The long history of argumentation studies cuts across the history of rhetoric, dialectic and logic. Argumentation studies appeared as autonomous field only after the Second World War; it is nevertheless possible to note inflections during this short history.

1. The long history: dialectics, logic, rhetoric

Greek and Latin Antiquity ­— From the perspective of classical disciplines, argumentation studies are related to logic, “art of thinking correctly”; to rhetoric, “art of speaking well and addressing a group”; and to dialectics, “art of interacting well, articulating one’s intervention and thought with those of others”. This triad is the basis of the system in which argumentation was conceptualized, from the time of Aristotle until the late nineteenth century. Argumentation is seen as a theory of convincing reasoning in ordinary language. The central issues are argument scheme theory, and validity and soundness theory, depending on the quality of the premises and the reliability of the principles used to derive conclusions from these premises. S. Dialectic; Logic; Rhetoric.

Modern Times — Walter Ong has commented upon the decline of dialectical practices (1958) since the Renaissance, the reduction of rhetoric to figures of speech and considerations of literary style, and the critique and rejection of the Aristotelian logic as an exclusive or essential instrument of scientific thought. New scientific methods based on observation and experimentation, making increasing use mathematics, are looked for.

Late nineteenth, early twentieth century — At the end of the nineteenth century rhetorical argument is delegitimized as a source of knowledge. Logic is formalized and becomes a branch of mathematics. The tradition of argumentation studies remains active in law and theology.

2. A symptom: the titles

In French, until the publication of Perelman & Olbrechts-Tyteca’s Treatise on Argumentation, the books entitled Argumentation were pamphlets containing arguments about specific topics, not theoretical books about argumentation in general, as shown by their complete titles:

1857 – Discussion About Etherization Considered from the Standpoint of Medical responsibility — Argumentation. By Marie Guillaume Alphonse Devergie.
1860 – Arguments on Administrative Law of the Municipal Administration. By Adolphe Chauveau.
1882 – The Issue of Water Before the Medical Society of Lyon. Argumentation in Response to Mr. Ferrand. By Mr Chassagny. P.-M. Perrellon.
1922 – Argumentation of the Polish Proposal About the Border in the Industrial Section of High-Silesia.

The substance and field of the argument is specified by an additional subtitle: argumentation on, about … The title Argumentation corresponds to modern titles such as “An Essay on —” or “Thesis”; it refers to a textual genre. Thus, it seems that the emergence of the genre “[Theoretical work on] Argumentation” came with the disappearance of the genre « Argumentation [on —]« .

In English – Toulmin’s book “The Uses of Argument” (1958) comes apparently in a traditional line of books titled “Argument”. Some of these books offer “an argumentation” in support of a position, such as the following:

Yale C., Some Rules for the Investigation of Religious Truth; and Some Specimens of Argumentation in its Support, 1826.

Others are textbooks for composition and debate teaching:

Brewer E. C., A Guide to English Composition: And the Writings of Celebrated An- cient and Modern Authors, to Teach the Art of Argumentation and the Development of Thought, 1852
Foster, W. T., Argumentation and Debating, 1917.
Baird A. C., Argumentation, Discussion and Debate, 1950.
Lever R., The Arte of Reason, Rightly Termed Witcraft; Teaching a Perfect Way to Argue and Dispute, 1573.

The best known may be:

Whately R., Elements of Rhetoric Comprising an Analysis of the Laws of Moral Evidence and of Persuasion, with Rules for Argumentative Composition and Elocution, 1828.

In the first half of the twentieth century, many such books are published, where didactic purposes mingle with more theoretical considerations. But the work of Toulmin does not fit at all in this tradition, linked to the practices of the Speech Communication Departments or of the English Departments in the United States. No book of that kind is listed in his bibliography, and he quotes no work coming from the field of rhetoric.

Actually, Toulmin and Perelman both break with a modern tradition and establish a new foundation in the treatment of the concept of argument.

3. 1958 and after: Constitution of the field of argumentation studies

3.1 A key date, 1958 

Chaïm Perelman, Lucie Olbrechts-Tyteca, 1958, Traité de l’Argumentation. La Nouvelle Rhétorique = 1969, The New Rhetoric — A Treatise on Argumentation.
Stephen E. Toulmin, 1958, The Uses of Argument.

These two titles are the best known in an impressive constellation of works that all help define, positively or negatively, the new field of argumentation studies.

— On “Public Relations”: a non rhetorical and non argumentative perspective on persuasion:

Vance Packard, 1957, The Hidden Persuaders.

— On the language of propaganda:

Sergei Chakhotine, 1939, Le Viol des foules par la Propagande Politique.
= 1940, The Rape of the Masses – The Psychology of Totalitarian Political Propaganda.
Jean-Marie Domenach 1950. La Propagande Politique [Political Propaganda]

— In law:

Theodor Viehweg, 1953, Topik und Jurisprudenz. Ein Beitrag zur rechtswissenschaftlichen Grundlagenforschung = 1993, Topics and Law. A Contribution to Basic Research in Law.

— On the rhetorical foundations of literature and Western culture:

Ernst Robert Curtius, 1948, Europäische Litteratur und Lateinisches. Mittelalter.
= 1953, European Literature and the Latin Middle Ages.

— An historical and systematic reconstruction of the field of rhetoric

Heinrich Lausberg, 1960, Handbuch der literarischen Rhetorik.
= 1998, Handbook of Literary Rhetorik. Foundation for Literary Study.

— A history of the adventures of dialectic and rhetoric at the time of the Renaissance

Walter J. Ong, 1958, Ramus. Method and the Decay of Dialogue.

3.2 Extended theories of argumentation

These theories have been developed since the 1970s, mainly in French:

— In a linguistic perspective:

Oswald Ducrot, 1972, Dire et ne pas Dire [To Say and Not To Say] — 1973, La Preuve et le Dire [Proving and Saying] — & al. 1980, Les Mots du Discours [The Words of Discourse] Jean-Claude Anscombre et Oswald Ducrot, 1983, L’Argumentation dans la Langue [Argumentation within Language]

— In a discursive and cognitive perspective:

Jean-Blaise Grize, 1982, De la Logique à l’Argumentation [From Logic to Argumentation]

3.3 The dialectical and critical approaches

Perelman & Olbrechts-Tyteca work is considered to be a revival of rhetorical argumentation, originating in Aristotle’s Rhetoric. Along the same line, Hamblin’s foundational work revived argumentation as a dialectical and critical thinking, based on concept of fallacies, and originating in Aristotle’s On Sophistical Refutations:

Charles L. Hamblin, 1970, Fallacies

3.4. The Pragma-Dialectical trend

From the 1980s on, Frans van Eemeren and Rob Grootendorst have developed the “Pragma-dialectical” approach. They recast the study of argumentation in terms of speech acts, linguistic pragmatics and a new conception of dialectic. They elaborated a powerful system of guidelines for the evaluation of arguments as a system of rules for the rational resolution of differences of opinion, S. Norms; Rules; Evaluation.

Frans H. van Eemeren & Rob Grootendorst, 1984, Speech Acts in Argumentative Discussions A Theoretical Model for the Analysis of Discussions Directed Towards Solving Conflicts of Opinion.
Frans H. van Eemeren & Rob Grootendorst, 1992, Argumentation, Communication, and Fallacies.
Frans H. van Eemeren & Rob Grootendorst, 2004, A Systematic Theory of Argumentation – The Pragma-Dialectical Approach.

Since 1986, every four years, a reference conference on argumentation is organized in Amsterdam. The series of Proceedings propose an up to date vision of the discipline (van Eemeren & al. (1987, 1991, 1995, 1999, 2003, 2006, 2010).

3.5 The Informal Logic trend

The “Informal Logic” of Anthony Blair, Ralph Johnson, Douglas Walton and John Woods connects argumentation studies to a logic and to a philosophy which take into account the ordinary dimensions of speech and reasoning. The focus is on the evaluation of the arguments and their educational applications in the development of critical thinking. The concept of argument scheme has been defined so as to integrate their corresponding counter-arguments, and developed on this basis a new approach to argument criticism.

Howard Kahane, 1971, Logic and Contemporary Rhetoric The Use of Reason in Everyday Life.
Ralph H. Johnson & J. Anthony Blair, 1977, Logical Self Defense.
Ralph H. Johnson, 1996, The Rise of Informal Logic.
Anthony Blair & Ralph H. Johnson, 1980, Informal Logic – The First International Symposium.
John Woods & Douglas Walton, 1989, Fallacies. Selected Papers 1972-1982.
Douglas Walton, Chris Reed & Fabrizio Macagno, 2008, Argumentation Schemes.
Anthony Blair, 2012, Groundwork in the Theory of Argumentation.

3.6 Argumentation and ordinary interactions

The Pragma-Dialectic and the Informal Logic schools of argumentation give special importance to dialog. The first papers integrating the perspective of conversation and interaction analysis are found in:

Robert Cox & Charles A. Willard (eds), 1982, Advances in Argumentation Theory and Research.
Jacques Moeschler (1985). Argumentation et Conversation. [Argumentation and Conversation] Frans H. van Eemeren & al. (eds), 1987, Proceedings of the [ISSA] Conference on Argumentation 1986.

4. Relations with other disciplines

The leading research programs maintain different relationships with the rhetorical, dialectical and logical heritage, as well as with language studies philosophy and education. The table below tries to give an idea of these links.

