1. Transductive reasoning

The concept of transductive reasoning is developed by Piaget ([1924], 185) to analyze the development of children’s intelligence. Transductive reasoning is characterized as the prelogical and intuitive way of thinking of the young child, which goes directly from an individual or a particular fact to another individual or particular fact, without the intermediary of a general law. According to Grize,

The young child who says, ‘It’s not afternoon because there was no nap’ is based on the daily experience of napping as an ingredient of the afternoon [reasons by transduction]”(1996, p. 107).

Transductive reasoning seems to be the product of a conditioned association “nap = afternoon”, which gives, by application of the scheme of the opposites: “no siesta = no afternoon”. From this perspective, napping is a defining feature of the afternoon.

Grize observes that adults are also likely to use this kind of reasoning:

When we say that we stopped at the traffic light because it was red, […] our thinking does not go through a general law of the kind: “any red traffic light implies stop” (ibid.).

In the latter case, the statement has the form of a “semantic block” (Carel 2011), “Answer because Stimulus”. Yet the adult does not apply the negation in the same way as the child; saying “it is not a red light since I did not stop” would be considered as a denial of reality.

However, it is said that a motorist deeply imbued with respect for the Highway Traffic Act refused to believe that he collided head on with another vehicle because he was driving down a one-way street, implying the material impossibility of a fact from its legal prohibition.

2. Two-term reasoning

In a very different context, Gardet and Anawati speak of, “two-term reasoning” which is characteristic of “a specifically Semitic rhythm of thought which the Arab mind knew how to use with a rare happiness of expression” (Gardet and Anawati [1967], p. 89). This type of reasoning seems to be similar in nature to transductive reasoning.

The ‘dialectical’ logic, connatural to the Arab genius, is organized according to modes of reasoning with two terms that proceed from the singular to the singular, by affirmation or negation, without a universal middle term. Should we say, as has sometimes been said before, that [this universal medium term], not explicitly understood, is nevertheless explicit in the reasoning mind? We don’t think so. Undoubtedly, two-term reasoning can be ‘translated’ into a three-term syllogism […]. Yet in the logical mechanism of thought, it is indeed the confrontation, by contrast, similarity or inclusion, of the two terms of the reasoning that gives the ‘proof’ its power of conviction. The universal middle term is not present in the mind, even in an implicit form. This is not about establishing a discursive proof, but about promoting a self-evident certainty. (Bouamrane & Gardet 1984, p. 75; my emphasis)

The Arab logician and theologian al-Sumnani has distinguished five rational processes, that is five argument schemes, which are characteristic of two-term reasoning. These five processes are based on

Findings, and then, by a movement of the mind operating either by elimination or by analogy from the same to the contrary, or from the same to the same. It is always a question of passing from the present, actual fact, the “witness” (shâhid) [the argument, cp], to the absent, (gha’ib) [the conclusion, cp]. There is no abstract search for a universal principle. (Gardet and Anawati [1948], pp. 365-367; my emphasis).

1. Schema

The word schema is used to refer to any kind of diagram used to represent and clarify the structure of an argumentative phenomenon: Toulmin’s schema, Convergent argumentation schema, etc.
Toulmin “layout of argument” is also known as “Toulmin Schema” or “Toulmin Argument Pattern” (TAP).

2. Argumentation scheme

An argumentation scheme is an abstract or generic representation of a series of concrete argumentations sharing the same structure (scheme).

3. Schematization

Natural Logic uses the term “schematization” to designate the succession of linguistic and cognitive operations through which a reality is given a linguistic expression by a speaker, S. Schematization.

Arguments can be approached on the basis of very different systems of rules.

— Rules expressing observational regularities.
— Rules expressing norms, imperatives, which are instrumental for argument evaluation.
— Rules as counsels to do things well, how to convince a person to believe or to do something.

1. General rules of interaction

1.1 Rules of interaction

Argumentative interactions in natural language follow the various systems of rules proposed for interaction in general, so for example, the rule of justification of non-preferred sequences is applied:

A dispreferred second part is a second part of an adjacency pair that consists of a response to the first part that is generally to be avoided, and which is likely to be marked by such features as delays, prefaces and accounts. (SIL, Dispreferred second part)

1.2 Cooperative principle

The principle of cooperation expresses not only what the participants actually do (observational regularity), but also what is reasonable for them to do (rational regularity).

1.3 Principles of politeness

The principles of linguistic politeness regulate talk relationships on the basis of the concepts of face and territory. In ordinary conversation, these rules might inhibit the development of arguments. The overriding concern to preserve the relationship means that contradiction is difficult to express and develop.

1.4 Language Sins

A set of commands related to the control of discourse has been developed by the theological tradition inspired by the Bible. The violation of any of these rules is characterized as a sin of language (Casagrande & Vecchio 1991), S. Fallacies as Sins of the Tongue.

2. Rules specifically attached to argumentative speech

2.1 Rules of the Place

Specific codes are attached to specific argumentative places. Parliamentary rules for example apply in Parliament; tribunal proceedings, or classroom interactions develop in line with their own specific regulatory conventions, S. Forum. These regulations are drawn up in accordance with a sui generis procedure and are applied by the competent authorities ruling in the given place. These rules frame the kind of local rationality which characterizes the “genius loci”, the spirit of the place.

In such places, the rules determine the topics to be dealt with, the procedures that will lead to a legitimate decision and conclusion, and the persons qualified to take the floor; they regulate the right to speak, the quantity of speech, and the succession of turns at speech. These rules might, for example prohibit overlaps and interruptions.

2.2 “The Rules of an Honorable Controversy”

Levi Hedge, in his Elements of Logick (1838), presents the following seven “Rules for Honorable Controversy”:

Rule 1. The terms, in which the question in debate is expressed, and the precise point at issue, should be so clearly defined, that there could be no misunderstanding respecting them.

Rule 2. The parties should mutually consider each other, as standing on a footing of equality in respect to the subject in debate. Each should regard the other as possessing equal talents, knowledge, and desire for truth, with himself; and that it is possible therefore that he may be in the wrong and his adversary in the right.

Rule 3. All expressions which are unmeaning or without effect in regard to the subject in debate should be strictly avoided.

Rule 4. Personal reflections on an adversary should in no instance be indulged.

Rule 5. No one has a right to accuse his adversary of indirect motives.

Rule 6. The consequences of any doctrine are not to be charged on him who maintains it, unless he expressly avows them.

Rule 7. As truth, and not victory, is the professed object of controversy, whatever proofs may be advanced, on either side, should be examined with fairness and candor; and any attempt to ensnare an adversary by the arts of sophistry, or to lessen the force of his reasoning, by wit, caviling, or ridicule, is a violation of the rules of honorable controversy.
(Hedge, 1838, pp. 159-162)

Some of these rules sound familiar. Rule 5 corresponds to the accusation of having a hidden motive: “You agree with this proposal not because you approve it but to please the director”. Rule 6 is original, and refers to the problem of the hidden agenda, or even of conspiracy, S. Pragmatic argument. Disputes can be said to be “honorable” in both the intellectual and social sense. This system reintroduces what is socially acceptable in a situation where the participants will not spontaneously apply the common rules of cooperation and politeness. Such considerations join the rhetorical problematic of the prepon and the aptum (Lausberg [1960], § 1055-1062).

