one comes to believe that the thief acted under force majeure, the conclusion that this person is punishable must be withdrawn. This conclusion may be reinstated again if still later on one comes to believe that the force majeure was caused by the thief’s own fault. Notice, however, that whether the thief is really punishable does not depend on what one is justified in believing. It depends on all relevant rules and facts, whether they are taken into account in human decision making, or not.

An argument and its conclusion are said to be defeasible, if it is possible that new information invalidates the argument and its conclusion. We have seen that reasoning with rules of inference is defeasible. It would, however, be wrong to extrapolate the defeasibility of reasoning with a rule of inference to the defeasibility of the operation of its underlying constitutive rule. Constitutive rules are assumed to take all relevant facts into account, while rules of inference only operate on beliefs that are actually held and that may be incomplete.
Summarising, we can state that although all rules may be amenable to exceptions, this amenability only gives rise to defeasible reasoning in the case of rules of inference, because only rules of inference operate on beliefs, which may be incomplete. Constitutive rules operate  on the facts of the world, which cannot be incomplete. Therefore the operation of constitutive rules is not defeasible.[iv]

4. Propositional logic as an ontological theory
Logic is the theory of valid reasoning. As such it does not presuppose a notion of logical validity. The traditional characterisation of a logically valid argument as an argument which cannot have a false conclusion if its premises are true is just one interpretation of what logical validity amounts to.
In my opinion an argument is logically valid (in contrast to other forms of validity such as legal validity, or the validity of moves in chess) if its conclusion is sufficiently justified by its premises. There are in principle different possible ways in which the premises of an argument may justify the conclusion, and the traditional notion of logical validity provides one of them. In the rest of this paper I will exploit the distinction between constitutive rules and rules of inference to argue that the traditional notion is not the most attractive interpretation of logical validity. In my argument I will take propositional logic as a starting point, although another variant of deductive logic would do just as well.
I assume that all propositions express a state of affairs, and that logical relations between propositions are reflected in ontological relations between the corresponding states of affairs. To facilitate my argument I use the notational convention that propositions are represented by capitals, while the corresponding states of affairs are represented by corresponding lowercase letters. For instance, the proposition P expresses the state of affairs p, and the proposition P & Q represents the state of affairs p & q.

A proposition is then true and expresses a fact, if and only if the corresponding state of affairs obtains. So there is a perfect parallel between the truth of propositions and the obtaining of the states of affairs expressed by them. As a consequence the truth-functional definition of the logical operators has repercussions on the ontological level:
(1) the state of affairs ~p obtains, if and only if the state of affairs p does not obtain;
(2) the state of affairs p & q obtains, if only if the states of affairs p and q both obtain;
(3) the state of affairs p _ q obtains, if and only if either the state of affairs p, or the state of affairs q obtains, or if both states of affairs obtain.
These ontological repercussions of the definitions of the logical operators illustrate the constitutive function of meaning rules. The rules that define the meanings of the logical operators create connections between states of affairs. These relations are in their turn reflected in the truth values of the propositions that express them. In deviation from the traditional view, I assume that the definitions of the logical operators have primarily effects on the ontological level, and only indirectly on the truth values of propositions.

Since the ontological relations between states of affairs are reflected in relations between truth values of the propositions that express these states of affairs, it is possible to use the knowledge about these ontological relations to make inferences about the truth values of these propositions. For instance, since the state of affairs that it is both raining and the sun shines obtains if the states of affairs that it is raining and that the sun shines both obtain, it is possible to infer from the propositions ‘It is raining’ and ‘The sun shines’ the proposition ‘It is raining & The sun shines’.
Given the relations between the states of affairs expressed by these propositions, it is not possible that the premises are true while the conclusion is false. This is impossible because it is impossible that if the states of affairs that it is raining and that the sun shines both obtain, the state of affairs that it both raining and the sun shines does not obtain. The impossibility on the ontological level is primary, and the impossibility on the level of truth-values is derived. Similarly the use of the rule of inference that it is justified to conclude that ‘It is raining & The sun shines’ from ‘It is raining’ and ‘The sun shines’, or in general, to conclude ‘P & Q’ from ‘P’ and ‘Q’, derives its justification from the ontological effects of the meaning rules for the word ‘and’, respectively the operator ‘&’.
On this view propositional logic is nothing else than a study of the ontological implications of a particular set of meaning rules, that is the meaning rules for the so-called ‘logical words’. In a similar way, it is possible to study the implications of other sets of meaning rules. For instance one can study the ontological implications of the meaning rules for the words ‘quadrangle’, ‘rectangle’, and ‘square’. These rules also have implications for the truth values of some propositions given the truth values of other propositions. For instance, given the usual meanings of the words ‘rectangle’ and ‘square’, the proposition that squares are rectangles is (necessarily) true.
The same counts for the meaning rules for the words ‘obligated’, ‘permitted’, and ‘forbidden’, that make it necessary that the proposition that this act is permitted is true given the truth of the proposition that this act is not forbidden. In fact the study of the ontological implications of these meaning rules has been conducted under the name of ‘deontic logic’.
In general it holds that all constitutive rules have ontological implications which may be studied in the form of a logic. Propositional logic and predicate logic, deontic logic, logics of events and logic of actions (dynamic logic) are all examples of this kind of research. All of these logics can be seen as ontological theories, cast in the form of theories of valid reasoning.
Should we conclude, then, that logic is really a form of ontology instead of a theory of valid reasoning? In my opinion this conclusion is not inevitable. It is possible to develop a logic as a general theory of valid reasoning that is not the study of the ontological implications of constitutive rules. In the next section I will argue for another form of logic, the topic of which is the study of how rules of inference operate.

5. The operation of inference rules
In section 2 I have argued that constitutive rules have corresponding inference rules that allow us to make inferences that match constitutive relations in the world. Propositional logic as a theory of what we may infer deals with the rules of inference that correspond  to the meaning rules for the logical operators. It seems a little restricted, however, to limit the study of valid reasoning to the implications of just a small set of meaning rules. As a general theory of valid reasoning, logic should have something to say about all arguments, not only those based on the meanings of a few so-called ‘logical words’.
The traditional idea that logic should abstract from the topic of arguments is still a good one. It should however be realised that a logic that deals with the meanings of just a few words does not abstract from the topic of the argument. A theory that deals with the meanings of the words ‘and’, ‘or’, ‘not’, and the construction ‘if …then …’ is just as restricted as a theory that deals with the meanings of the words ‘quadrangle’, ‘rectangle’, and ‘square’. This last theory is not called ‘logic’, however, but rather ‘geometry’.
How, then, is it possible to have a theory about valid reasoning in general, which is not confined to the study of the meanings of some special category of words? I think that such a general theory of valid reasoning should study the logical behaviour of inference rules. There are many kinds of inference rules, some of them counterparts of constitutive rules, others based on empirical evidence or the result of legislation, and again others are based on statistical laws. The contents of these rules are domain (or field-)dependent. In the previous section I have argued that this is even the case for the inference rules of propositional logic. To the extent that they are domain dependent, inference rules are less interesting from a logical point of view. What is interesting is the role of these rules in making valid inferences. If it is possible to develop a general theory about the operation of inference rules in making valid inferences, such a theory would be a worthy candidate for the label of ‘logic’.

Is such a theory possible? In my opinion, the answer to this question is affirmative. Well-known examples of such theories, although not advertised as such, are Toulmin’s theory about the lay-out of arguments (Toulmin 1958), and Naess’ work on reasons for and against a conclusion (Naess 1976). Less known are the thesis of Verheij (Verheij 1996), and my own work on ‘Reasoning with Rules’ (Hage 1997). In the following paragraphs I will give an outline of the last work, in some respects amended and rephrased as a theory about the function of inference rules. There is, however, no room for argument here, and consequently my presentation must be rather apodictic, more an illustration of what is possible than an attempt to convince sceptics. Those who are interested in the subject are referred to Hage 1997.
I consider rules of inference as rules that indicate which inferences are allowed. Since inferences are a kind of transitions between (hypothetical) beliefs, rules of inference may also be characterised as rules that indicate which beliefs may justifiedly be held, given that some other beliefs are held. As a consequence, the theory about the operation of inference rules is part of the more general theory of rational belief, which is in its turn part of the theory of rational action.
A rule of inference is used by some person, if this person is disposed to believe the conclusion of the rule if (s)he believes the conditions of the rule. Using a rule of inference is a form of mental behaviour that is subject to evaluation from the point of view of rationality. The standards for this evaluation are part of a theory of rational action, the contents of which are for the most part not a logical issue. For instance, it is not a logical issue to determine whether it is justified to use a rule of inference if that rule was recommended by Sherlock Holmes in ‘A Study in Scarlet’. It is, however, a logical issue to the extent that facts to which the standards are to be applied may be the conclusions of inferences which in turn may be evaluated by logical standards. Inference rules justify inferences either if it is justified to use them, or if they are used in the sense described above, and this use is not irrational.
An inference rule makes the facts that satisfy its conditions into reasons for believing its conclusion. In most case it suffices to have one reason to believe something. For instance, if I read in the newspaper that there was an accident on the highway, I have a reason to believe that there was an accident on the highway. This reason normally suffices to be justified in believing that this accident occurred.
Sometimes, however, we have both reasons to believe something and to believe something different. For instance, if my friend lives near the place where the highway accident was reported to happen, and my friend tells me that he was home all day and did not notice anything of an accident, I am not fully justified to believe the newspaper report anymore. Somehow I must balance two reasons, one for believing that there was an accident and one against believing it.

There may even be more reasons. For instance if my friend’s wife heard unusual noises that might be explained by a crash, this would be an additional reason to believe in the crash. But if she did not notice anything either, this is an additional reason against believing in the crash.
There may also be ‘meta-reasons’ that deal with the absolute and the relative weight of the reasons. For instance, if I know that my friend uses to work very concentratedly and seldom notices anything of what happens on the highway, the reason that he did not notice anything is not strong on an absolute scale. If I happen to know that my friend makes a custom of contradicting the newspaper, this is a reason to give the reason based on his testimony a smaller weight relative to the reason based on the newspaper report.
If there are both reasons for and against believing a conclusion, it is justified to believe the conclusion either if one is justified in believing that the reasons for believing the conclusion outweigh the reasons against believing it, or if one actually holds that belief and the belief that it was otherwise would not be justified. Notice that the logic of inference rules as such does not say anything about what are reasons or how they should be weighed. This is all ‘field dependent’, to use Toulmin’s phrase.
The conclusion that some set of reasons outweighs another set of reasons will itself be based on a rule that deals with the relative weight of sets of reasons. A legal example of such a rule would be that a set of reasons outweighs another set of reasons, if there was an authoritative court decision in which this relative weight was assumed.

