ISSA Proceedings 1998 – A Critique Of The Dialectical Approach: Part II

ISSAlogo19981. Introduction
This paper is part of a project designed to explore the nature of the dialectical approach in argumentation theory, its relationship to other approaches, and its methodological fruitfulness. The main motivation underlying this project stems from the fact that the dialectical approach has become the dominant one in argumentation theory; now, whenever a given approach in any field becomes dominant, there is always the danger that it will lead to the neglect or loss of insights which are easily discernible from other orientations; this in turn may even prevent the dominant approach from being developed to its fullest as a result of the competition with other approaches.
In a previous paper (Finocchiaro 1995), I undertook a critical examination of two leading examples of the dialectical approach. I argued that Barth and Krabbe’s (1982) demonstration of the equivalence of the methods of axiomatics, natural deduction, and formal semantics to formal dialectics works both ways, so that the former acquire the merits of the latter, and the latter the limitations of the former. I also argued that Freeman’s (1991) demonstration that the structure of arguments as products derives from the process of argumentation is insufficiently dialectical insofar as it involves a conception of dialectics in which dialogue is easily dispensable, and insofar as it suggests that argument structure is rooted more in an evaluative process than in a process of dialogue between distinct interlocutors.

In this paper I plan to examine the ideas of other authors who have written on or have used the dialectical approach. I shall use as a guide the following three working hypotheses suggested by the just stated conclusions reached in my previous paper. The first is the claim that if one takes the point of view of formal dialectics, the formal dialogical approach is not essentially different from the monological approach, but rather the two approaches are primarily different ways of talking about the same thing. The other two working hypotheses involve informal rather than formal dialectics. The second working hypothesis is that perhaps there are two versions of the informal dialectical approach, depending on whether one emphasizes the resolution of disagreements or their clarification. The third working hypothesis is that the dialectical approach is fundamentally a way of emphasizing evaluation, a way of elaborating the evaluative aspects of argumentation.[i] These are working hypotheses in the sense that I shall be concerned with testing their correctness, namely with determining whether they are confirmed or disconfirmed by other actual instances of the dialectical approach. Since I shall be examining only examples of the informal dialectical approach, I will be dealing primarily with the second and third working hypotheses.

2. Johnson on the Dialectical Approach
In their paper entitled “Argumentation as Dialectical,” Blair and Johnson (1987: 90-92) claimed that to say that argumentation is dialectical involves four things:
1. we should emphasize the process as well as the product;
2. the process involves two roles, that of questioner and that of answerer;
3. the process begins with a question or doubt, perhaps only a potential question or doubt; and
4. argumentation is purposive activity, in which there are two purposes corresponding to the two roles.

