1. Introduction
The Discourse-Historical Approach (DHA), pioneered by Ruth Wodak (see Wodak, de Cillia, Reisigl & Liebhart 1999; Wodak & van Dijk 2000; Wodak & Chilton 2005; Wodak & Meyer 2006; Wodak 2009), is one of the major branches of critical discourse analysis (CDA). In its own (programmatic) view, it embraces at least three interconnected aspects (Wodak 2006, p. 65)[i]:

1.’Text or discourse immanent critique’ aims at discovering internal or discourse-internal structures
2. The ‘socio-diagnostic critique’ is concerned with the demystifying exposure of the possibly persuasive or ‘manipulative’ character of discursive practices.
3. Prognostic critique contributes to the transformation and improvement of communication.

CDA, in Wodak’s view (ibid.),
is not concerned with evaluating what is ‘right’ or ‘wrong’. CDA … should try to make choices at each point in the research itself, and should make these choices transparent[ii]. It should also justify theoretically why certain interpretations of discursive events seem more valid than others.
One of the methodical ways for critical discourse analysts to minimize the risk of being biased is to follow the principle of triangulation. Thus one of the most salient distinguishing features of the DHA is its endeavour to work with different approaches, multimethodically and on the basis of a variety of empirical data as well as background information.

One of the approaches DHA is using in its principle of triangulation is argumentation theory, more specifically the theory of topoi. In this article, I will be concerned with the following questions: how and in what way are topoi and, consequentially, argumentation theory, used in DHA as one of the most influential schools of CDA? Other approaches (e.g. Fairclough (1995, 2000, 2003) or van Leeuwen (2004, 2008), van Leeuwen & Kress (2006)) do not use topoi at all. Does such a use actually minimize the risk of being biased, and, consequentially, does such a use of topoi in fact implement the principle of triangulation?

2. Argumentation and CDA
Within argumentation theory, Wodak continues (2006, p. 74),

‘topoi’ or ‘loci‘ can be described as parts of argumentation which
belong to the obligatory, either explicit or inferable premises. They are the content-related warrants or ‘conclusion rules’, which connect the argument or arguments with the conclusion, the claim. As such, they justify the transition from the argument or arguments to the conclusion (Kienpointner, 1992: 194).

We can find the very same definition[iii] in The Discursive Construction of National Identity (Wodak, de Cillia, Reisigl & Liebhart 1999, p. 34), in Discourse and Discrimination (Reisigl & Wodak 2001, p. 75), in The Discourse of Politics in Action (Wodak 2009, p. 42), in Michal Krzyzanowski’s chapter “On the ‘Europeanisation’ of Identity Constructions in Polish Political Discourse after 1989”, published in Discourse and Transformation in Central and Eastern Europe (Galasinska and Krzyzanowski 2009, p. 102), and in John E. Richardson’s paper (co-authored with R.Wodak) “The Impact of Visual Racism: Visual arguments in political leaflets of Austrian and British far-right parties” (manuscript, p. 3), presented at the 2008 Venice Argumentation Conference[iv]. In addition to the above definition, Richardson (2004, p. 230) talks of topoi “as reservoirs of generalised key ideas, from which specific statements or arguments can be generated”.
One could wonder about the purpose and the (ontological) status of the two definitions: are topoi “content-related warrants” or are they “generalised key ideas“? Because warrants are much more than just ideas; they demand much more to be able to secure the transition from an argument to a conclusion than just being “generalised ideas”, namely, a certain structure, or mechanism, in the form of an instruction or a rule. While ideas, or generalised ideas, lack at least a kind of mechanism the warrants seem to possess in order to be able to connect the argument to the conclusion. Let us proceed step by step.

3. How topoi are found…
In the above-mentioned publications, we get to see the lists of the(se) topoi. In the chapter “The Discourse-Historical Approach” (Wodak 2006, p. 74), we read that “the analyses of typical content-related argument schemes can be carried out against the background of the list of topoi though incomplete and not always disjunctive”, as given in the following table:
1. Usefulness, advantage
2. Uselessness, disadvantage
3. Definition, name-interpretation
4. Danger and threat
5. Humanitarianism
6. Justice
7. Responsibility
8. Burdening, weighting
9. Finances
10. Reality
11. Numbers
12. Law and right
13. History
14. Culture
15. Abuse.

In Richardson (2008, p. 4), we have exactly the same list of topoi, but this time they are characterised as “the most common topoi which are used when writing or talking about ‘others’“, specifically about migrants.
In The Discourse of Politics in Action (Wodak 2009, p. 44), we get the following list of “the most common topoi, which are used when negotiating specific agenda in meetings, or trying to convince an audience of one’s interests, visions or positions“. They include:
1. Topos of Burdening
2. Topos of Reality
3. Topos of Numbers
4. Topos of History
5. Topos of Authority
6. Topos of Threat
7. Topos of Definition
8. Topos of Justice
9. Topos of Urgency

In The Discourse of Politics in Action, we can also find topos of challenge, topos of the actual costs of enlargement of the EU, topos of belonging, and topos of ‘constructing a hero’. Here the analyses of typical content-related argument schemes as found in discourse are not just carried out “against the background of the list of topoi”, but some parts of discourse “gain the status of topoi” (topos of the actual costs…). Thus, as far as the status of topoi is concerned, we seem to have got a bit further: there is not just a list of topoi that can serve as the background for the analysis; more topoi can be added to the list. And, presumably, if topoi can be added to the list, they can probably also be deleted from the list. Unfortunately, in the publications I have listed, we get no epistemological or methodological criteria as to how this is done, i.e. why, when, and how certain topoi can be added to the list, or why, when, and how they can be taken off the list[v].

The most puzzling list of topoi can be found in Krzyzanowski (2009, p. 103). In this article we get the “list of the topoi identified in the respective corpora” (the national and the European ones – IŽŽ). They are[vi]:
Topoi in the national corpus
1. Topos of national uniqueness
2. Topos of definition of the national role
3. Topos of national history
4. Topos of East and West
5. Topos of past and future
6. Modernisation topos
7. Topos of the EU as a national necessity
8. Topos of the EU as a national test
9. Topos of the organic work
10. Topos of Polish pragmatism and Euro-realism.

Topoi in the European corpus
Topos of diversity in Europe
Topos of European history and heritage
Topos of European values
Topos of European unity
Topos of Europe of various speeds
Topos of core and periphery
Topos of European and national identity
Topos of Europe as a Future Orientation
Modernisation topos
Topos of the Polish national mission in the European Union
Topos of joining the EU at any cost
Topos of preferential treatment.

How these topoi were “identified”, and what makes them “the topoi” – and not just simply “topoi” –, we do not get to know; Krzyzanowski just lists them as such. Is there another list that helped them to be identified? If so, it must be very different from the lists we have just mentioned. Maybe there are several different lists? If so, who constructs them? When, where, and, especially, for what purpose and how? Is there a kind of grid, conceptual or in some other way epistemological and/or methodological that helps us/them to do that? If so, where can we find this grid? And how was it conceptually constructed? And if there is no such grid, how do we get all these different lists of topoi? By casuistry, intuition, rule of thumb? Are they universal, just general, or maybe only contingent? Judging from the lists we have just seen, there are no rules or criteria; the only methodological precept seems to be: “anything goes”[vii]! If so, why do we need triangulation? And what happened to the principle stipulating that CDA “should try to make choices at each point in the research itself, and should make these choices transparent?”

All this leads us to a key question: can anything be or become a topos? And, consequentially, what actually (i.e. historically) is a topos? Before we try to answer these questions, let us have a look at how the above-mentioned topoi are used in the respective works.

4. … and how topoi are used
In Discourse and Discrimination (Reisigl & Wodak 2001, p. 75), as well as in “The Discourse-Historical Approach” (Wodak 2006, p. 74), we can find, among others, the following identical definition of the topos of advantage:

The topos of advantage or usefulness can be paraphrased by means of the following conditional: if an action under a specific relevant point of view will be useful, then one should perform it (…) To this topos belong different subtypes, for example the topos of ‘pro bono publico’ (‘to the advantage of all’), the topos of ‘pro bono nobis’ (to the advantage of us’), and the topos of ‘pro bono eorum’ (‘to the advantage of them’).

And then the definition is illustrated by the following example:
In a decision of the Viennese municipal authorities (…), the refusal of a residence permit is set out as follows:
Because of the private and family situation of the claimant, the refusal of the application at issue represents quite an intrusion into her private and family life. The public interest, which is against the residence permit, is to be valued more strongly than the contrasting private and family interests of the claimant. Thus, it had to be decided according to the judgement.

If a topos were supposed to connect an argument with a conclusion, one would expect that at least a minimal reconstruction would follow, namely, what is the argument in the quoted fragment? What is the conclusion in the quoted fragment? How is the above-mentioned topos connecting the two, and what is the argumentative analysis of the quoted fragment? Unfortunately, all these elements are missing; the definition and the quoted fragment are all that there is.

And this is the basic pattern of functioning for most of these works. At the beginning, there would be a list of topoi and a short description for each of them (some of the quoted works would avoid even this step): first, a conditional paraphrase of a particular topos would be given, followed by a short discourse fragment (usually from the media) illustrating this conditional paraphrase (in Discourse and Discrimination, pp. 75-80), but without any explicit reconstruction of possible arguments, conclusions, or topoi connecting the two in the chosen fragment. After this short theoretical introduction, different topoi would just be referred to by names throughout the book, as if everything has already been explained in these few introductory pages.

It is interesting to observe how the functioning of these topoi is described (especially in Discourse & Discrimination, which is the most thorough in this respect): topoi are mostly »employed« (p. 75), or »found« (p. 76), when speaking about their supposed application in different texts, but also »traced back (to the conclusion rule)« (p. 76) or »based on (conditionals)« (p. 77), when speaking about their possible frames of definitions. How topoi are “based on (conditionals)”, or “traced back (to the conclusion rule)”, and how these operations relate to argument(s) and conclusion(s) that topoi are supposed to connect is not explained.

Consider another interesting example, this time from Discourse of Politics in Action (Wodak 2009, p. 97). In subsection 4.1, Wodak examines the discursive construction of MEP’s identities, especially whether they view themselves as Europeans or not. At the end of the subsection, she summarizes:
Among MEPs[viii] no one cluster characteristics is particularly prominent; however, most MEPs mention that member states share a certain cultural, historical and linguistic richness that binds them together, despite differences in specifics; this topos of diversity occurs in most official speeches (Weiss, 2002). Among the predicational strategies employed by the interviewees, we see repeated reference to a common culture and past (topos of history, i.e. shared cultural, historical and linguistic traditions; similar social models) and a common present and future (i.e. European social model; ‘added value’ of being united; a way for the future). Moreover, if identity is to some extent ‘based on the formation of sameness and difference’ (topos of difference; strategy of establishing uniqueness; Wodak et al., 1993, pp. 36-42), we see this in the frequent referral to Europe, especially in terms of its social model(s), as not the US or Asia (most prominently, Japan).

In trying to reconstruct the “topological” part of this analysis, three topoi are mentioned: topos of diversity, topos of history, and topos of difference. Surprisingly, only the topos of history is listed and (sparingly) explained in the list of topoi on p. 44: “Topos of History – because history teaches that specific actions have specific consequences, one should perform or omit a specific action in a specific situation.” The absence of the other two should probably be accounted for with the following explanation on pages 42-43:
These topoi have so far been investigated in a number of studies on election campaigns (Pelinka and Wodak 2002), on parliamentary debates (Wodak and van Dijk 2000), on policy papers (Reisigl and Wodak 2000), on ‘voices of migrants’ (Krzyzanowski and Wodak 2008), on visual argumentation in election posters and slogans (Richardson and Wodak forthcoming), and on media reporting (Baker et al. 2008).

But in the study “on visual argumentation in election posters and slogans”, for example, the(se) topoi are not discussed at all; they are presented as a fixed list of names of topoi, without any explanation of their functioning, while the authors (Richardson and Wodak) make occasional reference to their names – and not to the mechanism of their functioning – just as Wodak does in the above example from The Discourse of Politics in Action.

Therefore, if a topos is to serve the purpose of connecting an argument with a conclusion, as the respective works emphatically repeat, one would expect at least a minimal reconstruction, but there is none. What we have could be described as referring to topoi or evoking them or simply mentioning them, which mostly seems to serve the purpose of legitimating the (already) existing discourse and/or text analysis, but gives little analytical- or theoretical-added value in terms of argumentation analysis.

When I speak of reconstruction, what I have in mind is at least a minimal syllogistic or enthymemetic structure of the following type. As an example, I am using another topic from The Discourse of Politics in Action (Wodak 2009, pp.132-142), namely the problem of EU enlargement, as discussed among MEPs:
1. If a specific action costs too much money, one should perform actions that diminish the costs. (Topos connecting argument with conclusion)[ix]
2. EU enlargement costs too much money. (Argument)
3. EU enlargement should be stopped/slowed down … (Conclusion)

5. Back to the foundations: Aristotle and Cicero
It is quite surprising that none of the quoted works even mention the origins of topoi, their extensive treatment in many works and the main authors of these works, namely Aristotle and Cicero. As mentioned earlier, the definition, borrowed from Kienpointner (mostly on a copy-paste basis), does not stem from Aristotle or Cicero either: it is a hybrid product, with strong input from Stephen Toulmin’s work The Uses of Argument, published in 1958. According to Aristotle, as with many of his commentators, topoi are supposed to be of two kinds: general or common topoi, appropriate for use everywhere and anywhere, regardless of situation, and specific topoi, in their applicability, limited mostly to the three genres of oratory (judicial, deliberative, and epideictic). Or, as Aristotle (Rh. 1358a31-32, 1.2.22) puts it: “By specific topics I mean the propositions peculiar to each class of things, by universal those common to all alike”

The Aristotelian topos (literally: “place”, “location”) is an argumentative scheme, which enables a dialectician or rhetorician to construe an argument for a given conclusion. The majority of Aristotle’s interpreters see topoi as the (basic) elements for enthymemes, the rhetorical syllogisms[x]. The use of topoi, or loci, as the Romans have called them, can be traced back to early rhetoricians (mostly referred to as sophists) such as Protagoras or Gorgias. But while in early rhetoric topos was indeed understood as a complete pattern or formula, a ready-made argument that can be mentioned at a certain stage of speech (to produce a certain effect, or, even more important, to justify a certain conclusion), most of the Aristotelian topoi are general instructions allowing a conclusion of a certain form (not content), to be derived from premises of a certain form (not content). That is why Aristotle could present them as a “list” (though it really was not a list in the sense DHA is using the term): because they were so very general, so very basic, that they could have been used in every act of speech or writing. This is not the case with the DHA lists of topoi we have been discussing above: these topoi cannot be used in just any situation, but in rather particular situations, especially the topoi “identified” by Krzyzanowski. They could be classified not as common topoi, but more likely as specific topoi, something Aristotle called idia, which could be roughly translated as “what is proper to…”, “what belongs to…”. Also, this “list” of Aristotle’s common topoi was not there for possible or prospective authors “to check their arguments against it”. This “list” was there for general use, offering a stock of possible and potential common topoi for possible and potential future arguments and speeches.

5.1. Some basic definitions
Here is a short schematic and simplified overview of how Aristotle defines the mechanics and the functioning of topoi and their parts in his Topics, a work that preceded Rhetoric. We have to start with a few definitions.

Problems – what is at stake, what is being discussed – are expressed by propositions. Every proposition consists of a subject and predicate(s) that belong(s) to the subject. These predicates, usually referred to as predicables, are of four kinds: definition, genus, property, and accident:

Definition is a phrase indicating the essence of something. (T. I. v. 39-40)
A genus is that which is predicated in the category of essence of several things, which differ in kind. (T. I. v. 32-33)
A property is something, which does not show the essence of a thing but belongs to it alone and is predicated convertibly of it. (T. I. v. 19-21)
An accident is that which is none of these things … but still belongs to the thing. (T. I. v. 4-6)

These are the theoretical and methodological preliminaries that lead us to topoi, not yet the topoi themselves! To be able to select subject appropriate claims, premises for concrete context-dependent reasonings from the pool of potential propositions, we need organa or tools. Aristotle distinguishes four:
The means by which we shall obtain an abundance of reasonings are four in number:
1. the provision of propositions,
2. the ability to distinguish in how many senses a particular expression is used,
3. the discovery of differences and 4) the investigation of similarities. (T. I xiii. 21-26)

Strictly speaking, we are not yet dealing with topoi here, though very often and in many interpretations[xi] the four organa, as well as the four predicables, are considered to be topoi (and in the case of predicables, maybe even the topoi).
In the Topics, Aristotle actually established a very complex typology of topoi with hundreds of particular topoi: about 300 in the Topics, but just 29 in the Rhetoric[xii]. Two of the most important sub-types of his typology, sub-types that were widely used throughout history, are:
a. topoi concerning opposites, and
b. topoi concerning semantic relationships of “more and less”.

For an understanding of how topoi are supposed to function, here are two notorious examples:
Ad a)
If action Y is desirable in relation to object X, the contrary action Y’ should be disapproved of in relation to the same object X.
This is a topos, as Aristotle would have formulated it. And what follows is its application to a concrete subject matter that can serve as a general premise in an enthymeme (topos cannot):
“If it is desirable to act in favour of one’s friends, it should be disapproved of to act against one’s friends.”

Ad b)
If a predicate can be ascribed to an object X more likely than to an object Y, and the predicate is truly ascribed to Y, then the predicate can even more likely be ascribed to X.
Once more, this is a topos. And what follows is its application to a concrete subject matter that can serve as a general premise in an enthymeme (topos cannot):
“Whoever beats his father, even more likely beats his neighbour.”

We should now be able to distinguish two ways in which Aristotle frames topoi in his Topics. Even more, topoi in the Topics would usually be twofold; they would consist of an instruction, and on the basis of this instruction, a rule would be formulated. For example:
1. Instructions (precepts): “Check whether C is D.”
2. Rules (laws): “If C is D, then B will be A.”

Instructions would usually check the relations between the four predicables (definition, genus, property, accident), and, subsequently, a kind of rule would be formulated that could – applied to a certain subject matter – serve as a general premise of an enthymeme.

What is especially important for our discussion here, i.e. the use of topoi in critical discourse analysis, is that though they were primarily meant to be tools for finding arguments, topoi can also be used for testing given arguments. This seems to be a much more critical and productive procedure than testing hypothetical arguments “against the background of the list of topoi”. But in order to do that, DHA analysts should:
1. clearly and unequivocally identify arguments and conclusions in a given discourse fragment,
2. show how possible topoi might relate to these arguments.

In the DHA works quoted in the first part of this article, neither of the two steps was taken.

This is how topoi were treated in the Topics. But when we turn from the Topics to the later Rhetoric, we are faced with the problem that the use and meaning of topos in Aristotle’s Rhetoric is much more heterogeneous than in the Topics. Beside the topoi complying perfectly with the description(s) given in the Topics, there is an important group of topoi in the Rhetoric, which contain instructions for arguments not of a certain form, but with a certain concrete predicate, for example, that something is good, honourable, just, etc.

With the Romans, topoi became loci, and Cicero literally defines them as “the home of all proofs” (De or. 2.166.2), “pigeonholes in which arguments are stored” (Part. Or. 5.7-10), or simply “storehouses of arguments” (Part. Or. 109.5-6). Also, their number was reduced from 300 in Topics or 29 in Rhetoric to up to 19, depending on how we count them.

Although Cicero’s list correlates pretty much, though not completely, with Aristotle’s list from the Rhetoric B 23, there is a difference in use: Cicero’s list is considered to be  a list of concepts that may trigger an associative process rather than a collection of implicit rules and precepts reducible to rules, as the topoi in Aristotle’s Topics are. In other words, Cicero’s loci mostly function as subject matter indicators and loci communes[xiii]. Or, in Rubinelli’s words (2009, p. 107):

A locus communis is a ready-made argument that, as Cicero correctly remarks, may be transferable (…) to several similar cases. Thus, the adjective communis refers precisely to the extensive applicability of these kind of arguments; however, it is not to be equated to the extensive applicability of the Aristotelian topoi /…/. The latter are “subjectless”, while the former work on a much more specific level: they are effective mainly in juridical, deliberative and epideictic contexts.

But being ready-made, does not mean that they prove anything specific about the case that is being examined, or that they add any factual information to it. As Rubinelli puts it (2009, p. 148):
… a locus communis is a ready-made argument. It does not guide the construction of an argument, but it can be transferable to several similar cases and has the main function of putting the audience in a favourable frame of mind.

This brings us a bit closer to how topoi might be used in DHA. In the works quoted in this paper, the authors never construct or reconstruct arguments from the discourse fragments they analyse – despite the fact that they are repeatedly defining topoi as warrants connecting arguments with conclusions; they just hint at them with short glosses. And since there is no reconstruction of arguments from concrete discourse fragments under analysis, hinting at certain topoi, referring to them or simply just mentioning them, can only serve the purpose described by Rubinelli as “putting the audience in a favourable frame of mind.” “Favourable frame of mind” in our case – the use of topoi in DHA – would mean directing a reader’s attention to a “commonly known or discussed” topic, without explicitly phrasing or reconstructing possible arguments and conclusions. Thus, the reader can never really know what exactly the author had in mind and what exactly he/she wanted to say.

6. Topoi, 2000 years later
Let us jump from the old rhetoric to the new rhetoric now, skipping more than 2000 years of degeneration of rhetoric, as Chaim Perelman puts it in his influential work Traité de l’argumentation – La nouvelle rhétorique.