There is one final issue, and that is that there may be exceptions to the use of an inference rule. Normally if the conditions of an inference rule are satisfied, the use of the rule is justified and there is a reason to believe its conclusion. There may be exceptional circumstances, however, in which the rule’s conditions are satisfied, but the rule should nevertheless not be applied. The constitutive rule that promises lead to obligations has its counterpart in the rule of inference that if somebody promised to do something, he may be assumed to have the obligation to do it. However, if Jane promised to marry John, it may not be assumed that Jane is under the obligation to marry John, if one believes that Jane mistook John for his brother. This illustrates that exceptions to constitutive rules have counterparts in exceptions to rules of inference.
It is amongst others this phenomenon, that rules of inference may have exceptions, that makes most arguments defeasible. A conclusion based on a rule of inference may lose its justification if it becomes known that an exception to the rule occurs. Another cause of defeasibility is that it becomes known that there are reasons against a particular conclusion that were not yet taken into account when the conclusion was drawn. These two causes of defeasibility are by and large identical to respectively undercutters and rebutters in the sense of Pollock (1987).

6. Conclusion
At the beginning of this paper I followed Searle in distinguishing between constitutive rules and regulative rules. I continued by assuming a broad notion of behaviour, to make behaviour include mental behaviour such as believing, drawing inferences, and using rules of inference. Rules of inference are on that view a special kind of regulative rules.
I contrasted constitutive rules to rules of inference. Where the former operate on the ontological level and create relations between types of facts, the latter operate on the level of belief and make that some beliefs may be held given that some other beliefs are held.
The distinction between constitutive rules and rules of inference was exploited in two arguments. It was first argued that the use of inference rules can be defeasible while the effects of constitutive rules are not defeasible. Second, it was argued that classical logics such as propositional logic and predicate logic are a kind of ontological theories, that study the ontological effects of a limited number of meaning rules. The derivations that they allow correspond to the ontological relations created by these meaning rules.
The final step in my argument was that it would be attractive to have a logic that deals with valid inference in general, and not only with the meanings of a small number of so-called ‘logical words’. I proposed to consider the theory about the operation of rules of inference as part of such a logic, and continued to give the outlines of such a theory.

NOTES
i. The expression ‘humanly conditioned fact’ stems from Weinberger’s paper ‘Facts and fact-descriptions’, in MacCormick and Weinberger 1986.
ii. Some rules of inference are the result of legislation, such as the legal rule that an official deed provides conclusive evidence for what is stated in the deed.
iii. I will leave the possibility of wrong beliefs out of consideration.
iv. This conclusion adds to and slightly deviates from my views as exposed in Hage 1997, where I did not attach sufficient consequences to the distinction between constitutive rules and rules of inference.

REFERENCES
Hage, J.C. (1997). Reasoning with Rules. An Essay on Legal Reasoning and its Underlying Logic, Kluwer Academic Publishers, Dordrecht.
MacCormick, D.N. and O. Weinberger (1986). An Institutional Theory of Law, Dordrecht e.a: Reidel.
Naess, A. (1966). Communication and Argument. Elements of Applied Semantics. London: Allen & Unwin.
Pollock, J.L. (1987). Defeasible Reasoning. Cognitive Science 11, 481-518.
Searle, J.R. (1969), Speech acts. An essay in the philosophy of language. Cambridge: Cambridge University Press.
Toulmin, S.E. (1958). The Uses of Argument. London: Cambridge University Press.
Verheij, Bart (1996). Rules, Reasons, Arguments. Formal studies of argumentation and defeat. Thesis Maastricht University.
Wittgenstein, L. (1922). Tractatus Logico-Philosophicus. Frankfurt a/M: Suhrkamp.

Throughout the High Middle Ages and into the Renaissance, the Inquisition was a continuing feature of the Christian world. To speak of the Inquisition as a singular institution is misleading, since inquisitions were undertaken by various authorities, episcopal or papal, working under varying legal systems and constraints. On its face an effort to ensure religious orthodoxy, it was from time to time overtaken by political concerns, both local and international; nor were purely personal vendettas completely irrelevant. Concerned at first with dualist heresies in southern France, it expanded its interests to cover witchcraft and Judiazing Christians, and later was an important front in the Catholic battle against Protestantism. The Protestants also had their inquisitions, though these were not as famous or institutionally developed as the Catholic ones.
But I will leave to others (e.g., Peters 1988; Lea 1955) the task of differentiating among the inquisitions of different times, places, and objectives. My purpose here is general enough that the more or less continuing features of the inquisitorial mode of jurisprudence will serve as a suitable basis for study. I intend to examine inquisitorial practices in the context of pragma-dialectics (van Eemeren & Grootendorst 1984; 1992; van Eemeren, Grootendorst, Jackson, & Jacobs 1993). Though I will say little that is new in detail about the Inquisition, my approach may possibly provide a coherent perspective on how the Inquisition accomplished what it did. My main purpose, however, is to illuminate an under-developed topic in the study of argumentation, disagreement space.

1. Disagreement Space
The idea of disagreement space appears as part of the project of reconstructing arguments (van Eemeren, Grootendorst, Jackson, & Jacobs 1993: esp. 95-102). The general task of reconstruction is to take what people actually say, and to “reconstruct” it, or understand it in an analytical way, for purposes of description and criticism.
People do not say everything they mean, and do not comment on everything they understand. By a close and disciplined examination of actual utterances, and what had to have been understood or meant for the statements to have served the communicative functions they did, analysts can specify the domain of interactive meaning, including all those background assumptions. As an example, three pages of conversation are expanded into about twenty pages of reconstruction, capturing understandings taken for granted, unstated connections among premises, implicit refutations, and so forth (van Eemeren, Grootendorst, Jackson, & Jacobs 1993: ch. 4).
Roughly speaking, disagreement space refers to all that could be argued about, all that needs to be filled in for a full analysis. Here is the defining passage:
Among the materials available to a participant in an argumentative discussion are the discourse itself and the surrounding context of practical activity. From these two components it will always be possible to infer an indefinitely large and complex set of beliefs, wants, and intentions that jointly compose the perspective of one’s partner. Any component of this perspective may be “called out” and made problematic within the discourse, if it has any sort of relevance to the underlying purpose of the exchange. When this occurs, the problematized element functions as a “virtual standpoint” in need of defense. Any reconstructible commitment associated with the performance of a speech act can function as a virtual standpoint when it is in fact reconstructed and challenged by an interlocutor. The entire complex of reconstructible commitments can be considered as a “disagreement space,” a structured set of opportunities for argument (van Eemeren, Grootendorst, Jackson, & Jacobs 1993: 95).

Now in the context of the Inquisition, what makes disagreement space interesting is that it doesn’t work properly. As I will show, quite a lot of the “beliefs, wants, and intentions” that were pointedly relevant in trials could not be “called out” and argued about. As a matter of fact, many of these argumentative components were quite explicit, but still were unavailable for controversy.
While I am confident that the authors would not be shocked to notice that many arguments are constrained in such a way as to prevent dialectical discussion, their treatment of disagreement space seems to imply otherwise. In the passage above, there is little hint that certain avenues of talk may not be allowed. Perhaps this is connected to the authors’ focus on discourse in these and other relevant sections. They even toy with the idea that disagreement space might be defined by the felicity conditions of the speech acts being expressed (p. 116, n. 7). Their attention has wandered away from the “surrounding context of practical activity” they mention, or at least has been diverted from any non-illocutionary sorts of practice.
They are entitled to focus anywhere they please, of course. Here, however, we will be looking at the argument’s context in a more institutional way, concentrating on how the external (i.e., non-argumentative) power of the Inquisition permitted the inquisitors to control disagreement space during the trials. This paper is intended as an expansion of the idea of disagreement space, and an exploration of how it can be controlled, and with what effects.

2. Inquisitorial Manuals and Procedures
Inquisitors were rarely trained to the vocation of inquisition. Many inquisitions were undertaken by the local bishop, who had many responsibilities and duties, the eradication of heresy being a pretty minor one (see Kieckhefer 1979). Inquisitors sent out from Rome, Avignon, or Madrid were most commonly Dominicans or, less often, Franciscans, who had distinguished themselves in their normal duties. Few people made a career of inquisitions, and few wanted to. To hold another’s life in one’s hands was an unhappy experience for a churchman, an exceptionally onerous duty; nor was it pleasant to confront heresy, witchcraft, or demonic inspiration face to face. All of these men were educated in Catholic theology, but few had any training in legal processes, either secular or ecclesiastical. Their intellectual orientation toward controversy was to find truth, not justice. A question was settled for them when they could trace an answer back to Scripture, papal bulls, or Patristic writings. Once understood, these could not be questioned; to do so was heresy.
So experienced inquisitors wrote manuals for the use of those who came later. These manuals gave the proper forms for summonses, admonitions, sermons, and sentences; they described the heretical beliefs one might encounter; they laid down and justified firm procedural requirements; and they gave advice on interrogation, torture, imprisonment, property confiscation, transcript preparation, sentences, and other practical matters. The earliest of these was the Processus Inquisitionis, emerging in 1248-1249 from the initial inquisitions in southern France (Anonymous 1980). This is a much briefer effort than those to come later, and is mostly confined to regularizing the formulae for the various legal documents. The first great manual was written in about 1323-1324 by Gui (1991), reflecting his further experience in the same region of France. This manual is an important document for scholars of heresy, because of its elaborately detailed descriptions of the leading heresies of the day (inquisitors typically destroyed any heretical writings they found, and few primary sources have survived). Gui also gives quite a lot of procedural detail. Perhaps the most mature manual is Eymeric’s, written in the late 14th century, dealing with his inquisitorial work in Aragon (summarized with enthusiasm in Walsh 1969: 94-112). This built upon Gui’s and other early manuals, and was an important resource for inquisitors in all Christian lands for several centuries. Although witchcraft had been of occasional interest to inquisitors from the early days, it became a preoccupation for inquisitors everywhere but Spain, beginning at the end of the 14th century and escalating in the mid-15th century. Kramer and Sprenger (1971) produced their infamous Malleus Maleficarum (Hammer of Witches) in about 1486, describing witchcraft in extraordinary, terrifying, and credulous detail. A somewhat more moderate, but still vituperative, manual was written by Boguet (1929), the chief judge in the district of St. Oyan de Joux of France, in 1590. Although not manuals per se, the Suprema of the Spanish Inquisition produced a series of instructions to inquisitors throughout its existence, notably in 1484, 1488, 1561, and 1568 (see Lea 1907).
Besides the manuals which specify how inquisitions ought to be conducted, quite a number of trial transcripts have survived. In those, one can see how the requirements and advice of the manuals are implemented. Conveniently available transcripts in English include those of Joan of Arc (Barrett 1931), a bizarrely heretical Italian miller named Domenico Scandella (Del Col 1996), the Salem witch trials (Boyer & Nissenbaum 1993; Trask 1992), several trials conducted by Jacques Fournier in southern France (Stork 1996), and a variety of inquisitions translated by Burr (1998). Although it may be a mistake to think so, my present view is that the trials essentially implement the manuals’ instructions, and serve as illustrative evidence rather than the primary sources on how the inquisitions were generally conducted. Consequently, I will not undertake a detailed study of any of the trials here, and will try to keep a broad perspective.
In looking at inquisitorial practice, I want to show how the inquisitors controlled disagreement space. For the most part, they constricted it to focus on the one key issue: whether the heretic’s soul could be saved. In a few respects, however, they insisted upon an enlargement of the disagreement space, requiring the accused to expand onto topics s/he resisted discussing.