In Hedge’s system, social control is the root of the imposition of co-operation. The rules for avoiding the sins of language originate from religion, S. Fallacies as sins of language. In the Pragma-dialectical system, the system of rules avails itself of communicational rationality, in the spirit of Grice, S. Cooperative principle.

3. Pragma-Dialectic rules and the re-conceptualization of fallacies

These rules define “A Code of Conduct for Reasonable Discussants” (van Eemeren, Grootendorst 2004, p. 190), for partners willing to rationally resolve their difference of opinion. A fallacy is defined as a violation of one of these “Ten Commandments for Reasonable Discussants” (id., 190-196), S. Fallacies (I):

Commandment 1, Freedom rule: Discussants may not prevent each other from advancing standpoints or calling standpoints into question

Commandment 2, Obligation to defend rule: Discussants who advance a standpoint may not refuse to defend this standpoint when requested to do so.

Commandment 3, Standpoint rule: Attacks on standpoints may not bear on a standpoint that has not actually been put forward by the other party.

Commandment 4, Relevance rule: Standpoints may not be defended by non-argumentation or argumentation that is not relevant to the standpoint.

Commandment 5, Unexpressed-premise rule: Discussants may not falsely attribute unexpressed premises to the other party, nor disown responsibility for their own unexpressed premises.

Commandment 6, Starting-point rule: Discussants may not falsely present something as an accepted starting point or falsely deny that something is an accepted starting point.

Commandment 7, Validity rule: Reasoning that in an argumentation is presented as formally conclusive may not be invalid in a logical sense.

Commandment 8, Argument scheme rule: Standpoints may not be regarded as conclusively defended by argumentation that is not presented as based on formally conclusive reasoning if the defense does not take place by means of appropriate argument schemes that are applied correctly.

Commandment 9, Concluding rule: Inconclusive defenses of standpoints may not lead to maintaining these standpoints, and conclusive defenses of standpoints may not lead to maintaining expressions of doubt concerning these standpoints.

Commandment 10, Language use rule: Discussants may not use any formulations that are insufficiently clear or confusingly ambiguous, and they may not deliberately misinterpret the other party’s formulations.

This system is inspired by the proposals of the Erlangen school for the definition of a rational “ortholanguage”, S. Logics for dialogue. In the spirit of Grice, these commandments introduce or impose cooperation where it would not be spontaneously practiced by the participants. The game is based on the notion of standpoint. It corresponds to a dialectical treatment of the difference of point of view, with a proponent affirming the point of view and responding to the attacks of an opponent who casts doubt upon it. Rule 9 recalls the aim of the game, that being to settle the difference of opinion either by eliminating the unsustainable opinion or by eliminating the doubt about a well-justified opinion.

Such a system of rules accounts for the validity judgments of the speakers (van Eemeren, Garssen, Meuffels 2009). It is also possible to identify the implicit rules to which the speakers refer for their evaluations based on observing their practices (Doury 2003, 2006).

4. On rules

Dialectic

Fallacies (I), Contemporary approaches

Fallacies (II), Aristotle’s foundational list

Fallacies (II), From logic and dialectic to science

Fallacies (IV), A moral and anthropological perspective

Argumentation (II): Key features and issues

Paradoxes of argumentation

1. Argument scheme

An argumentation scheme (argument scheme) is a discursive formula, a generic statement functioning as an argument rule, an inferring license. Concrete argumentations, or enthymemes are its actualization in specific passages.
The concept of an argumentation scheme (argument scheme) captures the specificity of the minimal concatenation of two statements (S1, S2) making up an argumentation (Arg, Concl). An argumentation scheme is essentially a specific kind of sentence connection, a special case of textual coherence and cohesion; that is to say, a general discursive inferential scheme associating an argument with a conclusion.

In Aristotle’s Rhetoric, all the argument lines are expressed as such generic statements, that can sometimes be formulated as proverbs or maxims. The saying, “if you can do the hard things, you can do the easy things as well” corresponds to the “from the biggest to the smallest” (a maiori ad minus) branch of the a fortiori scheme. Typical formulas, such as those proposed by Bentham “let us wait a little, the moment is not favorable” are also complete and perfectly adequate expressions of an argument scheme. S. Juridical arguments: three collections. This scheme can be specified in a discursive domain, S. A fortiori.

In the expression of the scheme, their characteristic indefinite components (subject, predicate) may also be expressed as variables. For example, the a fortiori scheme can be written as (according to Ryan 1984):

If <P is O> is more likely (more recommendable…) than <E is O>,
and <P is O> is false (not plausible, not recommendable)
then <E is O> is false (not likely, not recommendable).

And embodied in the following argumentation:

If teachers do not know everything, students know even less

In the same style, the scheme of the opposite is written as:

If <A is B>, then <not-A is not-B>.

Derived argumentation:

If I was of no use to you during my life, at least my death will be useful to you.

Such presentations should not be taken as a kind of “logical or semantic deep structure” of the scheme. Their unquestionable benefit is to clarify the reference of general terms.

2. Example: Argument scheme and argumentations on waste

To detect a scheme in a text is a key moment in argument analysis. But this identification is not easy, the key semantic components of the scheme being frequently disseminated in the text. How can we identify a scheme in a passage? Experts will say that they just recognize a scheme when they see it; but the non-specialist necessitates a methodical reconstruction, which can proceed along the following basic lines:

— First, an explicit definition of the topic is needed.
— Second, the passage must be clearly delimited.
— And finally, one has to show how the scheme can be projected upon the passage; that is, one has to establish a point-to-point correspondence between the scheme and the passage under analysis. In essence, these links consist in linguistic operations of equivalence and close reformulation.

This method can be illustrated by the case of the argument from waste, as defined and illustrated in Perelman & Olbrechts-Tyteca.

— The scheme:

The argument of waste consists in saying that, as one has already begun a task and made sacrifices, which would be wasted if the enterprise were given up, one should continue in the same direction. ([1958], p. 279)

— First derived argumentation:

this is the justification given by the banker who continues to lend to his insolvent debtor in the hope of getting him on his feet again in the long run. (Id., p. 279)

— Linguistic operations associating the argument to the scheme (bijective association Scheme – Argumentation)

Argumentation italics: arg. wording	Linguistic Operation italics: arg. wording; bold: AS wording	Scheme (AS) bold: AS wording
Implicit: a debtor is a person to whom the banker has already lent money	Lending money is a task; it can be a sacrifice	(Past:) one has already begun a task and made sacrifices
Insolvent debtor	Insolvent means that the previously lent money [will] be wasted	(Present:) which would be wasted if the enterprise were given up
The banker continues to lend	Continues to lend = continue in the same direction	(Decision:) one should continue in the same direction

The second enthymeme is more complex:

This is one of the reasons which, according to Saint Theresa, prompt a person to pray, even in a period of ‘dryness’. One would give up, she says, if it were not “that one remembers that it gives delight and pleasure to the Lord of the garden, that one is careful not to throw away all the service rendered, and that one remembers the benefit one hopes to derive from the great effort of dipping the pail often into the well and drawing it up empty”. (Id., p. 279)

— Linguistic operations associating the argument to the scheme (same conventions):

Argumentation	Linguistic Operation	Scheme
all the service rendered that is: (the great effort of) dipping the pail often into the well	rendered (presupposes) already begun a service (is) a task	one has already begun a task
the great effort of (dipping the pail often into the well)	the great effort (is a) sacrifice	and made sacrifices
in a period of “dryness”(1) driving it up empty	dryness — empty <=> no result	for no result
not to throw away	not to throw away <=>would be wasted	(present) which would be wasted if the enterprise were given up
prompt a person to pray	prompt to <=> urge to continue	one should continue in the same direction

 (¹) Traditional mystic metaphor for “no increase in faith” = no spiritual benefit.