0: no significant link

+: the number of stars indicates the importance of the link

5. Dialogues between main trend theories

The arrows represent commonalities, solidarities or affiliations between different schools

6. Argumentation studies, argumentation scholars:
How to name the field and its specialists?

The talk about of the “revival of the field of argumentation” in the fifties should be taken with precaution. Firstly, the expression is ambiguous: the talk is not about the field of argumentative practices; but about the theory of argumentation, the meta-language used to study this practice. Secondly, it is also slightly simplistic: although discontinuous, reflections on argumentation have been underway for more than two millennia, not half a century. The point is that, since the fifties, a learning community has formed around a vast and differentiated corpus of studies taking for object a set of practices directly characterized as argumentative.

How to designate a field of study, its object and its specialists? The situation is clear when each of these distinct realities is designated by a specific term. This is the case for example with the economists, specialists of economics, whose object is the study of economy (production and consumption of goods and services). But the term argumentation refers to both the object of study, as in “everyday argumentation”, and to the study itself, when, especially in the titles of books where “argumentation” shortens “theory of argumentation”.

The spectacular appearance of papers and books entitled “… Argumentation …” hides a deeper reality, the change in the disciplinary status of logic. All ancient books entitled Logic, dealing with the logic of terms, quantifiers, connectors, analyzed and non-analyzed propositions, etc., are actually theories, logic-based treatises on argumentation, as, for example the Port-Royal Logic, or The Art of Thinking ([1662]). Basically, we now use the word argumentation to refer to a field of study or to a theoretical book because, since the mathematization of logic in the late nineteenth century, the title Logic can only be used in the domain of formal logic, and is no longer available as referring to natural language argument. Exceptions are rare. In French, one can think of works such as the Elements of classical logic (François Chenique 1975, vol. I: The art of thinking and judging; t II. The art of reasoning), or especially Jacques Maritain’s Introduction to Logic ([1923]), which is perhaps one of the last books providing under the heading Logic a traditional “art of thinking”, inspired by neo-Thomist philosophy. This logic is, in this respect, the first in the series of “non formal”, “substantial”, “natural” logics… that flourished at the end of the last century; it is a treatise of argumentation as a theory of logical reasoning within natural language.

So we are left with the problem of naming the field by a single unambiguous term. Following the example of polemology, that is war studies, it might be argumentology. Along the same line, the corresponding professionals would be called argumentologists, a figure clearly distinct from that of the arguers. But these words sound jargon-ridden and slightly ridiculous. Anyway, usage will have the last say, and presently nobody seems to feel an urgent need for such words. Argumentology does not appear in the monumental and fundamental Proceedings on the Fourth International Conference of the International Society for the Study of Argumentation of 1999; one case in 2003, one in 2007; and no occurrence of argumentologist or any derivative name of that kind (van Eemeren & al. (eds.), 1999, 2003, 2007).

The explosion in theoretical questioning of the notion of argumentation at the end of the twentieth and beginning of the twenty-first centuries (van Eemeren & al. 1996; 2014), and the multiplicity of disciplines interested in the topic encourage the characterization of the domain according to an underlying system of key features, issues and orientations.

1. Key issues about the role of language

The following table proposes a possible organization of the field according to the role of language and the kind of speech situation which is given theoretical prominence. This hypothesis makes it possible to represent the various concepts of argument as a tree structure, where the nodal points correspond to research questions, or crossroad questions, which articulate the field. Such a representation illustrates that what could at first sight seem to be an arbitrary dispersion of options, in fact reflects the necessity of taking the complex range of argumentative situations into account. A vision of argumentation might be characterized as a structured choice between the various options opened by the following questions (other possible points of departure are suggested in §2).

Table : Key features and issues
about the role of language in argumentation

­­­

(1) Argumentation

(2) AS A THOUGHT ACTIVITY:
             Study of reasoning as a psycho-cognitive process

(3) AS A LINGUISTIC-COGNITIVE ACTIVITY

(5) Extended                                              

(7)  Form of language:       “ARGUMENTATION WITHIN LANGUAGE” 

(8) General form of discourse:       “NATURAL LOGIC”

(6) Situated

(9) monologue 

non polyphonic:       LOGIC AS AN ART OF THINKING   

polyphonic:       « BENE DICENDI” RHETORIC

(10) dialogue

without  turn-taking:       RHETORIC OF PERSUASION

 with turn-taking:       DIALOGUE LOGIC
                                          INTERACTION                      

(2) vs. (3) vs. (4): The cognitive, linguistic and multimodal dimensions of argument

Various general questions might be taken as points of departure, and each question would produce a different mapping of the field. This map is born of the general question: is argumentation basically a language activity or a cognitive activity — or both?

If argumentation were defined as a pure activity of thought, expressed in a perfectly transparent language, argumentation studies would correspond to a psychology of reasoning without language.

But, in the same way as everyday argumentation, mathematical thinking and scientific reasoning require a language. Language-based approaches to argumentation deal with the cognitive component within the linguistic component. Such approaches are compatible with various positions on the question of thinking and reasoning. Classical logic, Natural Logic, Informal Logic and cog-nitive approaches stress the articulation of thought and language in the argumentative activity.

Argumentation is unanimously considered to be a discursive practice. The consideration of still and moving images raises questions about how argumentative meanings are able to invest nonverbal semiotic supports. Research on argumentation in working situations also demands that we take the signifying intention steering both the action and the argument into account. In both cases, it is necessary to reconsider what exactly constitutes a well-built corpus within the field of argumentation.

(5) vs. (6) — Argumentation as a linguistic-cognitive activity: Extended or situated?

Should argumentation, as a linguistic-based cognitive process, be considered a local or a generalized phenomenon?

(7) vs. (8) — Extended argumentation: Saussurian langue or discourse?

Two different theories have extended the concept of argumentation to all linguistic activities, the theory of Argumentation within Language (Anscombre, Ducrot 1983) and the theory of argumentation as a Natural Logic (Grize 1982).

The former generalizes the concept of argumentation at the level of language (of Saussurian langue), whereas the latter enacts the same generalization at the level of speech (parole).

(7) Argumentation, as a condition on well-formed linguistic chain {E1, E2}:
S. Orientation

(8) Argumentation as a schematization of the situation

(9) vs. (10) — Situated argumentation: Monologue or dialogue?

If argumentation is limited to some characteristic forms of discourse, then in which kind of discourse is it best exemplified, in monological discourse, or in dialogue?

(11) vs. (12) — Monologue: Logic or rhetoric?

(11) Logic

(12) Bene dicendi rhetoric, S. Rhetoric

(13) vs. (14) — Dialogue: With or without turn-taking?

According to the externalization principle (van Eemeren and Grootendorst 1992, p. 10), dialogic theories consider either that dialogue is the basic form of argumentative activity, or that it is in the form of a dialogue that argumentative mechanisms of argument, can be most clearly seen.

Within this set of dialogic approaches, there are distinctions. Has the dialogue an exchange structure or not? Does the dialogue admit turns of speech? Do all the participants have equal possibility of taking the floor in the same conditions?

(13) Argumentation, a dialog without exchange structure: The rhetorical address

The rhetorical address is a special kind of dialogue, having a polyphonic structure; the voices of the others, especially the voice of the opponent, are re-built into the discourse of the speaker who holds the floor. The audience will give its answer only later and indirectly, as a judgment on the case or a decision on the policy.

(15) vs. (16) — A turn-taking dialogue: Dialogue logic or natural interaction?

In the case of a dialog in which there is a possibility of exchange, one of the two following poles will provide the appropriate baseline, 1) a logical approach to formal dialogues, or 2) an empirical approach to natural interactions.

(15) Argumentation, a formalized critical dialogue

Since the 1970s the Informal Logic and the Pragma-Dialectic theories have re-orientated argumentation studies by giving the priority to the study of argumentation as a kind of dialogue.

Dialectical critical theories of argumentation strengthen the constraints on the dialogue either by means of a system of rules designed to embody a rational standard, as in Pragma-Dialectic, or by means of a system of critical questions, as in Informal Logic. S. Norms.

 (16) Argumentation, a kind of ordinary interaction

Proto-argumentative activity is triggered by a lack of ratification by the addressee. Depending on the reaction of the interaction partners, conversational disorder might pass quickly, being absorbed into the flow of the on-going task they are engaged in. Otherwise, the interaction might develop into a fully-fledged argumentative situation. In all cases, the argumentative situation is basically ruled by interactional principles.

This vision is compatible with the ancient theory of “argumentative questions” (or stasis, or point to adjudicate).

For each of these points, the question is not which to adopt and which to exorcise, but to clearly articulate the contrast between the approaches they define.

2. Other points of departure

The above table develops from the question of language. Other questions might give rise to alternative maps of the field.

2.1 Kind of rationality?

Truth and rationality can be considered:

• As an attribute of a well-thought monological discourse, best exemplified in logic, as an art of thinking;
• As the consensus of the properly defined universal audience, within the prospect of a rhetoric of persuasion;
• As a social production, the result of a well organized critical dialog to reach the best possible true and rational answer in the course of a dialectical process;
• A a progressive construct, through a closer contact with scientific results, thought and method.

In complete opposition to these guidelines, generalized theories of argumentation maintain an agnostic perspective on rationality, and question the very possibility of reaching it through ordinary discourse.

2.2 Form or function?

Is argumentation (first, better) defined by its function or by its form? This question opposes two theoretical families, one focusing on persuasion, and the other focusing on the structural description and formal representation of argumentative episodes. These two starting points themselves give rise to symmetrical questioning: how to deal with functional aspects in the latter case? What are the structural criteria that ensure the descriptive adequacy of the in the former case?

