COLLECTION AND TYPOLOGIES OF ARGUMENTATION SCHEMES
Tradition has bequeathed us more or less systematized inventories of argument schemes:
Collections 2: From Aristotle to Boethius
Collections 3: Modernity and Tradition
Collections 4: Contemporary Innovations and Structurations,
and a number ries of questions about them:
— On their nature and number,
— Lists of argument schemes have been and are being compiled; but what is the unifying factor underlying these lists? Do they have a proper systematic organization? Are they amenable to some elementary headings (Blair 2012, Ch. 12 and 13)?
— Where do they come from? Are they recurring, remarkably stable structures that are taken up in argumentative discourses of all kinds? Or are they constructed from the a priori categories of the human mind?
— Are they logical, cultural, or anthropological entities? Are they culture-dependent?
— What kind of historical change, if any, can affect them? The question arises, when Toulmin, Rieke & Janik’s nine “forms of reasoning” of are compared with the Ciceronian and post-Ciceronian lists of topoi, see Collections (4) and Collections (2).
1. Categorizing arguments: collections and typologies
A class is a set of entities; basically, a typology is a class divided into different subtypes; the same class can admit different subtypes, differently organized, see classification. A catalog can be considered as a single-level typology.
A typology of arguments is a set of topics or argument schemes that connect the argument to the conclusion. Argument typologies include from ten to several dozens of argument schemes.
To categorize a speech segment (an individual, level 0) as a “pragmatic argument” is to identify in this segment the characteristic features that define the pragmatic argument. This operation is itself argumentative, and obeys the rules of argumentation by definition, see Nomination; Definition; Argumentation Scheme.
The idea of argument types, the possibility of making inventories of these types, and of giving these inventories an internal structure in order to build a “typology of topics”, is the very foundation of the theory of rhetorical argumentation. Walter Ong sees these typologies of argumentations as engaged in a perpetual movement of renewal and redefinition:
As the general intellectual tradition changes, the active associative nodes for ideas change, and classification changes too. Revising the tradition has been a common phenomenon in antiquity, when Aristotle differed from the sophists in the list of topics he proposed, Cicero from Aristotle, Quintilian from Cicero, Themistius from all these, and Boethius from all of them again and from Themistius as well. The revision continues in our day with Professor Mortimer Adler’s “Great ideas” (augmented beyond their original hundred), and with such articles as Père Gardeil’s very helpful study of the lieux communs in the Dictionnaire de théologie catholique, where, after reporting Melchior Cano’s description of the loci (which he notes are taken at times verbatim from Agricola) and Cano’s organization of theological loci, Gardeil proposes, in true topical tradition, a still better classification of his own. (Ong 1958, p. 122)
There are many lessons to be learned from this passage. First, it provides us with a definition of topics as “active associative nodes for ideas”, as they have been theorized since the birth of rhetoric in the context of the theory of argumentation in discourse. The particular interest of this passage lies in the description of the taxonomic trap. In order to put an end to the irritating proliferation of typologies, one might be tempted to propose a new and final one, thus bringing everyone to agreement — but, in the end, it seems that an additional typology has been added to an already overloaded list, thus exacerbating the very evil, it was supposed to remedy. This observation can be read as an ironic historical counterpoint to the works that, in that year, 1958, revived reflection on argumentation schemes.
2. The place of collections in the theories of the argumentation
The question of argument schemes plays a key role in some theories of argumentation while in others it is either redefined, or plays only a marginal role.
(i) The question of argument types does not arise in Anscombre and Ducrot’s theory of Argumentation within Language. The concept of topos is defined as a semantic link between predicates. It follows that the number of topoi is extremely large, even uncountable, while classical theories enumerate less than one hundred topoi.
(ii) Grize’s “Natural Logic” is based on the concept of schematization. The operations of “reasoned organization”, or “shoring” essentially amount to the classical concept of a conclusion supported by an argument. The types of arguments correspond to the types of scaffolding. To my knowledge, this line has not been developed further. Grize focuses on inference, causation, explanation.
(iii) In Toulmin’s terminology, a type of warrant corresponds to a type of argument, as shown by Ehninger and Brockriede ([1960]). In addition, Toulmin, Rieke and Janik (1984) proposed a short collection of arguments, see Collections (4). The example illustrating Toulmin’s “layout of argument” corresponds to a very productive topic, the categorization of an individual.