3. Naming the argument schemes

Argument schemes are labeled according to their form or their content.

3.1 According to their specific domain and semantic content

Some famous arguments are named in reference to their precise content.

— The third man argument is an objection made by Aristotle to the Platonic theory of intelligible forms, as opposed to individuals. According to this objection, the Platonic theory implies an infinite regression. It may be seen as an argument from vertigo.

— The argument against miracles: the likelihood that the dead person was resurrected is lower than the likelihood that the witness is mistaken; so we can reasonably doubt that the dead person was resurrected (Hume, 1748, §86 “Of Miracles”). This formally refers to a hierarchy of the probable, and can be represented on an argumentative scale, S. Argumentative scale.

— The ontological argument infers the existence of God from the a priori notion of a perfect being, S. A priori; Definition.

3.2 Depending on their form and content

S. Collections (2): From Aristotle to Boethius
Collections (3): Modernity and tradition
Collections (4): Contemporaty innovations and structurations

On the use of Latin words and expressions, S. Ab Arguments, a/ad — e/ ex — 

3.2 Oriented labels

Usually, the label designating an argument specifies a form or content: the argument refers to the consequences (ad consequentiam), to authority (ab auctoritate), to the consistency of human beliefs (ad hominem), to emotion (ad passionem) or to any particular emotion (ad odium). The speaker may admit, without inconsistency, losing face and invalidating the argument he has just used, that he or she argues by the consequences, ad hominem, ex datis, upon a religious belief (ad fidem), or possibly from the number, ad numerum. These arguments can be assessed in a second, normative, stage.

Some other arguments involving the arguer are designated by oriented labels. An argument cannot be dubbed an appeal to stupidity, to superstition or to imagination without invalidating it; given the current vision of emotion as antagonistic to reason, referring to a passage as containing an appeal to emotion, from ad passiones to ad odium, amounts to a rejection of the argument. Such labels contain a built-in evaluation; there is some confusion between the levels of description and evaluation.

A call to faith will or will not be judged as fallacious depending whether or not one shares the beliefs of the speaker. In such cases, the theoretical language is biased, and normative action becomes ideological.

4. Typologies of argument schemes

A general typology of argumentation schemes is an organized collection of argument schemes. Collections of argument schemes are locally constituted as:

— The set of arguments locally exploited by a particular speaker, in a particular discussion, see « collections”.
— The set of arguments attached to a question, S. Script.

5. Argument schemes in discourse

The concept of the argument scheme anchors the study of argumentation in the material reality of speech and discourse. The capacity to identify an argument from authority, a pragmatic argument, etc. is an essential skill for the production, interpretation and criticism of argumentative discourse, S. Tagging.

Some works, such as the Thomas Aquinas Summa Theologica or texts such as Montesquieu “On the Enslavement of Negroes” can be described as dense and dry successions of arguments. Other texts are more fluid, and seem hardly reducible to circumscribed segments that could be plausibly described as an occurrence of an argument scheme.

The schemes are relatively under-determined by the linguistic expression; there may be several plausible analyses of the same text segment, some invalidating the argument, others not. This uncertainty should not be automatically seen as an indicator of the poor quality of the argument. In this respect, contextual considerations and the kind of editing given to the analyzed passage play a crucial role.

An argumentative text or interaction can be compared to a natural meadow, the most beautiful flowers corresponding to canonical argument schemes. But it is also necessary to wonder about what the dense plant tissue around these flowers is made of. To this end, interaction analysis, discourse analysis and text linguistics serve as crucial analytic instruments, which have to be adapted to the specificities of argumentation analysis. The “scheme approach” comes within a larger prospect, opening with the stance taken vis-à-vis the other’s discourses, the kind of argumentative situation they frame, the determination of general argumentative strategies, taking into consideration a whole range of semiotic phenomena. On a micro-level, one has to consider the operations producing the statements, as well as in their coordination: a good grammar book and a good dictionary are essential if one is to construct a good argument analysis, S. Argumentative Question; Indicator.

The study of schematizations is the defining objective of the Natural Logic developed by Jean-Blaise Grize, a student and subsequently a collaborator of Jean Piaget at the Research Center on Genetic Epistemology in Geneva. This logic is called “natural” as opposed to formal logic: on the one hand, it is a “logic of objects” (1996: 82) and a “logic of subjects” (Grize 1996: 96); on the other hand, it involves processes of thought that leave “traces” in natural discourse.
According to Grize, discourse is essentially argumentative, meaning that all utterances frame the world or the situation, along their subjectively relevant lines, to build a meaningful “schematization”. “Scheme” has here a totally different meaning from “argument scheme”, which would be called a “reasoned organization”, in Grize’s vocabulary, corresponding to the second-level phenomenon of sentence combination, whereas schematization is a first-level phenomenon, that of sentence production.

According to Grize’s favorite metaphor, to argue, is to “give to see” to the audience a situation as “spotlighted” by the speaker. As every speech throws some subjective lighting on the world, argumentation is inherent to speech. In Perelman’s terms, this operation consists in giving “presence” to an object (Perelman & Olbrechts-Tyteca, [1958], p. 116).

In this perspective, an argumentation is not necessarily a set of statements organized in line with the layout proposed by Toulmin. The influence of an argument and its rationality are not attached to a special kind of speech, or to the use of such and such specific “discursive techniques”, as suggested by Perelman & Olbrechts-Tyteca. Any statement, any coherent succession of statements, be it viewed as descriptive, narrative or argumentative, is indeed argumentative, insofar as it builds a point of view mapped into a meaningful schematization. Natural Logic is defined as the study of such schematizations, the cognitive counterpart of sentence construction.

This concept is adapted to a vision of arguing as story telling, as offering a coherent and detailed presentation of the world. This might be of some comfort to all students who find themselves disheartened by the difficulty of giving a dense account of extended texts or interactions in terms of argument schemes, even where these are supplemented by an extensive repertoire of figures of speech.
If persuading is defined as shifting the partner’s representations, and, accordingly, his or her behavior, then any informative statement, such as “It is 8 a.m.” is argumentative. If the addressee has to take the 7.55 train and is savoring a last coffee, thinking it is a quarter to 8, then, the information will dramatically change his vision of the immediate future. Natural Logic is also a theory of generalized persuasion, as just “focusing on the relevant aspect of reality”.

1. Schematization, a step-by-step process of meaning construction

Argumentation is traditionally defined as a combination of utterances. Natural Logic studies argumentation as a cognitive process evidenced in natural discourse, and manifested at every stage of discourse production, from the first elaboration of an idea to the combination of utterances, which is only the final stage of the argumentative process. Schematization corresponds to a representation embodied in a complex discursive unit,

Influencing the interlocutor is to try to modify his or her representations, by emphasizing some aspects of things, concealing others, proposing new ones, and all this by using appropriate schematization. (Grize 1990, p. 40)

Argumentation does not appear to be a chain of statements in a discourse. It emerges progressively at every stage of the production of the utterance, from the first operation of apprehension of content to the construction of a meaningful and therefore “reasoned” discourse. Any statement, any coherent succession of statements, whether or not it is traditionaly considered to be argumentative, narrative, or descriptive … , is indeed argumentative to the extent that it constructs a unique point of view, that is a “schematization”. This conception leads to reconsider all information as argumentation, tending to liken discursive meaning to argumentation, S. Argumentation (I); Argumentation (II).