2.3 Argumentativity, a binary or gradual concept?

For extended theories of argumentation, language (Ducrot) or discourse (Grize) are basically argumentative, S. Orientation; Schematization.

In the case of restricted theories of argumentation, however, some discursive genres (deliberative, epideictic, judicial) or, more broadly, certain kinds of discursive sequences are argumentative and opposed to other non-argumentative genres or other types of sequences. These definitions tend to consider that argumentativity is a binary concept: a sequence is or is not argumentative.

In reference to the language exchanged between partners defending contrasting positions, the argumentativity of a situation is not an all or nothing concept; various forms and degrees of argumentativity can be distinguished.

— A given linguistic situation begins to become argumentative when opposition emerges between two lines of speech, quite possibly without reference to each other, as in an argumentative diptych. This is most probably the basic argumentative structure, each partner repeats and restates his position. S. Disagreement. We can thus go beyond the opposition between narrative, descriptive or argumentative sequences. When a description or a narration is developed in support of an answer to an argumentative question, this narration or description should be considered as fully argumentative and evaluated as such.

— Communication is fully argumentative when the difference is problematized as an argumentative question, with the participants taking roles as proponent, opponent, or third party, S. Roles.

2.4 Central objects?

The various approaches to argumentation are characterized by the nature of their internal assumptions and external assumptions. The former correspond to the organization of the concepts postulated in the system, and the latter, to the kinds of objects taken into consideration. Both types of hypotheses are bound.

The extremities of the branches in any of the preceding “decision trees” represent a pole articulating theoretical views with specific “preferred” objects. To satisfy the requirement of descriptive adequacy each theory must combine its central objects with what it posits as peripheral objects. Decisions as to what is to be considered as central and as peripheral (derived or secondary) data, fall within the domain of external assumptions. Such choices are never self-evident and require justification. So, for example, the decision to give priority to dialogue or to take as reference monologal syllogistic discourse, correspond to two distinct external assumptions regarding the structure of the argumentation field, and clearly put to the fore quite different kinds of data.

This does not imply that second level (often annoying) facts and data are excluded, rather that all phenomena cannot be put on the same level; data must be ordered, and prioritized. In practice, the problem is to determine how the results established on the basis of central facts can be expanded to peripheral data.

Some major types of coupling of internal and external assumptions:

— Rhetorical argumentation, and planned monological speech.
— Dialectical argumentation, and conventionalized dialogues.
— Argumentation as orientation, and pairs of statements.
— Argumentation as schematization, and texts, etc.

Argumentation analysis has been intensely and specifically investigated since the post-second world war period (references infra).

The bi-millennial framework of logic as an “art of thinking” in natural language has been taken up and reworked in the new intellectual framework of the post-Fregean mathematical logic as a Substantial Logic, an Informal Logic, or a Natural Logic.

A new vision of argumentation as discourse orientation has been developed in the semantic theory of Argumentation within Language.

Ancient rhetoric has been reshaped into a New Rhetoric. Dialectics has been revisited in relation to pragmatics and speech acts theories, and expanded into a powerful critical instrument within the Pragma-dialectic framework.

The prospects of rhetoric and dialectic are now ubiquitous in contemporary studies and teaching programs on argumentation. The links between rhetoric, text linguistics and discourse analysis have been recognized and rearticulated.

The spectacular results obtained in interaction analysis have opened the immense field of everyday conversational interactions as a specific investigation domain, where argument as “dispute” intertwines with argument as “good reason”.

The different theories of argumentation developed in the late twentieth century are based on different visions and definitions of their objects, methods and goals. Given this diversity, and the apparent and real discrepancies between definitions, there is a real temptation of synthesis, that is, to look for a definition which, while not trivial, will restore order, unity, simplicity and consensus.
Experience shows, however, that many new definitions meant to supplant older ones, merely add to the existing lists, thereby further aggravating the problem that they were intended to solve.

Another solution could be to start with things as they are, that is, to admit that the field of argumentation studies does not develop in the hypothetical-deductive style, starting from an overwhelming “master definition” and deriving its consequences, but rather in a more empirical, data driven, manner.
In practice, this suggests that one can very well start with a corpus of definitions of the concept of argumentation in order to identify the points of consensus and divergence, while emphasizing the points of view that have proven to be the most fertile

1. Rhetorical argumentation, an instrument of persuasion

Socrates considers and rejects rhetoric as an enterprise in social persuasion through speech. He shares this definition with his opponents, in particular with Gorgias:

Gorgias — I’m referring to the ability to persuade by speeches judges in a law court, councilors in a council meeting, and assemblymen in an assembly or in any political gathering that might take place. (Plato, Gorgias, 452e; p. 798)

Socrates — Well, then isn’t the rhetorical art, taken as a whole, a way of directing the souls by means of speech, not only in the law courts and on other public occasions, but also in private? (Plato, Phaedrus, 261a ; CW, p. 537)

This defines the common use of the word rhetoric in ancient Greece, what people call rhetoric.
Now what rhetoric is, in its substance — or lack of substance — is another story:

By my reasoning, oratory is an image of a part of politics. (Plato, Gorgias, 463d; CW, p. 807)

Politics is defined as the craft of addressing “the soul » (ibid, 464b, p. 808), and rhetoric is disposed of as an unsubstantial “image”, an eidolon, a counterfeit of politics. Socrates unreservedly condemns rhetorical discourse aimed at persuasion, as a lie, an illusion, a manipulating enterprise, antagonistic to truth-seeking philosophical discourse.
This unqualified and irrevocable condemnation of rhetoric as a fake is at the root of the popular negative meaning of the word, and this obviously includes argumentative rhetoric as well. The criticism of rhetoric is part of the field of rhetoric, and the same applies to the field of argument.

Aristotle positions rhetoric not as a counterfeit but as “the counterpart of dialectic” (Rhet, I, 1, 1354a1; RR p. 95) and defines it as an empirical techne, a craft, oriented towards the study of specific cases:

Rhetoric may be defined as the faculty of observing in any given case the available means of persuasion (Rhet, I, 2, 1355b25; RR, p. 105).

Cicero follows this functional definition:

Cicero Junior: — What is an argument?
Cicero Father — A plausible device [probabile] to obtain belief.
Cicero, Part., II, 5; p. 315

Crassus — As becomes a man well born and liberally educated, I learned those trite and common precepts of teachers in general; first, that it is the business of an orator to speak in a manner adapted to persuade. (Cicero, De Or., I, XXXI; p. 40)

Likewise, the “New Rhetoric” of Perelman & Olbrechts-Tyteca focuses on persuasion:

The object of the study of argumentation is the study of the discursive techniques allowing us to induce or to increase the mind’s adherence to the theses presented for its assent. ([1958], p. 4; italics in the original)

By focusing on “discursive techniques” and on “the mind’s adherence”, this definition re-builds argumentation studies on the same basis as those of the Aristotelian argumentative rhetoric, persuasive speech. It re-connects contemporary understanding of argumentation with the experience gained throughout two millennia.

Thesis, mind, presented, assent, discursive techniques: this definition articulates the core concepts of what could be called “the argumentation movement” as a vision of man and discourse in modern democratic societies.

— The claims are theses. This is a philosophical term; the issues covered by argumentative interventions are complex and high level, “the most rational” (id., p. 7). The Treatise keeps its distances from everyday argument and minds: it does not address the ignoramus, and more: “there are beings with whom any contact may seem superfluous or undesirable…” (id., p. 15).

— These theses are presented to and not imposed on the audience.

— Moreover, they are presented to the audience’s mind, that is to say to men and women endowed with a choice and decision-making capacity; and living under social conditions that allow them to fully exercise this capacity.
This action upon minds can be opposed to the manipulation of souls and bodies: souls with their capacities of emotion and sensibility / sensitivity to romantic or mystical appeals; bodies which can be forced to march or vibrate in unison under a musical mantra or image.

— The assent results from an explicit judgment of a free and conscious mind. Assent can be given or withdrawn. Expressing one’s assent is opposed to producing a response under the causal pressure of a stimulus.

— Finally, argumentation is a discursive technique, that is, a form of speech in which speakers can practice and improve.

— The Treatise does not deal with fallacies, but the evaluation of argument is a key issue of the book. The sound criticism and evaluation of arguments is not a matter for the orator, but for the partner audiences, particular and universal.

2. Argumentation as a way to deal with stasic situations

The Rhetoric to Herennius by an unknown author of the first century BC (formerly attributed to Cicero) articulates argumentative rhetoric with the key concept of stasis. In court, the contradiction brought by one party to another party determines the “point to adjudicate” and produces a stasis, which defines an argumentative situation:

The Point to adjudicate is established from the accusation and the denial, as follows: Accusation: ‘You killed Ajax.’ Denial: ‘I did not.’ The point to adjudicate: Did he kill him?
(To Her., I, 17; p 53)

Argumentation can be thus defined in general as an institutional instrument developed institutionally to deal with and settle stasic situations. S. Argumentative Question.

3. Argumentation as “substantial logic” and default reasoning

According to Toulmin’s “layout of argument”, the argumentative passage is defined by its structure.

— A speaker puts forwards a Claim, based on Data oriented by general rules or principles, the Backing, and the Warrant, defining the monologal assertive component of argumentation.

— The Claim is defeasible under certain Rebuttal conditions, expressed by a Modal affecting the Claim. This reservation component refers to a dialogic and critical approach of argumentation.