3. Inquisitorial Constriction of the Disagreement Space
Inquisitions did not begin until the judges were fairly certain of the accused’s guilt. Denunciations were received and witnesses were interviewed. Evidence might well accumulate for years before the accused was called to answer. The issue in the trial was not, as 20th century Westerners might assume, whether or not the person was guilty; that was assumed. The issue was whether the sinner could be reconciled to the Church: whether s/he was contrite and willing to undertake penance (which might take the form of wearing a yellow heretic’s cross, making pilgrimages, undertaking service on the seas, or enduring prison; for the most part, only relapsed or unrepentant heretics were burnt). Consequently, professions of innocence or claims that acts were not heretical were out of order, regardless of whether the accused thought these were legitimate issues. Nor was it permissible to challenge the Inquisition’s procedures or authority, for this constituted heresy in itself. The Inquisition used a number of practices to constrict disagreement space, and we will explore these in this section.

3a. Anonymity of Witnesses
As early as the Carcasonne manual (Anonymous 1980), names of witnesses were withheld from the accused, and this practice continued throughout the Inquisition’s history. The stated reason for this is that the Inquisition feared retribution on the witnesses, and this was not a fictional concern (e.g., Del Col 1996: xc-xcii; Le Roy Ladurie 1978). Eymeric warned that the accused might try to evade the Inquisition by intimidating witnesses (Given 1997: 93-95). Witness anonymity was also the practice in some secular courts of the day (Peters 1988: 64). When defense attorneys were permitted, the lawyer generally had a right to see the Inquisition’s evidence, sometimes including the names of witnesses. However, the names would typically be disordered, and irrelevant names possibly included, to prevent any effective argument against them (Lea 1907: v. 3, 49). And the fact that some of the witnesses had died, their testimony surviving them by many years, made cross-examination impossible in any case.
For our purposes, the main consequence of all this is that it closes off a whole line of defense. The accused could not effectively argue that the denunciations were inaccurate or personally motivated. True, the accused was asked for a list of mortal enemies, and if the witnesses happened to appear on that list, the Inquisition would make genuine inquiries about the quality of the testimony. But since the actions at issue might have occurred many years before, and since the accused might not even know the time and place of the alleged acts (even the specific charge might be withheld during this initial phase: Lea 1907: v. 3, 39), the list of mortal enemies was at best a shot in the dark (Haliczar 1990: 76; Lea 1907: v. 3, 68-69).

3b. Control of Witnesses
Not anyone could give evidence. However, the qualifications for witnesses were quite different, depending on whether they had evidence for or against the accused. Even witches could give evidence against other witches (Boguet 1929: arts. 3 and 58). In France the inquisitors heard evidence from children, heretics, criminals, and accomplices (Given 1997: 15). In Spain, prosecution witnesses could be disqualified only for mortal enmity. Children, Jews, slaves, family, and excommunicates were all permitted to be witnesses against the accused; however, for the defense, no family, no Jews, no Moors, no New Christians (converted Jews), and no servants were allowed (Lea 1907: v. 2, 536-540). Defense attorneys were not permitted to advise the accused to call witnesses in defense, anyway (Lea 1907: v. 3, 69).
These tactics obviously gave the Inquisition considerable control over what could be placed in evidence. The trial began on the Church’s terms, and contrary discourse was difficult to introduce.

3c. Document Control and Other Intimidation
One of the striking features of the Inquisition, and one that has made it so attractive to modern historians, is its records (e.g., Ginzburg 1980; 1983; Le Roy Ladurie 1978). Statements and testimonies were carefully recorded and stored in such a way as to make indexing possible. Naturally, these records were secret, and so this resource was unavailable to the accused. Materials from one trial could lead to dozens of others, sometimes many years later; or a trial could be generated out of minor points uncovered in several earlier investigations (see Given 1997: ch. 1). Defenses could not be constructed in the same way. These documents could be very intimidating to the often illiterate accused, and inquisitors were trained to make use of this reaction. They sometimes read out a witnesses’ statement to the accused, to force out a confession (Given 1997: 40). Sometimes, too, they only pretended to be reading, or would flip through irrelevant pages and sadly remark that the defendant must be lying (Lea 1955: v. 1, 416-417).
This was all done in order to coerce a confession, which was always the objective of everything the inquisitor did. Confession was required if the accused’s soul were to be saved. Other forms of intimidation were also used, to the same end: excommunication, imprisonment, threat of torture, and actual torture. Lea (1955: v. 1, 422) even reports a case in which the inquisitor got the accused drunk in order to obtain a confession and list of accomplices. We see here a funneling of discourse: anything from the accused that was not a confession was essentially irrelevant, and anything that led to confession was eternally justified.

3d. Defense Attorneys and Other Spies
In the early centuries of the medieval Inquisition, defense attorneys were not allowed. They were, however, permitted to witchcraft defendants by the end of the 15th century. The Spanish Inquisition’s provision of defense attorneys for the poor came close to being an innovation in legal practice (Lea 1907: v. 3, 42-43; Ginzburg 1983: 125, also reports a case of an attorney being appointed for a poor defendant in Italy, in the 17th century). However, French and English courts as late as the 16th century did not permit defense attorneys at all (Haliczer 1990: 78). Defense attorneys had to be approved as to their character and attitude by the inquisitors (Kramer & Sprenger 1971: part 3, question 10), and one would not be appointed if he were, for example, litigious.
The possibility of an effective defense was rather slim, partly for reasons already given. Only after responding to the charges would the Spanish defendant be permitted to have an attorney at all (Lea 1907: v. 3, 42). By 1522 in Spain, the defense counsel was not permitted to communicate with relatives of the accused, eliminating any hope of their knowing the accusation and being able to find favorable witnesses (Lea 1907: v. 3, 48). The defense could do certain things: call witnesses as to the accused’s Christian character or the mortal enmity of other witnesses, deny that s/he did the act at a given time and place, plead for mitigation (on grounds of youth, insanity, ignorance, grief, drunkenness, etc.), or try to recuse the judges (Lea 1907: v. 3, 56-63). Haliczer (1990: 77) gives an example of a good, thorough, and apparently effective defense of a New Christian in 1521, and while others instances of vigor can be found, they are rare.

Attorneys had to please the tribunal in order to keep working (Haliczer 1990: 75). The lawyers were always themselves at risk, for an energetic defense might result in the attorney himself being prosecuted for protection of a heretic, or for impeding the Inquisition (Lea 1907: v. 3, 43). By 1562, even if the accused found his/her own attorney, the official defense attorney was prohibited from communicating with him. “The advocate thus became one of the officials of the tribunal, duly salaried and working in full accord with the inquisitors” (Lea 1907: v. 3, 46). His main task was to advise the accused to confess, and throw himself/herself on the tribunal’s mercy. Defense attorneys believed that spontaneous confession would result in more lenient punishments (Haliczer 1990: 64).
So, rather than a vigorous advocate, the accused got an extension of the inquisitor, someone who would explain how hopeless one’s case is, and how one ought to confess promptly and fully, in hopes of long term salvation and short term peace of mind. Nor were defense attorneys the only such agents. Inquisitors commonly supplied cell mates to inform against the accused, or would eavesdrop on prison conversations, either personally or through the guards (Lea 1955: v. 1, 416-417; Boguet 1929: art. 18).
In all of this, we see further constriction of the disagreement space. The sorts of defenses 20th century Westerners are used to, were essentially forbidden, and all that argumentative opportunity lost. In the place of a defense, the Inquisition supplied more and more opportunity for confession, the only desirable sort of talk from a defendant.

3e. Reflexive Arguments
The initial assumption that the accused was guilty colored the meaning (to the Inquisition) of everything the defendant might say. On certain topics, to deny a charge was to prove it. In an earlier paper, I called this reflexive argumentation (Hample 1997). One illustration in that essay was from Joan of Arc’s trial. Accused of heresy, she was asked to justify her actions, including the obviously sinful ones of wearing men’s clothing and not obeying her parents. Her explanation was that her Voices instructed her to do these things. Since Satan is wily and can assume the form of angels and saints, her answer proved that she was willingly being influenced by the Devil, and thus self-evidently a heretic. Her only available answer to the charge proved its truth.
This was a much more common problem for defendants than that one example suggests. To deny one’s heresy was, in general, to prove it. “Persistent denial of guilt and assertion of orthodoxy, when there was evidence against him, rendered him an impenitent, obstinate heretic, to be abandoned to the secular arm and consigned to the stake” (Lea 1955: v. 1, 407). To the inquisitors, this made perfect sense: the accused was guilty, after all (Eymeric always refers to the accused as the criminal, or the guilty one; Walsh 1969: 107). Denials were proof, not of innocence, but of resistance to the Church Militant. Those who refused to confess were simply sent back to prison, and admonished to examine their consciences more fully; thus, some prisoners lived out their days in jail without their trials being concluded or even properly begun.
If defense was unlikely, denial was pointless. Even at the level of simple assertives, the Inquisition exercised its power to constrict the disagreement space.