Grize defines Natural Logic as “the study of logical-discursive operations that make it possible to construct and reconstruct a schematization” (1990, p. 65); “Its task is to account for the operations of thought allowing a speaker to construct objects and to predicate upon them at will” (1982, p. 222).
The concept of schematization defined as a “[discursive representation], oriented towards an addressee, of what the author conceives or imagines of a certain reality” (1996, p. 50), “of what it is all about” (1990, p. 29). A schematization is a discourse that focuses the listener’s attention upon a “micro-universe” given as “an accurate reflection of reality” (id., p. 36), which constructs or “structures” (id., p. 35) a synthetic, coherent, stable meaning. The purpose of schematization is “to show something to someone” (Grize 1996, p. 50; my emphasis); “to schematize […] is a semiotic act: it is to give to see” (id., p. 37; my emphasis). The object of Natural Logic is the study of the operations constructing such images.
The functioning of schematization is particularly clear in classical argumentative situations, when a discourse directly confronts a counter-discourse; the same reality is given two antagonistic descriptions:

S1 — These replacement workers, you will pay them with the strikers’ money!
S2 — Not the strikers’ money, the taxpayers’ money.

2. Operations constructing a schematization

Natural Logic postulates the existence of “primitive notions”, of a pre-linguistic nature (Grize 1996, p. 82), linked with the culture and the activities of the speakers. These pre-notions are the place of “cultural pre-constructions”, i.e., received ideas and current, accepted ways of doing things. The language “semantizes” these primitive notions turning them into “objects of thought” associated with words (Grize 1996, 83).
Schematization operations are anchored in these “primitive notions” (id., p. 67) and are constructed by a series of operations; “primitive notions” are actually noted by words between brackets. The following sequence is formed of the primitive image and fuzzy notions /fuzzy/ and /image/:

It’s unfortunate that the edge of the image is blurry, and it needs to be corrected. (Ibid.)

This construction follows these steps:

(a) The process of discourse construction begins with the selection of relevant primitive notions, to produce the objects of discourse; here “image, edge of the image” as well as the predicative pair “to be blurred, not to be blurred”. The objects thus schematized will evolve with the development of the discourse, S. Object of discourse.
(b) Then, the operation of characterization produces “contents of judgments” that is predications, and these are accompanied by modalizations, carried out on the objects of discourse. Here, the content of judgment is, “that the edge of the image be quite blurry”.
(c) A subject then asserts (positively or negatively) the preddication, and produces a statement, “it is unfortunate that the edge of the image is quite blurry”.
(d) Operations of configuration then connect several utterances and so build a discursive chain, “a reasoned organization”. The preceding statement for example, is connected to another statement, “this must be corrected”, which is produced according to the same mechanism:
It’s unfortunate that the edge of the image is blurry, and it needs to be corrected.

These different linguistic-cognitive operations can be likened to the vision of language and mind developed by the philosophy of traditional logic, S. Logic.

(a) Apprehension of content by the mind;
(b) Predication, constituting unasserted propositions;
(c) Judgment, expressed in an assertion, which can be true or false;
(d) Concatenation of judgments, i.e. discourse construction.

The aim of this approach is to emphasize that all operations relevant to the genesis of the utterance have an argumentative import. Argumentation is as much a sentence construction process as a sentence connection process.

3. Shoring

The concept of shoring developed in Natural Logic is defined as,

a discursive function consisting, for a given segment of speech (whose dimension can vary from a simple statement to a group of statements having a certain functional homogeneity), to accredit, to make more likely, to reinforce, etc. the content asserted in another segment of the same discourse. (Apothéloz & Miéville 1989, p. 70)

This concept corresponds to the classical problematic of argumentation as a composition of statements, a statement-argument supporting a statement-conclusion. To refer to the same phenomenon, Natural Logic also uses the expression “reasoned organizations”:

Many statements are made merely to support, to shore up the information given. This is part of the general process of argumentation, and allows us to envisage more or less extensive blocks of discursive sequences as reasoned organizations. (Grize 1990, p. 120)

The study of reasoned organizations is an instrument for the study of representations, defined as “a network of articulated contents” (id. p. 119-120). It should be emphasized that, for Natural Logic, the reasoning process is not limited to the combination of utterances but includes the whole dynamic process of structuring the utterance, whether it will function as argument or conclusion in a reasoned organization.

4. Schematization and communication

Schematizations refer to a particular communication situation. They are the product of “the activity of speech [which] is used to construct objects of thought” (1990, p. 22); these objects being part of a dialogue where they are used “as shared references for interlocutors” (ibid.). The communication situation envisioned is intended to be “essentially dialogical in nature” (1990, p. 21), but it is actually analogous to that of rhetorical address. It never considers the possible interactions between the respective schematizations of the participants.

By [dialogal] I don’t mean the interweaving of two discourses, but the production of a speech between two parties, a speaker [orator] … addressing a listener. Admittedly, in most texts, the listener remains virtual. This, however, does not alter the basic problem: the speaker constructs the speech according to his or her representations of the listener, simply, if the listener is present, he or she can actually say, “I do not agree” or, “I do not understand”. But if the listener is absent, the speaker must indeed anticipate his or her refusals and misunderstandings. (1982, p. 30)

Persuasion is given up, “the speaker can only propose a schematization to his or her audience, without actually ‘transmitting’ it” (ibid.).

5. “Logic of Contents” (Grize) and “Substantial Logic” (Toulmin)

Grize defines his Natural Logic in relation to formal logic:

Alongside a logic of form, a formal logic, it is possible to envision a “logic of contents”, that is, a logic taking into account the processes of thought, the development and interconnection of these contents. Formal logic based on propositions accounts for the relations between concepts, while Natural Logic proposes to highlight the construction and interconnection of the notions. (Grize 1996, p. 80)

This “logic of contents” might remind us of Toulmin’s “substantial logic”, S. Layout. But, unlike Toulmin, who characterizes argumentation as an arrangement of statements without discussing their internal structure, Grize considers that argumentation begins with the basic operation producing the statement itself.

---

 

The correlative concepts of argument scale and laws of discourse have been developed by Ducrot (1973) in the framework of the theory of Argumentation within Language to describe the grammar of co-oriented arguments.

Argument scale translates “échelle argumentative”, word for word “argumentative₁ scale”. Argument scales strictly deal with argument₁ “good reason” or premise for a conclusion, not with argument₂, “dispute”, S. To argue.

1. Cooriented arguments on a scale

An argumentative class is defined as follows:

A speaker places two statements p and p’ in the argumentative class determined by an utterance r if he considers p and p’ as arguments for r. (Ducrot [1973], p. 17)

S: — Your great grandmother spent time in The Two Maggots, she dressed in black, she read Simone de Beauvoir, she was a true existentialist!

S presents three convergent arguments co-oriented towards the conclusion “She was a true existentialist” (a mid-twentieth century popular philosophy). These arguments correspond to features borrowed from the stereotype of what existentialists are and do. S. Categorization.

The word class refers to an unordered and non-hierarchical set of elements. There is no reason to think that “spending time in The Two Maggots” (an existentialist Parisian café) is considered by S as a stronger or weaker argument than “reading Simone de Beauvoir”.