In his latest paper, Johnson (1996: 103-15) speaks more generally of a pragmatic approach and restricts the dialectical component to just one of three elements, the others being the teleological and the manifestly rational. The most basic feature is that argumentation is teleological in the sense that its aim is rational persuasion. For Johnson, the dialectical aspect of argumentation now becomes largely a consequence of the fact that it aims at rational persuasion. For now by dialectical Johnson means that argumentation must include answering objections and criticism. His own words are worth quoting: “That argumentation is dialectical means that the arguer agrees to let the feedback from the other affect her product. The arguer consents to take criticism and to take it seriously. Indeed, she not only agrees to take it when it comes, as it typically does; she may actually solicit it. In this sense, argumentation is a (perhaps even the) dialectical process par excellence)” (Johnson 1996: 107). Johnson then goes on to argue that, because argumentation is teleological and dialectical, it needs to be manifestly rational; that is, not only must it be rational, but it must
be so perceived by the participants.
It is beyond the scope of the present remarks to discuss Johnson’s account more fully. Here, the main thing I want to stress is his conception of the dialectical nature of argumentation. It obviously refers to a critical or evaluative element. He seems to be saying that arguing for a conclusion has two aspects: that of providing reasons and evidence in support of the conclusion, and that of taking into account counter-arguments and counter-evidence. Moreover, since this taking into account can take the form of either refuting the objections or learning something from them, it is clear that what is involved is not merely negative criticism of the objections but also positive evaluation, as the case may be.
Although Johnson’s notion of the dialectical is clear, there is an aspect of his discussion which is not so clear. The difficulty stems from the fact that he plausibly finds it useful to distinguish argument and argumentation, and on the basis of this distinction he seems to say that what is dialectical is argumentation, not argument. In his own words:
Although it seems clear that if the process of arguing is to achieve its goal, the arguer must deal with the standard objections, it is not clear that we would be wise to take this same view of the argument itself – else a great many arguments (which many times fail to deal with objections) would ipso facto have to be considered defective – this consequence seems unduly harsh [Johnson 1996: 104-5].
The issue here is whether we want to make dialectics – or evaluation in my terminology – an integral part of the process of arguing. Perhaps this issue could be described as involving two versions of the dialectical approach, in a strong and in a weak sense. The strong dialectical approach would make the evaluation of objections an essential part of the process of arguing, whereas the weak dialectical approach would make it only a part of a complete evaluation of an issue or claim. This is reminiscent of my distinction between the weak and strong dialectics discussed in my earlier paper.
Be that as it may, my conclusion here is that Johnson’s account is such as to support my working hypotheses, primarily the one about the evaluative nature of dialectics, and secondarily the one about the existence of two versions of the dialectical approach.

3. An Example of the Pragma-Dialectical Approach
My next example of a dialectical approach is Snoeck Henkemans’s (1992) account of complex argumentation. I take her work to be an excellent application and elaboration of the pragma-dialectical approach of the Amsterdam school. Examining her work can also serve here as a good substitute for examining the general framework of van Eemeren and Grootendorst’s approach because she deals with a relatively concrete and specific problem. The aim of her doctoral dissertation (Snoeck Henkemans 1992) was to give a pragma-dialectical analysis of complex argumentation, and in particular of the difference between multiple and coordinatively compound argumentation. Having used these terms, I should give some terminological clarification.
By complex argumentation is meant argumentation where a conclusion is supported by more than just a single reason, either in the sense that two or more reasons are given to support the conclusion, or in the sense that the reason which directly supports the conclusion is itself in turn supported by another reason. When two or more reasons support the same conclusion, the reasons may be completely independent of one another or inter-related to some extent. Snoeck Henkemans, following the Amsterdam school, speaks of “multiple” argumentation when the two or more reasons are completely independent. This case corresponds to what other scholars call convergent or independent reasons. When the two or more reasons are inter-related, she speaks of “coordinatively compound” argumentation; this corresponds to what others call linked, interdependent, cumulative, or complementary. When a reason that supports the conclusion is itself supported, she calls this case “subordinatively compound” argumentation; it corresponds to what others call serial structure or chain arguments. As if such terminological confusion were not enough, it ought to be remembered that the Amsterdam school also speaks of a “standpoint” to refer to a conclusion, and of an “argument” to refer to a reason.