Topoi are characterised by their extreme generality, says Perelman (1958/1983, pp.112-113), which makes them usable in every situation. It is the degeneration of rhetoric and the lack of interest for the study of places that has led to these unexpected consequences where “oratory developments” against fortune, sensuality, laziness, etc. – which school exercises were repeating ad nauseam – became qualified as commonplaces (loci, topoi), despite their extremely particular character. By commonplaces we more and more understand, Perelman continues, what Giambattista Vico called “oratory places”, in order to distinguish them from the places treated in Aristotle’s Topics. Nowadays, commonplaces are characterised by banality, which does not exclude extreme specificity and particularity. These places are nothing more than Aristotelian commonplaces applied to particular subjects, concludes Perelman. That is why there is a tendency to forget that commonplaces form an indispensable arsenal in which everybody who wants to persuade others should find what he is looking for.

And this is exactly what seems to be happening to the DHA approach to topoi as well. Even more, the works quoted in the first part of the article give the impression that DHA is not using the Aristotelian or Ciceronian topoi, but the so-called »literary topoi«, developed by Ernst Robert Curtius in his Europaeische Literatur und Lateinisches Mittelalter (1990, pp. 62-105, English translation). What is a literary topos? In a nutshell, already oral histories passed down from pre-historic societies contain literary aspects, characters, or settings, which appear again and again in stories from ancient civilisations, religious texts, art, and even more modern stories. These recurrent and repetitive motifs or leitmotifs would be labelled literary topoi. “They are intellectual themes, suitable for development and modification at the orator’s pleasure“, argues Curtius (1990, p. 70). And topoi is one of the expressions Wodak is using as synonyms for leitmotifs (2009, p. 119):
“In the analysis of text examples which were recorded and transcribed I will first focus on the leitmotifs, which manifest themselves in various ways: as topoi, as justification and legitimation strategies, as rules which structure conversation and talk, or as recurring lexical items …”

This description and definition may well be dismissed as very general or superficial, but in The Discursive Construction of National identity, where 49 topoi are listed (without any pattern of functioning[xiv]), we can also find (p. 38-39) locus amoenus (topos of idyllic place) and locus terribilis (topos of terrible place). These two topoi have absolutely nothing to do with connecting arguments to conclusions, but are literary topoi per excellence, formulated and defined by E. R. Curtius[xv]. To clarify this: there is nothing wrong with literary topoi, their purpose just is not connecting possible arguments to possible conclusions.

For the New Rhetoric (Perelman/Olbrechts-Tyteca1958/1983, p. 113) topoi are not defined as places that hide arguments, but as very general premises that help us build values and hierarchies, something Perelman, whose background was jurisprudence, was especially concerned about.

Perelman has made some very interesting and important observations regarding the role and the use of topoi in contemporary societies. He argued that (Perelman/Olbrechts-Tyteca1958/1983, p. 114) even if it is the general places that mostly attract our attention, there is an undeniable interest in examining the most particular places that are dominant in different societies and allow us to characterize them. On the other hand, even when we are dealing with very general places, it is remarkable that for every place we can find an opposite place: to the superiority of lasting, for example, which is a classic place, we could oppose the place of precarious, of something that only lasts a moment, which is a romantic place.

And this repartition gives us the possibility to characterize societies, not only in relation to their preference of certain values, but also according to the intensity of adherence to one or another member of the antithetic couple.

This sounds like a good research agenda for DHA, as far as its interest in argumentation is concerned: to find out what views and values are dominant in different societies, and characterize these societies by reconstructing the topoi that underlie their discourses. But in order to be able to implement such an agenda – an agenda that is actually very close to DHA’s own agenda – DHA should dismiss the list of prefabricated topoi that facilitates and legitimizes its argumentative endeavour somehow beforehand (i.e. the topoi are already listed, we just have to check our findings against the background of this list of topoi), and start digging for the topoi in concrete texts and discourses. How can DHA achieve that?

7. Toulmin: topoi as warrants
Curiously enough, the same year that Perelman and Olbrechts-Tyteca published their New Rhetoric, Stephen Toulmin published his Uses of Argument, probably the most detailed study of how topoi work. I say “curiously enough” because he does not use the terms topos or topoi, but the somewhat judicial term “warrant”. The reason for that seems obvious: he is trying to cover different “fields of argument”, and not all fields of argument, according to him, use topoi as their argumentative principles or bases of their argumentation. According to Toulmin (1958/1995, pp. 94-107), if we have an utterance of the form, “If D then C” – where D stands for data or evidence, and C for claim or conclusion – such a warrant would act as a bridge and authorize the step from D to C (which also explains in more detail where Manfred Kienpointner’s definition of topos draws from). But then a warrant may have a limited applicability, so Toulmin introduces qualifiers Q, indicating the strength conferred by the warrant, and conditions of rebuttal (or Reservation) R, indicating circumstances in which the general authority of the warrant would have to be set aside. And finally, in case the warrant is challenged in any way, we need some backing as well.

It is worth noting that in Toulmin’s diagram, we are dealing with a kind of “surface” and “deep” structure: while data and claim stay “on the surface”, as they do in everyday communication, the warrant is – presumably because of its generality – “under the surface” (like the topos in enthymemes), and usually comes “above the surface” only when we try to reconstruct it. And how do we do that, how do we reconstruct a warrant?
What is attractive and useful about Toulmin’s theory is the fact that he is offering a kind of a guided tour to the centre of topoi in six steps, not just in three (as in enthymemes). All he asks is that you identify the claim or the standpoint of the text or discourse you are researching, and then he provides a set of five questions that lead you through the process.

If we revisit our semi-hypothetical example with the topos of actual costs of enlargement (Wodak 2009, pp. 132-142):
1. If a specific action costs too much money, one should perform actions that diminish the costs. (Topos connecting argument with conclusion)
2. EU enlargement costs too much money. (Argument)
3. EU enlargement should be stopped/slowed down… (Conclusion)

and expand it into the Toulmin model, we could get the following:
Claim: EU enlargement should be stopped/slowed down …
What have you got to go on?

Datum: EU enlargement costs too much money.
How do you get there?

Warrant: If a specific action costs too much money, one should perform actions that diminish the costs.
Is that always the case?

Rebuttal: No, but it generally/usually/very often is. Unless
there are other reasons/arguments that are stronger/
more important … In that case the warrant does not apply.
Then you cannot be so definite in your claim?

Qualifier: True: it is only usually… so.
But then, what makes you think at all that if a specific action costs too much money one should perform actions …

Backing: The history of the EU shows…
If the analysis (text analysis, discourse analysis) would proceed in this way[xvi] – applying the above scheme to concrete pieces of discourse each time it wants to find the underlying topoi – the lists of topoi in the background would become unimportant, useless, and obsolete. As they, actually, already are. Text mining, to borrow an expression from computational linguistics, would bring the text’s or discourse’s own topoi to the surface, not the prefabricated ones. Even more, Toulmin’s scheme allows for possible exceptions, or rebuttals, indicating where, when, and why a certain topos does not apply. Such a reconstruction can offer a much more complex account of a discourse fragment under investigation than enthymemes or static and rigid lists of topoi.

8. In place of conclusion
If DHA really wants to follow the principle of triangulation, as described in the beginning of the article, to make choices at each point in the research itself, and at the same time make these choices transparent, taking all these steps in finding the topoi in concrete texts would be the only legitimate thing a credible and competent analysis should do. If DHA wants to incorporate argumentation analysis in its agenda, that is, not just references to the names of concepts within argumentation analysis.

NOTES
[i] The complete and more detailed version of the paper can be found in Lodz papers in Pragmatics, vol. 6, n. 1/2010. Electronic version is available at: http://versita.metapress.com/content/v4477v71x19p/?p=f734650e911642299c0c906ff9982d14&pi=0.
[ii] Apart from the italicized use of topoi as terminus technicus all emphases (italics) in the article are mine (IŽŽ).
[iii] It should be noted that Kienpointner’s definition is a hybrid one, grafting elements from Toulmin (1958) onto Aristotelian foundations.
[iv] The paper was recently published in Critical Discourse Studies 6/4 (2009), under the title “Recontextualising fascist ideologies of the past: right-wing discourses on employment and nativism in Austria and the United Kingdom”. In this article, I am referring to the manuscript version.
[v] Let alone the fact that there is no theoretical explanation why there should be lists at all, or how we should proceed when checking the possible argument schemes “against the background of the list of topoi”.
[vi] These lists may look like recipes, but this is the way the authors present them.
[vii]It is interesting to observe that in his plenary talk at the CADAAD 2008 conference (University of Hertfordshire), Teun van Dijk emphasized: “CDA is not a method, CDA is not a theory … CDA is like a movement, a movement of critical scholars.” But then he added: “And they will use all the methods we know in various domains and schools of discourse analysis (see: http://www.viddler.com/explore/cadaad/videos/4/; 5th and 6th minute).” “Anything goes” should therefore be interpreted and understood in a much more narrow sense, namely, as “any method goes”. In other words, if a particular scholar or a particular school is using a certain method, the rules and principles of this chosen method should be followed.
[viii] Members of the European Parliament (IŽŽ).
[ix] It is worth noting that each topos can usually have two “converse” forms, and several different phrasings. Therefore the phrasing of this topos could also read: “If a specific action costs too much money, this action should be stopped”, depending on the context, and/or on what we want to prove or disprove (i.e. put forward as an argument).
[x] An important and more than credible exception in this respect is Sara Rubinelli with her excellent and most thorough monograph on topoi, Ars Topica. The Classical Technique of Constructing Arguments from Aristotle to Cicero, Argumentation Library, Springer, 2009.
[xii] See Rubinelli 2009, pp. 8-14.
[xii] The 29 topoi in the Rhetoric cannot all be found among the 300 topoi from the Topics. There is a long-standing debate about where these 29 topoi come from, and how the list was composed. Rubinelli (2009, pp. 71-73) suggests that their more or less “universal applicability” may be the criterion.
[xiii] This is probably due to the fact that Cicero was selecting and using loci in conjunction with the so-called stasis theory, or issue theory. What is stasis theory? Briefly and to put it simply, the orator has to decide what is at stake (why he has to talk and what he has to talk about): 1) whether something happened or not; 2) what is it that happened; 3) what is the nature/quality of what happened; 4) what is the appropriate place/authority to discuss what has happened. And Cicero’s loci “followed” this repartition.
[xiv] Instead, we can read (p. 34): “In place of a more detailed discussion, we have provided a condensed overview in the form of tables, which list the macro-strategies and the argumentative topoi, or formulae, and several related (but not disjunctively related) forms of realization with which they correlate in data.”
[xv] For a succinct description of locus amoenus, see Wikipedia: http://en.wikipedia.org/wiki/Locus_amoenus.
[xvi] Our sample analysis is, of course, purely hypothetical. Concrete analysis would need input from concrete discourse segments.

REFERENCES
Aristotle (1989). Topica (Transl. by E.S. Forster). Cambridge, Massachusetts: Harvard University Press.
Aristotle (1991). Art of Rhetoric (Transl. by J.H. Freese). Cambridge, Massachusetts: Harvard University Press.
Cicero, M. T. (2003). Topica (Transl. by T. Reinhardt). Oxford: Oxford University Press.
Curtius, R. E. (1990). European Literature and the Latin Middle Ages. New Jersey: Princeton University Press.
Fairclough, N. (1995). Critical discourse analysis: the critical study of language. Harlow: Longman.
Fairclough, N. (2000). Discourse and Social Change. Cambridge: Polity Press.
Fairclough, N. (2003). Analysing Discourse. Textual Analysis for Social Research. London/New York: Routledge.
Galasinska, A. & Krzyzanowski, M. (Eds.) (2009). Discourse and Transformation in Central and Eastern Europe. Basingstoke: Palgrave Macmillan.
Kienpointner, M. (1992). Alltagslogik. Stuttgart-Bad Cannstatt: Frommann-Holzboog.
Krzyzanowsky, M. (2009). On the ‘Europeanisation’ of Identity Constructions in Polish Political Discourse after 1989. In: A. Galasinska & M. Krzyzanowski (Eds.), Discourse and Transformation in Central and Eastern Europe (pp. 95-113), Basingstoke: Palgrave Macmillan.
Perelman, C. & Olbrechts-Tyteca, L. (1983). Traité de l’argumentation. La nouvelle rhétorique. Bruxelles: Editions de l’Université de Bruxelles.
Reisigl, M. & Wodak, R. (2001). Discourse and Discrimination. Rhetoric of Racism and Antisemitism. London/New York: Routledge.
Richardson, J. E. (2004). (Mis)Representing Islam: the racism and rhetoric of British Broadsheet newspapers. Amsterdam: John Benjamins.
Richardson, J. E. & Wodak, R. (2008). The impact of visual racism: Visual arguments in political leaflets of Austrian and British far-right parties. (Manuscript. Paper presented at the 2008 Venice Argumentation Conference).
Rubinelli, S. (2009). Ars Topica. The Classical Technique of Constructing Arguments from Aristotle to Cicero. Berlin: Springer.
Toulmin, S. (1995). The Uses of Argument. Cambridge: Cambridge University Press.
van Leeuwen, T. (2004). Introducing Social Semiotics. London/New York: Routledge.
van Leeuwen, T. (2008). Discourse and Practice. New Tools for Critical Discourse Analysis. Cambridge: Cambridge University Press.
van Leuween, T. &Kress, G. (2006). Reading Images. The Grammar of Visual Design. London/New York: Routledge.
Wodak, R. (2009). The Discourse of Politics in Action. Basingstoke: Palgrave Macmillan.
Wodak, R. & Chilton, P. (Eds.) (2005). A New Agenda in (Critical) Discourse Analysis. Amsterdam/Philadelphia: John Benjamins.
Wodak, R. & Meyer, M. (Eds.) (2006). Methods of Critical Discourse Analysis. London: Sage.
Wodak, R. & van Dijk, T. (Eds.) (2000). Racism at the Top. Klagenfurt: Drava.
Wodak, R., de Cillia, R., Reisigl, M. and Liebhart, K. (1999). The Discursive Construction of National Identity. Edinburgh: Edinburgh University Press.

1. Introduction
This paper offers a theoretical framework for describing the structure of arguments directed against epistemic authority. The interest of the previous studies concentrates mainly on the argumentations presented by the expert and on the argument ex auctoritate (Walton 1997). In this paper a different question is addressed, namely the structure of arguments directed against the expert’s arguments. A specialist’s reputation that can grant him the status of an epistemic authority depends on two factors, i.e. on what he actually argues and how the recipients react to his arguments. The reaction is direct when a recipient refutes an argument in a discussion with the author, and indirect when a recipient discusses these arguments with other persons, not with the author. Since reputation and refutation are strictly connected (Dascal 2001) negative reactions may contribute to the erosion of expert’s status of authority.

The expert is treated as a kind of personal warrant of the quality of arguments he formulates during his professional activity. Indeed, he has got a name and personal recognizability thanks to the quality of his argumentations in professional settings. Therefore, refutation of his argumentations may take a form of criticisms directly against him in his social function of a personal epistemic warrant. Such criticisms may even adopt the form of ad hominem, when some personal characteristics of the expert are introduced as premises of arguments against his arguments.

The paper is organised as follows. Section 2 offers a theoretical framework for describing the ad hominem. It combines the rhetorical categories of logos and ethos with three concepts stemming from the reputation studies: exceptionality, trustworthiness and solidity. Section 3 relates the concept of exceptionality to the two meanings of ‘expert’. In Section 4, the ad hominem in the Hamblian sense is discussed in relation to the trustworthiness. Section 5, instead,  is dedicated to the interrelation between the ad hominem in the Lockean sense and the solidity component. For the clarity of the presentation, the expert will be referred to as ‘he’, while the critic as ‘she’.

2. Theoretical framework
With respect to the human factor in argumentation, there are two main lines of research: the argumentation analysed with or without a human agent (Reygadas 2003).

The mainstream studies on argumentation seem to privilege the research on apersonal norms which assure the quality of argumentation. Virtually any reference to human agents involved in the argumentation has been treated for a long time as fallacy or at least as heterogeneous intrusion with respect to the logical rules and principles. However, the studies in the last decades have led to the identification of many nuances, some of them regarded as non-fallacious.

Another trend conceives of argumentation as relation, i.e. a consequential social act (Tindale 1999). It is claimed that the structure of arguments should be examined in relation to complex social, psychological and emotive factors. Such factors influence the arguers and nuance the interpretation of the arguments. In the authentic settings of communication, the human agents are held responsible for their own arguments and they are expected to be responsive to the arguments formulated by others. Not all de homine remarks are to be treated as fallacious ad hominem, only those which are used as irrelevant premises for an argument.

The first formulation of ad hominem argument has been traced already in Aristotle (Walton 2001: 209). Subsequent important developments appear in the work of Galileo and Locke (discussed e.g. in Finocchiaro 1980). Ad hominem is classified as one of the fallacies of irrelevance (Hamblin 1970: 41). In the standard treatment (van Eemeren and Grootendorst 1993), three realizations of ad hominem are pointed out: the abusive variant, the circumstantial variant and tu quoque variant. Modern studies explore various aspects of this argument, mainly concentrating on factors which differentiate fallacious from non-fallacious uses.  Perelman (1969) regards ad hominem not as an error, but as a necessary condition for successful argumentation. Walton (1997) offers a detailed account of realizations of ad hominem, many of which are non-fallacious. For Johnstone (1959, 1978), ad hominem (in the Galileo’s and Locke’s definition) is regarded as the very basis of philosophy, in which the philosopher as author of his own argumentative universe is the expected warrant of the coherence of the expressed views. The complexity of ad hominem explains why sometimes it can be conceived of as a fallacy and sometimes as a valid argument.

The interest of studying the ad hominem in relation to the expert’s argumentation is due to the fact that the expert, with his unique intellectual capacities proved in various instances of argumentation, is perceived as an embodied warrant of knowledge. The key concept is ‘being perceived as’: it relates the research on expert argumentation with the reputation studies.

According to an early definition, “A good reputation consists in being considered a man of worth by all, or in possessing something of such a nature that all or most men, or the good, or the men of practical wisdom desire it” (Aristotle, Rhetoric 1361a 8). This formulation underlines the role of evaluation by others (“being considered”) in the emergence of good reputation. However, it only hints at the objects of evaluation (“worth”, “possessing something”) and it does not address the mechanisms involved in the emergence of good reputation.

The modern studies on reputation in different professional settings list factors such as exceptionality, reliability, solidity, fitting the expectations, responsibility, and trustworthiness (for recent contributions, see Klewes and Wreschniok eds. 2010).  Some of these concepts partially overlap (e.g. solidity and fitting the expectations, which in the subsequent analysis will be treated as equivalent). Three of them seem to be essential for a good reputation (Wieseneder and Cerny 2006). The first one is exceptionality: a person differs positively from others and he is indeed  perceived by them as outstanding. The second one is trustworthiness: a person continues repeating positively valued actions and people actually perceive him as doing so. The third one is solidity: a person performs his role within the framework of the professional and ethical expectations and is actually perceived as such. The good reputation is therefore based on the paradox of being different from others, i.e. exceptional, and fitting the expectations of others, i.e. deserving trust and behaving as expected. The definitions of the essential factors of the good reputation are very broad, since they refer to the activity in any profession. In what follows, these three concepts will be examined in relation to a peculiar case: the activity of argumentation performed professionally by an expert and to the criticisms in the form of ad hominem.

Since knowledge is communicated through texts, the interrelation between “apersonal” and “personal” aspects of expert argumentation may be fruitfully described in terms of rhetorical logos and ethos. Logos is a term loaded with many philosophical, epistemological and rhetorical meanings. For the purposes of this paper, logos, a rational component within persuasive communication, is defined as roughly equivalent to the argumentation presented by the author. In other words, it represents “apersonal” view on argumentation and allows for examining arguments in terms of structures and rules.

Ethos, treated as a “personal” and, according to Aristotle, emotive component of argumentation, is far more complex. For the goals of this paper, it is useful to distinguish its three facets. The first one is the ethos in the strict sense of the term, i.e. based on what is actually argued in the text (see below, Section 3). The second one combines the textual information, i.e. ethos proper, with the referential information about the author in the extratextual world. It  concerns both the features independent from the author (such as ethnicity or gender) and the features at least partially dependent on him (e.g. academic credentials, intellectual commitments; see below, Section 4). The third facet regards a kind of “capitalised ethos”, i.e. a global evaluation based on the body of texts ever authored by the expert (see below, Section 5).

3. Expert’s exceptionality
Within the common use of the term ‘expert,’ there are two partially overlapping meanings. One is the term ‘expert’ in an institutional sense: after having acquired appropriate academic credentials, anyone can be called an ‘expert’ (or a ‘specialist’, or a  ‘professional’).

The other is the term ‘expert’ granted to relatively few knowledgeable persons in virtue of the professional recognition of their outstanding performances. The two notions cannot be directly opposed, since any active member of a professional community, i.e. an expert in the institutional sense, aims at gaining good reputation as an expert in the qualitative sense. However, the recognition of the quality of his performances is not always unanimously shared by all the members of his community. The exceptionality, i.e. the basic element of good reputation, is therefore required to become an expert in the qualitative sense of the term.