(iv) The concept is central to the New Rhetoric of Perelman & Olbrechts-Tyteca, as well as for Pragma-Dialectic and Informal Logic, see Collections (4).
3. Dimension of the classification: Number of argument schemes
Classical lists of argument schemes tend to propose a relatively large number or argument schemes. Aristotle’s Rhetoric offers a set of twenty-eight schemes, plus some “lines of argument that form the spurious enthymemes” (Rhet., II, 24; RR, p. 379); plus, some rules taken from the Topics. Cicero’s Topica lists a dozen schemes, and Quintilian’s Institutio Oratoria twenty-five. Boethius handed down fifteen forms on to the Middle Ages, see Collections (2).
The Dupleix’s Logic (1607) and Bossuet’s Logic (1677), may probably be regarded as representative, in modern times, of this classical tradition. The former contains fourteen schemes and the latter twenty.
Other modern typologies are quite different: Locke [1690] proposes a typology — if it can be considered as such — consisting of four elements to which Leibniz [1765] adds one. Locke’s scientific world however, is extremely different from, and antagonistic to the rhetorical world of the classics.
Bentham enumerates thirty-one argumentative formulas for the field of political arguments.
In contemporary times, Conley counts “more than eighty different argument types” in Perelman & Olbrechts-Tyteca Treatise (Conley 1984, p. 180-181) see Collections (4).
4. Forms of collections
In the Rhetoric, Aristotle presents a catalog of twenty-eight randomly listed topoi.
Perelman & Olbrechts-Tyteca have constructed a clearly organized four-level typology of the various “techniques of argumentation”
— A segment of speech (an individual, level 0) can be categorized as a “pragmatic argument”; that is, this segment presents the essential features that define the pragmatic argument (Level 1).
— Level 1 arguments are grouped into a super-category; for example, a “pragmatic argument” is classified as an “argument based on the structure of reality” (Level 2).
— Level 2 arguments are grouped into the class of the “techniques of association”, (Level 3), one of the two kinds of “techniques of argumentation” (Level 4, top level).
5. Foundations of collections
The collections of argument schemes can be organized in different ways.
(i) From the point of view of their contribution to the growth of scientific knowledge, inconclusive arguments are contrasted with compelling arguments. The latter are generally equated, in modern times, with mathematical demonstration and scientific proof. In Locke’s words, they “bring true instruction with [them] and advance us on our way to knowledge” (Locke [1690], Ch. 17, § 19-22), see Collections (3). Person-centered arguments are irrelevant from this point of view. The same might be said of those arguments that play only with the guilty pleasures of natural language and the nuances of interpersonal relations.
(ii) In terms of their linguistic form, metonymic arguments based on a relationship of contiguity, can be distinguished from the metaphoric arguments based on a relationship of similarity. This distinction reflects the opposition between the arguments “establishing the structure of reality” (analogy type) and those “based on the structure of reality’ (causal type) (Perelman & Olbrechts-Tyteca [1958] p. 261; 350), See Collections (4).
(iii) From the point of view of their productivity. The productivity of an argument scheme depends on the number of actual arguments (enthymemes) derived from it. Intuitively, some topics are very productive. One might think for example of those based on the twin argument schemes of categorization and definition; or arguments based on causal or analogical relations, or from the contraries, etc. Others, including the argument from waste are less productive. Other argument schemes may be more common than their name suggest, such as the argumentative exploitation of syzygies.
(iv) From the point of view of their legitimating power. A good example of organizing argument schemes according to their strength is given by the hierarchy of legal and theological arguments in the Arab-Muslim culture and religion, such as proposed by Khallâf ([1942]). He distinguishes between ten sources, ranked according to their degree of legitimacy. The most legitimate forms are those based on the Qur’an and the Tradition. Those with the weakest degree of legitimacy are, “the laws of the monotheistic peoples”, and, perhaps surprisingly given the present situation, “the opinions of the Prophet’s companions”, in that order. Arguments presented at the time of the emergence of Islam are given the least weight in the hierarchy of arguments. Such was the situation in 1942; it has changed significantly with the rise of Salafism.