Two utterances p and q belong to the same argument scale (for a given speaker in a given situation) “if the speaker considers that p and q are both arguments for the same conclusion r (they belong therefore to the argumentative class of r), and if he considers that one of these arguments is stronger than the other” (Ducrot, [1973], p. 18).

The following scale represents a situation where q is stronger than p for the conclusion r.

The situation where the speaker considers that “reading Simone de Beauvoir” is a stronger argument than “spending time in The Two Maggots” for the conclusion “to be a true existentialist” is represented as follows:

 

The scales where the force of the arguments p and q is determined solely by the speaker, are called relative, S. Force.

Scales for which the gradation is objectively determined are called absolute, for example the scale of the cold:

2. Laws of discourse

The functioning of argument scales is regulated by four laws: Lowering Law,  law of Negation, law of Inversion, and law of Weakness.

2.1 Lowering law

According to this law “in many cases, (descriptive) negation is equivalent to less than” (Id, p.31).

Negation is asymmetric; it does not exclude just a point on the argument scale, but the whole zone including the denied argument and all arguments which are potentially stronger. Denying an argument which is positioned at a higher point on a given scale implies the affirmation of the lower argument.
Let’s consider the argument scale determined by a positive answer to the argumentative question “should we invite him to our hunting party?”, under the presupposed context “we are ourselves a group of decent hunters”.

In such a context, “he or she is not a good hunter” means, “he or she is a poor hunter”, not “he or she is a first rate hunter”.

The statement “he or she is not a good hunter, he or she is a first rate hunter” (stress on good and first class) involves a very particular form of negation, whereby an earlier statement is refuted, S. Denying. The stronger argument is necessarily expressed, while the weaker argument remains implicit in the unmarked use of negation.

2.2 Law of weakness

According to this law, “if a sentence p is fundamentally an argument for r, and if, on the other hand, when certain conditions (in particular contextual conditions) are met, it appears as a weak argument (for r), then it becomes an argument for not-r (Anscombre and Ducrot 1983, p. 66):

He’s a good hunter: he killed two partridges last year

In particular, the weak argument must be presented in isolation, and not in conjunction with conclusive arguments. Grice’s principle of exhaustiveness can also account for this fact: an isolated weak argument will be interpreted not only as weak, but also as the best possible, which results in the rejection of the attached conclusion, and consequently, in a binary situation, as a good reason to go for the opposite conclusion, S. Cooperation.

From an interactional point of view, putting forward a weak argument might also serve a positive purpose, serving to open a discussion and clarify the positions of the participants.

2.3 Law of negation

The law of negation posits as a regularity that, “if p is an argument for r, not-p is an argument for not-r” (Ducrot 1973, p. 27). If “the weather is nice” is an argument for “let’s have a walk”, then “the weather is not nice” is an argument for “let’s stay at home”. This law corresponds to the argument by the opposite (corresponding to the paralogism of negation of the antecedent).

The following example combines the law of weakness with the law of negation; a weak argument for a conclusion is reversed as a strong argument for the opposite conclusion:

After the Second Iraq War, which began in 2003, Saddam Hussein, former President of the Republic of Iraq, was tried and executed in 2006. Some commentators felt that the trial had not been conducted fairly, and considered that the trial was so rigged that even Human Rights Watch, the largest unit in the US human rights industry, had to condemn it as a total masquerade.
Tariq Ali, [A Well-Orchestrated Lynch], 2007[1].

According to the author, the Association Human Rights Watch generally approves decisions in the interests of the United States. So, the fact that they approve the sentence is a weak argument for the conclusion “the sentence is fair”. In this case, the fact that even the association has condemned the decision (like other persons or associations more inclined to criticize the United States) is a strong argument for the conclusion that the sentence is unfair.

Inversely, a weak refutation of r reinforces r. This strategy falls within the general framework of the paradoxes of argumentation.

2.4 Law of inversion

If p’ is stronger than p with respect to r, then not-p is stronger than not-p’ with respect to not-r. (Ducrot 1973, p. 239; 1980, p. 27)

— “Leo has a Bachelor’s degree” and “Leo has a Master’s degree” are two arguments for “Leo is a qualified person”.
— “Leo has a Master’s degree” is a stronger argument that “Leo has a Bachelor’s degree” for this conclusion: under normal circumstances, we can say:

Leo has the Bachelor degree and even a Master’s degree.

While “Peter has a Master’s degree and even a Bachelor’s degree” is incomprehensible. One can say, “he has a thesis, and even a Bachelor’s degree”, but with some irony on the value of diplomas S. A fortiori. If one wants to argue against Peter, to show that he is insufficiently qualified, one will say:

Peter does not have a Master’s degree, let alone a Bachelor’s degree.

The negation turns the weakest argument for qualification into the strongest argument for the lack of qualification.

Argument scales can express the argument a fortiori: “He doesn’t have a Bachelor’s degree, a fortiori he doesn’t have a Master’s degree”.

---

[1] Tariq Ali, Un Lynchage bien orchestré [A well-orchestrated lynching]. Afrique-Asie, feb. 2007.

1. Question as interrogation

A question may be a sentence “attempting to get the addressee to supply information” (SIL, Question), using the specific morphemes and syntactic transformation attached to the interrogative form.

— The fallacy of many questions or loaded question is one of the six Aristotelian linguistic fallacies, S. Fallacy (I). A loaded question is a question about a complex statement, containing several implicit statements. The loaded question presupposes the truth of these underlying statements, which may be disputed by the recipient of the question. Such questions are said to be oriented, S. Orientation.

— Rhetoric uses a series of common place ontological questions to gather information.

— A rhetorical question, in the traditional sense of the term, re-frames the argumentative question as a question admitting a self-evident answer, S. Argumentative question, §4

2. Question as problem

A question can also be the subject of a discussion, an “issue; broadly: a problem” (MW, Question). It doesn’t necessarily have an interrogative form.

An argumentative question represents the discursive confrontation generating an argumentative situation. Such a question does not refer to a quest for information, but to a problem. S. Argumentative question.

---

 

The arguments from wealth and from poverty are two subspecies of the argument from authority. Special weight can be given to the word of the wealthy — because wealthy, as well as to that of the poor — because poor. The Rich and the Poor are then believed on their word, and their words are exploited as an argument from authority by a speaker, who validates a position by putting it in the mouth of a rich or a poor man, S. Authority; Common place. Both arguments are extremely common and equally formidable.

Argument of wealth, or “top people” argument

The argument from wealth is the substrate of a family of discourses elaborating upon the key topic:

She is rich, therefore what she says is true; I consider her advice to be authoritative; she made the best financial decision; she has an extraordinary artistic taste, as proved by the value of her collections — I vote for her!

This argument easily extends from the rich to the upper classes and the ruling class, the most glamorous and lucrative professions, etc. It could be called “the top people” argument.

Argument of poverty: appeal to “people down there”

The argument of poverty is symmetrical to the argument of wealth. It validates a speech by the authority derived from poverty, “the poor are right”:

The poor are good, because they who have no money, and who has no money has no vice; they are not corrupt; what they say is authentic; they are the repositories of common sense; their opinions are basically sound.

As the argument of wealth, the argument of poverty extends beyond the poor to all “the people down there”, that is the exploited proletariat, the dominated, the lower 10%… as well as to country people, who live close to nature (naturalistic argument), or to the tramp as a wise philosopher … Truth comes out from their mouths, as it comes from the children’s mouth.