The combination of an assertive and a refutative components into an “argumentative cell”, both linguistic and cognitive, defines reasonable-rational discourse.

This Toulminian complex layout is often reduced to the main parts of its assertive component “Data, Claim”,

Slavery was abolished, why not prostitution? I do believe in the progress of civilization.
When snakes come out, it’s going to rain. We know that from experience.

Toulmin makes no reference to rhetoric. But as Bird has pointed out (1961), with his warrant and backing, Toulmin has “re-discovered” the more than two-thousand-year-old concept of topic, fundamental to the rhetorical theory of argument.
This approach is entirely compatible with a class of classical definitions of rhetorical argument, such as the following,

Cicero Senior — I take it that what you desire to hear about is ratiocination, which is the process of developing the arguments. […]
Cicero Junior — Clearly that is exactly what I require.
Cicero Senior — Well then, ratiocination, as I said just now, is the process of developing the argument; but this process is achieved when you have assumed indubitable or probable premises from which to draw a conclusion that appears in itself either doubtful or less probable.
Cicero, Part., XIII, 46; p. 345-347; my italics

How to make the doubtful a little less doubtful? Like Toulmin, Cicero sees argumentation (“ratiocination”) as a technique to reduce uncertainty.

4. Argumentation as saying and schematizing

According to Jean-Blaise Grize,

As I understand it, argumentation considers the interlocutor not as an object to manipulate but as an alter ego with whom a vision has to be shared. To work on him means to try to change the various representations attributed to him, by highlighting certain aspects of things, hiding others, proposing him new perspectives, and all this with the help of an appropriate schematization. (Grize 1990, p. 40)

Arguing consists in schematizing, or framing the situation for the interlocutor.
Such a generalization extends the concept of argumentation over the whole act of saying something to somebody:

Arguing amounts to putting forward some assertions that we choose to compose in a discourse. Conversely, asserting (saying) amounts to arguing, simply because we choose to say and put forward some meanings rather than others. (Vignaux 1981, p. 91)

This vision of saying as essentially a rhetorical argumentative activity has deep roots in the rhetorical tradition.

It may be compared with what Quintilian presents as the essence of rhetorical argumentation:

The art of speaking well. (IO, II, 15, 37)

This famous formula is often quoted in Latin, rhetoric is the “ars bene dicendi”; the definition is complemented by the definition of the orator as “a good man speaking well”.
Argumentative rhetoric becomes the legislative technique of persuasive speech, guaranteed by the quality of the speaker, S. Ethos.
This vision of rhetoric constitutes the backbone of the classical humanities

Compared with Grize — who, to my knowledge, never quotes Quintilian, no more than Toulmin referred to the classical science of topoi — the only difference is that Quintilian stresses the educative dimension of rhetoric, whereas Grize simply analyzes argumentation as found in natural discourse.

This line of thought generalizes rhetoric to all forms of controlled expression, thus founding a Rhetorik der Sprache (Kallmeyer 1996), a “rhetoric of speech”.

5. Argumentation as orientation

Anscombre and Ducrot’s theory of Argumentation within Language is based on the fact that, in natural language, the argument as a statement is linguistically linked to the conclusion, defined as the following statement:

A speaker argues when he presents a statement S1 (or a set of statements) as intended to make acceptable a new one (or a set of new ones), S2. Our thesis is that there are linguistic constraints governing this presentation. For a statement S1 to be given as an argument supporting a statement S2, it is not sufficient that S1 gives reason to admit S2. The linguistic structure of S1 must also meet certain conditions to be able to constitute, in a speech, an argument for S2. (Anscombre & Ducrot 1983, p. 8)

This approach results in a redefinition of the concept of topos, as a semantic link between two predicates, S. Topos in Semantics.

By re-defining the argumentative constraint as an inter-statements linguistic constraint, Anscombre and Ducrot generalize the concept of argumentation as a property of the linguistic system (langue and not parole “speech”, as defined by de Saussure).

S. Orientation; Argumentative scale.

6. Argumentation between monologue and dialogue

Argument seems to be a mode of discourse which is neither purely monologic nor dialogic. (Schiffrin 1987, p. 17)
[I have defined argument as] a discourse through which speakers support disputable positions. (Id., p. 18)

Schiffrin’s work is not primarily devoted to argument. This succinct definition, however, perfectly express the mixed character of the argumentative activity.

7. Argumentation, a discourse submitted to a rational judge

Argumentation is a verbal and social activity, aiming to strengthen or weaken the acceptability of a controversial point of view from a listener or reader, advancing a constellation of proposals to justify (or disprove) that view before a rational judge. (van Eemeren & al. 1996, p. 5)

This definition summarizes the rhetorical and dialectical positions. It re-defines the position of the third party, the judge, not as an empirical, institutional figure, arguing on the basis of the legal corpus of law and jurisprudence shaped by history and sociology, but instead as a normative rational figure, arguing on the basis of a set of independently defined rational principles, S. Norms; Evaluation and Evaluators.

8. Guidelines adopted in this dictionary

(i) An argumentative situation is defined in the Ad Herennium style: a complex dialogic situation opened by an argumentative question.

(ii) An argumentative question is a question to which the arguers (the debaters) give argued answers, possibly both sensible and reasonable, but incompatible, organized in pro- and a contra-discourse.

(iii) These answers express the conclusions (points of view) of the arguers about the issue. The elements of pro- and counter-discourse which support these conclusions have the status of argument for their respective conclusions.

(iv) Argumentative situations come in a variety of degrees and types of argumentativity, according to the kinds of relationship established between the pro- and counter- discourses and to the interactional and institutional parameters framing the exchanges.

Points (i) to (iv) define the external argumentative relevance, as the relevance of a conclusion for a question.

(v) An argumentation, in the monologic sense is defined as the “argumentative cell”, as represented in Toulmin’s layout.
In the broad sense, the word argumentation covers all the verbal and semiotic activities produced in an argumentative situation.

(vi) An argument is an implicit or explicit combination of statements supporting a conclusion.

(vii) The internal argumentative relevance, as the relevance of an argument for a claim is defined in relation to an argument scheme.

1. Argument

The word argument is used in different domains, in grammar, logic, literature, and argumentation, with quite different meanings.

— In logic and mathematics, the arguments of a function f are the empty places x, y, z… characterizing the function; the independent entities (variables) organized by the function.

— By analogy, in grammar, the verb plus its subject and object(s) can be considered the counterpart of a function. To give for example, corresponds to a predicate governing three arguments “x gives y to z”; to love to a two-argument predicate, “x loves y”. By substituting adequate phrases (i.e., respecting the semantic relationship characterizing the verb) for each of these variables, we form a proposition^@: “Adam gives Eve an apple”, S. Proposition.

— In literature, the central argument of a play or a novel corresponds to the plan, the summary, or the guiding principle of the plot. With this meaning, the word argument is morphologically and semantically isolated; argument as “summary” bears no relation to conclusion, nor to to argue, argumentation.

2. Argument and argumentation

The words argument and proof are used to translate the Greek word pistis and the Latin word argumentum.

2.1 Argument ~ argumentation

By synecdoche, argument often means argumentation: “let the best argument prevail!”

2.2 Premise, data, argument

— In logic, the premises of the syllogism lead to a conclusion. The premises are propositions expressing true or false judgments. The conclusion is a proposition which is different from the premises and which is derived exclusively from their combination, without the surreptitious introduction of implicit background information into the reasoning, S. Syllogism. A premise is not an argument but a constituent of an argument; the argument is constructed by combining the two premises.

— In argumentation, the conclusion is derived from an item of information combined with an inferential topic. The situation is the same in Toulmin’s layout of argument, where the data becomes an argument when combined with an often implicit system warrant / backing. The word argument is routinely used to refer to the data element as the head of such combinations.

— In analytical and immediate inferences, the conclusion is derived directly from a single statement, which is an argument in itself. The conclusion is derived from the form or the semantic contents of the statement argument, S. Proposition.

Argument and conclusion are correlative terms. The “argument — conclusion” relationship is expressed, more or less accurately by expressions such as those listed below. If necessary, “is” may be replaced by “is presented as such by the speaker” (as in line 1, etc.).

2.3 Argument: true, probable, plausible, accepted, conceded…

A statement is considered (or presented) as a certain truth and may function as an argument on very different bases.

— The argument conveys a well-known fact, an intellectual self-evidence, S. Self-Evidence.

the heat of the wax dilates the pores and pulling up is thus less painful (Linguee)

— The partners have explicitly agreed on the statement, for example as part of a (quasi-) dialectical agreement:

We agree that now Syldavia cannot leave the Eurozone, so we can place further requirements upon them.

— The speaker has chosen his argument from those considered to be true by the audience, even if he or she has personal doubts about its validity, S. Ex datis:

You think that Syldavia will never leave the Eurozone, so…

— A simple fact: the statement is challenged neither by the opponent nor by the audience.

The audience’s acceptance of stable statements that may serve to support the conclusion, is always precarious. The opponent’s belief in the truth of a given statement is even less stable. The choice of what will be considered a valid argument is thus a strategic choice which will change in view of the circumstances, S. Strategy.

Challenging the argument — If the argument is disputed, it must itself be legitimized. As part of this operation, the argument takes the status of a claim put forward by the proponent and supported by a series of arguments. These new arguments serve as sub-arguments supporting the overarching claim, S. Linked argument; Epicheirema. If no agreement can be reached on any statement, things can, theoretically, go back indefinitely and the debate may continue without end. The risks associated with such “deep disagreement” should not be considered to invalidate argumentation as a useful social tool to deal with social incompatibilities, as far as third parties play their role in well-regulated settings.