3f. Imposition of Theoretical Frames
To this point, I have only considered ways in which the Inquisition restricted the accused. In some ways, however, it also reduced what the inquisitors themselves were able to think, say, and hear. The problem is again the manuals, along with the other elements of an inquisitor’s education. Witchcraft, for instance, was a real thing, because the manuals said it was. Heresies had to be recognized, and so testimony had to be fit to the manuals’ descriptions of earlier heresies.
Ginzburg (1983) describes the difficulties that Fruilian inquisitors had when confronted with the unprecedented benandanti (doers of good) in the 16th and 17th centuries. The benandanti said that they left their bodies to do battle with witches several times a year, with the harvest at stake. They also said that they could interfere with witches’ spells, and often saved children from ensorcered deaths. Were they witches? If they were, could there be such a thing as a good witch? They said they fought for Christ; they made no pacts with the Devil; they did no evil. But as time went on, the inquisitors focused more and more on those elements of benandanti practice that resembled witchcraft: out of body night flights, animal guides, facility with spells, and so forth. More and more, they pressed the benandanti peasants, and they described these practices in inquisitorial terms in the public sentencing sermons. By the mid-17th century, the Inquisition had completely assimilated the benandanti to witchcraft. Significantly, the later testimonies of the accused fit that frame as well, although the early trials reveal little, if anything, like black magic.
Sullivan (1996) argues that a similar thing happened to Joan of Arc, as regards the identities of her Voices. Prior to her trial, she only spoke of her revelations as being from God, and early in her condemnation trial, she was no more specific. But the inquisitors pressed her, and she finally, reluctantly, identified the Voices as being from Saints Michael, Catherine, and Margaret. By the end of the trial, Joan gives these attributions more and more spontaneously. Sullivan says that this resulted from the inquisitors’ insistence that the Voices had to have been from God, an angel, a man, or a devil: Joan had to choose, and so she did. We see here, as we did with the benandanti, how the inquisitors’ frame can constrict both the judge and the accused.
Peters (1988: 20-21) says that the early inquisitors in southern France made sense of what they were learning, by fitting the testimony into the ancient heresies of Manichaeism, Gnosticism, and Donatism, theologies whose adherents’ bones had turned to dust a thousand years before. Theoretical frames can be sturdy structures, no less constricting on the theorist than on his/her human data.

3g. Rules of Evidence and the Evaporation of Witchcraft
We have already seen how the Inquisition established rules of evidence and procedure that made guilt almost unavoidable. The confession was the foundation of the Inquisition. Once a full confession was in hand, further inquiry was halted, and the judge moved on to the sentence. Such uncritical acceptance of this sort of evidence could pose problems, however. Although the great witch hunts had many causes and were sustained by many cultural currents (see Russell 1972), one contributing element must have been the inquisitors’ willingness to believe the confessions they heard and coerced. When the Inquisition decided to treat the confessions more critically, the persecutions ended.
The Spanish Inquisition, alone among those in the Christian world, decided early on that witchcraft was not real, and therefore that they would not prosecute it. This is an important story, told in detail by Henningsen (1980). The Spanish began with the same witchcraft concerns as other nations, but in Spain, the skeptics won out. One of them, Alonso de Salazar Frias, was appointed the third member of the Logrono tribunal in June of 1609. Salazar took to heart the Suprema’s somewhat critical instructions, and actually undertook scientific tests of witches’ claims. He sought out actos positivos (i.e., witches’ actions that could be tested by the testimony of non-witches), he tried to determine if witches who said they had attended the same aquelarre told the same story about it (they were actually taken to the spot individually, and asked where the Devil sat, etc.), he searched out and investigated the ointments said to be used by the witches (discovering that the earlier inquisitors had been tricked into thinking that they had found authentic ointments; he also witnessed a witch consuming a magical powder, with no apparent effect) (Henningsen 1980: 295-301). Based on Salazar’s report, the Suprema essentially declared that witchcraft was no longer to be punished in Spain, and even permitted confessed witches to abjure their earlier confessions without penalty (normally withdrawing a confession of heresy would be self-evidence of perjury) (Henningsen 1980: 371-376). Witchcraft thus disappeared from Spain because the Inquisition would no longer entertain evidence about it.

A similar thing happened in regard to the Salem witch trials, a Protestant inquisition. These trials stopped when the Governor, on advice from clergy, determined that the Devil could impersonate innocent people, so that testimony that a person had been seen or touched while engaged in witchcraft was no longer sufficient or even on point (letter, Gov. William Phips to Earl of Nottingham, 21 February 1693, in Boyer & Nissenbaum 1972: 120-122). This simple change in the rules of evidence eliminated the possibility of prosecution, and so ended the Salem persecutions.
Just as rules of evidence could be used to constrict the disagreement space in ordinary heresy trials, they could also make certain sorts of heresy impossible, shrinking the disagreement space in another way. Perhaps I should remind readers that my point in this paper is not to show that the Inquisition was an irrational institution (modern historians now judge that it was actually somewhat advanced in comparison to secular courts; Peters 1988; Del Col 1996: xxvii). Rather, I want to explore how disagreement space can be regulated by means of resources external to the immediate discourse.

4. Inquisitorial Enlargement of the Disagreement Space
I was drawn to this topic by the realization that the Inquisition functioned in large part by restricting the accused’s opportunities for argument, as I have shown. However, the Inquisition also increased the scope of discourse in two respects. These were not topics that the accused necessarily wanted to discuss, nor were they even topics that the Inquisition knew existed.
The first enlargement occurs because of the Inquisition’s wish that the trial begin with a confession, even before the charges are heard. The inquisitor’s first question would be, “Do you know why you are here?” (Gui 1991). A witness who claimed not to know might well not be informed, but instead be sent off to prison to meditate. The idea was, of course, to have a spontaneous confession, indicating true contriteness and a strong desire to be reconciled to the Church. A noticeable result, however, was that inquisitors often learned things they had not expected to hear. This was a useful resource in pursuing both the instant and later cases, and systematically enlarged the disagreement spaces for those defendants. In fact, it may have created disagreement spaces for some.
The second enlargement is also connected to the Inquisition’s model of a satisfactory confession. To confess a sin, one must hate it, and must see that it is an evil aimed at Christ’s Church. Therefore, it is not enough to confess one’s own involvement. A true confession will also name everyone else known or suspected of guilt by the accused, for their souls are in jeopardy. To withhold the identities of one’s accomplices was to make only a partial confession, putting the accused at risk of being held to be impenitant (the penalty for which was burning at the stake). These accomplice lists were one of the Inquisition’s main resources in rooting out heresies.

5. Implications for Argumentation Theory
The point of this paper has been to explore disagreement space, as it is affected by institutional power. Hutchby (1996) has done a similar sort of thing in his excellent study of how a radio talk show host controls the topic on his call-in program. By choosing what elements of the caller’s talk to regard as arguable, by reining the caller back when s/he tries to move to a new subject, by having the last word, and by other means as well, the host strongly influences what we are here calling the disagreement space. These possibilities all derive from the host’s institutional status, and his allied control of the radio show’s technology. They are, however, all implemented in the actual talk, and it is there that Hutchby finds his evidence. I have not here undertaken anything like conversation analysis of inquisitorial trials, but I have little doubt that a Hutchby-like analysis would succeed, even though we do not have verbatim transcripts.

Both Hutchby’s work and the present analysis demonstrate that disagreement space is, in practice, not equally available to all parties. Another way of saying this is that the theoretical disagreement space (containing all the beliefs that an analyst sees as potentially arguable) is not the same as the disagreement space as it exists in a real, situated argument. While I do not read van Eemeren, Grootendorst, Jackson, and Jacobs (1993) as having taken much notice of this, I doubt that they would object to this conclusion. I think that this is all connected to a more fundamental idea in pragma-dialectics, the rules for critical discussion (van Eemeren & Grootendorst 1992). These are the norms that need to be respected for a good dialectical argumentation. To violate one is to commit a fallacy. Here is the list:
1. Parties must not prevent each other from advancing standpoints or casting doubt on standpoints.
2. A party who advances a standpoint is obliged to defend it if the other party asks him to.
3. A party’s attack on a standpoint must relate to the standpoint that has indeed been advanced by the other party.
4. A party may defend his standpoint only by advancing argumentation relating to that standpoint.
5. A party may not falsely present something as a premise that has been left unexpressed by the other party or deny a premise that he himself has left implicit.
6. A party may not falsely present a premise as an accepted starting point nor deny a premise representing an accepted starting point.
7. A party may not regard a standpoint as conclusively defended if the defense does not take place by means of an appropriate argumentation scheme that is correctly applied.
8. In his argumentation a party may only use arguments that are logically valid or capable of being validated by making explicit one or more unexpressed premises.
9. A failed defense of a standpoint must result in the party that put forward the standpoint retracting it and a conclusive defense in the other party retracting his doubt about the standpoint.
10. A party must not use formulations that are insufficiently clear or confusingly ambiguous and he must interpret the other party’s formulations as carefully and accurately as possible (van Eemeren & Grootendorst 1992: 208-209).

These were all systematically and intentionally violated by inquisitorial practices (working through the proof of this would require another whole paper, so I will leave it as an exercise for the reader: enough evidence should be available in the present essay). This suggests very close connections among institutional power, control of disagreement space, violation of the rules for critical discussions, and systematically fallacious discourse. The pragma-dialectical school has concentrated its energies on the analysis of discourse, trying to avoid both cognitive and macro-sociological issues. It has achieved a lot with this strategy. However, it now seems likely that pragma-dialectics can broaden its own domain of applicability if it wishes, and engage in analysis and critique of larger social institutions, based upon how they affect disagreement space.