The adage vox populi vox dei, “the voice of the people is the voice of God”, which underlies the ad populum argument, is grounded in the argument of poverty and in that of number.

These arguments are different from the appeal to money, or the wallet argument, attached to the argumentation by punishment and reward, S. Threat and Promise.

In an argumentative exchange, the participants are part of a complex system of roles and characters, according to which they speak and act. Some of these roles are general; others are specific to the argumentative situation.

1. General interactional roles (not specific to argumentation)

1.1 Roles attached to the “participation framework” (Goffman)

The concept of a participation framework details and clarifies the traditional concept of a verbal exchange between a speaker and one or many listeners. The participation format is defined as a relation with two complex speech structures, the production format and the reception format (Goffman, 1981). These concepts are instrumental to the analysis of all argumentative interactions, from rhetorical addresses to everyday argumentative interactions. They are relevant to the analysis of the ethos, and the polyphonic structure of the argumentative text.

• Reception format (id., pp. 141-142)

The people who can actually hear the words said by a speaker occupy various statuses in relation to these words.

— The addressed participants are the people to whom the words are openly directed; the pronoun you refers to the addressed participant(s). Everyday group conversations show that successfully addressing a specific person may demand complex maneuvers.

— The ratified participants are the members of the group constituted around the ongoing speech event. They may be addressed or non-addressed. To get the floor in a discussion, one must be a ratified participant. Ratified, non-addressed participants may be addressed in the further development of the speech event.

In a codified dialectical exchange, the opponent is the only participant both ratified and addressed. Both participants alternately hold the floor. The referee of the debate, if there is one, is a ratified participant, who will be addressed only as a resource if a crisis looms, or during a planned slot in order to move forward, to evaluate and conclude the debate. If the debate is open to a broader audience, the members of the audience are ratified participants but not addressed participants.

In a classical rhetorical address, the audience is ratified and addressed. The difference with the dialectical situation is that the audience has no official right to the floor; nonetheless, it may applaud and boo its reactions (Goffman 1981).

— Overhearers and eavesdroppers. Any people passing within earshot are non-ratified participants. Overhearers accidentally hear the sounds and words of the conversation, possibly without even listening. Eavesdroppers intentionally spy on the conversation.

• Production Format

Speech is produced by the speaker. Goffman (1981) and Ducrot (1980) have independently shown that the speaker should not be considered to be a unified entity but as a complex articulation of different discursive personae; in Goffman’s words the Animator, the Author, the Figure and the Principal (id., p. 144; p. 167).

— The Animator (Goffman) is the talking machine, physically producing the discourse. In the same function, the Speaker is re-defined by Ducrot as “the empirical being” to which all the external determinations of speech can be attributed, “the psychological or even the physiological process which is at the origin of the utterance, the actual intentions, cognitive processes that have made the statement possible” (Ducrot 1980, p. 34).

The counterpart in the reception format of this talking machine is the hearing machine, that is the listeners, the whole range of ratified and non ratified participants, as persons who physically hear the speech and listen to it or not (Ducrot 1980, p. 35).

— The Author selects the thoughts expressed and the words to encode them. A speaker reading a book or quoting another person is the Animator of the words taken up without being their Author (Schiffrin 1990, p. 242). The pronoun I refers to the Author of the speech (except in quoted speech).

— The Figure corresponds to the discursive self-image of the Author, that is, the ethos.

— The Principal is “(in the legalistic sense) someone whose position is established by the words that are spoken, someone whose beliefs have been told, someone who is committed to what the words say … a person acting under a certain identity, in a certain social role” (Goffman 1987, p. 144). “The same individual can very quickly alter the social role in which he is active, even though his capacity as animator and author remains constant” (id., p. 145). The same Author can address a student as a teacher, as an adult, as a citizen, as a New Yorker, etc. Defined as “someone who believes personally in what is being said and takes the position that is implied in the remarks” (id., p. 167), the Principal plays a key role in the polyphonic space, as the person taking the responsibility of what is said (Ducrot’s énonciateur, 1980, p. 38). S. Interaction, Dialogue, Polyphony.

For example, in the following statement:

The weather is nice (V1), but I must work (V2)

the Author stages two voices:

— In V1, the voice of a person arguing that a nice weather is a good reason to have a walk.
— In E2, the voice of a person arguing that having to work is a good reason to stay home

The decisive point is that, the Principal identifies to voice V2; that is, the argument E1 is dropped, and the argument E2 validated.

There is no intrinsic superiority of argument E2 over E1. The speaker authors E1 and E2, and, as a principal, acts upon E2, not E1.

To sum up, “the Animator produces talk, the Author creates talk, the Figure is portrayed through talk, and the Principal is responsible for talk” (Schiffrin 1990, p. 241).

1.2 Roles attached to different types and genres of speech

To account for the variety of discursive genres, one must introduce new roles, such as narrator and narratee for storytelling; expert and profane for explanation; proponent, opponent and third party for argumentation (see below).

Interactional genres bring their share of professional or occupational roles: seller and customer for shop interactions; teacher and students in classroom interactions; physician and patient for therapeutic interactions, etc.

1.3 Socio-interactional roles

Linguistic roles combine with a set of social roles, in which we distinguish (after Rocheblave-Spenlé [1962]):

— Global social roles: gentleman, cool guy, cheerleader, troublemaker…
— Biosocial roles: young/old, male/female/transgender…
— Social class roles: bourgeois, aristocrat, white or blue collar…
— Professional roles: farmer, trader, truck driver…
— Community roles: religious believer, member of a trade union, a political party, a sport team…
— Family roles: husband, wife, child, father, uncle …

2. Argument-acting roles: Proponent, Opponent, Third party

The argumentative situation is defined as a three-pole situation, that is to say a three-role situation: Proponent, Opponent and Third Party. Each of these poles corresponds to a specific discursive modality, a discourse of proposition, supported by a proponent, a discourse of opposition, supported by the opponent, and a discourse of doubt or questioning, which defines the Third Party position.

2.1 Proponent and Opponent

The terms proponent and opponent are defined in dialectical theory, which frames argumentation as a game between these two partners, S. Dialectic. From an interactive perspective, the argument becomes dialectical when the third party is eliminated and each actor is assigned a role (“you will be the proponent, and I the opponent”) that must be assumed during the whole “dialectic round” (Brunschwig 1967). The elimination of the third party goes hand in hand with the expulsion of rhetoric and the constitution of a system of objective-rational norms. In a figurative sense, one could say that the third party is then replaced by Reason or Nature, in other words by the rules of truth.

If we take a rhetorical view of argumentation, the argumentative game is defined first as an interaction between a proponent, the speaker, and a third party, who is the silent audience to be persuaded. Opponent and counter-discourse are not absent, but are consigned to the background.

By getting into a discussion, participants acknowledge the fact that none of them has enough power or authority to decide on the matter at stake, and that they are engaged in a problematic situation.

2.2 Third Party

Considering that the argumentative question as a full systemic component of argumentative interaction emphasizes the role of the third party. This figure materializes what is publicly at stake and the contact between the contradictory discourses.

In its prototypical form, the argumentative situation appears as a situation of interaction between the speech of the proponent, the counter-discourse of the opponent, both mediated by a third, mediating, interrogative discourse. The third party stabilizes or manages the question, and decides upon the external relevance of the participants’ interventions. The argumentative situation, as embodied in an exemplary way in public adversarial exchanges is therefore a three-role situation. Basic argumentative situations such as political debates and cases being tried before the courts, are tripole. Argumentative speech is systematically multi-addressed, the addressee is not only, or not necessarily, the adversary-interlocutor, but in one case the public about to cast their vote, and in the other, the judge about to pronounce a verdict.