3. Claim, thesis, conclusion, point of view, standpoint

In argumentation, the conclusion is also called the claim, or point of view. A philosophical conclusion is often called a thesis, S. Dialectic. The set of conclusions drawn from complex data at the end of an abduction process can be a full-blown theory, S. Abduction.

3.1 Point of view, viewpoint, standpoint

In the socio-political domain, a point of view is an “opinion”, possibly justified by arguments. The pragma-dialectical program is aimed at reducing, resolving, or eliminating differences of opinions. The corresponding expressions “resolving… differences of conclusions, claims, thesis…” are not in use.
An argument as a point of view, an opinion, a perspective… conveyed in just one sentence is a very special case. Points of views and opinions are generally expressed in complex discourses, supported by equally complex argumentative sub-discourses. The expression point of view can be used to refer to a whole speech, including the point of view and the good reasons supporting it.
In ordinary language, the concept of point of view organizes the perceptual reference system of the speaker:

On the other side of the hedge, was a gardener.
On the other side of the hedge, we saw a road.

In one case, the speaker is outside the garden, in the other in the garden. The concept of point of view used in argumentation is strongly metaphorical. It frames the argumentative situation according to the visual metaphor of a spectator within a landscape, which would be the reality, inaccessible as such, if not represented on a map.
The spectator’s vision provides a slice of reality restructured according to the laws of perspective. The reality referred to by the point of view is only so with regard to a, by definition, unstable focus. In this sense, a point of view is either questionable as it functions as blinkers; or valuable, because it protects one from the objectivist illusion produced by consensus, and from the paranoia of absolute knowledge.

An affirmation corresponds to a point of view if it is brought back to one subjective source, while absolute truth, or vision, is independent of any source, or has a universal, absolute source.
The point of view is an inescapable starting point. Points of view are comparable and assessable. We cannot be without perspective-point of view, yet we are able to define a better point of view; change our point of view, and multiply our points of view. In order to eliminate differences in points of view, one would have to eliminate subjectivity, or the plurality of voices, and de-contextualize the discourse. Scientific discourses do that routinely, but, as far as argumentative discourse seeks to deal with human affairs, involving (legitimate) interests, values, and their affective correlates, argumentation analysis cannot align itself with scientific language without changing the nature of its objects and objectives. The radical elimination of points of view would require the resurrection of the absolute rational Hegelian subject, or of the objective and omniscient narrator of nineteenth century novels.

3.2 Conclusion

The opening section of a discourse is its introduction, its closing section its conclusion. The argumentative conclusion is distinct from the material conclusion ending an intervention. The argumentative conclusion can be stated, or repeated, in any part of speech, at the beginning or at the end, or both.

The argumentative conclusion is defined in correlation with the argument (see Table above). In an argumentative monological text, the conclusion is the assertion according to which the discourse is organized; towards which it converges; in which its orientation materializes; the intention which gives the discourse its meaning, and the ultimate core of the text obtained by condensing it.

The conclusion is more or less detachable from the arguments supporting it. Once we have reached the conclusion that “probably, Harry is a British citizen”, we can, by default, act on the basis of this belief. But, as far as the modal probably expresses clear reservations on the whole inferential process, the claim will remain revisable if conditions change. The “fire and forget” principle[1] does not work well in argumentation. The conclusion is never fully detachable from the speech used in its construction.

A statement S becomes a claim in the following dialogical configuration

(1) — S is put forward by a speaker (as something essential for him, or merely anecdotal)
(2) — S is not ratified by the addressee: not preferred second turn
(3) — S is re-asserted, possibly reformulated by the speaker
(4) — S is explicitly rejected by the dialogue partner (re-statement not ratified: disagreement ratified)
(5) — Emergence of pro- and contra-arguments.

At stage (3), the disagreement emerges. At stage (4) the disagreement is ratified as such, a stasis is formed, and S is now a Claim put forward by the first speaker. At stage (5), the stasis begins to develop

Stage (1) is not a dialectical “opening stage”. The speaker does not necessarily intend to open a dispute. Non-ratification can occur at any time in an interaction, and may concern any foreground or background statement, S. Denying; Disagreement. In other words, being a claim is not a property of a statement, but is attached to the treatment of a statement in an interactive configuration.

[1] “(Of a missile) able to guide itself to its target once fired.” (EOD, fire-and-forget) (11-08-2017)

1. The English words

1.1 To Argue ­

The verb to argue has two different accepted meanings which will be referred to, respectively, as to argue₁ and to argue₂:

— To argue₁: “to put forth reasons for or against; debate”
— To argue₂: “to engage in a quarrel; dispute: We need to stop arguing and engage in constructive dialogue (tfd, Argue).

The morphological, syntactic, and semantic differences between these meanings are crucial and clear.

Morphology

The word argumentation derives from to argue₁ via argument₁; it refers only to a speech in which a conclusion is supported by good reasons.

Syntax

— To argue_1 is followed by a that clause: “A argues that P”; P is the claim.
— To argue_2 is followed by a double indirect complementation, “A argues with B about Q”. Q is neither A‘s nor B‘s claim, but refers to the issue of the dispute.

Semantics

— To argue_₁ means “to give reasons” (MW, Argue), and refers to a semiotic activity (verbal and co-verbal).

— To argue_₂ means “to have a disagreement a quarrel, a dispute” (ibid.), and refers to the broad field of interactions ranging from a lively discussion to outright pugilism, as shown in the following passage, in which the detective Ned Beaumont questions an informant, Sloss:

Ned Beaumont nodded. ‘Just what did you see?’
‘We saw Paul and the kid standing there under the trees arguing’
‘You could see that as you rode past?’
Sloss nodded vigorously again.
‘It was a dark spot’, Ned Beaumont reminded him. ‘I don’t see how you could’ve made out their faces riding past like that, unless you slowed up or stopped.’
‘No, we didn’t, but I’d know Paul anywhere,’ Sloss insisted.
‘Maybe, but how’d you know it was the kid with him?’
‘It was. Sure it was. We could see enough of him to know that’
‘And you could see they were arguing? What do you mean by that? Fighting?’
‘No, but standing like they were having an argument. You know how you can tell when people are arguing sometimes by the way they stand’
Ned Beaumont smiled mirthlessly. ‘Yes, if one of them’s standing on the other’s face.’ His smile vanished.
Dashiell Hammett, The Glass Key, [1931][1].

1.2 Argument ­

The noun an argument inherits the two meanings of to argue; an argument₁ is a “good reason”, an argument₂ is a “dispute”, possibly containing argument₁. Grimshaw’s book, Conflict talk. Sociolinguistic investigations of arguments in Conversation (1990), exclusively deals with arguments₂ “dispute”, not at all with arguments_1, “good reasons”.

A third, specific, meaning adds to these two inherited meaning, argument₃, as “the abstract, the theme, the subject matter” (of a literary work, etc.).

“Argument is War” — Lakoff and Johnson have proposed the famous equivalence “argument is war”:

Let us start with the concept argument and the conceptual metaphor argument is war. This metaphor is reflected in our everyday language by a wide variety of expressions:

Your claims are indefensible.
He attacked every weak point in my argument.
His criticisms were right on target.
I demolished his argument. […]

“We can actually win or lose arguments” (1980, p. 4)

Lakoff and Johnson refer to the “concept argument”. If the preceding conclusion is correct, there is not just one but two concepts of argument. To argue₂ and argument₂ may be associated with a kind of war; but what about argument₁ and to argue₁?

If interlinguistic comparisons can tell something about words used as concepts, note that, in French, the first series of metaphors easily translates word-for-word; but the expression “we can actually win or lose arguments”, does not. The words to argue, argument, and argumentation have clearly recognizable counterparts in French or Spanish, or in the Romance languages at large:

French argumenter, argument, argumentation
Spanish argumentar, argumento, argumentación

This graphic illustration of the proximity of these words certainly favors the internationalization of the concept. Yet there are deep differences between their respective meanings, which can be roughly represented as follows:

The French word argument and the Spanish word argumento never refer to a dispute. The field of argumentation studies develops from the shared meaning of argument₁, “good reason”.

This shows that the meaning of to argue₂, argument₂ in a language is independent of the concept referred to by the family to argue₁, argument₁, argumentation.

1.3 Argumentative

According to the Merriam-Webster Dictionary, the adjective argumentative shares the two meanings of its morphological base, argument: “controversial” and “disputatious” (MW, Argumentative). The Merriam-Webster Learner’s Dictionary, however, is more categorical (MWLD, Argumentative):

Argumentative: tending to argue; having or showing a tendency to disagree or argue with other people in an angry way: quarrelsome.

an argumentative person
he became more argumentative during the debate.
an argumentative essay.

In this dictionary, argumentative will be attached by default to the family “argumentation”, thus a semantically derived of argument₁ “good reason”, unless contextually clear or otherwise specified. An argumentative essay will be taken as “an essay developing an argumentation”; if referring to “a polemical essay”, its quarrelsome character will be explicitly mentioned.

2. Differential orientations:
the French words arguer, argutie

In French, from a morphologic point of view, the verb arguer is the basic verb from which all the argu- words derive:

arguer   → un argument      → argumenter       → une argumentation, etc
an argument”     “to argue”          “an argumentation”, etc.