REFERENCES
Anonymous (1980). A manual for inquisitors at Carcasonne, 1248-49. In: E. Peters (Ed.), Heresy and Authority in Medieval Europe: Documents in Translation (pp. 200-206). Philadelphia: University of Pennsylvania Press. This is the Processus Inquisitionis.
Barrett, W. P. (trans.) (1931). The Trial of Jeanne d’Arc. London: Routledge.
Boguet, H. (1929). An Examen of Witches. Trans. by E. A. Ashwin. NP: John Rodker. This is the Discours des Sorciers, originally published ca. 1590.
Boyer, P., & Nissenbaum, S. (Eds.) (1993). Salem-Village Witchcraft: A Documentary Record of Local Conflict in Colonial New England. Boston: Northeastern University Press.
Burr, D. (1998). Medieval Sources in Translation. http://dburr.hist.vt.edu/.
Del Col, A. (1996). Domenico Scandella Known as Menocchio: His Trials before the Inquisition (1583-1599). Trans. by J. Tedeschi & A. C. Tedeschi. Binghamton, NY: Medieval and Rennaissance Texts and Studies.
Eemeren, F. H. van, & Grootendorst, R. (1984). Speech Acts in Argumentative Discussions. Dordrecht, Holland: Foris.
Eemeren, F. H. van, & Grootendorst, R. (1992). Argumentation, Communication, and Fallacies: A Pragma-Dialectical Perspective. Hillsdale, NJ: Erlbaum.
Eemeren, F. H. van, Grootendorst, R., Jackson, S., & Jacobs, S. (1993). Reconstructing Argumentative Discourse. Tuscaloosa, Alabama: University of Alabama Press.
Ginzburg, C. (1980). The Cheese and the Worms. Trans. by J. Tedeschi & A. Tedeschi. Baltimore: Johns Hopkins University Press.
Ginzburg, C. (1983). Night Battles: Witchcraft and Agrarian Cults in the Sixteenth and Seventeenth Centuries, trans. by J. Tedeschi & A. Tedeschi. New York: Penguin.
Given, J. B. (1997). Inquisition and Medieval Society: Power, Discipline, and Resistance in Languedoc. Ithaca, NY: Cornell University Press.
Gui, B. (1991). The conduct of the inquisition of heretical depravity. In: W. L. Wakefield & A. P. Evans (Eds.), Heresies of the High Middle Ages (pp. 373-445). New York: Columbia University Press. This is Part V, the procedural part, of Gui’s Practica Inquisitionis Heretice Pravitatis, dating from about 1323-1324.
Haliczer, S. (1990). Inquisition and Society in the Kingdom of Valencia 1478-1834. Berkeley, CA: University of California Press.
Hample, D. (1997, August). Reflexive argument: Joan of Arc and other victims. Paper presented to the AFA/NCA Summer Conference on Argumentation, Alta, UT.
Henningsen, G. (1980). The Witches’ Advocate: Basque Witchcraft and the Spanish Inquisition. Reno, NV: University of Nevada Press.
Hutchby, I. (1996). Confrontation Talk: Arguments, Asymmetries, and Power on Talk Radio. Mahwah, NJ: Erlbaum.
Kieckhefer, R. (1979). Repression of Heresy in Medieval Germany. Liverpool: Liverpool University Press.
Kramer, H., & Sprenger, J. (1971). Malleus Maleficarum. Trans. by M. Summers. New York: Dover. Originally published, ca. 1486.
Lea, H. C. (1907). A History of the Inquisition of Spain, 4 vols.New York: MacMillan.
Lea, H. C. (1955). A History of the Inquisition of the Middle Ages, 3 vols. New York: Harbor Press. Reprint of 1887 edition.
Le Roy Ladurie, E. (1978). Montaillou: The Promised Land of Error, trans. by B. Bray. New York: Braziller.
Peters, E. (1988). Inquisition. New York: Free Press.
Russell, J. B. (1972). Witchcraft in the Middle Ages. Ithaca, NY: Cornell University Press.
Stork, N. (1996). The Inquisition Record of Jacques Fournier Bishop of Pamiers 1318-1325.
http://www.sjsu.edu/depts/english/Fournier/jfournhm.htm.
Sullivan, K. (1996). “I do not name to you the voice of St. Michael”: The identification of Joan of Arc’s voices. In: B. Wheeler & C. T. Wood (Eds.), Fresh Verdicts on Joan of Arc (pp. 85-112). New York: Garland.
Trask, R. B. (1992). “The Devil Hath Been Raised”: A Documentary History of the Salem Village Witchcraft Outbreak of March 1692. West Kennebunk, ME: Phoenix.
Walsh, W. T. (1969). Characters of the Inquisition. Port Washington, NY: Kennikat Press. Originally published, 1940

‘Reported speech’ (RS) in journalistic texts has been the object of linguistic investigations in the last decade. It has taken place in the greater frame of studies of speech and thought presentation (Slembrouck, 1992, Waugh, 1995, Semino et al., 1997), or approached from a more (journalistic) discourse point of view (Geiss, 1987, Zelizer, 1989), with a special interest for the speech-act verb introducing the reported speech (Geiss, 1987, Gruber, 1993, May, 1995).
In an often quoted study of French journalistic discourse, Waugh (1995) highlights the presence of what she calls a combined form of ‘indirect-direct’ speech, of high frequency in the French press, and particularly in Le Monde. Example (1), taken from another broadsheet Le Figaro, is illustrative of such a form:
(1)
Il avait demandé au premier ministre <au nom de la transparence>, de lever le secret-défense, et expliqué, évoquant François Mitterrand, que cette <triste histoire> ne correspondait pas <à l’idée> qu’il se faisait <de lui sur ce plan> (Le Figaro, 9 avril 1997, ‘Les écoutes brouillent les socialistes’)
The relative sizes of indirect and direct stretches may vary considerably from case to case; furthermore, this ‘partial quoting’ procedure is not restricted to an indirect speech frame, but is also frequent with forms of NRSA (Narrative Report of Speech Act), as pointed out by Semino et al. (1997) for English reporting. Semino et al. pay special attention to these small stretches of direct quotation, which they call ‘Q-forms of speech presentation’[ii], and stress on the one hand their importance and convenience in journalistic writing, on the other their potential to lead to bias:
‘They allow the reporter to foreground selected parts of the original utterance without having to provide a lengthy quotation. They achieve vividness and precision without sacrificing the need for brevity. Clearly, such forms also lend themselves to partial or slanted representations of other people’s voices, since the original speaker’s words are embedded, both grammatically and semantically, within the reporter’s own discourse.’ (31)
This grammatical and semantical dependency, combined with the limited length of discourse, makes it difficult to refer to these forms as direct speech. I will argue that it also does not do credit to the special function these partial quotes often have. Looking back at example (1), we see that quotations in the French press are marked by low, double quotation marks and  italics.

In headlines, subheadings and introductory paragraphs no italics are used, which leads to a potential ambiguity. The quotation marks could then refer to either a real quote or to a special use or meaning attached or given to the words quoted. In the last case, they would be so-called ‘scare quotes’. This potential ambiguity is essentially present when the quote is limited to one or two words:
(2)
[h23] Soupçonné de <favoritisme>, le maire de Nantes et président du groupe PS à l’Assemblée plaide la <bonne foi>
(3)
(sh26) Le président sud-africain critique le <gendarme du monde> qui <n’a pas de morale>

In (2), only the second fragment of discourse between Q-marks was confirmed as a real quote. In (3) ‘Gendarme du monde’, referring to the United States, is used by the journalist in the text of the article between quotation marks but without italics. It is also used by Nelson Mandela in his declaration (or at least translated as such), so that it could be given both functions.

In this paper[i], I aim to show that partial quotations in headings, especially those limited to one or two words, can use their identity of form with ‘scare quotes’ to maintain a welcome ambiguity. When the text of the article confirms that the words have been said, the ‘evaluative’ or ‘attitude’ function of partial quotations, as described by Weizman (1984), seems to be generally present, staging a particular form of polyphony (Bakhtine, 1973, Ducrot, 1984). The process of the ‘entextualization’ of small stretches of discourse (Bauman & Briggs, 1990, Silverstein & Urban, 1996), through decontextualization and recontextualization, as observed in a corpus of headings from Le Monde, will reveal some aspects of the relation between journalistic writers and social power.

Quoting in journalistic discourse
Zelizer (1989) states that approaching quoting practices can allow us to ‘understand the large arena of journalistic behavior, and its relationship to society’ (370). Quoting is the one device that reflects directly on sourcing/evidentiality, hence it increases the authority, or legitimacy, of statements into which quotes are embedded. On the one hand, quotes lend factuality, precision, credibility and truthfulness[iii]. To quote is also to lend authority to largely unspecified sources behind the news: ‘By relying on quotes from those who are supposedly ‘involved’, journalists emphasize a posture of technical neutrality’ (373), but, Zelizer adds, ‘journalists might use quotations in a way to reinforce different aspects of their own authority’(374). So, on the other hand, quotes are testimonies to the journalist’s interface with events, ‘actors’ and institutions. Quoting can also be seen as a testimony to the journalist’s ‘creative’ handling of someone else’s discourse. Mouillaud & Tétu (1989) reminds us that what the modern press is using as raw material (‘matière première’) for its writing is most of the time not a factual reality, but an already existing discourse, social and often institutionalized. So journalistic practice is essentially a practice of ‘entextualization’, that is ‘the process of rendering discourse extractable, of making a stretch of linguistic production into a unit – a text – that can be lifted out of its interactional setting’ (Bauman & Briggs, 1990:73). Following Bauman and Briggs, I will consider decontextualization and recontextualization to be two aspects of the same process of entextualization: ‘Because the process is transformational, we must now determine what the recontextualized text brings with it from its earlier context(s) and what emergent form, function, and meaning it is given as it is recentered’ (75).

Recontextualization will lead to polyphony, as it will present the ‘voices’ of the speaker and the journalist in various degrees of distinction or fusion. ‘To decontextualize and recontextualize a text is [thus] an act of control, and in regard to the differential exercise of such a control the issue of social power arises’, say Bauman & Briggs (1990:76), with a reference to Bourdieu.

Let us look at this process when applied to journalistic quoting. In the two phases of the process, we can globally describe the writer’s various degree of exercising his freedom. In the first place, selecting a quote will be done according to criteria of relevancy and saliency in relation to the topic of the news, and depending on the speaker being himself either the topic, or being just involved, as actor, in the event that constitutes the topic. On that level the freedom of the writer is maybe the most limited, as his choice is a result of the newsworthiness of the information and the newspaper’s genre.
In editing the original speech, designating the speaker and qualifying his speech act, the journalistic writer[iv] acts quite freely, within stylistic and ethical constraints. The choice of a reporting mode, direct speech, indirect speech or any other forms, will most probably be related to the content of the utterance, and the goal of the reporting; designating the speaker(s) will follow other criteria, relative to the source of the quoted discourse, institutional or private, open or secret, the status of the source and the content of the discourse. The choice of the speech act verb could be traced to the illocutionary nature of the original speech act, to some prosodic information perceived directly by the journalist (or taken over from other news reports) or to inferences from the content, not to mention personal stylistic preferences.
The preceding remarks do not mean that we can always, as much as we would like to, discover why and how the writer made his choices. Even going back to the original discourse and comparing it to the reporting does not univocally clarify these choices.[v] In the case of partial quotes, especially the very short ones, even a ‘simple’ reader of the news can sometimes wonder why the journalist has chosen to quote such a limited stretch of discourse, as a longer quote would have been more informative on the subject and contributed more to the credibility and liveliness of the report.