In contrast to the categorical assertions and denials of the proponent and the opponent, the third party may appear as a softer and undecided character. In reality, it is the third party who refuses to accept either of the opposing proposals or points of view, asking for more arguments, remaining doubtful and leaving the question open, in the name of making an informed decision. In this sense, and in accordance with the most classical concept of argumentation, the judge is the prototypical figure of the third party. In the third party position are all ratified participants of the argumentative situation who consider that the argumentative forces are balanced, or, more subtly, that even if one seems to prevail, the other cannot be considered to be null. In philosophy, the radicalization and reification of this position is elaborated as methodological skepticism.

Once the third party and the argumentative question have been accepted as key elements of the argumentative exchange, the proponent and opponent may be granted full responsibility for their speeches. One may answer, “No!” and the other “Yes, of course!” without either of them being systematically accused of manipulation, bad faith or other kinds of fallacious speech.

Institutions may stabilize the argumentative roles and their attribution to individuals. In an ordinary interaction, the argumentative roles correspond not to permanent roles but to footings in the sense of Goffman (1987, chapter 3), in particular, in that they are labile. In the same speech turn, the same person can take the role of both the third party and proponent in relation to an issue (affirming a position while expressing a certain degree of doubt about it), or speak as a Proponent on an issue and as Opponent on another.

3. Argument-actors [Fr. actants] and Actors [Fr. actors]

The individuals engaged in the argument are the physical participants, or actors of the argumentative situation. When clarification is needed, the expression argument-actor (Fr. actant, borrowed from semiotic theory), may be used to refer to the three basic argumentative roles, Proponent, Opponent, Third Party.

Any actor can occupy successively each of the three argument-acting roles (Fr. rôle actantiel), according to all the possible paths. For example, an actor may abandon his or her discourse of opposition in order to develop a discourse of doubt, switching from the argument-acting role of opponent to that of third party. An argumentative issue remains unsolved as long as the contradiction survives, even if some actors change their point of view. If two actors swap their argument-acting roles, that is to say, if they convince each other, the issue remains open.

In the case of an argumentative alliance, or co-argumentation, the same argument-acting position can be occupied simultaneously by several actors; that is to say, by several individuals producing co-oriented interventions. The study of the argument should focus as much upon co-oriented interventions as upon anti-oriented interventions.

 

The distinction between argument-actors / actors makes it possible to revisit the famous and strangely prized slogan “argument is war”, accompanied with its family of connected bellicose metaphors (Lakoff, Johnson 1980). Argument_2 may be a kind of war, luckily with a more limited number of casualties, but argument_1, that is argumentation, is not. The opposition between discourses, that is between argument-actors [Fr. actants], is not necessarily confused with possible collaborations or oppositions between people, that is, between actors [Fr. acteurs]. Argumentative situations are confrontational only when the actors identify themselves with their argumentative roles. In the most obvious case, that of internal deliberation, the same actor may progress quite peacefully go through all the argument-acting positions. If a group deliberates upon a question involving their common interest, it fortunately happens that the associated members will together examine the various facets of the problem, that is to say the different possible answers to the question and the arguments that support them. During this process, they systematically occupy the different argument-acting positions, without clear identification with one of these positions, and without necessarily transforming this process into a war between the actors. The argumentative situation is not inherently polemical; but it certainly can be so when the features defining the identity of the participants are involved and are put at risk in the discussion.

Ancient argumentative rhetoric is grounded in the natural speaking competence. This natural skill is developed through conceptualization and practical exercises concerning general or social issues. Such a rhetoric combines linguistic, interactional and citizenship’s competencies.

1. The rhetorical address

The rhetorical address corresponds to discourse in its traditional sense, that is to say, “that which, in public, treats a subject with a certain method, and a certain length” (Littré, [Discourse]); a discourse is a “formal and orderly and usually extended expression of thought on a subject.” (W., Discourse). This concept of discourse has nothing to do with the concept of discourse as defined by Foucault (1969, 1971) or Pêcheux (Maldidier, 1990). Moreover, this meaning of discourse does not appear among the six meanings considered by Maingueneau in his founding presentation of “French discourse analysis” (1976, p. 11-12).

A rhetorical address is a speech delivered by a speaker or orator to an audience, along the following lines.

— The orator deals with an urgent issue of general interest, typically he or she aims to influence an ongoing decision-making process developing under certain time constraints. Classical rhetoric considers an orator, addressing an audience. In reality, a full rhetorical situation is a situation of choice, involving as many orators or voices as there are possible choices.

— The speech is a relatively long, planned monologue composed of a set of speech acts constructing a unified representation, supposed to lead to action.

— It is produced in the context of discursive competition taking place between different speeches of mutual opponents, supporting incompatible proposals. The rhetorical address is given in a space of contradictory discourse, where all interventions are positioned in view of one another. Even if the speaker tries to erase all traces of the counter-discourses that surround him or her, the speech is nevertheless structured by the competing discourses.

— This speech is delivered to an audience, composed of all the people who will play a role in the decision-making process relevant to the matter in hand. The audience is divided in regard to what the right decision would be; it includes staunch supporters and opponents of each proposal, as well as undecided people, S. Roles. The focus traditionally put on persuasion suggests that the orator focuses upon those who doubt and question, more than upon the determined opponents. The job is to remove doubt, to create and lead the opinion, S. Logos, Ethos, Pathos.

The rhetorical audience is both lowered and magnified. It is lowered, because it is defined by its lack of knowledge, its indecision and dissension. But within the New Rhetoric framework at least, the audience is also magnified as a critical instance, somewhere on the way to achieving a universal, deeply rooted and justified consensus.
Argumentative rhetoric has theorized, codified, evaluated and stimulated this kind of public communication, which was the only kind of public address possible before the appearance of the radio, cinema, television and the internet. Its theoretical object, the circulation of contradictory speeches within a decision-making group, remains well-defined. S. Argumentation (II); Persuasion.

2. The rhetorical catechism

At least until the modern age, rhetorical argumentation was the backbone of teaching and education in the Western world. In the Middle Ages, rhetorical argumentation served as one of the three arts of speech constituting the trivium (grammar, logic, rhetoric), propaedeutic to the quadrivium (geometry, arithmetic, astronomy, music).

For pedagogical purposes, rhetoric has constructed a standard self-representation of both the production process of the address, and its product, the address as delivered to the audience:

— A five-step production process, invention, disposition, speech, memory, pronunciation.

— Three genres of discourse, deliberative, judicial, epideictic.

— Three actors: the rhetorical interaction is functionally three-poles. It brings together “the speaker who wants to persuade, the interlocutor who must be persuaded, and the opponent whom he must refute” (Fumaroli 1980, p. 3).

— Three discursive means of pressure focused on the transformation of the audience representations and desire for action. The speaker seeks:

To inform and teach, by his or her logos, that is the logic of the narrative and the argumen
To please and attract by his or her style, that is the self-image, or ethos, projected in the speech.
To move to action, through the pathos.

— According to the tradition, the acts aimed at producing these effects are concentrated in the strategic moments of the discourse:

The introduction is the ethotic moment.
The narration and the argumentation are ruled by the logos.
The conclusion is the pathemic, emotional moment, through which the speaker hopes to wrest the final decision.