But arguer_F must be set apart; to argue matches argumenter_F, nor arguer_F. There is a semantic discontinuity between arguer_F and argumenter_F. When S1 says:

S:   — Pierre argumente en faveur de P, “Peter argues that P”

S recognizes that Peter does give arguments. When he or she says:

S:   — Pierre argue que… “Peter argues_F that…”

S just quotes the argumentative discourse of Peter without taking a position on the validity of the arguments he offers, and even suggesting that they might be fallacious. In a democratic or republican newspaper the construction:

the extreme right argues_F that…

introduces an argumentation considered as weak or invalid. That is, the verbs arguer_F and argumenter_F have opposing orientations. The former values discourse as argumentative, whereas the latter suggests that it posits only pseudo-arguments.

Quibble may translate in French as argutie_F, a word derived from arguer_F:

These people are the manipulated agents of subversion, performing instructions and rehashing quibbles [“répétant des arguties”].

Arguer_F and argutie_F are only used occasionally. Arguer_F might be replaced by argument_F between quotation marks. So a pro-wind farm group quotes the arguments of its opponents, the anti-wind farm group, as follows:

Let’s look at some of the ‘arguments’ put forward by anti-wind farms
(Complete example, S. Convergent argumentation)

The concept of arguments, and argumentation studies, benefit from the strong positive orientation that the words argument and argumentation have in ordinary language. The case is the same for the word and the concept of dialogue, S. Interaction, Dialogue, Polyphony.

[1] Quoted after Dashiell Hammett, The Four Great Novels. Picador, 1982. P. 725-726.

An apagogic argument is a form of argument by the absurd, which argues that unreasonable interpretations of the law must be rejected:

The apagogic argument assumes that the legislator is reasonable and could not have admitted an interpretation of the law that would lead to illogical or unfair consequences. (Perelman 1979, p. 58)

It parallels the psychological argument, presupposing that the legislator is rational and benevolent, V. Absurd; Juridical arguments.

According to Alexy, the apagogic argument is one of the four types of arguments prevailing in law, the others being the arguments by analogy, a contrario (opposites) and a fortiori, (1989, quoted in Kloosterhuis 1995, p. 140).

The rhetoric of figures defines the antithesis as an opposition between two terms (words or phrases) of opposite meanings, entering into parallel syntactic constructions. The argument scheme of the opposites materializes discursively as an antithesis.

1. Antithesis as argumentative diptych

An argumentative situation emerges with the appearance of a point of confrontation ratified as such, a stasis. It develops into a diptych, characterized by the confrontation of two schematizations, that is to say two sets of descriptions, narrations and argumentations supporting two opposing conclusions. At this stage, the two discourses develop at cross-purposes, without explicitly taking this opposition into account, S. Stasis. This elementary argumentative situation corresponds to a discursive antithesis.

Such a confrontation might be taken up in a structured monologue juxtaposing the two sides of the issue. Such a monologic diptych features an “antiphony”, that is two voices putting forward incompatible arguments with respect to the same issue. This is typically seen when an individual having a vested interest in an issue engages in inner deliberation, and oscillates between two points of view, acting actually as a third party. This situation is elaborated as a dilemma whose anti-oriented horns are articulated by an and:

I admire your courage and I pity your youth.
Corneille, Le Cid 2, 2, verse 43. Quoted by Lausberg [1960], §796

When the speaker clearly identifies with one of the two voices, the balance of the two voices is broken in favor of one of the positions. The and dilemma transforms into a but opposition, overcoming the antithesis:

… but I pity your youth; so I won’t accept your challenge to duel.

2. Antithesis, figure and argument

The following argumentation is structured by the scheme of the opposite:

(D1) He is submissive to the privileged; I would not like to confront him in a weak position.

exactly as the self-argued description:

(D2) He is submissive to the privileged and powerful, and hard with the weak.

Whereas in (D1), the second member of the scheme “he must be hard with the weak”, remains implicit, (D2) corresponds to a complete expression of the topos. But the two discourses are based on the same mechanisms, the argumentation is “valid” or acceptable insofar as the portrait sounds “true”; both are “convincing”. Description and argument are rooted in the same figure or scheme.

1. Terminology

Structural analogy connects two complex domains, each articulating an indefinite and unlimited number of objects and relationships between these objects. It combines intra-categorical analogy (a property of objects) with proportional analogy (a property of relations). One could also speak of formal analogy (the areas have the same shape) or borrow the mathematical term “isomorphism”, S. Intra-categorical analogy; Proportion.

The expression “physical analogy” refers to the relationship between two objects when one is a replica of the other. The concept covers different phenomena, such as the relationship between a model and its original, or the relationship between a prototype and the object to be manufactured. The reasoning based on the model or prototype is then applied to the original.

Structural analogy is involved in the two following situations.

(i) A, B, C … are similar ­— To establish if the complex objects or domains A, B, C are similar, one has to compare their components and the relations between them. The conclusion of this investigation will be a claim such as “A, B, C… are similar”; “A, B, are indeed similar, but C is something different”, etc.

One may ask if the 1929 Great depression, the Lost Decade of Japan during the 90s, and the Argentinian Crisis in 2001 share some significant characteristics. The whole purpose of the investigation may be to establish a typology of economic crisis, without — as far as possible — drawing on preconceived ideas of how politicians will use the conclusions of this investigation.

The areas are symmetrical from the viewpoint of the investigation, which does not favor one of the areas over the others, but only focuses on their relationships.

(ii) A is similar to B — A contrario, the importance of the previous situation appears when the series involves the 2008 crisis. Given the actuality of this last crisis, it will certainly be tempting to see if we can “learn lessons” from the previous crises and to apply them to the 2008 case, with the intention of making provisions for the current situation. If the proponent uses the analogy 1929 ~ 2008 to predict a third world war, her opponent can rebut the inference by showing that the domains are not similar, and that it is therefore impossible to rely on the first instance, in 1929, to make inferences about something about what will happen in 20** and after (see farther).

The difference in status between the two areas is expressed in different ways. In his analysis of the metaphor, Richards opposes Tenor and Vehicle (1936); Perelman & Olbrechts-Tyteca speak of Theme and Phore ([1958], p. 501). A simple way to name these domains may be comparing domain / compared domain; or, in view of the analysis of argument, Resource domain / Target domain.

The argument by analogy works on the asymmetry of the compared areas; that is why these two areas will be designated, when necessary by the letters of alphabets, R, as Resource field and Π (capital Greek letter “pi”),   the Problematic field, targeted by the investigation. The field R is the source or the Resource on which the arguer relies to make changes in the Targeted area Π, or to derive from R certain consequences about Π. In other words, the Resource field R is the argument domain and the Targeted field Π is the conclusion domain. The two fields are differentiated from epistemic, psychological, linguistic and argumentative perspectives.

— In epistemic terms, the Resource field is the best-known area; the Target field is the area under exploration.
— In psychological terms, intuition and values operating in the Resource field are put to work in the Target field.
— In linguistic terms, the Resource field is well covered by a stabilized, well-known and easily spoken language; the Target domain is not.
— In practical terms, we know what to do within the Resource field whereas in the Target domain, we do not.

2. Explicative analogy

In the well-known analogy proposed by Ernest Rutherford between the atom and the solar system, the Resource field is the solar system, the Target field is the atom:

the atom is like the solar system.

This is a didactic analogy, intended to provide a first intuitive understanding of the atomic structure, taking advantage of a (supposed) better understanding of the solar system. The asymmetry of the areas is obvious: the Resource field, the solar system, has been known and understood for a long time. The Targeted field, the atom, is new, poorly understood, inaccessible to direct perception, enigmatic.

The explanatory analogy retains some educational merits even when partial. A comparison is not identification, and two systems can be compared simply in order to identify the limits of the comparison, that is, the irreducible specificities of each field, cf. infra, §6.

The analogy has explanatory value in the following situation:

In the world Π, the proposition π is poorly understood. In a world R, there is no debate over r. Π is isomorphic to R (structural, systemic analogy). The position of π in Π is the same as that of r in R. So, the knowledge, images, obligations… attached to r are now transferred to π; π  is now slightly better understood; we know how to do with π.

The analogy relationship integrates the unknown on the basis of the known. As causal explanations, explanations by analogy break the insularity of the facts.

The analogy is an invitation to see and handle the Problem through the Resource. The Resource domain is considered to be a model of the Target domain. The relation of the domain under investigation to the Resource domain is treated like that of the domain of investigation to an abstract representation of this domain. Otto Neurath uses a maritime metaphorical analogy to explain his vision of epistemology:

There is no tabula rasa. We are like sailors at sea, who must rebuild their ship without ever bringing her to a dock to be disassembled and rebuilt it with better items. (Otto Neurath, [Protocol Statement],1932/3.[1])

The analogy can be translated word for word: “There is no ultimate foundation of knowledge from which we could, without any presuppositions, re-build the whole of our present knowledge.” This resource is extremely powerful; the image could also be applied to social life: “There is no ‘good explanation’ (meaning “good discussion of our disagreements”) that permits reconstructing a damaged relationship and re-start from scratch.”

3. Arguments based on structural analogy

In ordinary situations, analogy is used argumentatively, as in the following case:

— In the world Π, we are in a difficult situation; what should we do? Should we accept or reject perspective π?
— But we know for sure what happened in a world R.
— Fortunately, Π is isomorphic to R (structural, systemic analogy); if necessary we can argue for that.
— The position of π in Π is the same as that of r in R.
—So we can act, in world Π, on the basis of the knowledge, images, obligations… attached to r — That is to say, we can now decide about π.