In the case of the professional activity consisting of producing well-argued texts, the essence of the exceptionality is the rhetorical ethos. The complexity of the concept is underlined by Aristotle: “Persuasion is achieved by the speaker’s personal character when the speech is so spoken as to make us think him credible. We believe good men more fully and more readily than others: this is true generally whatever the question is, and absolutely true where exact certainty is impossible and opinions are divided. This kind of persuasion, like the others, should be achieved by what the speaker says, not by what people think of his character before he begins to speak. It is not true, as some writers assume in their treatises on rhetoric, that the personal goodness revealed by the speaker contributes nothing to his power of persuasion; on the contrary, his character may almost be called the most effective means of persuasion he possesses” (Aristotle Rhetoric 1356a).

In the strict meaning of the term, the rhetorical ethos is a textual construction: “when his speech is delivered in such a manner as to render him [i.e. the speaker] worthy of confidence”. Aristotle underlines that this conviction should emerge from the text and not from the extratextual information. The best illustration of ethos in written texts is the practice of the blind review. In such a case, no extratextual information about the author is available: the only cues emerge from the quality of arguments. An outstanding quality of argumentation in a text permits to evaluate the author as possessing exceptional intellectual abilities. The rational, i.e. logos, creates the emotional, i.e. ethos, converted into a sort of social gain assuring the respect among the recipients.

Since ethos is in a way a byproduct of logos, in order to damage ethos it is necessary to undermine logos by a direct attack against the arguments themselves. Such a criticism apparently does not have a form of ad hominem, since it is the argumentation, and not its author, that is at stake. However, although formulated in ad rem way, such criticisms are abusively made with an ad hominem intent, aiming to destroy the author’s credibility as an expert. The typology of such attacks is very rich; since it has been discussed elsewhere (see Załęska 2008), it will not be reminded here due to space limitations.

4. Expert’s trustworthiness
In the case of expertise, the trust is the essence of the relation between epistemic authority and authoritee, based on the cognitive asymmetry. The non-experts in a particular field of competence are constrained to trust the expert because they lack the necessary field knowledge to evaluate his arguments. The expert, on his side, is expected to observe the ethical standards of intellectual honesty, formulating a possibly unbiased judgment, based only on available evidence transformed in relevant premises and conclusions.

Two of  Walton’s (1997) critical questions regard the trustworthiness. One of them – the backup evidence question (“is the expert opinion based on evidence?”) – addresses it indirectly. This formulation may be linked to the requirements of the rhetorical logos. The trustworthiness question addresses it directly: “is E personally reliable as a source?”.

The meaning attributed to trustworthiness within the reputation studies does not limit itself to the relation of cognitive trust. It is defined instead in behavioral terms, formulated very broadly in order to fit the description of any profession. A trustworthy (or reliable) person is the one that continues repeating positively valued actions and is actually perceived as such. This definition of trustworthiness underlines the temporal and iterative aspect of habituation that both confirms previous expectations and enhances further ones (see below, Section 5).

When the answer to the Walton’s trustworthiness question (especially if negative) is inserted into the premises of the argument itself, it is treated as an ad hominem in its most popular definition, i.e. when “a case is argued not on its merits but by analyzing (usually unfavorably) the motives or the background of its supporters or opponents” (Hamblin 1970, p. 41). The ad hominem (1) concerns mainly the expert’s extratextual features, i.e. his being “personally reliable as a source”. In a way such a criterion of extratextual ethos appears also in the Aristotle’s definition quoted above in which he underlines that “it is not true […] that the personal goodness revealed by the speaker contributes nothing to his power of persuasion” and that “his character may almost be called the most effective means of persuasion he possesses”.

There are three interrelated meanings of trustworthiness relevant for the ad hominem referred to an expert. One, the extratextual ethos, regards the stereotypical  trustworthiness ascribed to the groups to which he belongs (e.g. nation, gender, profession). The stereotypical motivations attributed to members of these groups stem from the ideas about their trustworthiness. Such an information gives material for the circumstantial version of ad hominem which overlaps with the genetic fallacy of the polluted source.

The two other meanings emerge within the textual universe. The first one is textual ethos as described above (see Section 3): the trust is due to ethos which emerges through the quality of arguments. If each text raises trust, it is likely to generalize into the second kind of trustworthiness, i.e. as a lasting attribute of an author.

The introduction of the extratextual ethos as overt or covert premise of an argument is fundamental for the ad hominem (1). The first group of examples regards the prejudice (in the etymological sense of prae iudicium, i.e. ‘before judgment’) or stereotypes, which, on the one hand, are regarded as useful heuristics to a preliminary categorization, but, on the other hand, usually exemplify the fallacy of hasty generalization. The author is perceived as biased often due to the biases of his opponents who insist on analyzing (usually unfavorably) his motives or his background. Biased perceptions of ingroups and outgroups revolve about attributions of positive or negative intentions, beliefs, motivations and predetermine one’s trustworthiness even before he starts to speak. These judgements may concern the sociological categorizations into nations, professions etc., or psychological ones, attributing stereotyped intentions to different groups of people.

In the aggressive version of ad hominem (1), a hint at motives and backgrounds stigmatized in various historical periods presents the author as unworthy of trust (e.g. as in the case of Einstein, discredited by Nazis as a Jew). This realization of ad hominem aims at depriving the expert of the right to argue and/or at destroying his arguments as stemming from an allegedly polluted source.

In the apparently non aggressive version, a favorable interpretation of motives and backgrounds imposes instead the relation of trust based on the extratextual information (e.g. she is your professor, so she is right). It is one of the realizations of the inverse ad hominem, i.e. ex auctoritate. Sometimes it is merely a manifestation of an informal probabilistic reasoning: it is indeed probable, and only probable, that a professor is a knowledgeable person. However, the introduction of such pieces of information about the author of the argument is treated usually as fallacious due to irrelevance of a general categorization regarding extratextual features for the structure of a concrete argument.

The second group of examples involves local, insiders’ knowledge about the particular experts who do act in particular social settings and have particular interests. The information used in such an ad hominem is therefore concrete (e.g. he is an independent expert vs. he is an expert hired by a firm X) which renders the particular categorization more probable. However, it is always only hypothetical and probabilistic. In reality, an expert hired by a firm X can offer an unbiased, professional judgment. The adequacy of the evaluation of expert’s trustworthiness depends on individual contextual knowledge and on the evaluator’s perspicacity and objectivity. Indeed, the evaluation may range from imputation of quite imaginary motives up to the unveiling of an actual hidden agenda.

The uncertainty of this type of judgment makes it difficult to include such an extra-textual information about the person into schemes of valid argumentation. However, a dismissal of background knowledge regarding the complexity of human intentions, interests, characters in favour of the pure face-value of the arguments sometimes would be a theoretical and practical naiveté. The appeal to such a standard is namely evoked by some pseudo-scientists, propagandists, politicians, spin doctors or scammers who are likely to have problems with trustworthiness. They often claim that their utterances should be interpreted only against the face-value of the arguments they present. As a rule, indeed, nicely expressed and well sounding arguments seem very appealing if interpreted without an examination of the intents, biases and actions of their authors.

The distinction between general, approximate knowledge and detailed information about a peculiar case is not polar. It is rather a combination of some general, preliminary categorizations with the circumstances of a peculiar case. Such a correlation confirms the already formulated observations that not all ad hominem are fallacious, and that sometimes the skepticism towards the author allows for appropriate evaluation of his argument.

The author’s professional trustworthiness, as meant in the reputation studies, is mediated through the way in which the reader perceives author’s professional competence as constructed in the text. If the reader:
1. interprets that the argumentation is reasonable, based on evidence, without any (perceptible) bias,
2. believes that, until contrary evidence appears, by default all the authors act in bona fide,
3. decides to accept the face-value of the arguments, she may infere that the expert is trustworthy.

The criticisms and the consequent discredit may occur at any of these three levels. If the argumentation is not valid, the counterarguments ad rem suffice to dismiss it; the author’s textual ethos is damaged but collaterally. Instead, in the second and in the third parameter listed above, the skepticism appears when the reader has doubts concerning the expert’s good faith. The opponent may voice the divergence between textual ethos,  which emerges from the face-value of the arguments presented, and the supposedly relevant extratextual ethos, e.g. ascription to categories of people who are not trusted, for instance, propagandists. A remark such as what he says sounds too well to be true, I don’t believe him illustrates this case of ad hominem. Particularly appealing, almost irresistible arguments, sometimes raise trust. However, they can also provoke distrust if they are interpreted as reminiscent of manipulation techniques which only mimic the valid arguments. The allusion to the untrustworthiness of the author and his text is created by evoking a general category of people (propagandists, scammers, manipulators) who try to create well-sounding arguments while realizing their hidden agenda. The opponent may also throw discredit on the expert by showing that what he claims in his publications is not adequate to what actually happens, or, in more ad hominem way, that what the author used to claim in the textual universe is at odds with his own decisions in the extratextual universe. The growing split between what the author affirms and what are his personal decisions within the same area (e.g. the economic decisions) may be interpreted in terms of the lack of trustworthiness.

5. Expert’s solidity
Solidity is not a concept used within the argumentation theory, although some of its features are involved in the discussion of other notions. Within the reputation studies, solidity in the most general sense means that a person is able to perform his role within the framework of the professional and ethical expectations, behaving in a coherent and responsible way. In the case of an expert, the solidity refers to a peculiar aspect of expert’s professional activity, i.e. formulation of arguments in professional texts. Thus, solidity regards the capacity to formulate correct and coherent argumentations consistent with the author’s declared epistemic commitments. Using the criterion of solidity, the recipients treat the expert as a living link between all the argumentations he has ever authored. Even if the expert, while developing his competence, changes his views, abandons the obsolete concepts, refines his first intuitions and searches for better solutions, it is expected that he will not contradict himself by arguing for standpoints which are at odds with what he had claimed previously.

The responsibility for the coherence of views ever expressed is the basis for the argument ad hominem (2), whose first formulations are attributed to Galileo and Locke (Finocchiaro 2005): “to press a man with consequences drawn from his own principles or concessions […] is […] known under the name of argumentum ad hominem” (Locke 1959 [1690]: 2, 411).

The criteria evoked above underline the importance of both the authorship and the readership. On the one hand, it is expected that the author is aware of the consequences of his arguments and controls the unity of the created argumentative universe. On the other hand, only an inquisitive and critical audience, which takes the arguments seriously and is competent enough to detect discrepancies, is able to influence the  author.

The author’s “own principles or concessions” may be considered at the intra-textual or inter-textual level. The intra-textual level regards the argumentative coherence as examined within one text. It could therefore be seen as a sub-case of the rhetorical ethos, in a way collateral to the rhetorical logos (see above, Section 3). The  author himself creates expectations: he autonomously formulates the claim, chooses the facts he regards as relevant to the case, transforms them into argumentative premises and draws conclusions. Therefore, he is held accountable for the views he authored. It is expected that he will not contradict himself within a text. A historical example that fits nicely this case of ad hominem is Socrates’ maieutic method: he elicited from the interlocutors public epistemic commitments and afterwards he confronted them with the argumentative consequences of their own principles and concessions formulated within such a conversation. The intra-textual level has only a partial impact on the author’s reputation, since one text is only a small contribution to the overall evaluation of the author as an expert.

Far more important for the expert’s solidity is the inter-textual level, embracing the whole body of his texts. What changes is the scale of argumentation for which the author is held responsible: not the one exposed in a single text, but the argumentation produced over a longer period of time in different texts. Such a lasting coherent argumentative activity is likely to give the author renown.

If it the unity of argumentative universe is barely noticed and/or if important discrepancies are detected, the author may be regarded as expert only in virtue of his institutional credentials, but not his professional activity. Let’s consider a case of an author who in several books develops a theory claiming A. Afterwards, without any explanation or disclaimers, he starts to affirm that not A. Such an articulation of the contradictory views, if noticed, cannot remain without consequences for his professional reputation.

If instead the author’s work is characterized by an overall coherence with the declared epistemic commitments, it contributes to the perception of author’s solidity and to his reputation of an outstanding expert. The non-contradictoriness of arguments within the whole of his professional activity projects positively on the expert’s reputation, since it suggests that the expert made judiciously his original epistemic commitments, based on a correct interpretation of evidence, and, therefore, he does not need to negate any part of his work based on them.

The solidity may be viewed as a kind of “capitalised ethos”, i.e. a generalisation of ethos which emerges from logos in each of the examined texts. Such a “capitalised ethos” is partly a rhetorical achievement resulting from the quality of texts, but partly a result of complex social mechanisms which create one’s  reputation (Załęska 2010). It is a kind of  “post-iudicium”, or a motivated opinion after having read the texts,  which is likely to become a “prae-iudicium” when reading a new text written by the author.

As stated above, the solidity in the sense of the non-contradictoriness and coherence of the views endorsed by the same author is not an absolute value. It is connected with the adequacy of the views or theories which should be supported by the factual evidence. Therefore, the requirement of solidity as one of the elements of good reputation is composed of three elements:
1. Expert argued x;
2. Expert still argues x;
3. Expert has been substantially right from the beginning.

There are two main types of opponent’s reaction likely to bring the author into disrepute. In the first one, the opponent underlines that:
1. Expert argued x;
2. Expert still argues x;
3. Expert has been substantially wrong from the beginning (or from a certain point in his argumentation). Such a criticism reveals unnoticed errors or even the fallacy of inconsistent commitment. In this case, the expert is unaware of the argumentative consequences of his own commitments, principles and concessions until the opponent attacks them directly, and, directly or indirectly, attacks their author.

In the second case, the opponent underlines that:
1. Expert argued x;
2. Expert does not argue x any more;
3. Expert has been substantially wrong at the beginning. An appropriate illustration is the case of an expert economist to whom it is reminded that he used to argue for the superiority of the communist system over the capitalist one, and now he is arguing for the superiority of the capitalist system over the communist one, contradicting in this way his own original expert commitments. The cases in which an expert overtly abandons an erroneous view and embraces the one which corresponds with the reality more correctly, seem less discreditable, although not without consequences for the author’s reputation.

Due to the strict interrelation between personal solidity and the argumentative coherence, the qualification of the criticism by the opponent in terms of ad hominem or not, seems to depend on the linguistic ways of expression.

When the linguistic expressions are explicitly directed towards the person, in the form of meta-textual cues (now, as a liberal, you say A, but it is not coherent with what you had claimed before adopting a conservative stance, namely -A), the opponent clearly discredits the author, pressing him with the consequences of his own principles and concessions. In other words, the opponent requires expert’s responsibility for the views expressed in the etymological sense of response ability, i.e. the capacity to give an answer.

However, the same counterarguments concerning “consequences drawn from… principles or concessions” may be formulated in an ad rem way, as if they were authorless and as if the flaws in argumentation did not have any influence on the evaluation of the expert (e.g. the proposition A is in contradiction with the proposition B). In other words, ad hominem (2) seems not to be the case of substance, but of the linguistic framing of the counter-argumentation.

The realization of ad hominem in Locke’s sense fits the tu quoque variant of the argument: it is reminded to the expert that he also shared a certain view at a certain moment of his argumentative activity.

Expert’s solidity in the construction of epistemic universe of argumentation interplays with the referential adequacy of this epistemic universe to the facts. The obstinacy to defend his original epistemic commitments if they are at odds with the available evidence is fatal for the expert’s professional reputation.

The gravity of the consequences for the expert’s compromised solidity depends on the intellectual atmosphere of a certain historical period. The critical attention for the work of others and respect for the intellectual seriousness increase the possibility of discredit. The pluralism of ideas and the acceptance of different standpoints, instead, makes such inconsistencies barely noticed.

6. Conclusions
Taking into account the “personal” dimension extends the vocabulary and concepts applied to describe the argumentation. Exceptionality, trustworthiness and solidity, with correlated notions such as intentions, beliefs, biases or personal interests problematize the straightforward relation between the face-value of the arguments and their function in interpersonal communication.

The potential material for clear cases of ad hominem does not stem from the textual ethos, in a way collateral to logos, but from two other interrelated sources. One is the extra-textual ethos, i.e. what the arguer is and/or does and/or supposedly thinks or intends.  The other is the “capitalised ethos”: what is said about the author, in relation to what he actually argued in previous instances of argumentation. Both are interpreted as two different kinds of ad hominem: the first emerging within the extra-textual universe, the latter – within the inter-textual universe. As shown above, there is a relationship, although not straightforward, between the two discussed versions of ad hominem and  the trustworthiness and solidity as conceived of within the reputation studies.

The trustworthiness is realisable only on the interpersonal level. In the prototypical examples of ad hominem, the extra-textual information regarding social or psychological categorizations predetermines the perception of argumentation. However, there are also the peculiar cases in which the author is brought into disrepute through a peculiar interpretation of two factors. On the one hand, there is the face-value of the arguments which sound well and are expected to grant trust. On the other hand, the authors that formulate the arguments are ascribed to a group that by definition is mistrusted.

The solidity is based on the expectation that the expert functions as a personal warrant of the non-contradictoriness of the argumentations he formulates. The criticisms which have an impact on reputation consist mainly of pointing out to the major contradictions and discrepancies between different argumentations presented by the same expert. The Lockean version of ad hominem evokes therefore intellectual standards of argumentation, holding the author responsible for his words within the whole of his argumentative activity.

Although the concepts elaborated within the argumentation theory and reputation studies share certain common features, there are collocated at different levels of generality. What the argumentation theory defines in relation to the cognitive activity, the reputation studies refer to any professional activity. The argumentation theory and the fallacies that it points out to regard mainly the structure of single arguments. The reputation studies, instead, are concerned with approximate evaluations regarding long-lasting patterns of behaviour, such as series of recognized argumentations.

The study of relation among the structure of the arguments and their social impact is fundamental in order to gain insight into complexities of the actual argumentative practices. It could explain a paradox, so aptly captured by Aristophanes, of apparently correct arguments excluded due to their source: “Even if I believe you, I do not believe you”.

REFERENCES
Aristotle Rhetoric. Retrieved from http://www2.iastate.edu/~honeyl/Rhetoric/
Dascal, M. (2001). Reputation and refutation: Negotiating merit. In E. Weigand, & M. Dascal (Eds.), Negotiation and Power in Dialogic Interaction. Amsterdam: Benjamins.
Eemeren, F. H. van, & Grootendorst, R. (1993). The history of the argumentum ad hominem since the seventeenth century. In E. C. W. Krabbe, R. J. Dalitze, & P. A. Smith (Eds.), Empirical Logic and Public Debate. Essays in Honour of Else M. Barth (pp. 49-68). Amsterdam, & Atlanta: Rodopi.
Finocchiaro, M. (1980). Galileo and the Art of Reasoning. Rhetorical Foundations of Logic and Scientific Method. Boston: Reidel.
Finocchiaro, M. (2005). Valid ad hominem arguments in philosophy. Johnstone’s Metaphilosophical Informal Logic (2001). In: M. Finocchiaro, Arguments about Arguments. Systematic, Critical and Historical Essays on Logical Theory (pp. 277-291). Cambridge: Cambridge University Press.
Hamblin, C. L. (1970). Fallacies. London, & Methuen: Rpt. Newport News, Va.: Vale.
Johnstone, H. W. Jr. (1959). Philosophy and Argument. University Park: Pennsylvania State University Press.
Johnstone, H. W. Jr. (1978). Validity and Rhetoric in Philosophical Argument. University Park, Pa.: Dialogue Press of Man & World.
Klewes, J., & Wreschniok, R. (Eds). (2010). Reputation Capital. Building and Maintaining Trust in the 21st Century. Berlin – Heidelberg: Springer – Pleon.
Reygadas, P.(2003). A non-propositional approach to emotions in argument. In F. H. van Eemeren, J. A. Blair, C. A. Willard, & A. F. Henkemans (Eds.), Proceedings of the Fifth Conference of the International Society for the Study of Argumentation (pp. 893-898). Sic Sat, Amsterdam.
Locke, J. (1959 [1690]). An Essay Concerning Human Understanding. A. C. Fraser (Ed.). New York: Dover.
Perelman, Ch. (1969). Le Champ de l’Argumentation. Bruxelles: Presses Universitaires de Bruxelles.
Tindale, Ch. (1999). Acts of Arguing: a Rhetorical Model of Argument. Albany: SUNY Press.
Walton, D. N. (1998). Ad Hominem Arguments. Tuscaloosa: The University of Alabama Press.
Walton, D. N. (1985). Arguer’s Position: A Pragmatic Study of Ad Hominem. Attack, Criticism, Refutation, and Fallacy. Westport, Conn.: Greenwood Press.
Walton, D. N. (1997). Appeal to Expert Opinion: Arguments from Authority. University Park, Pennsylvania State University Press.
Walton, D. N. (2001). Searching for the roots of the circumstantial ad hominem. Argumentation 15(2), 207-221.
Wieseneder, S., & Cerny, T. (2006). Reputationsmanagement. Carl Hanser Verlag: München-Wien.
Załęska M. (2008). Il troppo stroppia. Le critiche dell’expert writing nelle discipline umanistiche e sociali. In U. Reutner, & S. Schwarze (Eds) Le Style, c’est l’Homme. Unité et Pluralité du Discours Scientifique dans les Langues Romanes (pp. 205-226). Frankfurt am Main – Berlin – Bern – Bruxelles – New York – Oxford – Wien: Peter Lang.
Załęska M. (2010). Stratégies d’accréditation et de discréditation dans le discours académique. A paper presented at the conference Communication and Argumentation in the Public Sphere 4 (CAPS 4), University “Dunarea de Jos”, Galati (Roumanie), 13-16.05.2010.