Partial quotation and evaluation
Weizman (1984) makes a distinction between two functions of quotes: a reliability function, toward the addressee/reader to assure him of the reliability of the reporter, and an attitudinal function that refers to a whole variety of attitudes of the reporter towards the utterance, the quoted source or the event, ‘from reservation to ironic rejection’ (41). Attitude utterances do have one common feature, she writes,‘the incompleteness of the utterance in quotation marks’ (42). Weizman describes other possible markers of this attitudinal function: the presence of objective and subjective affirmations in a mixed form of reported speech, for instance indirect speech with a quote, the presence of ‘emotionally-loaded words’ within the quotation (connotations), the length of the discourse unit within the quotation marks, etc.

Her findings are confirmed by Gruber (1993). In his analysis of evaluative devices in journalistic reports, Gruber describes explicit and implicit forms of evaluation on the part of the journalist. An explicit evaluation can take the form of scare quotes, an implicit one, the form of a short quote. In agreement with Weizman, he concludes:
‘The smaller the reported discourse unit is, the more the function of the quotation marks shifts from “reliability function” to the “attitude function”’ (472)

The empirical study whose results I will present in the following section is based on and limited to one French broadsheet, Le Monde, and to a special type of discourse, the headings of articles. The choice of Le Monde is motivated by the frequency of quotations, and of partial quotations, in articles (See Waugh, 1995, Mouillaud & Tétu, 1989). In comparison with such other broadsheets as Le Figaro, or Libération, most events are covered by Le Monde, in many more words and quotes.[vi] The choice to analyse headings is related to the leading function of headings, since reading headings is what we usually do before reading the article itself or moving on to another part of the newspaper. Often, our knowledge of facts is based on a superficial reading of headings, as can be some of our opinions on facts, events and people. A personal experience was also essential in choosing this particular corpus. Reading, in the winter of 1998, the following headline in Le Monde:

(4)
[h53] Martine Aubry <se donne du temps> pour la politique de la ville

my first impression was of a biased heading, as, to my eyes, it seemed to be giving the wrong impression of someone known to be a hard-working, very professional (woman) politician. In this subjective, and possibly illogical reaction of mine, the use of quotation marks without italics played a part. Finding out that, in the text of the article, this attributed act or attitude was a wish (‘elle souhaite <se donner du temps>’) partly confirmed the biased impression I got at first reading.

The data: Partial quotes in headlines and subheads[vii] in Le Monde 47 issues of Le Monde provided 142 examples of headings with 160 partial quotes (most had one quote, 11 had between 2 and 5 quotes) of the following structure[viii]:
– an explicit designation of the quoted speaker (where the speaker could refer to a person or an institution)
– a speech-act verb
– a partial quote, equal to or smaller than a clause

The selection of the headings was partially random: during the period October 97-March 98, 3 to 4 issues per week of Le Monde were systematically searched for headings with partial quotes. The first observation was that partial quotes in headings were closely related to a certain topic or domain: only 3 of the 142 headings did not have a political topic (one religious topic, 2 referred to ‘faits divers’/‘news in brief’). 117 referred to French politics or affairs with a political dimension, 21 to international politics. In 90% of the cases the speaker was an important actor on the political stage; in the remaining 10%, it was a ‘second-rank’ actor who played an important role in the represented event or affair, with important political issues in the present time (financial and political scandals) or in the past and present (Papon trial). In Mouillaud and Tétu’s terms, the newspaper is handling the voice that is speaking not as a ‘source’ but (following Greimas’s terminology) as a ‘actant’, whose ‘saying’ is ‘doing’ (le dire est un faire): ‘L’actant a un intérêt stratégique à produire un discours et celui-ci peut être considéré comme une intervention dans le jeu des interactions de l’espace public’(131), they write, referring to Habermas.

Two basic structures were found in the headings, illustrated by (5) and (6):
(5
(h25) Jacques Chirac a indiqué que la France soutiendrait les <efforts> tunisiens
(6)
(h12) M.Chirac insiste sur la nécessité d’un dialogue social <efficace et confiant>

The first, a form of indirect speech with a quote, was found in 24 of the 142 occurrences. The second, in the terms of Semino et al. (1997) a NRSATQ, that is a narrative report of speech-act with Topic and Quotation, was found in the remaining 118 occurrences. There seems to be a preference for the second structure, which shows a greater integration of the different voices heard. In 7 of the occurrences, it was the speech-act verb that appeared between quotation marks, as in:
(7)
(h62) M.Pasqua <dénie> à M. Chirac le droit de ratifier le traité d’Amsterdam sans référendum

Prior to any analysis, the nature of the quotation marks had to be verified. In other words: was the quote in the headine a real quote, that is, could it be found as quote in the text?

Only 3 quotes could not be found in the article, 2 of these were found in another article on the same page. 7 recurred in the article as ‘scare quotes’, most but not all in a quotation. One of them illustrated, in retrospective, a case of referential ambiguity,
(8)
(h92) Amnesty International dénonce la <terreur en Casamance> as the discourse between quotation marks was found to be the title of a written report by Amnesty International[ix].

Only 1 quote was found in a canonical form of indirect speech, with neither quotation marks nor italics. 126 of the 160 partial quotes were found as partial quotations in the text. So, in my (limited) data, the real quotes were largely dominant, and the ambiguity of the marks, especially with short quotes, was more subjective than could have been thought originally. However, this does not say anything yet about the attitude function of the quotes.

Analysis
In order to cast light on the functioning of partial quotes in headings, I chose to look first at the quote, that is at what was quoted and the relation between the length of the quote and the content.
In the second place, I looked at the frame in which the quotation was placed in the heading. Besides the type of speaker – as said above, it was dominantly a political actor -the verb used by the journalist or editor was going to be of some importance, as has been shown in different recent studies.
Studying political bias in news magazines, Geis (1987) examined among several linguistic modes of bias, the ‘attribution bias’ that concerns the choice of verbs of reported speech. In a data base of 534 cases of reported speech, Geis found 489 occurrences of verbs and 133 different verbal items, a great number of which were what he called ‘high volatility’ verbs, verbs that tend to vary widely in how they are perceived by subjects. Usually those verbs are not standard verbs of reported speech and tend to have a negative affect, with the following potential consequences:
‘It should be clear that journalists can enliven their reports by using affectively charged verbs of reported speech or verbs that are high in volatility, but they pay a price, which is that they will be a corresponding increase in potential for bias (the negative affect of volatile verbs) and a corresponding decrease in clarity.’(130)
Geis’s findings are confirmed by empirical studies by Gruber (1993) and May (1995). Burger (1997) found a similar variety of speech-act verbs in French journalistic discourse: about 300 different verbs in a data of 1300 occurrences of reported speech in Le Monde. The most neutral verbs were used with DS; the marked verbs, more action-describing verbs than strictly speech-act verbs, were found with IS, NRSA and other (mixed) forms, confirming observations made for English/American (Geiss, 1987, May, 1995) and for German (Gruber, 1993) journalistic texts. As the studies mentioned here show, the great diversity of verbs is not only the result of a stylistic choice (bringing variation) but reflects the possible ways offered to the journalist to transmit either an evaluation by the speaker cited of the fact or event he was referring to, evaluation taken over by the journalist, or an evaluation by the journalist of the speaker or of the event/fact that is reported. In the case of headings, we could add an editorial decision, that is to stress some aspect of the information, or to arouse the interest of the reader and invite him to read further.
Finally, as the quotes of the headings were in about 95% of the cases found in their ‘original’ frame in the article, it was tempting to compare heading and text, in order to describe the process of de/recontextualization.[x]

The Quote
67 of the quotes consisted of one-word quote (33 Nouns, 16 Verbs, 14 adjectives, 4 adverbs). The rest presented two or more words of which 61 were NP’s, 17 VP’s, and only 10 were clauses such as
< alors que la croissance est de retour>
One-word quote

Out of context, the adjectives and nouns that were quoted have a clear positive or negative meaning. There are two potentially neutral adjectives (personnelle, plurielle[xi]) as opposed to 13 clearly positive/negative ones: idéologique, validée, impartiale, équitable, etc. Nouns show a similar distribution out of context: 10 had a positive meaning/connotation; 16 were negative and only 5 could be said a priori neutral.

As examples:
– écoute +[xii]
– dialogue +
– débat +
– complot –
– erreur –
– instructions N
– repentance +
– efforts +
– insultes –
– infractions –
– arrogance –
– barbarie –

In context, the positive affect could be nuanced or denied as in the following examples:[xiii]
(9)
(h5) La direction du PCF [French communist party] reconnaît un
<débat> interne
(10)
(h94) L’accusé [Papon] a, une nouvelle fois, mis en avant ses <efforts> pour sauver des juifs

The quotations in examples (10) and (11) could be qualified as polyphonic: we hear the voice of the PCF and of Papon, and also the voice of the journalist that seems to express his doubts about the possibility of a debate within the PCF and about the nature of Papon’s effort to save the lives of Jews. (10) and (11) are examples of evaluation by the journalist, of the implicit kind that Gruber mentioned in his study.

The verbs presented as quotations are also mostly polarized:
regretter, bloquer, approfondir, se donner du temps, regarder en face, dénier, (vouloir) dialoguer, dialoguer, etc.

7 out of the 16 found in the data had a double function as quote and speech-act verb:
(11)
(h9) H.Emmanueli <regrette> la hausse des taux de la Banque de France
(12)
(h62) M.Pasqua <dénie> à M.Chirac le droit de ratifier le traité d’Amsterdam sans référendum
(13)
(h84) E.Balladur <dialogue> avec les électeurs sur l’emploi

Note that this double function gives a ‘performative’ dimension to the predicate. It has also the same effect as the one described for nouns and adjectives. It illustrates the existence in the quoted utterance of two voices, the second one, that of the journalist seeming to express a reserve or doubt about the sincerity, the well-foundedness or the reality of the (verbal) act, or, on the contrary, to emphasize the sincerity, the well-foundedness and the reality of the verbal act.