This argumentative operation argues that “if the domains are analogous, so are their corresponding elements and the relations between them”, which may prove true or false under further investigation. The analogy gives us something to think about, but proves nothing; the conclusion projected upon Π may be false or ineffective.

4. From analogy to metaphor and back

A language is attached to the Resource domain. For example, ​​the human body is referred to in a language that may be incomplete and fairly incoherent, but commonly understood, the language of the flow of organic matter, of popular physiology, of good health and sickness, life and death. This language synthetizes and builds a common intuition of the body. Other unfamiliar areas are not equipped with such a dense, effective and functional language. The analogy projects the language of the Resource area, the human body, onto the Problematic field, the society. As a result, the target can be problematized in a familiar, non-controversial language; so that social convulsions can be discussed and a cure found. The analogy is an invitation to see the problem through the lens of the resource; full metaphorization enables us to forget the glasses.

The following apologue is based on the analogy “society is like a body”, as expressed in the metaphorical expression “social body”. Note the explicitness of the vocabulary of analogy in the final commenting section.

The senate decided, therefore, to send as their spokesman Menenius Agrippa, an eloquent man, who was also accepted by the plebs as being himself of plebeian origin. He was admitted into the camp, and it is reported that he simply told them the following fable in primitive and uncouth fashion. ‘In the days when all the parts of the human body were not as now agreeing together, but each member took its own course and spoke its own speech, the other members, indignant at seeing that everything acquired by their care and labour and ministry went to the belly, whilst it, undisturbed in the middle of them all, did nothing but enjoy the pleasures provided for it, entered into a conspiracy; the hands were not to bring food to the mouth, the mouth was not to accept it when offered, the teeth were not to masticate it. Whilst, in their resentment, they were anxious to coerce the belly by starving it, the members themselves wasted away, and the whole body was reduced to the last stage of exhaustion. Then it became evident that the belly rendered no idle service, and the nourishment it received was no greater than that which it bestowed by returning to all parts of the body this blood by which we live and are strong, equally distributed into the veins, after being matured by the digestion of the food.’ By using this comparison, and showing how the internal disaffection amongst the parts of the body resembled the animosity of the plebeians against the patricians, he succeeded in winning over his audience.
Titus Livius, The History of Rome, Vol. 1, Bk 2; between 27 and 9 BC.[2]

The resource does not necessarily preexist its use in an analogy. An analogy can create ex nihilo a self-evident resource, as in the following analogy, proposed by Heisenberg in 1955. The danger mentioned in the first line refers to the cold war era, and the resource term is “a ship built with such a large quantity of steel and iron that its compass, instead of pointing to the North is oriented towards the iron mass of the ship.” Note that, once again, there is no clear-cut frontier between structural analogy and metaphor. Heisenberg refers to the situation he imagines as a metaphor; and in the next line, he uses a construction expressing an analogy: “humanity is in the position of a captain…”.

Another metaphor might make such a danger even clearer. By the seemingly un­limited growth of its material power, humanity is might be compared to a captain whose ship has been built out of such a large quantity of steel and iron that its compass, rather than pointing to the North, orients towards the huge iron mass of the ship. Such a ship would get nowhere. It would be blown off course and led in circles.

But back to the situation of modern physics: we must admit that the danger exists only if the captain does not know that his compass no longer responds to the magnetic force of the earth. By the time he understands this, the danger is already halved. Because the captain who, not wishing to turn around, wants to achieve a known or unknown purpose, will find a way to steer the boat, either by using new modern compass that does not react to the iron mass of the boat, or by steering in relation to the stars as sailors once did. It is true that the visibility of stars does not depend on us, and perhaps today do we see them only rarely. Despite this, our awareness of the limits of our hope in progress supposes the desire not to go in circles, but to achieve a goal. Once recognized, this limit becomes the first fixed point which allows a new orientation.
Werner Heisenberg, [Nature in Contemporary Physics], [1955][3]

5. Structural analogy as an epistemological barrier

Analogy is fertile to stimulate discovery and invention, useful for teaching and popularizing knowledge. Yet it becomes an epistemological obstacle when the proposed explanation by analogy seems so clear and satisfying that it hinders further research:

For example, blood flow like water. Canalized water irrigates the ground, so blood should also irrigate the body. Aristotle was the first to assimilate the distribution of blood from the heart to the body with the irrigation of a garden by canals (De Partes Animalium, III, v, 668 a 13 et 34). Galen did not think otherwise. But to irrigate the soil, it is ultimately to get lost in the soil. And here is exactly the main obstacle to a proper understanding of blood circulation.
Georges Canguilhem, [The Knowledge of Life], 1951.[4]

The systematic rejection of analogy as an instrument for knowledge is grounded in such observations.

6. Refutation of structural analogies

6.1 Vain analogy

In an explanation, the explanation (explanans) must be clearer than the thing to explain (explanandum). Analogical explanation must also satisfy this condition, and if the resource area is even less well known than the area under investigation the analogy does not help in the understanding of things.

The analogy is also vain when used to impress the audience and display the grandstanding of the speaker as familiar with the Resource domain. Gödel’s theorem is used extensively for this purpose (Bouveresse [1999]).

6.2 False analogy

An argument by analogy can be rejected by showing that there are critical differences between the Resource domain and the Target domain, prohibiting the projection of the former upon the latter so that no lesson can be learned from the supposed Resource domain. In the following passage for example, it is argued that the comparison of the 2008 and 1929 crisis is marred by the facts that the present situation in Germany has nothing to do with its situation after 1918 and the coming years. Furthermore, it is argued that there is nothing similar to Hitler and Nazism in the European landscape in 2009:

Jean-François Mondot — Does the economic crisis weaken our civilization? We sometimes hear intellectuals and columnists making analogies with the 1929 crisis that led to World War II.

Pascal Boniface — We often make the mistake of thinking that history repeats itself, and so make very risky comparisons. Russia bangs his fist on the table, everybody immediately talks about the Cold War. An economic and financial crisis erupts on Wall Street, and immediately an analogy is drawn with 1929, the suggestion being that Hitler could come to power as a result of these difficulties. Yet the political circumstances are obviously very different, insofar as no great country is now humiliated as Germany was after 1918, leaving it wishing to take revenge. This comparison is easy to make, but it has no basis, neither strategic nor intellectual.
Pascal Boniface, [The clash of civilizations is not inevitable], 2009.[5]

6.3 Partial analogy

Partial analogy (“misanalogy” Shelley, 2002, 2004) is an analogy that has been criticized and recognized as limited. The two domains cannot be equated. Nonetheless, partial analogy still has a pedagogical use, as seen in the case of the analogy between the solar system and the atom (cf. supra §2):

A central body: the sun, the nucleus of the atom.
Peripheral elements: the planets, the electrons.
A central mass much larger than peripheral masses: the mass of the sun is larger than the planets; the mass of the core is larger than that of electrons. —etc.

Differences (analogy breaks):

The nature of the attraction: electrical for the atom, gravitational for the solar system.
There are identical atoms, each solar system is unique.
There may be several electrons in the same orbit, whereas there is only one planet in the same orbit. — etc.
The fact that the limits of analogy are precisely known prohibits any automatic transposition of the knowledge gained in one field into the other field.

6.4 Reversed analogy

A conclusion C1 has been established for a Target resource on the basis of an analogy drawn from the Resource domain R. The opponent argues that the same analogy drawn from the same domain R leads to another conclusion C2 about the same Target domain, that is incompatible with C1 (“disanalogy” Shelley, ibid.). These two contradictory conclusions prohibit the use of the Resource domain to argue in the Target domain.

This is particularly effective because the opponent concedes to playing on her adversary’s home ground. The opponent accepts and examines more closely the analogy advanced by the proponent, in order to neutralize his or her conclusions. This strategy is exploited in the refutation of argumentative metaphors.

Argument: ­­— This area lies at the heart of our discipline.
Refutation: — That’s true. But disciplines also need eyes to see clearly, legs to move in, hands to act, and even a brain to think.
Other refutation — That’s true, but the heart can very well keep beating preserved in a jar.

A supporter of hereditary monarchy speaks against universal suffrage:

Argument:— An elected president, that’s absurd, we do not elect the driver.
Rebuttal: — Nor are there natural born drivers.

Both sides enact the same metaphorical field. This form of rebuttal has the strength of an ad hominem refutation, based on the own beliefs of the speaker: “You are your own refuter”.

Counter-analogy — As with any argument, one can oppose an argumentation by analogy by putting forward a counter-argumentation (an argumentation whose conclusion is incompatible with the original conclusion). This counter-argumentation can be of any kind, including another argument by analogy, taken from another Resource domain; an analogy equilibrates another analogy:

Argument:   — The university is (like) a company, so …
Rebuttal:     — No, it is (like) a daycare, an abbey …

[1] Otto Neurath, “Protokollsätze”. Erkenntnis 3 (1932/3), p. 206. Quoted in A. Beckermann “Zur Inkohärenz und Irrelevanz of Wissensbegriffs”. Zeitschrift für Philosophische Forschung 55, 2001. P. 585.
[2] Trans. by Rev. Canon Roberts; Ed. by Ernest Rhys. J. M. Dent & Sons, Ltd., London, 1905. Quoted from; http://mcadams.posc.mu.edu/txt/ah/Livy/Livy02.html. No pag. (11-08-2017)
[3] Quoted after Werner Heisenberg (1962) La Nature dans la Physique Contemporaine. Paris: Gallimard, 1962. P. 35-36.
[4] Quoted after Georges Canguilhem, La Connaissance de la Vie. Paris: Vrin, 1965. P. 26-27.
[5] Pascal Boniface, “Le clash des civilisations n’est pas inévitable”. Interview by J.-F. Mondot, Les Cahiers de Science et Vie, 2009. www.iris-france.org / Op-2009-03-04.php3] (09-20-2013)

Intra-categorical analogy draws on the relationship between individuals belonging to the same category. For a definition of the concept of category, the categorization process of individuals; the organization of categories in classifications and the corresponding forms of syllogistic reasoning, S. Categorization and Nomination.