1. The emphasis on agreement
It is almost a truism in argumentation studies that productive disagreement must be grounded in agreement. Shared understandings of the goal, shared commitment to particular procedures, and shared adherence to basic truth-claims are thought to be necessary in order for arguers to engage each other rather than to talk past each other. Among the many writers who offer some version of this postulate are Perelman and Olbrechts-Tyteca (1969: 65), who say, “The unfolding as well as the starting point of the argumentation presuppose indeed the agreement of the audience. … from start to finish, analysis of argumentation is concerned with what is supposed to be accepted by the audience.” In a similar vein, Ehninger (1958: 28) wrote, “Debate is not a species of conflict but of co-operation. Debaters … co-operate in the process of submitting a proposition to rigorous tests. … They believe … not so firmly that they are unwilling to put their convictions to a severe test and to abide by the decision of another concerning them.” These underlying beliefs about purpose and mode of procedure are agreed to by all disputants. Brockriede (1975: 182), identifying indicators of argumentation, includes among them “a frame of reference shared optimally.” Argument is pointless, he suggests, if two people share too much in their underlying presuppositions, but it is impossible if they share too little.  And MacIntyre (1984: 8) notes the impossibility of reasoning with one another when there are no shared standards to undergird rational talk. These are only four representative examples.

It is not hard to see why there would be so much agreement on the need for agreement. First, as Aristotle acknowledged, we do not argue about matters that are certain.  But claims that are not self-evident must be evaluated by reference to some standards to determine whether they are strong or weak, better or worse.  Second, though, neither the foundationalism of traditional philosophy nor the universal standards of formal logic and mathematics encompasses ordinary argumentation. So consensus of the arguers about standards becomes the substitute for formal validity.

2. Deep disagreement
But what happens when this underlying stratum of agreement is, or is thought to be, lacking? Then any claim advanced by one arguer can be challenged by the other, in a potentially infinite regress, because there is no point at which the interlocutor, by virtue of his or her own prior commitments, is obligated to accept any standpoint. This state of affairs was first characterized by Robert J. Fogelin (1985) as deep disagreement. Each arguer’s claims are based on assumptions that the other arguer rejects. Deep disagreement is the limiting condition at which argumentation becomes impossible. Most discussions of deep disagreement assume that it is a relatively rare occurrence that hardly denies the utility of argumentation for enabling ordinary arguers to resolve their disagreements peacefully. And because many discussions of argumentation presume a dialogue framework, deep disagreement is often dismissed as if it had no serious consequences beyond the immediate dialogue participants.

Both of these assumptions are dubious: the first because of the growth of fundamentalism and the second because deep disagreement has been found politically useful.  The past generation has seen the increased appeal of fundamentalism within many of the world’s major religious traditions – ultra-Orthodox Judaism, evangelical Christianity, and radical Islam. Fundamentalism rejects the modernist assumption of human fallibility and the resulting tolerance of diverse viewpoints. Fundamentalists believe that it is possible to know God’s will for sure. God has made it clear, and the Divine Word can be read and understood by anyone willing to try. Deviation from God’s word in order to demonstrate tolerance to misguided others is not only unnecessary but perverse, implicating the righteous in the sins of the godless.

Because of the conflict between fundamentalism and modernism (or, even more so, postmodernism), many disagreements are understood by one side in moral and religious terms and by the other in pragmatic and secular terms. This is true not only with respect to matters of personal identity and rights, such as abortion, feminism, and gay rights, but increasingly to issues ranging from taxation and fiscal policy, to protection of the environment, to theories of criminal justice and penology. Even when shorn of an obviously religious dimension, public discussions of health care, economic stimulus, and financial regulation seem with increasing frequency to devolve very quickly to bedrock assumptions about the rights of the individual and the role of the state, assumptions on which agreement seems impossible. So advocates on either side of these issues talk increasingly to the like-minded, and the belief that argumentation can be used productively to resolve differences is hollowed out and withers. The difficulty may be more pronounced in the U.S. because of the greater influence of fundamentalism there. Yet from what I read about the immigration issue, the economic integration of the EU, and the question of whether religion has a public role, it seems that Europe is moving in the same direction.

The second assumption also is questionable. If deep disagreement is politically useful, it may affect all who are interested in the policy that is at issue. This has happened in the United States particularly over the past twenty years. The minority party often has seen more advantage in simply opposing the administration in power than in working cooperatively to solve problems.  They have behaved as if the two parties were in a state of deep disagreement, and this produces an impasse in public deliberation. Issues will be unsolved or will be settled by numbers, money, or force, rather than by reasoned discourse.

If anything, this tendency has become more pronounced since the election of Barack Obama. Republicans in the Senate and House of Representatives have voted almost unanimously against most of the president’s initiatives, delaying or obstructing their passage and making it necessary for Obama to make old-fashioned political deals to hold the Democrats together. This may not be a true case of deep disagreement, although it is argued as if it were. When Obama has incorporated into his legislation initiatives that Republicans previously had supported, they have changed stance and voted against them. They have portrayed Obama’s center-left positions as “socialism” and have seen the contest as one between extending the reach of government and protecting the liberty of the people – ostensibly a sharp clash between incompatible world-views. The Obama administration has not been the unique object of such partisan division, although it does seem to be more extensive and systematic than under either George W. Bush or Bill Clinton.

If deep disagreement is prevalent and consequential, then argumentation studies should pay more attention to it. Nearly a decade ago, Nola Heidlebaugh (2001: xi) explored these concerns in depth. As she posed the question, “Without consensus on standards of reason, how can we have good public argument? And without the eloquence and enriched conversation of good public argument, how can we reason together in order to reach consensus on the issues before us?” These questions give argumentation scholars an interest in exploring means to surmount deep disagreement and get deliberation back on a productive track.

3. Incommensurability: end or beginning of analysis?
Heidlebaugh observes that in a case of deep disagreement, the competing positions are incommensurable. They cannot be compared because they do not rely on the same rule-based way of making and legitimizing judgments. But if incommensurability makes further discussion impossible for the logician, she says, for the rhetorician the fun is just beginning. One or more of the arguers must find a way to transcend the deadlock and pursue the argument on another basis. As Heidlebaugh (2001:74) describes it, “the rhetor has to find something to say that will aid in solving a particular problem perceived by the rhetor.” Incommensurability is not something to be “cured” but a situation calling for practical wisdom. The arguer’s task is to discover “a particular vantage point from which new similarities and differences emerge,” because doing so “places value on discovering new things to say” (Heidlebaugh, 2001: 128). Although Heidlebaugh combs the tradition of classical rhetoric and claims that commonplaces, topics, and stasis offer resources for the task of invention, she does not identify particular strategies of transcendence. I would like to do that now, by way of speculation based in experience and in the analysis of case studies.

4. Possibilities for overcoming deep disagreement
I group these possible strategies in pairs under the headings of inconsistency, packaging, time, and changing the ground. Each of these moves reflects the assumption that advancing one’s own claim in an ordinary manner will be unproductive in breaking the impasse because it is not commensurable with the other’s standpoint. One must think in different ways about the clash between standpoints.

4.1. Inconsistency: hypocrisy and the circumstantial ad hominem
The first two moves attempt to get inside the opponent’s frame of reference and discredit it on grounds of inconsistency. They rely on the law of non-contradiction, that a soundly reasoned claim cannot be at odds with itself.

The charge of hypocrisy is that the advocate now maintains a position that is inconsistent with one he or she has maintained previously. In the absence of any explanation for the change, the reasonable implication is that the advocate is being hypocritical and represents only expediency, not principle.

In early 2010, some leading Republicans in the U.S. opposed more government funding to stimulate the economy because it would add to an already large budget deficit and swell the national debt. Many of the same Republicans, however, had voted for even larger deficits during the Bush administration, to support the costs of the war in Iraq or the prescription drug benefit for senior citizens, or as a consequence of tax cuts that were enacted without comparable spending reductions. A Democrat might respond to the Republican complaints about deficit spending as follows:
1. You are bothered by the deficit now.
1. But you were not bothered by it when your party was in power.
2. [There is no apparent explanation for the change in your position.]
3. Therefore you are a hypocrite. Your concern is not with the deficit but just political expediency. You just want to insulate yourself from the Tea Party supporters and to shore up your political base.
4. Therefore your argument is not sustained by any principle and should be rejected.
5. Since your standpoint cannot satisfy the consistency test and your standard is in conflict with mine, my standpoint prevails by the process of elimination.

Not all of these steps will be articulated explicitly, but these are the steps in the move. My standpoint is advanced not by my supporting it with additional reasons but by my demonstration that yours cannot withstand the test of consistency.

Of course, this strategic move is vulnerable. It depends on the unstated assumption that there is no apparent explanation for the change in position. People generally do not knowingly maintain inconsistent positions that will open them to the charge of hypocrisy, so the opponent will work hard to distinguish between the positions. It may be that deficit spending is justified for national security but not for economic stimulus. Or perhaps it is all right if it stimulated the economy by putting more money in individuals’ hands but not if it involves government spending. Or maybe it is acceptable if targeted to senior citizens but not if it supports the general population. Any of these explanations would need support, of course, but the burden of proof would be light precisely because we assume that advocates generally do not advance hypocritical claims.

Related to the charge of hypocrisy is the circumstantial ad hominem. This is not a personal attack on the opponent’s character. Rather, it is an assertion that the adversary’s expressed standpoints are at odds with his or her own behavior in a specific situation. On the commonplace belief that “actions speak louder than words,” the inference is that one’s actions reveal one’s true commitments far more than do one’s words (Walton 1998: 2-6,108-112). So the standpoint fails because it cannot be supported by the arguer’s own actions. Since my standpoint is the alternative to yours, mine prevails, again through residues. Johnstone (1959) has gone so far as to suggest that all valid philosophical argumentation is of this type.

Suppose that A is a lawyer for whom protection of civil liberties is a prominent value. A spoke out against the efforts during the Bush administration to expand the president’s powers in response to terrorism, believing that these measures unduly violated individuals’ rights to privacy. Yet A accepts an invitation to argue before the Supreme Court in defense of those expanded powers when the Obama administration seeks to retain them. “You must not really be committed to civil liberties,” a critic alleges, “when you abandon that commitment for a chance to appear before the Supreme Court to defend President Obama.” A’s actions reveal his true commitment – to the Obama administration – and discredit A’s professed commitment to civil liberties. That position having lost, the alternative position prevails by elimination: A thinks that defense of the nation against terrorists outweighs protection of civil liberties, at least with regard to the case at hand – the hierarchy that A’s interlocutor is trying to discredit.

As in the hypocrisy example, the opponent’s likely response will be to distinguish between the two situations, placing statements and actions on two different planes. He or she might oppose new restrictions on civil liberties and yet maintaining that removal of existing restrictions would convey to other nations the impression that the U.S. was weak. Or the opponent might want to keep the current restrictions because of trust that Obama will use them judiciously and as a last resort, trust that was lacking with respect to President Bush. If the adversary can succeed in distinguishing between the situation in which one made commitments and the situation in which one is called to the test, then the circumstantial ad hominem will lose its force and the perception of deep disagreement will be maintained.  Alternatively, the opponent might claim that he or she is just doing the job of a lawyer, seeing that each client receives the strongest possible defense.

4.2. Packaging: incorporation and subsumption
A second pair of strategies has to do with packaging arguments. One is incorporation, in which an advocate includes incommensurable arguments (and the proposals that accompany them) into a larger package. The success of this strategy depends upon a perception by both advocates that simply perpetuating the impasse is intolerable. Neither advocate is willing to concede but neither is willing to prolong the stalemate. The Obama administration attempted this approach in fashioning its health-care bill, when it incorporated some Republican proposals, such as “tort reform” to curtail lawsuits for malpractice. Obama’s supporters did not concede their own standpoints about the causes of health-care costs – indeed they maintained that “tort reform” would address only a very small part of the problem – but they included some degree of “tort reform” in the bill so that Republicans could act consistently with their professed principles and still support health care reform.

This effort clearly failed, and the failure exposes the difficulty with the strategy of incorporation. Both advocates must desire to overcome the impasse. In this case, passage of health-care legislation was not an important priority for the Republican opponents unless it could be passed on their own terms. Even though tort reform was part of the bill, they did not have enough incentive to swallow other elements of the bill that they found objectionable.  Some actually preferred to vote against the bill while others, noting that the administration wanted desperately to get a bill passed, could hold out to see whether their hard-line stance would yield even more concessions.

Related to incorporation is subsumption, a strategy which seeks to subsume both of the irreconcilable standpoints within a larger frame. One advocate initiates the move, inviting the other to cooperate. The standard form of the argument would be something like this:
6. Our positions X and Y appear to be incommensurable.
7. If you support X, you should support Z because it will advance the cause of X.
8. If I support Y, I should support Z because it will advance the cause of Y.
9. So we can subsume the disagreement about X and Y under our agreement on Z.

The difference between incorporation and subsumption is that incorporation aims only to overcome the impasse in arguments whereas subsumption also aims to develop positive identification with the common term Z.

The abortion controversy offers an interesting example of an attempt at subsumption. The controversy between “pro-life” and “pro-choice” quickly reaches an impasse; the competing standpoints reflect incommensurable world-views and differ on such basic questions as whether we are in control of our own bodies. But arguers may be willing to subsume these differences under the question, How can we best prevent unwanted pregnancies? Both sides have an interest in this question, because it will reduce the circumstances under which the moral dilemma of abortion presents itself.  As a practical matter, it might work.

Then again, the phrase “as a practical matter” is a warning signal. The dispute between “pro-life” and “pro-choice” does not take place on the ground of practicality but as a matter of principle. One can imagine the dispute playing out almost the same way regardless of whether the two sides support a program to reduce unwanted pregnancies. Either side could accept the reduction of unwarranted pregnancies as well and good, taking that benefit off the table, and then immediately revert to its standpoint rooted in incommensurable principles and world-views.

Incorporation and subsumption can be combined. A famous example is the U.S. Senate debate over the Compromise of 1850, originally presented as an omnibus bill to resolve all outstanding disputes over slavery. Incompatible goals were somewhat incorporated into a package, but these individual actions were subsumed under the rubric of finality. Those on either side could see the appeal of settling the controversy, regarding every square inch of U.S. territory, once and for all.  Both political parties committed themselves in their 1852 election platforms to the Compromise of 1850 as the final resolution of the controversy. Yet the compromise was vulnerable. Over time each side could (and ultimately did) think it gave up more than it gained, suffering a raw deal. This is approximately what happened during the years leading to the American Civil War.

4.3. Time: Exhaustion and urgency
The third pair of strategic moves deploy time and timing as a way to break the argumentative impasse. One such move is the appeal to exhaustion. Cases of deep disagreement can remain in an impasse for some time. Eventually, one party may decide that the duration of the controversy has become disproportionate to its importance and try to entice the other to move on. The original disputants may even have passed from the scene, and their successors may be less disposed to carry on the fight. Or time may have passed the controversy by as the consequences of either participant’s position have diminished. Or the impasse may itself become uncomfortable because “life’s too short” to obsess over it. For any of these reasons, one party may try to convince the other that the time has come, not necessarily to resolve the deep disagreement but at least to set it aside and move on.

Something like this attitude motivated the late Israeli Prime Minister Yitzhak Rabin in the early 1990s to make overtures toward peace negotiations with the Palestinians. Bitter enmity over the years had exacted a terrible toll. The Palestinians had not become Israel’s friends, but as Rabin pointedly noted, one does not need to make peace with one’s friends.

Like some of the other moves, the pitfall of this one is that it depends upon a mutual state of exhaustion. The party making the argument must convince the other to feel the same way. Otherwise one arguer may see the other’s appeal to exhaustion as a confession of weakness. If the non-exhausted party will just hold on, the other may lose heart and give up the fight. This is about what happened in the case of the Vietnam war.

More often than appealing to exhaustion, though, advocates will appeal to urgency caused by a crisis in order to get beyond a deep disagreement. The suggestion is that while deep disagreement is a luxury to be tolerated during normal times, we cannot afford it now; time is of the essence and the severity of the situation demands a prompt response.

During the fall of 2008, the U.S. financial system was threatened with implosion, with major repercussions likely around the world. To avert disaster, the Bush administration advocated massive infusions of cash and loan guarantees in order to restore confidence in the U.S. economy. These proposed “bailouts” were castigated by many in Bush’s own party who were convinced of the resilience of an unaided free market. Even President Bush acknowledged that he was uncomfortable with the measures he was proposing and that in normal times he would not suggest them. But the belief that a major crisis was looming required him to set his ideological commitments aside. Not so for many Republicans in the House of Representatives.
Not prepared to accept that the U.S. faced financial meltdown, they initially defeated the proposed bailout. Only when the stock market plunged in response did they reassess their position and pass a modified version of the bailout bill.

Recognizing a state of affairs as a crisis is in the eye of the beholder. If one party holds out and refuses to regard the situation as a crisis, the argument from crisis will be ineffective and perhaps even counterproductive. On the other hand, the perception of a situation as a crisis is a powerful impetus to action. This perhaps is the reason that White House Chief of Staff Rahm Emanuel reportedly said, “never let a crisis go to waste.”

4.4. Changing the ground: Interfield borrowing and frame-shifting
The final pair of moves may be the most ambitious in that they focus on shifting the ground on which the deep disagreement takes place. One such move is what Willard (1983: 267-270) called interfield borrowing. Willard observes that argument fields have distinctive standards of evidence and modes of reasoning, but also observes that many disputes cannot be assigned uniquely to a particular field. Euthanasia, for instance, is both a scientific and a moral issue, but scientists and moralists will be likely to see the question differently. Deep disagreement will result unless one set of advocates is willing – for the sake of the argument – to invoke the other field’s standards for the purpose of defeating the adversary on his own terms. With respect to accounting for human origins, for example, moralists might “borrow” the scientific understanding of evolution and then attempt on scientific grounds to reduce evolution to the status of an unproved theory. Or, conversely, the scientist may take on the persona of a moralist in order to contend that a Biblical account of creation is not at odds with judgments regarding evolution.

The point of “borrowing” from another field is to put both sides of the argument onto the same plane and then to discredit the “other” field on its own terms. But the borrower never will be as knowledgeable as the person who genuinely occupies the field from which the advocate borrows. The second party can find reasons that the borrowing is not genuine or fair, or allege that the borrower has a stereotyped and limited notion of the other party’s field.

The other strategic move related to changing the argumentative ground is frame-shifting, in which one party will seek to move the argument from one context or frame of reference to another. The famous Lincoln-Douglas debates of 1858 offer an interesting example. The central issue was whether it was right or wrong to permit slavery to spread into new territories. Lincoln believed that it was wrong because slavery itself was wrong and it made no sense to say that it was right to expand what was wrong. His standpoint was defended with a substantive moral argument (Zarefsky 1990). But for Douglas the real question was who should decide whether slavery was right or wrong. It was a complex moral question on which good people disagreed, and he did not presume to make the decision for the people who actually would go to the territories and live with the results. Accordingly, he championed “popular sovereignty” and his standpoint was buttressed by a procedural argument. The substantive and procedural positions were incommensurable. This may be why arguments about the morality of extending slavery occupied such a small portion of the debate time. Instead the two candidates disputed about, among other things, what the nation’s founders would have done about the issue if they were alive. The candidates thus shifted the debate from a moral frame to a historical one. Here there could be shared standards, because both men venerated the founding fathers and both believed that their insight could inform present deliberations. And there could be argument, because the question could not be answered conclusively. The founders never were confronted with the question at hand, so one would need to infer their likely position from statements made and actions taken on other topics over the years.

Frame-shifting was helpful to the Lincoln-Douglas debates because both candidates could accept the surrogate frame, each believing that it worked to his advantage. But this is not always the case. The advocate who tries to shift the frame of reference might encounter resistance. For example, Lincoln or Douglas could have insisted that historical speculation was an irrelevant distraction from the issues of the moment. Or the candidates might have experienced deep disagreement about what was the relevant historical evidence or whether it was being understood correctly.

5. Two case studies
It should be noticed that each of these eight strategies for moving beyond deep disagreement is an available option with probative force but that none is assured of success. Like all rhetorical moves, they must be adapted to the particular situation. Sometimes an advocate will be able to show that they fit well and sometimes another advocate will succeed in showing them to be inapplicable. This will be clear from two brief case studies, one a success and the other a failure.

5.1. Johnson on education
In the U.S., elementary and secondary education traditionally has been seen as a responsibility of state and local governments and of the private sector. While there have been some exceptions, such as federal subsidies for schools located near military bases that add to their enrollment, general federal aid to education did not become government policy until the 1960s even though a majority of legislators and of the population supported it. Part of the reason was that supporters were divided on the question of whether federal aid should be extended to religious schools. Some said that to do so would be to dissolve the separation between church and state, creating an establishment of religion in violation of the U.S. Constitution. Were such a provision in the aid to education bill, they would oppose the legislation, even though they supported federal aid to education in principle. But it was no solution simply to keep religious schools out of the bill, because other legislators were convinced that omitting it would be discriminatory, denying equal protection of the laws to those families who sought a religious education for their children. Their tax money would be used to support education but they would be unable to receive the benefit. This, some legislators said, was interference with the free exercise of religion – also a violation of the U.S. Constitution. Meanwhile, the minority who opposed federal aid to education under any circumstances hardly needed to defend their standpoint since supporters of federal aid were in deep disagreement over a subsidiary question.