One of Snoeck Henkemans’s (1992: 85-99) main accomplishments is to examine how these various structures result from various kinds of dialogue in which the proponent is involved in answering various kinds of criticism. In particular, multiple argumentation results when the proponent accepts some criticism of a premise and offers a new reason for the conclusion. Subordinatively compound argumentation results when the proponent tries to answer criticism of the acceptability of a premise. Coordinatively compound argumentation results when the proponent tries to answer criticism of the sufficiency of a premise. The case of criticism of the relevance of a premise generates subordinatively compound argumentation in which a reason is given for the unexpressed premise linked to the explicit reason.[ii]
This analysis is for the most part interesting, intelligent, and plausible. But I want to offer some critical observations. First, I would say that the upshot of Snoeck Henkemans’s analysis is to show primarily that and how complex argumentation is an attempt to overcome criticism of the conclusion, understanding that the criticism may be actual or potential. Now, I believe this thesis to be essentially correct, but it seems to me that it advances the evaluative approach more than the dialectical one. That is, it tends to show how argumentation is essentially a form of evaluation. I do not deny the presence of the dialectical element in the sense of dialogue, but I wish to stress that the purpose of the dialogue is to elicit evaluation. Thus, if the evaluation can be elicited by the proponent’s imagining of potential objections, then the dialogue is not essential. Of course, one may then speak, and the proponent of the dialectical approach do speak, of an internal dialogue, but that is just a manner of speaking.
Another striking aspect of Snoeck Henkemans’s analysis is that it exploits the notions of acceptability, sufficiency, and relevance of a reason or premise. In a sense what she is doing is to take these notions as relatively unproblematic, and to analyze complex argumentation in their terms. Although this is valuable, there is a difficulty here stemming from the fact that it is not always clear whether a given criticism is directed at the acceptability, or the sufficiency, or the relevance of a premise. This in turn implies that, despite its theoretical elegance, this theoretical framework is not too useful as a practical instrument for the analysis and understanding of actual argumentation.

A related difficulty stems from the artificiality of the dialogical situations examined. These dialogues are artificial in the sense that they are too atomistic. That is, like other proponents of the dialectical approach, Snoeck Henkemans tends to consider dialogues where the interchange involved bits of discourse that are too small to be realistic. The more realistic situation is one where the basic unit of discourse in a dialogue is already an instance of complex argumentation and the interlocutor’s criticism is itself another complex argument. To determine how the two relate requires that we begin with a non-dialogical analysis of each discourse, along the lines of what proponents of the dialectical approach would label a structural approach. This suggestion will be illustrated presently.
The critical conclusion suggested here is that Snoeck Henkemans’s analysis is not primarily dialectical but evaluative insofar as it is correct, and it is inadequate insofar as is is primarily dialogical.

4. Walton on the Dialectical Approach
In his latest book entitled Argument Structure: A Pragmatic Theory, Douglas Walton (1996) offers many insights which are beyond the scope of the present paper. One line of argument is, however, directly relevant; it is found in the first two chapters. There, Walton seems to argue that the dialectical approach is needed in order to properly distinguish argument from reasoning on the one hand and from explanation on the other.
He begins by admitting that argument is a special case of reasoning, namely reasoning which fulfills the probative function consisting of premises supporting a conclusion. But he claims that such probative reasoning must be viewed in a dialectical context. Doing this requires understanding that the probative function can be fulfilled in several different types of dialogue: critical discussions, negotiations, inquiry, deliberation, quarrels, and information seeking. In Walton’s own words, “what is characteristic … in all these contexts, is the existence of a proposition that is unsettled, that is open to questioning or doubt, and open to being settled by a dialogue exchange between (typically) two parties” (Walton 1996: 26).
Similarly, in regard to the distinction between argument and explanation, Walton aims to improve the best textbook definitions by adding a dialectical element. He regards as basically right the criterion advanced by Copi and Cohen (1990) which says the following about an expression of the form “Q because P”: “If we are interested in establishing the truth of Q and P is offered as evidence for it, then ‘Q because P’ formulates an argument. However, if we regard the truth of Q as being unproblematic, as being at least as well established as the truth of P, but are interested in explaining why Q is the case, then ‘Q because P’ is not an argument but an explanation” (Copi and Cohen (1990: 30). Walton objects that this applies only to critical discussions, and that in order to generalize the test one must ask two questions about the proposition at issue, namely:
1. Does the respondent doubt it or disagree with it, implying an obligation on the part of the proponent to support it with premises that provide reasons why the respondent should come to accept it as a commitment?
2. Is the proposition one the respondent is prepared to accept (or at least not to dispute), but desires more understanding of why it is so, or lacks clarification about it? [Walton 1996: 62]