Two-words or more quotes
NP’s showed a similar classification to the nouns, with a clear positive or negative interpretation/connotation (37 positive/negative) or a neutral or undetermined meaning (24). This last group is rather greater than in the one-word group. This would indirectly confirm Weizman’s hypothesis: the longer the fragment of discourse between quotation marks is the less marked, positively or negatively it appears to be. Examples are:
– signe fort +
– respect de l’autre +
– seconde équipe N
– la cohérence permanente +
– sauvagerie extrêmement froide –
– mythe gaulliste +/-?
– situation ubuesque –
– bonne foi +

‘Mythe gaulliste’ is an example of ambiguous value: ‘mythe’ can refer to a construction that has no reality, that is, lies or delusions. It can also refer to or be an expression of the grandeur of someone or something.
The use of quotation (scare) marks around a small unit such as an adjective, a noun or even a noun phrase captures the reader’s attention immediately. The positive or negative connotation is (potentially) reviewed in the light of the context, mixing information derived from the way the speaker is referred to, the topic, the addressee and the speech-act verb. ‘Any of these ingredients can have an effect on how the citizenry will evaluate such a report’ says Geis (126). However, if the reader’s view on the topic and the speaker can have a bearing on his interpretation of the report, the evaluation of the affect of the verb (out of context) will not be influenced as much by personal political opinions. Let us now look at the verbs.

The verbal frame
There were 67 different speech-act verbs in the data. Canonical ‘speech-act’ verbs represent about 30% of the verbal items and also 30% of the occurrences. The other verbs are less neutral speech-act verbs, and tend to become/to be action-describing verbs. These observations confirm what has been said about introductory verbs of reported discourse in journalistic discourse.

The most frequent verbs are:
type tokens
1. estimer (que) 12
2. dénoncer 11
3. vouloir + verb 9
4. accuser 6
5. souhaiter (que) 6
6. appeler à 5
7. affirmer (que) 5
8. assurer (que) 5
‘
Just looking at this small list makes the problem of the verb classification obvious. Only one could be said neutral, estimer (que), though this neutrality seems to weaken in NRSATQ structures (for instance in:‘X estime <scandaleux> le comportement du premier ministre’). All the other verbs are susceptible of different interpretations, and hence categorisations. For Mouillaud & Tétu, one way to assess the epistemic judgment (‘le croire vrai’) is to modalise the verb phrase in the main clause, which they describe as happening in 3 major ways:
‘
– distance. Ex: affirmer, prétendre
– agreement. Ex: souligner, faire ressortir
– neutrality. Ex: dire, déclarer
‘
Such a classification (illustrated with a few examples) is not based on an empirical study, and interpretations of, for instance ‘affirmer’, could vary according to the speaker or the context.[xiv]  Geis proposes a classification of verbs of positive/negative affect and low/high volatility on the basis of an experiment. Verbs could also be classified according to the illocutionary act they perform, if this act could be consensually determined. The problem is that the verb’s value is mostly determined out of context, or, in Geis’s study, in a constructed context that is the same for all verbs submitted to a native speaker for evaluation; then, this interpretation has to be confronted with the context of the data, leading to adjustments of the first evaluation or ambiguity judgments.[xv] As no cross-analysis of the data will be proposed here, I will limit myself to a global classification of verbs and to some examples, mainly to illustrate the diversity:
– ‘neutral’ speech-act verbs/verbs of ‘thinking’
croire (que) 1
déclarer (que) 5
décrire 1
demander 4
dire 1
estimer (que) 12
évoquer 2
….
I found 26 verbs in that category. As to the others (41), a first group could be qualified as verbs of argumentation:
approuver 2
contester 1
démentir 1
dénier 1
exclure 1
rejetter 1

A second group would be the group of action verbs:
accuser 6
appeler à 5
attendre 1
condamner 1
conseiller 1
critiquer 1
défendre 2
dénoncer 11
exhorter 1

ATQ occurrences were globally marked verbs; speech-act verbs used in a ISQ structure tend to be more neutral, with few exceptions such as ‘prétendre que’ which is a explicit evaluation by the journalist:[xvi]
avouer que 1
affirmer que 2
assurer que 4
déclarer que 1
estimer que 8
prétendre que 1

From text to heading
Three features will be taken into consideration in the comparison between the quote and its context in the heading and the quote in the text: the designation of the speaker, the introductory verb and the quote itself.
39 headings were found in their article without significant changes. 103 headings showed some transformation. Following the natural process of reading, I will go from heading to text.

The speaker
There were two major types of changes around the designation of speaker, from general (in the heading) to more specific in the article, from important to less important:

(14)
[sh20]: Lors d’une soirée souvenir à Paris, dont l’émotion a été avivée par la coïncidence avec le procès Papon, Algériens ayant subi la répression et Français présents sur les lieux des massacres ont raconté la <sauvagerie extrêmement froide> des policiers a(rticle): <C’était d’une sauvagerie extrêmement froide>, témoigne J-L Péninou, journaliste….

In most cases, we could speak of a metonymic substitution that emphasizes the status of the speaker, qualitatively or quantitatively speaking.

The verbs
The quote was found in the article
– without a speech-act verb, in a Direct Discourse quote, or similar forms such as ‘Propos’ or ‘Interview’[xvii]34
– with the same S-A verb 25
– with a different S-A verb 75

The changes in verbs are related to the form of RS that is used in the text:
exclure (NRSAT) becomes expliquer que with ISQ
démentir becomes affirmer with DS and negation

or to a different speaker and/or a different voice:
(15)
[h35] les Etats-Unis accusent Israël de pénaliser les <intérêts américains dans la région>
a: le chef du gouvernement israélien (…) devait s’entendre dire clairement que sa politique porte désormais <atteinte aux intérêts américains dans la région>

As expected, given the ‘leading function’ of headings, most changes show a stronger, less nuanced formulation in the heading than in the text. For instance, approuver becomes refuser de polémiquer, saluer becomes mêler compliments et conseils fermes, or as in the following example:
(16)
[h70] M. Jospin salue <l’approfondissement de la démocratie> au Maroc
a: Le premier ministre a indiqué qu’il voit <avec intérêt le constant approfondissement de la démocratie marocaine> conduit <sous l’impulsion> de Hassan II.

The quote shows different kinds of modifications:
– an argumentatively significant reduction
(17)
[h19]: <L’Etat ne saurait tout faire> pour la coopération, a déclaré M. Josselin
a: Charles Josselin a déclaré, qu’en matière de coopération, <l’Etat seul ne saurait tout faire>
(18)
[h77]t: A la Havane, le pape exhorte Cuba à <s’ouvrir au monde>
a: Il devait surtout marteler: <Puisse Cuba s’ouvrir au monde et le monde à Cuba>
(19)
[h110]t: M. Suharto appelle les Indonésiens à <se serrer la ceinture>
a: < Portons le fardeau ensemble>, a-t-il ajouté lors d’une cérémonie. <En tant que nation, nous devons nous serrer la ceinture>
– a transformation from an active voice (with agent) in the heading to a passive voice in the article
(20)
[h48]: M. Seillères dénonce la <brutalité> du premier ministre
a: E-A. Seillière affirme, dans un entretien au Point du samedi 29 novembre, que la réduction du temps de travail a été <imposée au pays avec une extrême désinvolture et une parfaite brutalité>.

In the heading of (21), X calls Y to do P; in the article, X says that Y knows that he (Y) has to do P.
(21)
[h122] Dans l’Oise, Jean-François Mancel appelle le FN à faire <partie de la droite de demain>
a: <… le Front national a enlevé tout ce qui peut nous hérisser sur le plan des valeurs. Ils se rendent compte qu’ils doivent devenir une partie de la droite de demain>

The distance marked by the journalist towards the quote and the speaker in the heading can become explicit in the article:
(22)
[sh94] L’accusé a, une nouvelle fois, mis en avant ses <efforts> pour sauver des juifs, en faveur desquels il affirme avoir mené sept types différents d’interventions
a: Le plaidoyer tient .. d’une litanie récapitulative, convoi par convoi, de ce que l’accusé considère comme des <efforts> en faveur des juifs, qu’il qualifie d’<incessants> et <soutenus>”.

In example (13), the quotation was perceived as polyphonic and potentially ironical:
(13)
[h82] Edouard Balladur <dialogue> avec les électeurs sur l’emploi In the article, ‘dialogue’ appears as a noun in a long quote:
a: Son ambition est, selon ses propres termes, de <changer la politique et de restituer à l’échange, au dialogue, toute la place qu’ils doivent…>

In the rest of the text, it is mentioned several times that E. Balladur (a former prime minister) listened: ‘Pendant trois heures, il a écouté…, Edouard Balladur a donc écouté…’, and it is announced that, a week or so later, ‘il sera en mesure de répondre, c’est-à-dire de s’engager sur certaines demandes..’. On the one hand, the journalist put stress on ‘dialogue’ in the heading, as a result of its double function; on the other, in the article, we find a recurrent mention of the speaker listening and announcing his answers for later.
There is no way to know with certainty if the use of ‘dialogue’ in the headline was intended with a certain irony, but this possibility cannot be excluded.

Polyphony of a third kind
Two voices are heard in the headings with partial quotes: the speaker and the journalist/editor’s voices. In the headings with short quotes, limited to one or two words, a third ‘voice’ can be heard, which seems to be generated by the combination of the words quoted and one element of the context, the speaker, the topic, the addressee, or something else. This voice speaks in simplified and stereotypical terms, staging the ‘actors’ in a easily recognizable way. It refers to the collective memory, through collocations and associations of terms. I will first give some examples, with a brief explanation. The first word is the one quoted in the heading; the second word or group of words is involved in the third voice’s ‘discourse’:
débat – PCF
(The French Communist Party is renowned for having until very recently refused to debate about its history and its inheritage)

idéologique – La Droite (about Gauche)
(The accusation of being ideological or acting in an ideological manner is mainly directed from the Right to the Left, but also against extreme Right that can been accused of ideological behaviour)

complot – Patronat vs./Gouvernement/Syndicats
(Illustration of a dominant feature of French political life, which favours confrontation as opposed to consensus)

impartial – Justice/France
(One of many recurrent discussions)

rénovateur – Le Pen
(For most readers, these cannot be equated)

exclusion – Chirac
(One of the catch-words of the presidential campaign of Chirac, one word that is said to have made him win the presidential elections, and two years later, lose the general elections. The word ‘exclusion’ to refer to the general situation of the Jews in France during the Second World War, as was the case in the heading, is somewhat anachronistic)

souveraineté – France
(No comments needed about that one)

mensonges/efforts – Papon
(Opinion shared by most French)

This third ‘voice’ would be the voice of public opinion’s, the voice of the ‘doxa.’[xviii] That this voice is heard mostly in/through the short quotes brings us back to the fundamental question: what is so special about those partial quotes and how can we describe their function, as opposed to the function of canonical forms of reported discourse, direct speech or indirect speech?