So matters stood at an impasse until the ascendancy of Lyndon Johnson to the presidency of the United States. Johnson successfully engaged in frame-shifting. He urged that the matter be seen not as aid to either secular or religious schools, but to children (Dallek 1998:197). His proposal involved aid formulas that were based on the number of children in a jurisdiction whose families had incomes below the poverty line. Figuratively, the children would take the aid to whatever school they attended. In practice, schools acted as agents for the children, applying for aid based on their number of qualifying children. This reformulation of the issue, shifting the frame, satisfied both groups who previously were at an impasse. Both sides could view the reformulated proposal as consistent with their strongly held convictions.

5.2. Zarefsky on abortion
My second case study has a less salutary result, particularly since it involves me. Some years ago I produced an audio- and videocourse on argumentation for commercial sale (Zarefsky 2005). In one of the early lectures I made the point that argumentation presumes uncertainty because there is no need to dispute matters that we know for sure. One of my examples was that there was no way to know for sure when human life began; I said that this was a major reason that the abortion controversy was so intractable.

Some time later I received a group of nearly identical letters from several home-schooled teenagers in Minnesota. The letters took strong exception to my statement that there was no way to know when human life began. Of course there is, they replied. Everyone knows that human life begins at conception; it says so in the Bible. They quoted what they thought were applicable Biblical verses. So abortion is murder, they told me. Some people apparently believe that it is acceptable for society and the government to condone murder of the unborn.  That’s why there is a controversy.

I could have ignored these letters, but I wanted to recognize their serious and respectful tone. So I wrote the students back.  I tried interfield borrowing – specifically, to use the Bible, their source of privileged evidence – to argue that the origin of human life was uncertain. I quoted passages from Exodus saying that if a man struck a pregnant woman and she died, the man would be punished for murder. If the woman lived but miscarried, there was a lesser penalty limited to monetary damages. The fetus was valued less than a living person. Here was evidence, I said, that challenged their view that the Bible regarded abortion as murder. My goal, remember, was not to deny their claim outright but only to argue that its status was uncertain, because the point at which human (as distinct from animal) life began was itself uncertain. It seemed like a relatively weak burden of proof and I thought I had shouldered it.

I was surprised when I received a reply not from the students but from their teacher. She thanked me for writing to the students but complained that I was misleading them. Her translation of the Exodus text distinguished between the expulsion of a live fetus and the death of the fetus on the womb. She said that monetary penalties applied in one case but capital punishment was warranted in the other. Since my translation did not make this distinction, she said, it was erroneous if not fraudulent, and for the sake of my own enlightenment I should obtain a better text and recant my heresy. She prayed for my soul. (I note in passing that she did not ask or seem to care what my text was.)

I am not a sophisticated Biblical scholar, but I think the problem here is that the original Hebrew verb is ambiguous with respect to whether the fetus is expelled alive or dead. I have some reason to think that my translation was more authoritative than hers, since it reflects usage conventions at the time the Biblical text was redacted. But all I was trying to establish was that the matter was uncertain and hence a fit and necessary subject for argument.

At this point I abandoned the discussion. My correspondent’s attack on my source without ever knowing what it was suggested to me that her world-view would brook no uncertainty. Counter-evidence would be dismissed in advance so that the argument was self-sealing.  This was a case of fundamentalism vs. modernism. My position depended at its root on uncertainty; hers on certainty; and there seemed no way to bridge the two. My effort at interfield borrowing was unsuccessful because in her view I could not establish my bona fides within her field.

Now perhaps I did the wrong thing. Maybe I should have tried harder, whether by defending my choice of text, or trying to find a passage in her own translation that worked against her claim, or perhaps even looking for different ground than the authority of the Bible. But I thought such efforts would be futile, I had other things to do, and  so I left the discussion agreeing to disagree. I would not change the statement in my lecture that when human life began was uncertain, and she would not abandon her conviction that this statement in my lecture was inaccurate. Remaining at an impasse was a harmless outcome for an interpersonal dialogue between two individuals. As I have suggested above, though, it is not so innocuous when multiplied many times over and when it affects social policy as well as individual judgment.

6.  Conclusion
In models of dialogical argument, the outcomes generally affect only the individual arguers. In models of rhetorical argument, however, there is a third party, an audience that is affected by the exchange. As Schmitt (2010: 10) recently wrote, “The consequences of this apocalyptic rhetoric and all-or-nothing politics fall on the rest of us when government can’t act.” The audience is ill served by continued deep disagreement. Its demand to advance the discussion can put external pressure on the disputants to overcome their impasse. Currently in the U.S., audience dissatisfaction with stalemated political argument is widespread. But it is manifested in an unsophisticated and, in my view, unhelpful way: as largely indiscriminate right-wing populism symbolized by the Tea Party and its demands to “take our government back.” It has unleashed a widespread prejudice against incumbent office-holders and a political discourse in which inexperience is exalted as a virtue. This popular prejudice of the moment stymies efforts to work collaboratively for compromise solutions, because that represents consorting with the enemy. And fear of being accused of such treachery further deepens the sense of fundamental disagreement between the dominant U.S. political parties. But there is a sizeable if underrepresented middle ground consisting of people who also are unhappy with the current impasse but who are unwilling to yield to the oversimplification and further polarization exemplified by Tea Party supporters. They are the ones who must be aroused to demand that our political discourse move past the polarization of deep disagreement to recover the tradition of deliberation through public argument. Some of the strategic moves I’ve discussed here, if skillfully executed, might be means to accomplish that goal.  At least they are places to start.

REFERENCES
Brockriede, W. (1975). Where is argument? Argumentation and advocacy, 9, 179-182.
Dallek, R. (1998). Flawed giant: Lyndon Johnson and his times, 1961-1973. New York: Oxford.
Ehninger, D. (1958). Debating as critical deliberation. Southern communication journal, 24, 22-30.
Fogelin, R.J. (1985). The logic of deep disagreements. Informal logic, 7, 1-8.
Heidlebaugh, N.J. (2001). Judgment, rhetoric, and the problem of incommensurability: Recalling practical wisdom. Columbia: University of South Carolina Press.
Johnstone, H.W., Jr. (1959). Philosophy and argument. University Park: Pennsylvania State University Press.
MacIntyre, A. (1984). After virtue. 2d ed. Notre Dame, Ind.: University of Notre Dame Press.
Perelman, Ch., & Olbrechts-Tyteca, L. (1969). The new rhetoric: A treatise on argumentation. Notre Dame, Ind.: University of Notre Dame Press. (Originally published in French in 1958.)
Schmitt, M. (2010, July/August). Boring politics, please. The American prospect, 21, (6), 10.
Walton, D. (1998). Ad hominem arguments. Tuscaloosa: University of Alabama Press.
Willard, C.A. (1983). Argumentation and the social grounds of knowledge. Tuscaloosa: University of Alabama Press.
Zarefsky, D. (1990). Lincoln, Douglas, and slavery: In the crucible of public debate. Chicago: University of Chicago Press.
Zarefsky, D. (2005). Argumentation: The study of effective reasoning. Chantilly, Va.: The Teaching Company. [Audio and video].

1. Lessons on Teaching Argumentation from Science Education[i]
Teaching argumentation has an obvious entry point in most educational systems through science courses and teaching science. As editors of a recent edited volume summarize: “ … there is an increasing emphasis on resting the science curriculum on a more appropriate balance between science process and citizenship skills, and factual or content knowledge of science. The main rationale for the inclusion of argumentation in the science curriculum has been twofold. First, there is the need to educate for informed citizenship where science is related to its social, economic, cultural and political roots. Second, the reliance on evidence has been problematised and linked in the context of scientific processes such as investigation, inquiries and practical work.” (Erduran and Jiménez-Aleixandre 2008, p. 19). These curricular reforms – most often connected to NOS (Nature of Science) or SSI (Socio-Scientific Issues), and CT (Critical Thinking) discussions in science education – recognize the need for the explicit teaching of argumentation, and the importance of developing students’ existing argumentative skills.

The curricular reforms, however, have rarely born the fruits that supporters and enthusiasts have expected, and that curricular descriptions demand. The results so far are somewhat discouraging with respect to NOS, SSI, and CT, and to the more general argumentative skills. They show that effective teaching of argumentation in science classes is not without difficulties: “Only a minority of people progress to the final, evaluative epistemology, in which all opinions are not equal and knowing is understood as a process that entails judgment, evaluation and argument.” (Zohar 2008, p. 256). One can argue that the curricular expectations are set too high, and do not take the cognitive development of students fully into account. Setting realistic desiderata, however, runs into methodological difficulties. The fact that the results of high-achievers is more informative of the one end of the ability spectrum than the result of weak students (Voss, Segal, and Perkins 1991) is one of the problems that need to be addressed. At present it appears that: “Some desiderata concerning epistemological understanding are never reached by a large percentage of students. This is a serious problem that most curriculum-development has to face and tackle.” (Garcia-Mila and Andersen 2008, p. 39). But whether the cognitive constraints of the students or the didactical ineffectiveness of the educational system is the (main) culprit for the rather disappointing results, is hard to tell. Didactics can surely improve, as, despite the efforts at the level of international policies about the science curriculum, “the systematic uptake of argumentation work in everyday science classrooms remains minimal” (Erduran and Jiménez-Aleixandre 2008, p. 20).

Didactic effectiveness is affected by many factors, starting from the theoretical frameworks used in science classroom, through the approaches utilized to reconcile the critical attitude with the authoritative image of science (Donnelly 2002, Zemplén 2007), to the management of the group dynamics in the classroom. Not providing an exhaustive list, this last mentioned aspect needs to be highlighted, as research indicates that: “Arguments by peers may be accepted more easily or defended more robustly according to group dynamics –the impact of social relationships within a group can have a bearing on the course of the argument” (Kolsto and Ratcliffe 2008, p. 123).

Recognizing the importance of social relationships implies that for optimally effective didactic interventions the teachers need to actively seek didactical settings that enhance the desired argumentative performances. The setting needs to take into account – among others – that there is an optimum of emotional involvement on the side of the students: too little of it precludes commitment and defence of positions, while too much of it is detrimental to the argumentative performance. Also, the teacher’s role has to depart from the traditionally authoritative image associated with science teachers; otherwise the students can easily assume that the teacher is not willing to change his/her position. (In this case why should they take part in argumentation?)

Understanding that many factors influence the success-rate of teaching argumentation, science educators have been focusing on framing the didactical situation in ways that are conducive to developing argumentation-related skills (see e.g. Adúriz-Bravo et al. 2005). As an example, in the HIPST project (an European 7th Framework funded science education project the authors participated in) a special spatial allocation of reflective thinking in the classroom was proposed; in this “reflection corner” the students could make statements about science and the scientific method that could be challenged and debated in class. While these situational framing effects might be seen as lying outside the territory of argumentation theory, they clearly affect the argumentative performance: to what extent are students willing to take part in argumentative activity in the first place, to what extent do they utilize their already acquired argumentative skills, and to what extent are they learning how to change their positions as well as argumentative practices reasonably. Situational framing is, therefore, a key to successful teaching of argumentation, as, without creating the perception in the students that they are in a situation where (rational) argumentation is the right behavioural response, they will not even start to argue. Framing situations in certain ways is also important for maintaining the preferred attitude.

In most cases the framing is carried out via linguistic means. The teacher has to say utterances that have a specific regulative function with respect to the pragmatic situation: the students should engage in and continue with the argumentative activity, and not end up making jokes, start a fight, etc. This aspect of framing is linguistic, and has relevance for theories of argumentation, as we show below. We start by investigating the so-called appeal-framing and discuss its treatment in one specific theory of argumentation, the extended pragma-dialectical framework.

2. Can linguistic framing be normatively dubious?
In a recent article Daniel O’Keefe (2007) raised interesting questions concerning the relationship of argumentation studies with persuasion effect studies in psychology and elsewhere. He draws attention to cases, where arguers are using appeal framing; in these instances different “ways of expressing an appeal involve the same underlying substantive consideration” (O’Keefe 2007, p. 154). It is an established fact in social psychology that the different formulations of logically isomorphic contents might have a causal influence on the mind of the recipient beyond the causal effect of the information given. This extra persuasive effect of the speech act comes from the presentational device used and might affect the evaluation in specific directions (consider: Kahneman & Tversky 1986). Taking an example from O’Keefe, a medical expert might describe identical situations in various ways (O’Keefe 2007, p. 153, 155):

(a1) – success rate framing – “this surgical procedure has 90% survival rate”
(a2) – failure rate framing   – “this surgical procedure has 10% mortality rate”

It is reasonable to expect, that if two utterances have the same informational content then people will react with the same decision. No matter if a1 or a2 is presented, the reaction will be the same. But this expectation is false. We know from social psychology that recipients will more probably answer with an affirmative decision to a1 than to a2. This means that the decision is not only conditional upon the informational content.

The question for O’Keefe is whether we are normatively indifferent with respect to the choice of presentational formats or not. As he writes: “the common intuition would be that there is something wrong with knowingly and purposefully choosing one or another of these formulations” (O’Keefe 2007, p. 156). The reason identified behind this common intuition is that people are usually unaware of the fact that their choices are influenced by the way the information presented was framed. The use of appeal framing can therefore be regarded as manipulative. O’Keefe adds: “This way of putting things makes appeal framing look rather like a fallacy, at least in some traditional ways of thinking about fallacies. A long-standing characteristic worry about fallacies is that they lead an unsuspecting audience to be influenced in ways it otherwise would not have been.” (O’Keefe 2007, p. 157)

We think O’Kefee has a good point. And we also agree with him that from the point of view of argumentation theory, normative pragmatics or pragma-dialectics it is not easy to see for these cases what the problem would be with using this or that presentational format, or how the use of a framing device could generally be normatively dubious. In the pragma-dialectical theory, for example, if both a1 and a2 are uttered in the course of an argumentative exchange, then the analytical overview collapses these distinctions (due to their logical equivalence[ii]). There is, however, massive empirical basis for claiming that certain formulations that are logically seen as equivalent are in fact influencing participants in various ways, in situations where this difference in persuasiveness can result in radically different decisions. People trying to be reasonable arguers, when in need of making e.g. medical decisions, are more or less likely to accept a specific position depending on the appeal framing[iii].

As such, these cases may constitute anomalies (in a strong, Kuhnian sense) for certain normative theories of argumentation when rhetorical perspectives are incorporated into them. For this reason we now look at the possibility of finding a place for these framing effects in the notion of “strategic maneuvering”, as it has been used in the extended pragma-dialectical theory to unite dialectical and rhetorical insights. There are certainly other respectable and insightful accounts of argumentative discourse, but at present pragma-dialectics appears to be the most systematized and developed research program. Furthermore, the pragma-dialectical method of argument reconstruction is in accordance with the received logic-based accounts of critical thinking that prevail in contemporary approaches to science education, and the method has comparatively clear standards for both reconstruction, and (normative) analysis.

But, although we think that pragma-dialectics is a suitable framework to unfold the fruitful implications of the problems posed earlier, there is some conceptual work to be done before we can turn our full attention to reformulate our problem as the problem of effectively using presentational devices in a rhetorical and in a dialectical sense.

3. Strategic maneuvering, derailments, and appeal framing
In the last decade pragma-dialecticians have worked on incorporating rhetorical insights into their framework under the name of strategic maneuvering (henceforth SM). As they formulated: “The gap between dialectic and rhetoric can in our view be bridged by introducing the theoretical notion of ‘strategic maneuvering’ to do justice to the fact that engaging in argumentative discourse always means being at the same time out for critical reasonableness and artful effectiveness. […] strategic maneuvering refers to the continual efforts made in principle by all parties in argumentative discourse to reconcile their simultaneous pursuit of rhetorical aims of effectiveness with maintaining dialectical standards of reasonableness” (Eemeren and Houtlosser 2009, p. 4-5). According to the latest exposition (Eemeren 2010) the analysis of strategic maneuvers divides the rhetorical dimension into three inseparable aspects that are mutually attuned to each other: topical choices, adjustments to audience demand and presentational choices.

The presentational device aspect was earlier described as “the phrasing of moves in light of their discursive and stylistic effectiveness” (Eemeren and Houtlosser 2001, p. 152), and in the contemporary version this aspect is seen as ‘framing’. In Eemeren’s view “exploiting the possibilities of presentational variation in strategic maneuvering […] boils down […] to ‘framing’ one’s argumentative moves in a communicatively and interactionally functional way” (Eemeren 2010, p. 117). Although we posed a problem in the context of social psychology, as the presentational device aspect of strategic maneuvering ‘boils down’ to framing moves, incorporating insights from social psychology can contribute to the understanding (and possibly also to the normative regulation) of the presentational device aspect of the new pragma-dialectical framework[iv].

Let us return to the question whether the use of appeal framing (a kind of presentational device) is normatively problematic in the pragma-dialectical framework. In this theory a group of norms limit strategic maneuvers. No maneuver is allowed to violate the so called first order conditions, the (ten) dialectical rules worked out in the pragma-dialectical theory, presupposed as necessary for any reasonable discussion (Eemeren & Grootendorst 2004, p. 187-195). The extended pragma-dialectical theory also accounts for constraints linked to the specific institutional context (e.g. extra discussion rules in the court room) where the discussion takes place, but these are taken as specifications of the general first order rules (Eemeren 2010, p. 197). If a strategic maneuver does not comply with the first order rules, then it is classified as a derailment, and is normatively objectionable (fallacious). If it follows the track marked by these rules then it is a sound strategic maneuver.

On the one hand, it is hard to see how the appeal-framing scenarios we discuss could violate any of the first order conditions for a critical discussion[v]. On the other hand it is easy to imagine cases where the use of appeal framing fits the following loose definition of derailment, which states that “If a party allows his commitment to a critical exchange of argumentative moves to be overruled by the aim of persuading the opponent […] we say that the strategic maneuvering has got ‘derailed'” (Eemeren and Houtlosser 2009, p. 13). The reason behind these derailments is that people “also and perhaps even primarily [are] interested in resolving the difference of opinion effectively in favor of their case, i.e. in agreement with their own standpoint or the position of those they represent.” (Eemeren 2010, p. 39). So, a derailment occurs when the attempt to reconcile the two, in part, contradictory goals of arguers is unsuccessful, that is the “rhetorical aim has gained the upper hand at the expense of achieving the dialectical goal” (Eemeren and Houtlosser 2009, p. 5). If the cases of appeal framing we discussed can constitute derailments in SM, we have examples that raise interesting normative questions but which are not treated in the detailed exposition of strategic maneuvering.

How can we know whether there are cases of appeal framing where “strategic maneuvering has got ‘derailed'” in the above sense? In certain contexts the argumentative use of the kind of appeal framing discussed earlier can be considered as manipulative. (In such cases in the eyes of a pragma-dialectician the other party “allows his commitment to a critical exchange of argumentative moves to be overruled by the aim of persuading the opponent.”) Social psychology also knows of many cases where moves considered as manipulative produce a boomerang-effect[vi] as people act to protect their sense of freedom (Kruglanski & Higgins 2007, p. 267)[vii]. It is therefore possible that a party quits the kind of argumentative discourse preferred by the pragma-dialectical theory because this party identifies a move, a presentation device used by the other party as manipulative. A critical discussion can derail without violating the first order rules, as certain behavioural responses block the parties from reaching the dialectical aim of the discussion.

The discussed framing effects are achieved by presentational devices, but their contribution to reaching or not reaching the dialectical aims cannot be subjected to evaluation in the extended pragma-dialectical theory. Although they have a place in the analytic overview, the presentational devices used in a discourse can be effective or ineffective means of persuasion, but cannot be evaluated normatively. We think that this fact conjoined with the possible behavioural responses to framing raise interesting and possibly fruitful questions for the pragma-dialectical theory. How should we treat moves that can obstruct the dialectical aims, when the best current theory does not account for such obstructions? Or, if the uses of appeal-framing are not regulated by any norms in the pragma-dialectical theory, how can we say that they derail the SM?

4. From second order conditions to dialectical effectiveness
The pragma-dialectical theory has resources to overcome this problem. In his new book van Eemeren devotes a concise section to the so called higher order conditions of a critical discussion: “in order for people to be willing and ready and to have the opportunity for concluding a critical discussion, certain further prerequisites need to be fulfilled” (Eemeren 2010, p. 35). Parts of these prerequisites for a reasonable discussion are psychological, second order conditions. If these are not satisfied, then critical reasonableness cannot be fully realized in practice (Eemeren & Grootendorst 2004, p. 189). There is, however, no detailed discussion of these conditions, only their limited controllability is stressed: “Sometimes there are factors beyond the control of the arguers that hinder the adoption of the reasonable attitude toward discussion assumed in the code of behavior.” (Eemeren & Grootendorst 2004, p. 36). And: “To some extent, everyone who wants to satisfy the second-order conditions can do so, but in practice, people’s freedom is sometimes more or less severely limited by psychological factors that are beyond their control, such as emotional restraint and personal pressure.” (Eemeren & Grootendorst 2004, p. 189).

This suggests that there is a second way to hamper the realization of the dialectical goals, distinct from violating any of the first order discussion rules by committing fallacies. Second order conditions can be influenced negatively by presentation techniques without the violation of first order discussion rules, and therefore there is room for the discussed framing examples in the pragma-dialectical theory. Considering this, and in line with the loose formulation of derailment we have quoted in the previous section we suggest some terminological clarification.