It might seem as if there is an irreducible dialogical element here. This is especially true for those troublesome cases which have been advanced by various scholars as instances of reasoning which can be both arguments and explanations. However, Walton himself makes a number of qualifications the upshot of which is to suggest that the dialectical context is not that important after all, but may be mere window dressing on probative reasoning (for the distinction between reasoning and argument) and on the questionability of Q (for the argument-explanation distinction). In Walton’s own words:
Although this dialectical test focuses on the presumed attitude of the respondent (according to the evidence of the text of discourse in the given case), what is basic is the underlying type of conventionalized speech act and type of dialogue both participants are supposed to be engaged in. It is not the proponent’s, or the respondent’s, purpose that is the key to the argument-explanation distinction. It is the goal of the type of dialogue they are supposed to be engaged in, as a conventional type of social activity which has normative maxims and principles.
Explanation is one type of activity, argument another. But the key to testing in a given case is to look for the element of unsettledness … as indicated by the context of the discourse [Walton 1996:63].
My conclusion about Walton’s work is that his primary interest seems to be dialogues: to study their nature, structure, types, and so on. It is not surprising that such a study exhibits a deep dialectical component. Nor is it surprising that it leads Walton to study the relationship between dialogues and other things such as arguments, fallacies, and so on, and thus to study the dialectical elements of these other things. But such dialectical elements are things seen one when one is wearing dialogical glasses. One can choose to wear monological glasses, and then, for example, argument becomes probative reasoning, and the difference between argument and explanation becomes a matter of whether in “Q because P” the truth of Q is contextually problematic. This conclusion, of course, supports my first working hypothesis.

5. Examples of Concrete Argumentation
As a further test of my working hypotheses, I now want to examine some actual cases of argumentation. They are taken from The Federalist Papers, a work which is certainly well known as a crucial document of American history and as a classic of political theory, but which is largely unappreciated and little studied as a source-book of argumentation and material for argumentation theory. Yet, I would go so far as to say that it has few rivals in this regard as well.
There is no question, of course, that the context is one of a critical discussion, the main issue being whether not the U.S. Constitution should be ratified. The essays were written in 1786-1787, immediately after the constitutional convention in Philadelphia had written a constitution, which was then being considered for ratification by each of the original thirteen states. There is also no question of the dialogical, and to that extent dialectical, context in which pro-constitution arguments contained in The Federalist Papers were being advanced. However, to what extent the various ideas of the proponents of the dialectical approach are applicable remains to be seen.

Let us also readily admit that the authors of the federalist essays (Alexander Hamilton, James Madison, and John Jay) behave as good arguers in Ralph Johnson’s sense discussed above. That is, the federalists not only advance reasons and evidence favoring the ratification of the constitution, but they examine, criticize, and try to do justice to the objections and counter-arguments. But this same fact also shows that they are taking evaluation seriously, that they conceive their task of arguing for the constitution as involving inference, but also as involving evaluation. They know that to be effective they have to discuss the arguments on both sides, but rather merely “present” the arguments, they have to evaluate them. We can also agree with Johnson that this evaluative (or “dialectical”) requirement has to be used with care, and that there would be contexts in which it may be too harsh to apply it. A beautiful illustration of this problem is provided by what is perhaps one of the most ingenious of the federalist arguments, namely Madison’s argument that a large republic is more likely to controls the harmful effects of factions and the tendency for a tyranny of the majority.