1. From identity to intra-categorical analogy and circumstantial analogy

1.1 Individual identity

An individual is identical to itself (not similar nor resembling); it is not “more or less” identical to itself. This self-evidence establishes the principle of identity “A = A”.

1.2 Identity of indiscernibles

Two different individuals perfectly identical, for example products taken out of the same industrial production chain, are materially identical, that is perceptually indistinguishable. All that can be said of one can be said of the other; their descriptions coincide, they share all their properties, essential (categorical) or accidental.

Discernibility depends on the observer, the layman does not see any difference, and believes that “it’s all the same”, whereas the specialist will make crucial distinctions.

1.3 Intra-categorical analogy

Intra-categorical analogy is the relationship between the members of a category C. All members share, by definition, the characteristics defining the category. The phrase “another C” refers to another member of the same class C. Two beings belonging to the same category are identical for this category; a whale and a rat are identical from the point of view of the category “­— be a mammal”. This categorical identity is a partial identity, compatible with major differences; two beings of the same category are said to be analogous or similar. They are comparable in respect of their other non-categorical properties. Chicken eggs are all similar as eggs; an egg is identical to another egg; it is comparable to all other eggs in terms of freshness, size, color, etc. S. Comparison.

1.4 Circumstantial analogy

An individual a possessing the features (x, y, z, t), is similar to all individuals who have any of those features, whether it be an essential or accidental feature.

The descriptors of two objects define the point of view under which they are equivalent; two beings are similar if their descriptions overlap, contain a common part, which may or may not include all or some of their essential features. In other words, this common part generates a category, which may or may not make sense. One might speak of circumstantial analogy. Alice and a snake are identical from the standpoint of the category “— is a long-necked egg eater”, S. Definition.

2. Intra-categorical analogy as induction or deduction

Intra-categorical analogy can be reconstructed as an induction or a deduction:

2.1 As an induction

O is similar to P
P has the properties w, x, y, m
O has the properties w, x, y
So O probably has also property m.

From an overall judgment of analogy between two beings, based on the shared features w, x, y … we conclude that if one has the property m then the other most probably also possesses m. In other words, analogy is pushed towards identity.

2.2 As a deduction

O is similar to P
P has the property m
Conclusion: O probably has the property m.

O is similar to P. This means that they share a common set of features, and therefore belong to the category C defined by those features. In conclusion, as members of the same category C, O and P probably share other properties, among them m. This means that the predicate « — is like” is to be interpreted as a weaker form of « — is the same as”; analogy is seen as a weakened identity.

Deduction and induction are considered valid forms of reasoning. The purpose of the discussion about the possibility of reducing analogy to deduction or induction is to determine whether or not analogy is also valid as a form of reasoning. Reasoning by analogy is sometimes used to prove the existence of God, the ideological stakes of this issue are therefore high.

These formulations of the argument by analogy in the form of a dialectical syllogism are rather sterile because they do not emphasize the warranting operations, that contain all the interesting problems. The formulation of the conclusion not as a secure finding but as the product of a heuristic rule of thumb, however, is of great value. The conclusion should be written not as something “probable”, that is a kind of belief, but as a suggestion to do something:

It might be interesting to test P for property m.
It might be interesting to see whether O and P share other properties.

3. Arguments based on intra-categorical analogy

— Categories as a whole are structured according to their respective definition; two individuals belong to the same category if they have the same definition.

— Categories may be gradual, S. Rule of Justice.

— Categorical analogies may be restructured S. A pari; Definition (III). 

4. Refutation of categorical analogy

In one or other aspect, everything is like everything else, and analogies can be more or less “far-fetched”. Any rejected categorical analogy will be dubbed fallacious and denounced as a confusion, an amalgam (Doury 2003, 2006).

Intra-categorical analogy can be refuted by showing that the category created from those two beings is not based on essential features, but on some accidental property; in general, the generated class is deemed irrelevant. The nonsensical analogy “Chinese ~ Butterfly”, ironically discussed by Musil, illustrates the perils of circumstantial analogy, based on the arbitrary choice of a non-essential feature, here the “lemon yellow” color.

There are lemon yellow butterflies; there are also lemon yellow Chinese people. So, in a sense, butterflies can be defined as miniature winged Chinese people. Butterflies and Chinese people are symbolic of sensual pleasure. Here we can see for the first time a glimmer of a possible match, never considered before, between the great period of the moth fauna and Chinese civilization. The fact that butterflies have wings and not the Chinese people is only a superficial phenomenon. […] Butterflies did not invent powder: precisely because the Chinese have done it before them. The suicidal predilection for the lights of some nocturnal species is still an artifact of the past, which is difficult to explain in view of the daylight understanding of this morphological relationship between butterflies and China.
Robert Musil, [Spirit and Experience], [1921] [1]

The analogy relationship has difficulties with transitivity, S. Relation. Intra-categorical analogy is transitive: if A and B on the one hand, B and C on the other hand, are said to be similar because they possess the same essential features, A is thus similar to C. Circumstantial analogy is not transitive: nothing proves that if, on the one hand, the descriptions of A and B have common parts, and, on the other hand, the description of B and C have common parts, then the description of A and C will also have also common parts. Khallaf invokes a traditional analogy to criticize the concatenation of analogies:

A man is walking on the beach trying to find similar shells; once he finds a shell similar to the original, he throws away the original shell and goes on to find a seashell which resembles the second, and so on. When she has found the tenth shell, she should not be surprised to see that it is totally different from the first in the series. (Khallâf [1942], p. 89)

[1] Quoted in Jacques Bouveresse, Prodiges et vertiges de l’analogie [Prodigies and Dizziness of Analogy]. Paris: Raisons d’Agir, 1999. P. 21-22.

From an anthropological perspective, analogy is a form of thought that posits that things, people and events are reflected in each other. For analogical thinking, knowing is deciphering similarities; analogy unveils a world of secret links underlying reality, and generates a “cosmic feeling where triumph order, symmetry, perfection”, a closed world (Gadoffre & al. 1980, p. 50); thus conceived, analogy is the foundation of gnosis. From the perspective of the history of ideas, this form of thinking culminated in the Renaissance, when our “sublunary » world was, by analogy, mapped with the heavenly spheres, and with the divine world more generally.

In one of its manifestations, the doctrine of analogical correspondences validates the following type of argument:

Data: This plant looks like such or such part of the human body.
Conclusion: This plant has a hidden virtue, effective to cure the ills that affect the corresponding part of the body.
Warrant: If the shape of a plant is like a body part, then it cures ailments affecting that body part.
Backing: This is a divine provision.

This form of analogical thinking postulates that plants have hidden medicinal properties. The plant bears a divine signature, that is, a representation of the human body part that it can heal. This signature or “analogical sympathy” is a motivated signifier, a similarity or “resemblance” of the given body part. God, in his benevolence, has imposed this signature on particular plants in order to make them of use to us. A plant resembling the eyes, therefore might cure eye irritation.

Since the skin of the quince is covered with small hairs, it bears the “signature” of the hair, and eating the quince can make your hair grow. In the wording of Oswald Crollius [1609]:

Data: ‘This downy hair growing around quinces […] represents hair in some way.” (id., p. 41)
Conclusion: “So, their decoction makes hair grow, which fell because of the pox or another similar illness.” (ibid.)
Warrant: the healing power of plants “can be recognized more easily by the signature or analogical and mutual sympathy with the members of the human body with these plants than by anything else.” (id., p. 8)
Backing: “God gave an interpreter to each plant so that its natural virtue (but hidden in its silence) can be recognized and discovered. This interpreter can be nothing else than an external signature, that is to say a resemblance of form and figure, true indications of the goodness, essence and perfection thereof.” (id., p. 23)
Oswald Crollius, [Treatise on Signatures, or the True and Living Anatomy of the Big and the Small World]; [1609][1]

From this doctrine derives a research program for “those who want to acquire the true and perfect science of medicine”, “they should devote all their efforts to the knowledge of signatures, hieroglyphs and characters” (id., p. 20). Training will enable them to recognize “at first glance, on the surface of the plants, what faculties they are endowed with” (id., p. 9).

The knowledge of the medicinal properties of plants is acquired by learning how to read and understand the “discourse of nature”, that is to say, by mastering the signs scattered around the world. Such an analogical reading of the world is opposed to empirical causal investigation, which consists of observation and experience, practicing dissection or prescribing a concoction to the patient and then finding out if he or she is better, dead, or neither better nor worse. Analogical knowledge is a specific mode of thought, constitutive of magical thinking that substitutes for causal knowledge mysterious correspondences conveying influences, and bypasses the hierarchical system of categories organized according to genus and species, for which it substitutes a similarity network.

[1] Quoted after Oswald Crollius, Traicté des Signatures ou Vraye et Vive Anatomie du Grand et Petit Monde. Milan: Archè, 1976.