In cases where a move is not fallacious (i.e. no first order rule is violated) but results in an uncooperative behaviour of the other party (i.e. second order conditions are violated) we believe that it is sensible to classify these moves as derailments. Pragma-dialectical theory currently treats ‘fallacy’ and ‘derailment’ as co-referent[viii], but some of the definitional attempts suggest that derailment could be used for any move that hampers the full realization of critical reasonableness. Second order conditions currently play a marginal role in most discussions of the theory, even though their violations can also derail conversations.

This terminological differentiation has interesting consequences. During sound strategic maneuvering the parties want to realize their dialectical objectives to the best advantage of the position they have adopted. Strategic maneuvering that achieves the speaker’s rhetorical aim of winning without violating the dialectical standards of reasonableness is effective in reaching these aims.

Effectiveness can also be understood in a different sense when the autonomous causal effects that rhetorical devices can have on second-order conditions are investigated. If the cooperative behaviour of discussants is maintained then the use of presentational devices was effective with respect to the dialectical aims (i.e. maintaining the dialectical standards of reasonableness). As the aim of rational discussion in pragma-dialectical terms is the resolution of the difference of opinion on the merits, moves that hinder this aim are considered derailments. If, for example, a boomerang-effect occurs, an analyst can conclude (and in fact a participator often does conclude) that a specific speech act derailed the critical discussion. Those moves are effective in reaching the dialectical aims that do not hinder the resolution process. This sense of effective communication is a prerequisite of critical discussions.

As presentational devices (and the rhetorical dimension in general) can be used effectively (or not) in both senses, we will distinguish them as “rhetorical effectiveness” Er and “dialectical effectiveness” Ed[ix]:

Er: Effectiveness in the sense that the utterances of a party serve the advantage of the position held by that party (helps the party to win).
Ed: Effectiveness in the sense that the utterances of a party facilitate cooperative behaviour that is in line with the dialectical aims of the discussion.

As we have seen dialectical effectiveness (Ed) is conditional upon the limited controllability of the psychological processes or second order conditions. Nevertheless, a derailment-free discussion of the parties need not only follow the first-order rules, but also has to be dialectically effective.

5. Critical rationalism and the epistemic and didactical significance of dialectical effectiveness
The analyst can concentrate on either the rhetorical or the dialectical effectiveness of the presentational devices used when analyzing the rhetorical dimension of argumentative discourse. Certain types of argumentative discussions may provide reasons for focusing on either Er on Er. If the analyst believes that the arguers “in their assiduity to win the other party over to their side” (Eemeren & Houtlosser 2002, p. 142) neglect commitment to the critical ideal, then this can support the analytical decision to focus on Er and disregard Ed.
When arguers prioritize winning over the dialectical aims, as in forensic debates, the search for rhetorical effectiveness dominates the argumentative activity types. In other activity types (like rational/critical discussion) dialectical efficiency (Ed) is prioritized over winning (Er).

This distinction can be used to delineate argumentative activity types and can also come handy for those who believe that critical discussions can be epistemically valuable. Commitment to various epistemological positions, including critical rationalism (which gave rise to pragma-dialectics in the first place) entails commitment to dialectical effectiveness, and implies using moves that have a specific function. The function is to maintain the second order conditions necessary for achieving the dialectical aims. If a move has a specific function in an argumentative discourse, then it can be considered as part of the argumentation[x].

The extended pragma-dialectical approach functionalizes the rhetorical dimension independently of the resolution-oriented dialectical goals. What we tried to prove above is that there are dialectically important rhetorical aspects of discussions that cannot be evaluated in this framework. The functionalization of rhetoric in the new theory yielded mixed results. Rhetorical aspects could be seen as a) violating or being in conflict with dialectical rules or as b) sound maneuvers of the resolution-process (the actualization of a potentiality, in quaint parlance). Many dialectically functional audience- and persuasion-oriented aspects are left out if we draw the boundaries of evaluation here. Dialectical effectiveness pertains to second order conditions that need to be satisfied for critical reasonableness to be realized. A communicative move can be seen as (psychological) facilitator or hinderer of the resolution process. The function of the rhetorical aspect of such a move is to maintain the second order conditions. This rhetorical aspect of communicative moves can be evaluated through the notion of dialectical effectiveness in our view.

Let us try to unfold a scenario where such rhetorical aspects might be of significance. In the intellectual climate of the 17th century, scientists living in different countries who differed as regards religious, political and personal outlook, often openly professed to differences of opinion. They made contradictory claims about data (simple measurements), about the validity of inferences (whether a proposition has been demonstrated or not) and about the scientific method. Is it natural to assume that these people from different countries maintain a critical discussion over years and request copies of each others’ letters in case one is lost? What maintains the second order conditions of the participants in the debate? In one concrete example of the this scenario Isaac Newton writes several pages, full of precise descriptions of his different prisms, different measurements of image-lengths in different atmospheric conditions. This is the most detailed data about spectra (and prisms) available at the time, and therefore has scientific significance. From a rhetorical point of view the ethos of the meticulous observer Newton is established on these pages. From a pragma-dialectical point of view what function is assigned to these pages? To respond to four lines of a previous letter by Anthony Lucas, a Jesuit living in Liége? The answer is unnecessarily detailed, disproportionately long for the meaningful function we can ascribe to it and potentially irrelevant as a wider readership (and not Lucas) is addressed explicitly as audience (Zemplén 2008, p. 264.). But pages like these play an important role in maintaining second-order conditions.

Meticulous observers become trusted observers, and social historians have a host of other examples that these detailed descriptions functioned as trust-enhancing devices in the community of intellectuals in Early Modernity. This building up of trust is seen as a major impetus for the scientific revolution (Shapin 1994), and is also present in contemporary knowledge-production in many institutionalized forms. The ethos of the speaker therefore influences dialectical effectiveness.

A certain amount of trust is necessary for a critical exchange, and some aspect of this trust can be translated as the willingness of the discussant to entertain his fallibility. Entertaining fallibility can be conditional upon the trust in the knower. If I believe that my expressed opinion is the rationally most acceptable position available then I trust myself as a knower. If I believe this to characterize someone else’s position then I trust that person’s position on the issue. Fallibility in this sense is the measure of distrust towards a knower’s position. It is a prerequisite of critical discussion that the parties have some distrust towards themselves as knowers, and have some trust towards the other party as knower. Idealized models of symmetrical rational debate usually presuppose that the trust that positions receive is not affected by the trust in the proponent of that position as knower. One property of this debate-type is that if differences of opinion emerge then the models base the resolution-process on the consideration of the merits of argument. A critical rationalist in our view prefers this process to others and accepts that the trust in the proponent of a position as a knower itself has to be decided on the merits of argument if differences of opinion emerge with respect to this[xi].

Maintaining dialectical effectiveness in the process of argumentation is one behavioural property of (ideal) critical debaters, and so an ideal critical debater is dialectically effective. Dialectical effectiveness is furthermore required to realize dialectical goals, as we can only talk of a derailment-free resolution process of a critical discussion if dialectical effectiveness is a property of that discussion.

An example discussed earlier can be used to illustrate this point. A teacher teaching argumentation ‘has to say utterances that have a specific regulative function with respect to the pragmatic situation: the students should continue with the argumentative activity, and not end up making jokes, start a fight, etc.’ If students do not engage in critical discussion or break up the discussion, due to the peer pressure they experience then the teacher is not dialectically effective. This lack of dialectical effectiveness also implies that the dialectical goals have not been met, a characteristic of didactical interventions that teachers of argumentation try to avoid. But is anyone responsible for this dialectical ineffectiveness? Does a critical discussant have dialectical responsibility?

6. Dialectical responsibility and the didactical challenges of training critical discussants
Responsibility implies freedom of choice. Dialectical responsibility emerges when a party aims to be dialectically effective and is able to choose dialectically effective moves. To the extent that dialectical effectiveness of moves can be calculated such a party is responsible to pick dialectically effective moves. Dialectical effectiveness of the parties is a prerequisite to a critical discussion, and is therefore a key element of successful teaching of argumentative skills. Pragma-dialecticians appear to say something similar when they state that: “The fulfillment of the second-order conditions can be promoted by good training” (Eemeren & Grootendorst 2004, p. 37). Their didactic advice, however, is needlessly limited in our view. This training should encourage “reflection on the aims and merits of argumentation” as “compliance with second-order conditions can to some extent be stimulated by education that is methodically directed at reflection on the first-order rules and understanding their rationale.” (Eemeren & Grootendorst 2004, p. 37, 189)

As we argued, complience with “second order conditions” is conditional upon many factors. Reflecting on first-order rules and understanding their rationale prepares the arguers to use non-fallacious moves. But derailment-free argumentative activity also implies that the arguers are dialectically effective, their utterances facilitate cooperative behavior that is in line with the dialectical aims of the discussion.

If we think of any teaching situation, any kind of didactic intervention, where the aim is to develop skills for critical discussion, we can think of many ways to increase the dialectical effectiveness of the parties. In the opening section of this paper we mentioned a number of factors that influence argumentative performance in a school-setting. Group dynamics, optimal emotional involvement, and instructional strategy all influence the success of developing argumentation-related skills. These factors all have something to do with second order conditions, therefore compliance with second-order conditions can be stimulated in many ways.

In practical terms, this means that optimal learning (or testing) environments need to be designed to increase the dialectical effectiveness of the parties. As good arguers – or, more specifically, good critical discussants – are expected to behave in certain ways, some behavioural cues can be used to judge certain didactical settings preferable to others. If changes of opinion are seen as one such behavioral cue (as is generally assumed in science education), then didactical settings that induce this behavior are valuable in teaching critical discussion. If, for example, a researcher finds that discussion of issues in role-play “was the first [of all the studies we have conducted so far] in which changes of opinion were observed” (Simonneaux, 2008, p. 185), we can use this as an argument for designing learning environments that scaffold argumentative performance using role-plays. The teacher, in this example, seeks to create an environment conducive to (developing skills for) critical discussion within a classroom with the use of specific instructions.

Much of the empirical knowledge of social psychology can be used to improve the dialectical effectiveness. And remedies can likely be offered to common derailments that result in dialectical ineffectiveness. If the earlier discussed boomerang-effect is likely to deem the parties (and therefore the situation) dialectically ineffective, practical suggestions to decrease the likelihood of the boomerang-effect taking place are conducive to dialectical effectiveness. But this kind of knowledge comes with responsibilities. If a critical discussant has even limited / partial knowledge about the dialectical effectiveness of various communicative moves, then he has a responsibility to choose the dialectically more effective move. This move facilitates more / hinders less the resolution process that is the preferred epistemic route for a critical rationalist, so achieving dialectical effectiveness is a dialectical responsibility of critical rationalists. This perspective suggests that much work is to be done.

7. Conclusion
In this paper we showed that incorporating insights from social psychology can contribute to the understanding (and to the normative regulation) of the presentational device aspect of the new pragma-dialectical framework. During this investigation we developed the notion of dialectical effectiveness. Dialectically effective utterances of a party facilitate cooperative behaviour that is in line with the dialectical aims of the discussion. And any discussion that achieves these dialectical aims is also dialectically effective. This perspective opened up a position where the aim of a critical rationalist discussant matches the goal set for the critical discussion. This connection was used to introduce the notion of dialectical responsibility, and thus allowed the formulation of critical rationalist responsibilities with respect to the dialectical aims. We argued that these responsibilities stretch well beyond conforming to first order rules, and imply that for the successful training of critical discussants significant preparation may be required to maintain dialectical effectiveness.

NOTES
[i] The authors thank Jean H.M.Wagemans for fruitful discussions, Gábor Kutrovátz for commenting on the manuscript, and the anonymous reviewers for their helpful suggestions. The financial support from the HIPST project, the OTKA K 72598 grant, and the Bolyai postdoctoral scholarship (G.Z) is hereby acknowledged.
[ii] The ‘Reflective Judgment Model, for example, suggests that the reasoning skills of high school students do not display the ability to contrast evidence from different sources, to explicate criteria for decision making, etc (King and Kitchener 1994). More recent research reinforces this, and strong arguments have been made that inquiry and argument are the central skills of science education (Kuhn 2005).
[iii] This clearly holds for the standard version of the pragma-dialectical theory. For the extended one, the case is more complex. In this framework the reconstruction of utterance a1 and a2 is still isomorphic from a dialectical point of view. From a rhetorical perspective, however, it is not, as there is a difference in the persuasive effectiveness of the utterances. The extended pragma-dialectical approach functionalizes the rhetorical dimension independently of the resolution-oriented dialectical goals. As for strategic maneuverers the only limit for using rhetorical means is given in the pragma-dialectical norms, the normative evaluation of the rhetorical aspect of communicative moves remains a problem, as, to return to the point raised earlier, there is something normatively dubious in choosing this or that framing of the same content. What we are interested in is the conceptualization of this observation in the extended pragma-dialectical framework.
[iv] This can lead to moral issues in certain scenarios: if a doctor has any kind of interest in treating certain patients and not treating others, and knows how framing influences the response, then the doctor can influence the likelihood the patient opts for or rejects a certain treatment.
[v] It is to be noted here, that the sense in which Eemeren uses the term framing is narrower than as we use it, and basically refers to phenomena traditionally studied in stylistics. He divides the domain of presentational variation into two sub domains. Variations are possible in the language register and in the semantic dimension. We acknowledge that those kinds of framing effects that are highlighted in O’Keefe, are not obviously incorporated into this approach, but it is also true that there are no reasons for not to incorporate them either.
[vi] Here a proponent of the pragma-dialectical theory might cast some doubts and suggest that such a maneuver might be handled under by the 10th rule of dialectics, the norm that regulates language use. In (Eemeren & Grootendorst 2004, p. 195) the rule states that “discussants may not use any formulations that are insufficiently clear or confusingly ambiguous”. We think that as in examples like ours the informational content is sufficiently clear, this norm is insufficient to handle the problem.
[vii] In our view the boomerang-effect is a possible perlocution of the communicative move in the case of appeal framing, although there are no externaliseable commitments of the speech act performed that might contradict the pragma-dialectical norms. Similar effects, in this paper subsumed under the term “framing”, could suggest to some that the meta-theoretical principle of externalization put forward by the pragma-dialectical approach when studying argumentative statements is given up, and internalized positions are taken into account. This need not be the case. The contribution of social psychology depends on the extent that knowledge of this field can be utilized for scholars of argumentation-theory. Our contribution aims at finding room to incorporate novel kind of data into theories of argumentation, and not to develop in detail our position on externalization. The incorporation of empirical data from social psychology into models of argumentation requires further discussion not undertaken here.
[viii] One relevant argument that can be raised against such uses of experimental findings boils down to the general problem of extending generalizations that are invariant under certain interventions in the laboratory. Especially in the special sciences invariant regularities between variables are usually invariant only for certain values of the variables and for certain background conditions the careful investigation of which is carried out only when a research program starts to grow (see: Woodward 2003). So, further empirical support is likely to be acquired as empirical research informed by both social psychology and argumentation theory keeps growing in quantity and significance.
[ix] “All derailments of strategic maneuvering are fallacies in the sense that they violate one or more of the rules for critical discussion and all fallacies can be viewed as derailments of strategic maneuvering.” (Eemeren 2010, 198)
[x] In analytical philosophy, the term “dialectical effectiveness” (also referred to as “dialectical power”) is used differently: an argument is dialectically effective if it presents the audience with a piece of reasoning they can rationally accept. Our use discussed in Section 5. is not related to the epistemic validity of arguments, just as our use of “rhetorical effectiveness” is not related to certain rhetorical traditions using this term.
[xi] According to the functionalization principle of pragma-dialectics “an adequate description and evaluation of argumentation can only be given if the purpose for which the argumentation is put forward in the interaction is duly taken into account” (Eemeren & Grootendorst 1995, p. 133).
[xii] Consider also Lumer (2010), who argues that: „as long as the feature of argumentation that makes of it a dialectical activity, namely, its recursivity, is the warrant of its legitimacy as a persuasive device, dialectical conditions will happen to be regulative for any piece of discourse as a persuasive device. Finally, I also want to underline that, as a consequence of their recursivity, dialectical procedures are also tools for the evaluation of acts of arguing. Remarkably, on this account, such dialectical procedures amount to nothing but further argumentation.”

REFERENCES
Adúriz-Bravo, A., Bonan, L., Galli, L. G., Chion, A. R. & Meinardi, E. (2005). Scientific Argumentation in Pre-Service Biology Teacher Education. Eurasia Journal of Mathematics, Science and Technology Education 1 (1), 76-83.
Donnelly, J. (2002). Instrumentality, Hermeneutics and the Place of Science in the School Curriculum. Science & Education 11 (2), 135-153.
Eemeren, F. H. van (2010). Strategic Maneuvering in Argumentative Discourse. Amsterdam: John Benjamins
Eemeren, F. H. van & Houtlosser, P. (2009). Strategic maneuvering. In F. H. van Eemeren (Ed.), Examining Argumentation in Context: Fifteen studies on strategic maneuvering (pp. 1-23), Amsterdam: John Benjamins
Eemeren, F. H. van & Grootendorst, R. (2004). A systematic theory of argumentation. The pragma-dialectical approach. Cambridge University Press
Eemeren, F. H. van, & Houtlosser, P. (2001). Managing disagreement: rhetorical analysis within a dialectical framework. Argumentation and Advocacy 37, 150-157.
Eemeren, F. H. van, & Houtlosser, P. (2002). Strategic Maneuvering. Maintaining a Delicate Balance. In F. H. van Eemeren & P. Houtlosser (Eds.), Dialectic and Rhetoric. The Warp and Woof of Argumentation Analysis (pp. 131-159), Dordrecht: Kluwer Academic Publishers.
Eemeren, F. H. van & Grootendorst, R. (1995). The Pragma-Dialectical Approach to Fallacies. In H. V. Hansen & R. C. Pinto (Eds.), Fallacies: Classical and Contemporary Readings (pp. 130-145), The Pennsylvania State University Press
Garcia-Mila, M. & Andersen, Ch. (2008). Cognitive Foundations of Learning Argumentation. In S. Erduran and M. P. Jiménez-Aleixandre (Eds.), Argumentation in Science Education: Perspectives from Classroom-Based Research, Dordrecht: Springer Publishers.
King, P. M., & Kitchener, K. S. (1994). Developing Reflective Judgment. San Francisco: Jossey Bass.
Kolsto, S. D. & Ratcliffe, M. (2008). Social Aspects of Argumentation. In S. Erduran and M. P. Jiménez-Aleixandre (Eds.), Argumentation in Science Education: Perspectives from Classroom-Based Research, Dordrecht: Springer Publishers.
Kruglanski, A. & Higgins, E.T. (2007). Social Psychology: Handbook of Basic Principles (2nd edition). New York: Guilford Press
Kuhn, D. (2005). Education for Thinking. Cambridge: Harvard University Press.
Lumer, C. (2010). Pragma-dialectics and the function of argumentation. Argumentation, 24(1), 41-69.
O’Keefe, D. J. (2007). Potential conflicts between normatively-responsible advocacy and successful social influence: Evidence from persuasion effects research. Argumentation 21, 151-163.
Shapin, S. (1994). A social history of truth: civility and science in seventeenth-century England, Science and its conceptual foundations. Chicago: University of Chicago Press.
Simonneaux, L. (2008). Argumentation in Socio-Scientific Contexts. In S. Erduran and M. P. Jiménez-Aleixandre (Eds.), Argumentation in Science Education: Perspectives from Classroom-Based Research (pp. 179-199), Dordrecht: Springer Publishers.
Tversky, A. & Kahneman, D. (1986). Rational choice and the framing of decisions. Journal of Business 59, 251-278.
Voss, J. F., Segal, J. W. & D. N. Perkins. (1991). Informal reasoning and education. Hillsdale, N.J.: Lawrence Erlbaum Associates.
Woodward, J. (2003). Making Things Happen: a Theory of Causal Explanation. Oxford
Zemplén, G. Á. (2007). Conflicting Agendas: Critical Thinking versus Science Education in the International Baccalaureate Theory of Knowledge Course. Science and Education 16 (2), 167–196.
Zemplén, G. Á. (2008). Scientific controversies and the pragma-dialectical model: Analysing a case study from the 1670s, the published part of the Newton-Lucas correspondence. In F.H. van Eemeren & B. Garssen (Eds.), Controversy and Confrontation – Relating controversy analysis with argumentation theory (pp. 249-273), Amsterdam: John Benjamins.
Zohar, A. (2008). Science Teacher Education and Professional Development in Argumentation. In S. Erduran and M. P. Jiménez-Aleixandre (Eds.), Argumentation in Science Education: Perspectives from Classroom-Based Research, Dordrecht: Springer Publishers.

1. Introduction
This paper forwards the (presumably controversial) thesis that the use-value of empirically studying the conventional validity of the pragma-dialectical discussion rules (van Eemeren & Grootendorst 2004: 190-196) is heuristic. This thesis seems natural (to me), if the consequences of a particular theoretical commitment are appreciated: When treating argumentation that supports a descriptive standpoint with a normative premise (aka. a “value sentence”), and vice versa, pragma-dialecticians incur a commitment on the transition between “ought” and “is.” This commitment amounts to embracing the “naturalistic fallacy” as a discussion move that is never appropriate.