I will first propose a careful conclusion concerning the use of partial quotes in headings in Le Monde. More data and intersubjective verification of interpretations would be needed to confirm it. I will then try to summarize in a metaphor the function of partial quotes and their argumentative dimension.
The use of partial quotes in headings in Le Monde seems to be essentially limited to political articles and to quoting important ‘actors’ in the political field; consequently, partial quotes could be seen as an expression of the interface between journalist and establishment, and more particularly, of the power of the journalist to dispose of ‘public discourse’ as he ‘likes’. In the French situation, as described by Mouillaud and Tétu (1989), the political system and the journalistic system are in ‘close symbiosis’, which is not only political but also linguistic:
‘La véritable complicité en effet n’est pas seulement celle de la <sympathie politique>, elle est beaucoup plus profonde: en reprenant les petites phrases ou le silences, en attendant conférences de presse et communiqués, en fixant le regard des lecteurs sur les mêmes événements au même moment, la presse joue le jeu de qui produit les petites phrases, les conférences de presse, les communiqués.’ (40)

Mouillaud & Tétu describe quotation marks as ambivalent operators, whose function changes if they are perceived as belonging to the fragment of discourse or to the utterance that introduces the fragment. In the first case, the fragment is perceived as ‘full’ and the quotation marks represent the frame of a painting; in the second case, the fragment is perceived as ‘open’, and the quotation marks serve as a window frame[xix]:
‘Lorsque le fragment est perçu comme <plein>, il tend à appartenir à l’énoncé du journal; comme fragment <ouvert>, il donne sur un autre plan que celui du journal, le plan du discours d’origine; celui-ci est perçu comme s’il s’étendait au-delà de la citation, au-delà des limites du fragment. (…) Cette forme de citation est un hybride. (…) Sous cette forme (qui est la forme privilégiée du Monde), le journal est polyphonique: une chambre d’échos traversée de voix multiples, une étoile d’où partent des chemins divergents. Le regard ricoche sur les fragments guillemettés et se diffracte en de multiples éclats.’(1989:141)

Keeping this metaphorical frame, one could say that the recontextualization of the (reported) discourse can result in either a painting in a frame, that blends more or less well into the wall (the stretch of discourse is then perceived as ‘full’), or in a window (the discourse will then be perceived as ‘empty’). In headings, longer quotes are more like paintings, incomplete and retouched paintings in most cases; short quotes, limited to one or two words, are more like windows. The short quotes, in the headings I have looked at, referred to an original (quoted) discourse, but in such a way that they also conveyed an evaluation of the journalist that seemed to keep alive a public opinion on the speaker or on the event. The window allows a glimpse on the outside world, but can also reflect the reader’s own face, that is the beliefs and opinions that he shares with his social group, and, with the media’s help, is forced to keep on sharing. In a way, short quotes deprive the speaker even more of his discourse than we might think at first, as it is given back to the community of writers and readers, to the voice of the ‘doxa’.

NOTES
[i] This study represents part of a larger research on journalistic reporting being done at the Free University in which MA students have been participating since 1996. Cf. Burger (1997) and Mendonças Dias (1997).
[ii] Those RS-forms are almost restricted to journalistic written discourse. They are also quite rare in spoken discourse where they will be marked by a metadiscourse expression (‘I quote’, ‘his words’). The hand and fingers signs will be mostly associated with what is called ‘scare quotes’. See later in this paper.
[iii] Restrictions to that credibility function are often mentioned, as in Zelizer (1989:372): ‘Expectations held by the journalists, however, that news-discourse will clarify events to audiences through the story-telling devices they use (i.e. Carey, 1986) are undermined by journalistic reliance on quotes. Rather than clarify discourse, quoting practices blur its spatial and temporal parameters. News-quotes are generally anonymous (as in ‘experts said’) and uncentered (as ‘he said that she said that they said’). They are also recontextualized.’
[iv] The final text may show adjustments from the hand of the sub-editor or the editor; this will most probably be the case with headings.
[v] Mendonça Dias (1997) points out the differences between a transcript of a television programme with Jacques Chirac and its coverage inthree French broadsheets. A recent look at the choice of the introductory verbs (Torck, 6th International Pragmatics Conference, 1998) in the same corpus shows some possible influence of the context. For instance, the choice of ‘déplorer’ as speech-act verb, seems to be related to a repeated utterance ‘ce n’est pas bien’; ‘ajouter que’ to the presence in the cotext of words such as ‘également’, ‘deuxièmement’, etc.; ‘préciser’ could be induced by the occurrence of an interpolated clause in the original speech; ‘marteler impérieusement’ to multiple repetition of utterances as ‘ce que je veux’ and ‘on ne peut pas’, etc.
[vi] The coverage of Chirac’s television talk mentioned in note 5 was realized as 1 article and 29 occurrences of RS by Libération, 3 articles and 35 occurrences of RS by Le Figaro and 4 articles and 192 occurrences of RS by Le Monde.
[vii] Long articles often have, besides a headline and a subhead, an introductory paragraph usually printed in bold. Introductory paragraphs (called ‘chapeau’ in French) follow the same typographic rules (no italic). In the following I will refer to headings, though examples will be qualified as h or sh.
[viii] This selection does not cover all the possibilities of quoting in headings. There are many more quotes appearing in headings in the 47 issues of Le Monde I have prospected, occurring in structures suc as:
– pour X, selon Y (for X, according to X)
– no speaker is mentioned
Une politique du tout-répression: <Enfermez-les et jetez la clé>(5/2/98)
– a speaker is mentioned with a full quote:
Brigitte Engerer: <La sonorité d’un Cortot ou d’un Neuhaus, c’est cela la technique>(17/2/98) (Page Culture)
– there is a reference to the speaker, but no speech-act verb:
La <souffrance intime>de la femme du Prix Nobel italien Dario Fo (20/2/98)
[ix] We will see in a later section of this paper the frequency of metonymic substitution in headings. Example (8) does not fall under this category in my opinion, but is the result of a transformation of suppression applied to (something like) ‘Amnesty International dénonce dans un rapport intitulé “La terreur en Casamance” la situation (tragique) que connaît cette région du Sénégal’.
[x] I am implicitly making the assumption that the writing of the article chronologically precedes the writing of the title. In the analysis I will follow the ‘natural’ process of reading.
[xi] In the context of the heading, this adjective has a quite recent ‘new’ connotation. It has been used to refer to the left-‘coalition’ government of Lionel Jospin which includes communists, various kinds of socialists and ecologists (since June 1997). The ‘gouvernement <pluriel>’ could be interpreted as neutral or with a positive or a negative connotation depending on the reader’s own political opinions.
[xii] In the plural form, ‘les écoutes’ could have a negative connotation, as a French reader easily could relate this word to various scandals of the last ten years, in particular, ‘l’affaire des écoutes téléphoniques de l’Elysée’ when various journalists and well-known persons in the 80’s had their telephone conversations recorded and listened to on orders from the highest level. In the singular form, it would be associated with an open and productive attitude.
[xiii] As put by Geis (1987:131) in his study of speech-act verbs: “… a judgment about the affect of the verb of [these] sentences is not meant as a measure of the affect of the sentence that contains it”.
[xiv] In the occurrences of ‘affirmer (que)’, we find speakers as different (on a ‘credibility scale’) as Nelson Mandela, Pol Pot and Elisabeth Guigou, Minister of Justice.
[xv] An extensive study of contexts in which these verbs tend to appear would certainly be necessary in order to classify them. In my opinion, it would mostly cast light on the newspaper or journalist’s style, and would probably not allow generalisations to other types of texts.
[xvi] The speaker was ‘le régime de Phnom Penh’.
[xvii] With ‘Propos’ and ‘Interview’, there are usually no quotation marks (with the exception of the beginning or the end of the article) and no italics.
[xviii] The word ‘doxa’ (Barthes, 1975) would be appropriate, as it covers more than public opinion, that is, in Barthes’s words: ‘La Doxa, c’est l’Opinion publique, l’Esprit majoritaire, le Consensus petit-bourgeois, la Voix du Naturel, la Violence du Préjugé’ (51). Angenot (1989:894) describes a component of the doxa, the ‘idéologème’, that refers to expressions, collocations, set phrases, stereotypes, etc. The ‘idéologème’, in a moment of social discourse, is malleable, dialogic and polyphonic. Its meaning and acceptibility are the result of its migrations through discourse and ideology: ‘il se réalise dans les innombrables décontextualisations et recontextualisations auxquelles il est soumis’.
[xix] Note that it applies to quotations in general, and not specifically quotations in headings.

REFERENCES
Angenot, M. (1989). 1889. Un état du discours social. Québec: Ed. du Péambule.
Bakhtine, M. (1973). The dialogic imagination: four essays. University of Texas press.
Barthes, R. (1975). Roland Barthes par Roland Barthes. Paris: Seuil.
Bauman, R. & Briggs, Ch. (1990). Poetics and Performance as Critical perspectives on Language and Social Life. Annual Review of Anthropology, Vol.19, 59-88.
Burger, I. (1997). Les verbes introducteurs du discours rapporté dans Le Monde. Facteurs de variation de la ‘volatilité’ de ces verbes. MA-Thesis, Vrije Universiteit, Amsterdam.
Ducrot, O. (1984). Le dire et le dit. Paris: Minuit.
Geis, M. (1987). Political bias in News Magazines, The Language of politics. New York: Springer, ch.7.
Gruber, H. (1993). Evaluation devices in newspaper reports. Journal of Pragmatics 19, 469-486.
May, J.D. (1995). Argumentation in news discourse: Speech act reportage. SIC SAT 1995, Proceedings of the Third ISSA Conference on Argumentation, Vol. III, 571-587.
Mendonças Dias, M. (1997). Le discours rapporté dans l’écriture journalistique. Marqueur de l’information recueillie ou de l’information créée. MA-Thesis, Vrije Universiteit, Amsterdam.
Mouillaud,M. & Tétu, J.-F. (1989). Le Journal quotidien. Presses Universitaires de Lyon.
Sémino, E., Short, M. & Culpeper, J. (1997). Using a corpus to test a model of speech and thoughts representation. Poetics 25.1, 17-43.
Silverstein, M. & Urban, G. (Eds.) (1996). Natural Histories of Discourse. Chicago University Press.
Waugh, L.R. (1995). Reported speech in journalistic discourse: The relation of function and text. Text 15.1, 129-173.
Weizman, E. (1984). Some Register Characteristics of Journalistic Language: Are They Universals? Applied Linguistics, Vol.5, No.1, 39-50.
Zelizer, B. (1989). ‘Saying’ as collective practice: Quoting and differential address in the news. Text 9, 369-388.