3. Organizing the process

The process of constructing argumentative rhetorical discourse is traditionally described as involving five stages. The corresponding Latin words are mentioned in order to avoid possible confusion with the English words, of which they are false cognates.

(i) Inventio: Finding the arguments

“Invention [inuentio] is the devising of matter, true or plausible, that would make the case convincing” (Ad Her., i, 3). The inventio is the cognitive step corresponding to the methodical search for arguments, guided by the technique of “topical questions”, S. Common Places.
The Latin word inventio does not mean “invention” taken as a creation of something that did not exist before. The meaning is “to find, to discover” (Gaffiot [1934], Inventio).
Psycho-linguistic research on the production of written and oral discourse has taken over the reflection on the inventio techniques.

Rhetorical arguments are found on the basis of an exploration of reality, guided by a natural, substantial ontology. Religious argument has introduced a fundamental change in this vision. Good reasons are not statements expressing sense data or elaborated intellectual conceptions, but are sacred statements drawn from the foundational sacred text and, to a lesser extent, from the texts of tradition.

(ii) Dispositio: Planning the argumentation

“Arrangement [dispositio] is the ordering and distribution of the matter” (ibid.), that is, speech planning. Inventio and dispositio are the two cognitive stages of this process.

(iii) Elocutio: Expressing the argumentation

“Style [elocutio] is the adaptation of suitable words and sentences to the matter devised” (ibid.). The word style used in translation may evoke a superficial arrangement of the expression, but the elocutio is more than that, it corresponds to the “putting into language” of the arguments, to their semantization, corresponding to the whole linguistic expression.

The elocutio is characterized by four qualities, the grammatical correctness (latinitas), the clarity of the message (perspicuitas), the customization of the message to suit the audience (aptum) and the density and wealth of its expression (ornatus). A discourse may be rejected as defective on any of these levels, S. Destruction.

The English word elocution currently refers to “the skill of clear and expressive speech, especially of distinct pronunciation and articulation” (W., Elocution); elocution clearly belongs first to pronuntiatio, and only peripherally to elocutio, as expression and style.

(iv) Memoria: Memorizing the speech

The discourse must be memorized since it is intended to be delivered orally, without the use of paper documents or autocues. As the invention, the memory involves cognitive factors. The cultural import of this memorization work, which might seem anecdotal, was revealed by Yates (1966).

(v) Pronuntiatio: Delivering the speech

“Delivery [pronuntiatio] is the graceful regulation of voice, countenance, and gesture” (ibid.). The Latin word pronuntiatio refers not only to this physical process of speech production and modulation, but also expresses the idea of an ​​assertive speech: a pronuntiatio is a “declaration, announcement, proposal” (Gaffiot [1934], Pronuntiativus). The judge does not say or read the verdict, he or she pronounces it. Rhetorical tradition sees delivery as the moment of performance, and dramatization of discourse, requiring a special education of the body, the gesture and the voice. The orator, the preacher, the actor are under the same public performance constraints, although their techniques, social statuses and messages are quite different.

In short, finding arguments, ordering them, expressing them in writing: the rhetorical prescriptions are particularly suited to general academic essays. They seem clear and they are easy enough to teach ­— but, unfortunately, not so easy to put into practice.

In the Divisions of Oratory Art, Cicero has framed the concepts of ancient rhetoric as a succession of question-answers, “very similar to a catechism”, as Bornecque notes ([1924], p. VII). Rhetoric may have suffered from such allegedly pedagogical representation where everything has to be done and said by the book.

4. Textual organization of the speech

This process leads to a finished product, the speech delivered in a specific situation. It is articulated in parts, traditionally named:

Introduction (exordium)
Narration
Argumentation (a confirmation followed by a refutation)
Conclusion.

The argument is the central part of the speech. Contrary to a simplistic vision of discourse, there is no opposition between argumentation, narration and description. Argumentative narrations or descriptions, like literary narrations or descriptions, are made from a particular viewpoint.

5. Extensions and restrictions of rhetoric

Ancient argumentative rhetoric has been redefined on various dimensions.

— Restriction to its expressive dimension. Argumentative rhetoric can be oriented towards persuasive communication or the quality of expression.

— Generalization to its persuasive dimension. Nietzsche assimilates the rhetorical function to the persuasive function of language, S. Persuasion.

— Restriction to its linguistic dimension and liquidation of its cognitive dimensions. The apparent logic of the five components of rhetorical production was profoundly challenged in the Renaissance (Ong 1958). The three components related to thought (invention, disposition, memory) were separated from those related to language (expression and delivery). Inventio, the bones of argumentation, rejected out of rhetoric and language was no longer considered to be the fundamental moment of the discursive process. An orphan of the inventio, rhetoric redefined its object, moving away from social discourses to focus on literature and belles-lettres, and developing a passion for the autonomous study of the discourse variations and figures of style.

A language deprived of thought and a thought deprived of language: this orphaned rhetoric would become the target of violent attacks from Locke, S. Ornamental fallacies. In France, in the nineteenth century, Fontanier ([1827], [1831]) was the emblematic figure associated with this “restricted rhetoric” (Genette, 1970), in opposition to the so-called “general” rhetoric, which was revived by Perelman & Olbrechts-Tyteca (1958) — The question of a revival of an integral concept of rhetoric remaining a topos of rhetorical studies.

— Generalization along its linguistic dimension. A rhetoric restricted to figures of speech can itself be called “general”: this paradoxical expression corresponds to the “Group μ” approach in their General Rhetoric (1970). The problems of figures are taken up in a structuralist framework, and figures are reconsidered under the two basic dimensions of language, the syntagmatic and paradigmatic axes. Issues of argumentation, public speaking, interaction or communication, are not accounted for, nor indeed are the aesthetics of figures. This General Rhetoric was virtually the only concept of rhetoric to be considered in the French literature during the 1970s, and Perelman’s New Rhetoric occupied only a marginal position. Wenzel devoted an avenging paragraph to this “alarming” view of rhetoric (1987, p. 103; see Klinkenberg, 1990, 2001). The contrast with the status of rhetoric in the United States’ Speech and Communication Departments could not be greater.

— Extension to ordinary speech. The rhetorical approach can be extended to everyday forms of talk, insofar as they involve face management (ethos), data processing oriented towards a practical end (logos), and a correlative treatment of the affects (pathos) (Kallmeyer, 1996). The rhetorical trilogy can thus be regarded as the ancestor of the different theories of the functions of language (Bühler 1933, Jakobson [1960]), in a completely distinct theoretical atmosphere. This extension also retains a fundamental characteristic of rhetorical speech: to alter reality and participate in the structuration of ongoing actions. This view may resonate with Bitzer’s evocation of the dialogue between fishermen at work in the Trobriand Islands, and his definition of the “rhetorical situation” as involving a degree of “urgency”:

Rhetorical situations may be defined as complexes of persons, events, objects and relations presenting an actual or potential exigence, which can be partially or completely removed if discourse, introduced into the situation, can so constrain human decision or action as to bring about the significant modification of the exigence (Bitzer [1968], p. 5).

— Extension to any semiotic domain. Rhetoric naturally extends to the co-verbal and paraverbal signifiers. Moreover, any strategic implementation of a semiotic system can be legitimately regarded as a rhetorical practice; the rhetoric of painting, of music, of architecture, for example.

---