In Section 2.1, the aim, method and main result of the recent empirical investigation of van Eemeren, Garssen and Meuffels (2009) are presented. In Section 2.2, the discussion rules’ conventional validity is discussed. Vis à vis the explanation offered by the study’s authors – or so I admit –, the theory-internal purpose of this study remains rather unclear to me. After all, as stressed by the authors, the normative content of the pragma-dialectical theory is neither open to refutation by empirical data, nor to confirmation by such data (Section 3). Therefore, I claim, the theoretical value of this investigation is heuristic (Section 4). Section 5 comments on a tension between the level of measurement and the level at which measurement is reported.

2. Treating Conventional Validity Empirically
2.1 Aim, Method and Main Result
The aim is to determine “if and to what extent the norms that ordinary language users (may be assumed to) apply in judging argumentative discussion moves correspond to rules which are part of the ideal model of critical discussion” (van Eemeren, Garssen and Meuffels 2009: v; italics added). This means to study the rules’ intersubjective validity or – insofar as conventions are understood to normally remain implicit – their conventional validity (see van Eemeren and Grootendorst 2004: 56, fn. 35). In contrast, the rules’ problem validity cannot be studied empirically, but is a matter of expert agreement.

Four of the ten pragma-dialectical discussion rules are selected: Freedom Rule, Obligation to Defend Rule, Argumentation Scheme Rule, Concluding Rule. Based upon these rules, mini dialogues (of two to four turns) are created. On expert opinion, the last turn of these either is or is not a clearly fallacious discussion move (“multiple message design”). Under variation of domains/contexts (domestic, political, scientific), dialogues are presented to lay arguers – mostly younger students – in questionnaire form. This occurs under the normal precautions with empirical research (e.g., including filler items, in random order, controlling loadedness/politeness of examples, retesting items from previous studies); a sample size of 50 is typical. Refer to van Eemeren, Garssen and Meuffels (2009: 64f.) for examples. Hample (2010) and Zenker (2010) report further details; an accessible summary is Hornikx (2010). Notably:

“The third domain [the scientific discussion] was described as the scientific discussion in which – as was emphasized – it was not so much a matter of persuading others but of resolving a difference of opinion in an acceptable manner: Who is right is more important than with whom one agrees.” (van Eemeren, Garssen and Meuffels 2009: 66).

Participants were then asked to rate the reasonableness of the last move in a dialogue on a seven point Likert scale (1-7). Thus, for each dialogue and each subject, a reasonableness judgment value (RJV) becomes available. These RJVs are averaged – yielding an averaged reasonableness value (ARV) – , then assessed on measures of statistical significance (yielding, e.g., correlation coefficient, standard deviation, effect size).

This operationalizes reasonableness as a seven degree notion. One can now quantify the extent to which ordinary arguers’ responses are (in)consistent with the normative content of the four discussion rules as applied to some (mini-)dialogue. The value four (4) being the middle point, one reasons: If this rule, the violation of which generates these discourse fragments, is conventionally valid (to some extent), then fallacious fragments receive an ARV < 4 and non-fallacious fragments receive an ARV > 4. One compares whether the RJVs do, on average, fall within the expert predicted region.

Applied to four of ten rules, with the exception of the confrontation and the opening stages (van Eemeren, Garssen and Meuffels 2009: 224), the investigation is non-exhaustive in the following sense: In principle, violations of different rules (or of a subset of the same rules, but in a different discussion stage) might lead to different results. The ten rule version is a popularization of the more technical 15 rule set (van Eemeren & Grootendorst 2004: 135-157; Zenker 2007). How the 15 and the 10 rule set are related is not clear in detail. So, “four out of ten” or “x out of 15” rules have been studied. For a list of fallacies used, see van Eemeren, Garssen and Meuffels (2009: 223).

Under these reservations, the main result is that
“[T]he body of data collected indicate that the norms that ordinary arguers use when judging the reasonableness of discussion contributions correspond to a rather large degree with the pragma-dialectical norms for critical discussion.” (van Eemeren, Garssen and Meuffels 2009: 224)

This claim is based on the size of the effect obtained in comparing the ARVs for fallacious and non-fallacious discourse fragments.

2.2 Conventional Validity
Throughout the development of the pragma-dialectical research program, it has been contended that “[t]he [pragma-dialectical] rules (…) are problem valid because instrumental in the resolution process by creating the possibility to resolve differences of opinion” (van Eemeren, Garssen and Meuffels 2009: 27). They are considered instrumental to resolving a difference of opinion insofar as a violation of any rule is understood as a hindrance to this aim.

A further contention is normative in character: The pragma-dialectical rules should be conventionally valid, i.e., agreeable to lay arguers. This means, the rules’ content should not conflict with the norms that lay persons (i.e., those not specifically trained in the pragma-dialectical theory) can be construed to accept. This norm is regularly traced to Barth & Krabbe (1982: 21-22) or Crawshay Williams (1957).

Should these two books answer the question why it is important that the pragma-dialectical rules are conventionally valid, then this answer is hidden well. At any rate, neither van Eemeren, Garssen and Meuffels (2009) nor the comprehensive van Eemeren & Grootendorst (2004) offer much of an explanation. At the relevant places (known to me), it is stated that the rules should be conventionally valid, not why (e.g., van Eemeren, Garssen and Meuffels 2009: 27).

Perhaps an exception is a more detailed explanation in a 1988 article. From this, three quotes follow. These suggest that the conventional validity of discussion rules – understood as the acceptability or the acceptedness of some normative content by lay arguers – arises with insight into the rule’s pragmatic rationale. That is, the quotes are not inconsistent with an interpretation according to which intersubjective acceptance comes about through insight into problem validity.

“We believe that the process [of solving problems with regard to the acceptability of standpoints] derives its reasonableness from a two-part criterion: problem-solving validity and conventional validity (cf. Barth and Krabbe 1982: 21-22). This means that the discussion and argumentation rules which together form the procedure put forward in a dialectical argumentation theory should on the one hand be checked for their adequacy regarding the resolution of disputes, and on the other for their intersubjective acceptability for the discussants. With regard to argumentation this means that soundness should be measured against the degree to which the argumentation can contribute towards the resolution of the dispute [i.e., the degree of problem validity], as well as against the degree to which it is acceptable to the discussants who wish to resolve the dispute [i.e., the degree of conventional validity].” (van Eemeren & Grootendorst 1988: 280)

Pace stylistic changes (e.g., ‘dispute’ has been replaced by ‘difference of opinion’), this is in line with the 2004 presentation. Further in the same article:

“It may now be possible to make plausible that the rules are such that they merit a certain degree of intersubjective acceptability, which would also lend them some claim to conventional validity. [paragraph] The claim of acceptability which we attribute to these rules is not based in any way on metaphysical necessity, but on their suitability to do the job for which they are intended: the resolution of disputes [i.e., their problem validity]. The rules do not derive their acceptability from some external source of personal authority or sacrosanct origin. Their acceptability [i.e., their conventional validity] should rest on their effectiveness when applied [i.e., their problem validity]. Because the rules were developed exactly for the purpose of resolving disputes, they should in principle be optimally acceptable to those whose first and foremost aim is to resolve a dispute. This means that the rationale for accepting these dialectical rules as conventionally valid is, philosophically speaking, pragmatic.” (van Eemeren & Grootendorst 1988: 285; italics added)

Particularly the last sentence suggests (to me) that understanding the rationale of the pragma-dialectical rules brings about their acceptance. This interpretation seems to be consistent with that provided in van Eemeren and Grootendorst (2004: 187). That the rationale is pragmatic, I take to be irrelevant for providing some rationale for acceptance. It seems moreover uncontroversial (to me) that understanding the rationale for accepting them as conventionally valid presupposes understanding (learning) the pragma-dialectical rules. Similarly:

“The speech acts which are most useful to all concerned who share a certain goal, for example to resolve a dispute, possess a form of problem validity which may lead to their claim of conventional, intersubjective validity.” (van Eemeren & Grootendorst 1988: 289, n. 14)

Vis à vis these (less recent) quotes, and absent a more recent detailed explanation, it remains unclear (to me) why the pragma-dialectical rules should be conventionally valid independently of having being learned. One’s methodology may very well support the claim that they are (or not), but why begin?

If they are problem-valid (i.e., acceptable as a solution to a problem), then recognizing their problem-validity expectably brings about their acceptedness, and brings it about for this reason (cf. van Eemeren and Grootendorst 2004: 187). At any rate, the rules’ problem validity and one’s (cognitive) ability to appreciate their pragmatic rationale – are sufficient for acceptance (thus, for conventional validity). If so, how can being acceptable/accepted by those not trained in these rules be important for the theory?

It is trivial to state that the pragma-dialectical (or some other set of problem valid) rules cannot be effective in leading to dispute resolutions, unless at least two disputing parties de facto accept them (explicitly or implicitly). In one scenario, the pragma-dialectical rules being conventionally invalid means that problem valid rules are unaccepted by lay arguers (if the rules are problem valid). So, ceteris paribus, lay persons might not be expected to maintain a discussion (and obtain a result) which squares with the rules. Resolutions of differences of opinion would then perhaps be less expectable?

This author fails to see the upshot. Why demand (“should”) conventional validity independently of rule acquaintance?

I discount an otherwise important comment by Lotte van Poppel (personal communication). She points out that it might be less probable for the social aim behind the pragma-dialectical research program (improving argumentative praxis) to be reached, if the theory’s normative content turned out to be not accepted by lay arguers. This cannot merely relate to the exact formulation of said content; it must be more than a matter of style. If style did matter, why investigate conventional validity in an indirect way, rather than display the rule set and ask for assent? On this indirectness, see van Eemeren, Garssen and Meuffels (2009: 49f.).

Insofar as the comment then concerns the content, rather than various ways of formulating it (e.g., by avoiding/using technical terms): If lay arguers and expert judgment do not converge on the content of (a set of) problem valid rules – perhaps so be it! It remains unclear (to me) why one assesses (on a methodologically hardened measure) the distance between expert and a lay person judgment. Granted experts find the normative content problem-valid, what support does the content receive from convergence with lay arguer judgment? What doubt arises from divergence?

At this point, it does not help to learn that empirical data take on a special role. As the next section shows, distance between expert and lay person judgment appears to be of no immediate theoretical relevance.

3. The Special Status of the Results

3.1 Compare, not Test
Compared to applying and testing an empirical theory, the data obtained are special: “Empirical data can neither be used as a ‘means for falsification’ nor as ‘proof’ of the problem validity of the discussion rules” (van Eemeren, Garssen and Meuffels 2009: 27). Standardly, an empirical theory is tested against experience by applying it to a phenomenon (for which the theory is expected to account), in order to derive a prediction. In this case, the prediction is a judgment on the (non-)fallaciousness of some discourse item.

With A for antecedent, T for theory and P for prediction, applying an empirical theory may take the deductively valid form: A; T; (A & T) -> P; ergo P  (modus ponens). If the prediction, P, is born out – and A is not in doubt (!) –, then T counts as confirmed. Note, however that, on a deductive construal, such confirmation would instantiate a deductively invalid schema (affirming the consequent).

If the prediction is not born out (i.e., non P is true), and A is not in doubt, then – again, on a deductive construal – falsification instantiates a valid form (modus tollens). In deductive logic, however, only the negation of (A & T) follows from non P; to derive non T, A must be less retractable than T (see Lakatos 1978; Zenker 2009).

In contrast, the normative content of the pragma-dialectical theory is not tested against lay person judgments, but compared to them (van Eemeren, Garssen and Meuffels 2009: 27). This means, some discourse fragment, A, under application of the pragma-dialectical theory, T, may very well deductively imply a prediction, P: “This fragment is (not) fallacious.” That much is captured by ‘(A & T) -> P’. However, P and the lay person judgment con- or diverging does not (deductively logically) affect the theory.

The explanation offered in defense of this odd support behavior – vis à vis empirical theories, Lakatos might speak of “immunization” – builds on the contention that the pragma-dialectical theory offers norms rather than descriptions.

3.2 Normative vs. Descriptive Contents
The standpoint in van Eemeren, Garssen and Meuffels (2009) is: What lay persons do or do not accept can neither be turned against the theory in the sense of falsification, nor support the theory in the sense of verification. (Recall from above that falsification can be treated in deductive logic; verification requires a notion of inductive validity.) The explanation for this standpoint is comparatively brief.

“The presumption in all our empirical studies is that the discussion rules involved are problem valid; the focus is on their conventional validity. The status of the results of this empirical work is special: The empirical data can neither be used as ‘means of falsification’ nor as ‘proof’ of the problem validity of the pragma-dialectical discussion rules. In the event that the empirical studies indicate that ordinary language users subscribe to the discussion rules, it cannot be deduced that the rules are therefore instrumental. The reverse is also true: If the respondents in our studies prove to apply norms that diverge from the pragma-dialectical discussion rules, it cannot be deduced that the theory is wrong. Anyone who refuses to recognize this is guilty of committing the naturalistic fallacy, the fallacy that occurs when one inductively jumps from “is” to “ought.” (van Eemeren, Garssen and Meuffels 2009: 27)

On might take this quote to express a meta level assertions about the inferential relation between a set of normative and descriptive statements. In effect, the standpoint is: There is no deductive inferential relation. This standpoint also shows at object level when evaluating discourse items in which a descriptive standpoint is supported by value statements (normative premises).

“The combination of a descriptive standpoint and a normative argument always leads to an inapplicable argument scheme: The acceptability of a descriptive standpoint is after all independent of the values that are attached to the consequences of the acceptance of that outcome” (van Eemeren, Garssen and Meuffels 2009: 172).

Put more generally, “(…) whether something is true or not in a material sense does not depend on the question if we like it or not” (van Eemeren, Garssen and Meuffels 2009: 172). So, truths (“facts”) do not receive support from, nor can they be undermined by human (dis-)approval.

Pragma-dialectics, of course, is a normative theory. The discussion rules are claimed to be supported by achieving the theoretical value of problem validity. This value is achieved through systematically identifying hindrances to a resolution oriented discourse (aka. fallacies). Clearly, to claim problem validity of a normative theory is not to assert a norm, but a fact – if it is one. So, lay arguers endorsing norms (in)compatible with the pragma-dialectical ones does not (without committing a naturalistic fallacy) license a claim about the theory’s problem validity: Just as undermining norms by facts is considered fallacious, supporting facts with norms is considered fallacious.

These contentions indicate that the naturalistic fallacy is a theoretical commitment for pragma-dialecticians. This may surprise. After all, it has been recognized that “fallaciousness” depends on various conditions, to the point that “fallacies can have sound instances” is a meaningful assertion in some contexts. Pragma-dialecticians appear committed that this is not so in the cases discussed here.

3.3 The Theoretical Value of Inconsistency
To summarize the above: Facts (here: the reasonableness judgments of ordinary speakers) are impotent with respect to norms (here: the pragma-dialectical rules). On this background, why is the conventional validity of the pragma-dialectical rules under study to begin with? After all, in case the rules would be conventionally valid – and the claim is that they are to a rather large extent – this at most supports conditional claims, such as: If ordinary speakers accept normative contents, then these contents are not inconsistent with the normative content of the pragma-dialectical theory.

“Just as would be the case in corpus research, in our series of experiments the conventional validity of the pragma-dialectical rules is investigated not in a direct, but in an indirect sense. Due to the fact that discussion fragments that contain a fallacy are found to be unreasonable by normal judges, and fragments that do not contain any fallacies are deemed reasonable, we deduce that in the judgment of the fairness of argumentation the respondents concerned appeal, whether implicitly or explicitly, to norms that are compatible, or at least not contradictory, to rules formulated in the pragma-dialectical argumentation theory”. (van Eemeren, Garssen and Meuffels 2009: 49, italics added)

This indirectness comes about for the (above discussed) reason that, by the authors’ standards, a normative theory cannot be falsified by descriptive data, nor can its problem validity be confirmed by such data. Hence, consistency between the theory’s normative content and the content which speakers may be construed to rely on is rather useless for the theory. On the other hand, inconsistency between the theory and a lay-person judgment has no bearing on the theory either, but has heuristic value. Inconsistency informs on “what works” without specific training and what does not.

4. Heuristics
“Anomalies” forthcoming in this study should prove relevant for theoretical development. Most important, perhaps, context not only matters but counts. For example, participants judge an ad hominem fallacy to be as reasonable in a domestic as in a political context, but less reasonable than in a scientific context. Similarly, a direct personal attack in a scientific context is judged to be less reasonable than a tu quoque in the same context (ARV = 2.57; standard deviation 0.81 vis à vis 3.66; 0.86).

Normatively, that the reasonableness value should be similar or the same in all three contexts, and for both variants of the ad hominem in the same context, is a defensible claim. Note that nothing in the standard theory explains such a context-dependency.

When a standpoint enjoying presumptive status is supported in a fallacious manner, then participants tend to judge this move more leniently than when no such presumption is enjoyed. Normatively, this may not sit well with everybody. Moreover, there are (perhaps striking) differences in culture: some robust effects “break down.”

Without training, lay persons will normally not be able to reliably distinguish between a sound ad absurdum and a fallacious ad consequentiam argument. On the other hand, participants do reliably distinguish the legal principle according to which a presumption of innocence holds unless proven otherwise, suggesting that further legal principles may generate robust effects as well.

The “trickiness” of the mini dialogues may be varied in future work, to investigate the point at which variation in content produces effects. Discourse fragments in this study are conspicuously simple. Some “tweaking” towards realistic content should see rules “breaking down.” After all, also this study supports the claim that participants tend to be influenced by the content of a standpoint: If you assent to what is supported by fallacious means, you will judge such fallacies more leniently than you would, if you did not assent. Though perhaps understandable, even demonstrating such effects to depend on context would still register as unacceptable in some normative framework.

5. Data Reporting
A last point pertains to the tension between the level of measurement and the level of reporting measurements. As mentioned above, measurement occurs on a seven point scale: 1, 2, 3, 4, 5, 6, 7 (Likert 1932); very (un)reasonable marks the ends. Without further assumptions, this means that reasonableness judgments values are recorded at ordinal level. Here, one lacks distance information. It counts as unknown if the distance between 5 and 6, say, is the same as that between 2 and 3.

When reporting and statistically treating data, the assumption is that the distances are the same. This is needed. Otherwise, averaging – which yields fractions (e.g., an averaged reasonableness value of: 2:200/375 would be meaningless. Thus, data are treated as if they had been obtained at interval level. Deeply entrenched, the equi-distance assumption can be doubted in a particular case. The topic should make for a good case study on a scientific controversy. See Jamieson (2004), Carifo & Perla (2007) and Norman (2010) for both positions.

The standard report formats are the mean plus standard deviation. The mean is the sum of all measurement-values divided by the number of measurements. To indicate the spread of data points provided the mean, the standard deviation, s, is used (where x is a data value, x bar the mean, and n the number of measurements) (Figure 1).

Figure 1

The standard deviation is a widely accepted measure of dispersion. Yet, the value of s will not allow reconstructing the exact spread. Readers remain ignorant as to how many subjects showed what deviation in their reasonableness judgments. This makes data less useful for replication. By exactly how much individuals differed is hidden, since ARJs have replaced RJVs (see Section 2.1).

It suggests that the aim of the study was not to report the precise reasonable values assigned to artificial discourse items. Rather, the point was to show that, for the mini dialogues constructed (which suffer purposefully from near-triviality), theoretical prediction and averaged lay person judgment converge. Results strongly suggest that it is possible to construct examples which lay persons distinguish – on average and to a rather large extent – into fallacious and non-fallacious moves.

6. Conclusion
The theoretical purpose of comparing expert and lay person judgments concerning the reasonableness of rule-generated discourse fragments remains to be explicated. In the absence thereof, the naturalistic fallacy may count as a theoretical commitment for pragma-dialecticians. Whether this commitment needs additional justification would seem to depend on prior theoretical commitments.

Several examples of the heuristic value of the empirical investigation of the conventional validity of four of ten pragma-dialectical discussion rules were pointed out. On pains of having appeared critical, readers are reminded of two reviews (Hample 2010, Zenker 2010) praising van Eemeren, Garssen and Meuffels (2009). The study is highly relevant, irrespective of one’s theoretical background.

REFERENCES
Barth, E.M., & Krabbe, E.C.W. (1982). From Axiom to Dialogue: A Philosophical Study of Logics and Argumentation. Berlin: De Gruyter.
Carifo, J., & Perla, R.J. (2007). Ten Common Misunderstandings, Misconceptions, Persistent Myths and Urban Legends about Likert Scales and Likert Response Formats and their Antidotes. Journal of Social Sciences, 3, 106-116.
Crawshay-Williams, R. (1957). Methods and Criteria of Reasoning: An Inquiry into the Structure of Controversy. London: Routledge and Kegan Paul.
Eemeren, F.H. van, & Grootendorst, R. (1988). Rationale for a Pragma-Dialectical Perspective. Argumentation, 2, 271-291.
Eemeren, F.H. van, & Grootendorst, R. (2004). A Systematic Theory of Argumentation. The Pragma-dialectical approach. Cambridge: Cambridge University Press.
Eemeren, F.H. van, Garssen, B., & Meuffels, B. (2009). Fallacies and Judgments of Reasonableness. Empirical Research Concerning the Pragma-Dialectical Discussion Rules (Springer Argumentation Library, Vol. 16). Dordrecht: Springer.
Hample, D. (2010). Review of Eemeren, F.H. van, Garssen, B., & Meuffels, B. (2009). Argumentation, 24, 375-381.
Hornikx, J. (2010). Review of Eemeren, F.H. van, Garssen, B., & Meuffels, B. (2009). Information Design Journal, 18, 175-177.
Jamieson, S. (2004). Likert scales: how to (ab)use them. Medical Education, 38, 1212-1218.
Lakatos, I. (1978). The Methodology of Scientific Research Programs. Cambridge: Cambridge University Press.
Likert, R. (1932). A Technique for the Measurement of Attitudes. Archives of Psychology, 140, 1-55.
Norman, G. (2010). Likert scales, levels of measurement and the “laws” of statistics. Advances in Health Science Education, 15, 625-632.
Zenker, F. (2007). Pragma-Dialectic’s Necessary Conditions for a Critical Discussion. In Hansen, H. et al. (Eds.). Proceedings of the 7th Int. Conference of the Ontario Society for the Study of Argumentation (OSSA), Windsor, ON, CD ROM.
Zenker, F. (2009). Ceteris Paribus in Conservative Belief Revision. Frankfurt: Peter Lang.
Zenker, F. (2010). Review of Eemeren, F.H. van, Garssen, B., & Meuffels, B. (2009). Cogency, 2 (1), 149-165.

1. Introduction
In this paper we explore the argumentative function of reported speech in economic-financial newspaper articles. The present research is based on a corpus of articles of the three main daily Italian economic-financial newspapers: Il Sole 24 Ore, Italia Oggi and MF/Milano Finanza. Why are we interested in studying the relationship between reported speech and argumentative function of economic-financial news? The analysis of economic-financial newspaper articles previously carried out shows that the predictive speech act occupies a dominant position in the discourse structure of economic financial news (Miecznikowski, Rocci, and Zlatkova in Press). Being clearly oriented towards predicting events, the information demand in the journalistic discourse domain of finance differs significantly from other domains, such as editorials, sports, crime, whose informational interest lies in narrating or commenting past events.

The reader wants to know not what has happened, but also, more importantly, what is going to happen. The analysis also showed that the predictive speech acts and their supporting arguments are sometimes attributed to unnamed, but more often to named sources, such as financial analysts, money managers, bankers. Being geared towards the decision making of investors, financial discourse is overtly or covertly argumentative. These semantic and pragmatic features of economic-financial discourse make this genre particularly interesting for investigation. The frequent use of reported speech in this genre poses a challenge to argumentative reconstruction, because it is difficult to attribute the role of protagonist to the journalist who often seems to use reported speech strategically to avoid his/her personal commitment to either the standpoint or the argument. However, in this paper we argue that the distinction between different types of reported discourses and the distinction between different forms introducing them provide important cues for determining the functions of the reported segments in the journalist’s argumentation and ascertaining to what extent the journalist is committed personally to the stated claim.

2. Types of reported segment
For the present research we adopt a broad definition of reported speech that is a quotation of another’s discourse, the presence of another person’s words in the author’s discourse (Calaresu 2004, Smirnova 2009). The analysis of the corpus showed that the reported segment can be of two types: I) it is used to report an opinion; II) it puts forward an argumentation. The reported segment used to report an opinion can perform both non–argumentative and argumentative function. In the case of non-argumentative use, the journalist simply reports an opinion maintaining a clear distance with respect to what is said as illustrated in example 1.

1. Altre potenziali prede secondo Jason Goldberg, analista bancario di Lehman Brothers, sarebbero istituti cinesi, brasiliani, coreani e dell’Europa dell’Est, a cominciare dalla Russia. (Il Sole 24 Ore, 05.04.2006, doc. 9)
Other potential targets, according to Jason Goldberg, a banking analyst at Lehman Brothers, are Chinese, Brazilian, Korean, and Eastern European financial institutions, in addition to the Russian ones.

Here the journalist reports the opinion of the banking analyst Jason Goldberg about the potential financial institutions, without taking position or commenting on it. There are neither subjectifiers[i] in the co-text[ii], which indicate the stance of the journalist towards the expressed opinion, nor other markers by means of which we can infer the journalist’s position. Therefore, we cannot attribute the role of protagonist to the journalist. In such cases, the reporting of an opinion has merely informative and not argumentative function. Moreover, the choice of reported speech indicates the distance of the journalist from what is said, and what he/she undertakes no attempt to defend. In this example, the reported speech is introduced by an indirect glossed form of reported speech, analysed in the literature (Calaresu 2004, p.163) as a form expressing a clear distance of the speaker/writer with respect to the reported utterance. This form is characterized by an introductor which performs the function of “gloss” inside or on the margin of the citation (“Other potential targets, according to Jason Goldberg…”). Indirect glossed form of reported speech creates an unexpected dissociation between the author of the original discourse and who reports it. In fact, in the case of indirect glossed speech, a segment of discourse is interrupted by the introductor, signalling that the responsibility for the utterance is somebody else’s.

Beyond the non-argumentative use, the reported segment can perform an argumentative function as a basic argument from authority, as illustrated in example 2.

2. Il mercato italiano del vino sta uscendo dalla crisi. Lo afferma Vinitaly, il salone dei vini e distillati che aprirà le sue porte a Verona dal 6 al 10 aprile. (Italia Oggi, April 1, 2006 doc. 628)
The Italian wine market is overcoming the crisis. This was affirmed by Vinitaly, the salon of wines and spirits which will be open from April 6th to 10th.

Following the method suggested by Pragma–Dialectics (cf. Van Eemeren, Grootendorst, Snoeck Henkemans 2002), we will represent the argumentative structure graphically after every discussed example by showing which arguments support the standpoint and how these arguments are organized and combined.`(Figure 1)

Figure 1

The standpoint The Italian wine market is overcoming the crisis is supported by evoking the authority of Vinitaly. As we can see from the graphical representation, we have a case of single argumentation (Eemeren & Grootendorst 1992), with just one argument supporting the advanced standpoint. It is worth noticing that the reported segment is introduced by the anaphoric pronoun lo (“this”). In linguistics, the term anaphora is used to refer with a pronoun to an object that has already been introduced into the discourse by some other linguistic construction. In other words anaphora is the relationship between two linguistic elements where the interpretation of the one of the elements (called the anaphora) requires the interpretation of the other (called the antecedent) (Bazzanella 2005, p. 79). In our example lo (“this”) refers back to the situation described in the previous sentence, i.e. that The Italian wine market is overcoming the crisis. If we compare this way of introducing reported speech with a “classical” indirect form of reported speech: Vinitaly has affirmed that the Italian wine market is overcoming the crisis, we clearly notice the different position of the introductor. In the case of anaphoric use the introductor is postponed Lo afferma Vinitaly (“This was affirmed by Vinitaly”), whereas in the case of the classical indirect form of reported speech, the introductor precedes: Vinitaly has affirmed that […]. The rhetorical effect of the different position of the introductor has been widely discussed in the literature (Calaresu 2004). The strategy of the postponed introductor has the rhetorical function of a surprise effect; it means that the reader has to reinterpret what he has just read as the discourse of someone else, other than the journalist. The above-mentioned case differs significantly from cases where the introductor is put at the beginning and the reader immediately interprets the discourse as reported. In our corpus the use of the postponed introductor to introduce the argument from authority is frequently encountered in cases where the journalist endorses the reported opinion.

It emerges from the corpus analysis that the argument from authority can also be part of a complex structure used to support a standpoint advanced by the journalist as illustrated in example 3. We make a clear distinction between examples 2, where the journalist endorses what is said and examples such as example 3, where the journalist advances his/her own standpoint using an argument from authority to support it.

3. E a quel punto, è ipotizzabile – la maggioranza degli analisti tecnici e fondamentali è d’accordo – l’avvio di una fase laterale. Per questo gli investitori dovrebbero utilizzare i prossimi top per prendere profitto e iniziare la ristrutturazione dei portafogli. (Il Sole 25 Ore, April 10, doc. 170)
At this point, it is presumable that a sideway phase is about to start – the majority of the technical and fundamental analysts agree on that. For this reason, investors should use the next peak to make a profit and begin reorganizing their portfolios.

The argumentative structure of the example can be reconstructed as follows (Figure 2):

Figure 2

The standpoint advanced in example 3 is a directive speech act of recommendation: The investors should use the next peak to make a profit and begin reorganizing their portfolios. The standpoint is neither attributed to financial analysts nor to other sources. It is advanced by the journalist, so we can attribute to him/her the role of protagonist. It is worth noticing that the phrase per questo (“for this reason”), preceding the recommendation, is a typical argumentative indicator of the advancing of a standpoint. This standpoint is supported by subordinatively compound argumentation, where the defence itself is supported by a longer or shorter series of “vertically linked” single argumentation. Each of the arguments in the chain contributes to the defence of the standpoint by supporting the argument immediately above, and only the series as a whole contributes to its conclusive defence (Eemeren & Grootendorst 1992). In example 3, the specific standpoint: the investors should use the next peak to make a profit and begin reorganizing their portfolios is supported by an argument it is presumable that a lateral phase is about to start which serves as a substandpoint and in its turn is defended by an argument from authority the majority of the technical and fundamental analysts agree on that. It is worth noticing that the force of the argument from authority is further enhanced by the argument from consensus between technical and fundamental analysts, who usually are two divergent authorities. Technical and fundamental analyses refer to two different and often polemically contrasted stock-picking methodologies used for researching and forecasting the future growth trends of stocks.

Analysis of our corpus showed that, to enhance the credibility of the source, the journalist in economic-financial newspaper articles uses either professional characteristics of the source in order to present it as an authority in the domain, thus removing any possible doubts about his reliability (“Stefano Zoffoli, strategist of Julius Baer asset management […]” MF, April 14, 2006, doc. 24; “Adolfo Guzzini, president of Guzzini, turnover of 170 million Euros […]” Il Sole 24 Ore, April 25, 2006, doc.101) or he/she uses the argument from consensus to convince the reader about the credibility of what is said (“Even the most cautious analysts said that […]” Il Sole 24 Ore, April 20, 2006, doc. 27; “The majority of the technical and fundamental analysts agree on that” Il Sole 25 Ore, April 10, doc. 170). The combination of both strategies is also possible.

So far, we have discussed cases where the reported segment is used to report an opinion. Now we move to cases where the reported segment contains argumentation. Analogously to the cases discussed previously, also in the cases where the reported segment contains argumentation we distinguish between non-argumentative and argumentative uses. In the case of a non – argumentative use the journalist simply reports an argumentation, distancing himself from it as illustrated in example 4

4. Morgan Stanley sconsiglia invece di investire nel mercato del mattone reduce da quattro anni di crescita eccezionale. Il comparto è già in una chiara fase di frenata (Il Sole 24 Ore, April 2, 2006, doc. 47)
Morgan Stanley advised not to invest in the brick market after four years of exceptional growth. The sector is in clearly slowing down.

Here the journalist reports not only the advice of Morgan Stanley not to invest in the brick market but also the supporting argumentation that the sector is clearly slowing down. Since the journalist distances himself from the reported advice as well as from the argumentation supporting it, the role of protagonist cannot be attributed to him.
Differently from example 4, in example 5, the journalist endorses the reported argument giving rise to an argument from authority including the reported argumentation of the authority. Since the journalist endorses the reported argumentation the role of protagonist can be attributed to him.

5. Meno rosee le prospettive per i consumatori: secondo Browne il prezzo della benzina non potrà che salire data l’impennata del greggio. (Il Sole 24 Ore, April 26, 2006, doc.22)
The economic outlook for consumers is less bright. According to Browne, the
price of petroleum can only rise given the steep rise of crude oil.

Argumentatively, the standpoint the economic outlook for consumers is less bright is supported by a complex structure of argument from authority including the entire line of reasoning advanced by Browne. As has been argued by Smirnova (2009) in her paper on the argumentative function of reported speech in British newspapers, cases of pure appeal to authority are rare. In the majority of cases, we have a combination of an argument from authority with another type of argument. In this example the Brown’s authority supports the causal connection between the price of crude oil and the price of petroleum established in the major premise if the price in crude oil rises, then the price of petroleum rises (Figure 3).

Figure 3

Differently from example 5, where the journalist only endorses the reported standpoint, in example 6, the journalist advances his/her own standpoint and supports it by using a complex structure where the reported segment contains argumentation. We will discuss this example more in detail to demonstrate the contribution of the reconstruction of the argumentative scheme proposed by the Argumentum Model of Topics (see below) to the reconstruction of the argument structure.

6. Anche gli analisti più cauti puntano su nuovi rialzi: John Reade dell’UBS, li ritiene molto probabili, e Simon Weeks di ScotiaMocatta, nota che “sono in pochi a vendere” e ciò rende molto vicino il traguardo di 640$. (Il Sole 24 Ore, April 20, doc. 27)
Even the most cautious analysts predict new highs. John Reade from UBS, considers them very probable, and Simon Weeks of ScotiaMocatta, noticed that “only few people sell”, and this makes the goal of $ 640 very close.

The argumentative structure of the example can be reconstructed as follows (Figure 4):

Figure 4

The standpoint probably there will be new highs is supported by coordinative argumentation with an unexpressed major premise: If the most cautious analysts say that there will be new highs there will be probably new highs. The explicit premise even the most cautious analysts say that there will be new highs is supported by two independent arguments from authority, the first one is: John Reade considers them very probable and the unexpressed premise: Joan Reade is one of the most cautious analysts and the second one is: Simon Week argues that new highs are probable and the unexpressed premise: Simon Week is one of the most cautious analysts. In the second argument from authority, we have a reported argumentation of the source, as we can see from the graphical representation above: the reason why Simon Week argues that new highs are probable is based on economic causality that only few people sell. In order to explore in depth the relationship between standpoint and argument we use the Argumentum Model of Topics (henceforth AMT) , developed at the Institute of Linguistics and Semiotics, University of Lugano, in particular by Eddo Rigotti and Sara Greco-Morasso (Rigotti 2006, 2009a, 2009b, Rigotti and Greco-Morasso 2006). The AMT represents the reasoning chain underlying an argument and highlights both the logic and the pragmatic/contextual components of the argument scheme. It is made up of two syllogisms: one is the endoxical syllogism whose major premise is an endoxon, and the other is the topical syllogism, whose major premise is a maxim. A maxim is an implication of the form  p->q,  generated by a locus and which gives rise to an inferential process. An argumentative scheme (locus) from authority emerges from the reconstruction presented above. Using the AMT we build the “synergic” representation of an argument from authority (see figure 1 below) which allows us to distinguish, within the inferential structure of the argument, the two components mentioned previously. The specific standpoint here is probably there will be new highs. The maxim if a reliable authority said something, it is likely to be true is directly engendered from the locus from authority. In order for this maxim to generate the final conclusion, which coincides with the standpoint to be supported, the following minor premise is needed: the reliable authority said that there will be new highs. Such a premise however is not self-evident; it needs itself to be backed by another syllogistic reasoning, in this case anchored in an endoxon: Among all analysts, the most cautious ones are the most reliable. The datum, which is the factual statement constituting the minor premise of the endoxical syllogism is, the most cautious analysts, said that there will be new highs. This leads to the conclusion: the reliable authority said that there will be a new highs. This conclusion is “exploited” by the maxim (as indicated by the curved arrow in the diagram) to generate the final conclusion which coincides with the standpoint to be supported: probably, there will be new highs. The two syllogistic reasoning give rise to the complex inferential structure which is represented by a “Y-like structure” within AMT (fig.1). The two syllogisms have distinct, but at the same time complementary functions: the maxim is responsible for the inferential mechanism and defines the law, while the endoxon links the argument to a shared opinion in the community. So, we can say that the topical component ensures the inferential force, whereas the endoxical component ensures the persuasive effectiveness, but if the topical component is not combined with the endoxical component it remains a mere logical mechanism.

From the analysis of example 6 illustrated above, different degrees of complexity emerge:
1. we are dealing with multiplicity of sources – different authorities can be evoked in order to support the standpoint;
2. we can have an addition argument supporting the credibility of the source, e.g. the argument from consensus like in example 6 where the consensus between the less cautions and the most cautious analysts (“even the most cautious analysts say that there will be new highs”) is emphasized in order to boost the credibility of the source itself; and
3. we have the reporting of an entire line of argumentation of the source:
As it has been argued previously in the paper, the strategies of boosting the credibility of the source are highly used by the journalist when he/she endorses the standpoint advanced by the source or when he/she advances his/her own standpoint (Figure 5).

Figure 5.- Synergic representation

3. Conclusion

This paper explored the function of the reported segment with a particular focus on the journalist’s stance towards the reported statements in order to demonstrate that there is a constellation of indicators providing a sufficient basis for ascertaining to what extent the journalist assumes the role of protagonist, and that in many cases, the argumentative reconstruction is fully justified.

From the analysis carried out, it emerges that reported speech can perform both a non- argumentative and an argumentative function. The reported segment can have a purely informative function: the journalist simply reports an opinion, an argumentation maintaining a clear distance with respect to what is said; in this case he/she is not committed personally to any reported claim. Alternatively, the reported segment can perform an argumentative function. In the case of opinions (I), the journalist advances a standpoint, supported by an argument from authority. The reported segment may a) contain the standpoint itself, formulated by a third party but endorsed by the journalist; b) contain statements considered by the journalist as arguments for a standpoint expressed in his/her own discourse. Analogously, in case (II), the journalist a) either makes the cited speaker utter the entire line of argumentation he/she intends to put forth or b) expresses a standpoint in his/her own words, backed up by the argumentation contained in the cited segment. In both cases, the result is a complex argument from authority, including reported argumentation of different kinds (causal, pragmatic, symptomatic reasoning etc.).

Some correlations have been identified between the function and the form of reported speech. In the case of purely informative function, the reported speech is mostly introduced by an indirect glossed form, analysed in the literature as a form expressing a clear distance of the speaker/writer with respect to the reported utterance. When the reported segment performs an argumentative function, it is often framed by an indirect form with a postponed framing expression. The use of the postponed framing expression is frequently encountered in arguments from authority in which the journalist endorses the reported segment. The relationship between form and function of reported speech will be investigated more in detail in our future work.

NOTES
[i] For the purpose of this paper we are interested in subjectifiers such as boosters (e.g. infatti (‘indeed’), affatto (‘at all’), proprio, davvero (‘really’) and hedges like quasi (‘almost’), un po’ (‘a bit’), più che altro (‘rather’), or emotionally-connotated lexical items.
[ii] Co-text” is a commonly used term in Discourse Analysis. Co-text refers the words or sentences surrounding any piece of written (or spoken) text. (cf. Brown & Yule 1983).
[iii] Argumentative indicators are “words and expressions that may refer to argumentative moves, such as putting forward a standpoint or argumentation. The use of these argumentative indicators is a sign that a particular argumentative move might be in progress, but it does not constitute a decisive pointer” (van Eemeren, Houtlosser& Snoeck –Henkemans 2007:1)4
[iv] A definition of an endoxon is given by Aristotle: “opinions that are accepted by everyone or by the majority, or by the wise man (all of them or the majority, or by the most notable and illustrious of them)” (Topics, 100b.21)

REFERENCES
Bazzanella, C. (2005). Linguistica e Pragmatica del Linguaggio. Roma: Laterza.
Brown, G., & George, Y. (1983). Discourse Analysis. Cambridge: Cambridge University Press.
Calaresu, E. (2004). Testuali Parole: la Dimensione Pragmatica e Testuale del Discorso Riportato. Milano: Franco Angeli.
Conte, M – E. (1999). Condizioni di Coreferenza. Ricerche di Linguistica Testuale. Pavia: Edizioni dell’Orso.
Eemeren, F.H. van, & Grootendorst, R. (1992). Argumentation, Communication and Fallacies: a Pragma Dialectical Perspective. Hillsdale: Erlbaum.
Eemeren, F.H. van, Grootendorst, R., Jackson, S., & Jacobs, S. (1993). Reconstructing Argumentative Discourse. Tuscaloosa: The University of Alabama Press.
Eemeren, F.H.van, & Grootendorst, R. (2004). A Systematic Theory of Argumentation: the Pragma-Dialectical Approach. Cambridge: Cambridge University Press.
Eemeren, F.H.van, Grootendorst, R., & Snoek-Henkemans, F. (2002). Argumentation: Analysis, Evaluation, Presentation. Mahwah: L.Erlbaum.
Eemeren, F.H.van, Houtlosser, P., & Snoeck-Henkemans, F. (2007). Argumentative Indicators in Discourse. A Pragma-Dialectical Study. Dordercht: Springer
Miecznikowski, J., Rocci, A., & Zlatkova, G. (in Press). L’argumentation dans la presse économique et financière italienne. In L. Gautier (Ed.), Les discours de la bourse et de la finance. Forum für Fachsprachen-Forschung. Frank und Timme, Berlin.
Rigotti, E. (2006). Relevance of context-bound loci to topical potential in the argumentation stage. Argumentation, 20 (4), 519–540.
Rigotti E. (2009a). Whether and how classical topics can be revived within contemporary argumentation theory. In: F.H. van Eemeren & B. Garssen (Eds.), Pondering on problems of argumentation (pp.157-178). Dordrecht: Springer.
Rigotti, E. (2009b). Locus a causa finali. In G. Gobber, S. Cantarini, S. Gigada & S. Gilardoni (Eds.), Word meaning in argumentative dialogue (pp.559-577) Special issue of L’analisi linguistica e letteraria 2009/2. Milano: Vita e Pensiero.
Rigotti, E., & Greco- Morasso, S. (2006). Topics: the argument generator. Argumentation for financial communication. Argumentum eLearning module. Retrieved from www.argumentum.ch
Smirnova, A. (2009). Reported speech as an element of argumentative newspaper discourse. Discourse & Communication, 3(1), 79-103.