Madison’s own words are worth quoting:
The other point of difference is
a. the greater number of citizens and extent of territory which may be brought within the compass of republican than of democratic government; and it is this circumstance principally which renders
b. factious combinations less to be dreaded in the former than in the latter.
c. The smaller the society, the fewer probably will be the distinct parties and interests composing it;
d. the fewer the distinct parties and interests, the more frequently will a majority be found of the same party; and
e. the smaller the number of individuals composing composing a majority, and the smaller the compass within which they are placed, the more easily will they concert and execute their plans of oppression.
f. Extend the sphere and you will take in a greater variety of parties and interests;
g. you make it less probable that a majority of the whole will have a common motive to invade the rights of other citizens; or
h. if such a common motive exists, it will be more difficult for all who feel it to discover their own strength and to act in unison with each other.
i. Besides other impediments, it may be remarked that, where there is a consciousness of unjust or dishonorable purposes, communication is always checked by distrust in proportion to the number whose concurrence is necessary.
j. Hence it clearly appears that the same advantage which a republic has over a democracy in controlling the effects of faction is enjoyed by a large over a small republic – is enjoyed by the Union over the States composing it.
k. In the extent and proper structure of the Union, therefore, we behold a republican remedy to the diseases most incident to republican government [Rossiter 1961:83-84].

Suppose someone were to criticize this argument by objecting that it is flawed because it does not even mention the problem that, for example, the constitution (allegedly) violates the principle of the separation among branches of government (insofar as federal judges are appointed by the executive branch). The latter objection was, of course, an argument against ratification, and the federalists did answer it in another paper (No. 47). However, what would be the point of criticizing this particular argument for this reason? The only thing such a criticism would accomplish would be a reminder that there are other issues that need to be examined besides the advantageous effects of size in regard to factions and majorities. In other words, the criticism would remind us that the argument in question is not conclusive, that by itself it does not establish the conclusion beyond any reasonable doubt. But this limitation would be easily granted by the federalists; indeed, it is implicit in the context. Thus, we may say that the criticism would be too weak, almost worthless.

This passage is also a good illustration of the problem of distinguishing explanation and argument. For this purpose, let us begin by noting that the argument supports its conclusion by explaining how and why the situation it describes would come about from the situation described in the premises. The passage basically examines the effects of a republic’s size on the the composition and behavior of factions and majorities, arguing that a large size produces greater justice and less abuse of power. This is similar, though more complex that the two examples from Stephen Thomas which Walton discusses. I believe that unlike Thomas, Walton would regard the passage as an argument and not an explanation. And I would agree with Walton. Despite the presence of explaining in the arguing, we do not have an explanation. And we do not have an explanation because the context is such that the issue is precisely whether or not large size has this claimed beneficial effect. On the other hand, despite the debate over ratifying the constitution which is in the background, I do not think we need to appeal to any dialectical or dialogical principles to arrive at this interpretation of the passage.
Finally, the passage can also serve as an illustration of the relative merits of the “structural” and the dialectical approaches in analyzing the complex structure of an actual piece of argumentation. It might seem that the question whether the passage is an instance of single or multiple argumentation would be easiest. If we try to apply any dialectical principles of analysis, such as those of Snoeck Henkemans discussed above, the first thing we realize is that we need to have identified a conclusion. Next, we need to identify at least two other propositions, each of which in some sense supports the conclusion. Then the dialectical questions would be whether the proponent accepts criticism of one but not of the other(s), or is trying to answer criticism of the sufficiency of each premise. Now, in the passage quoted above, in order to make any progress at this point, we would have to consider the first full sentence (a-b) as a conclusion and the second full sentence (c-d-e) and the third full sentence (f-g-h) as being each single propositions supporting the first (despite the fact that they each contain three clauses); and then the dialectical questions could plausibly be answered by saying that each full sentence is open to a potential charge of insufficiency.Thus the second and third sentences constitute coordinatively compound reasons supporting the first. The fourth sentence (i) might be taken as anticipating criticism of the acceptability of the third one; thus the two of them constitute a “subordinatively compound” structure. In regard to the fifth (j) and sixth (k) sentence, the most natural thing to say would be that (j) is a further conclusion supported by (a-b) and (k) a further conclusion supported by (j). However, in Snoeck Henkemans’s dialectical terminology, we would have to say that (j) answers or anticipates a criticism of the acceptability of (k), and (a-b) answers or anticipates a criticism of the acceptability of (j). Such dialectical terminology might be taken to be passably adequate. However, I suspect that such terminology can be seen to make sense only after the fact, namely to justify an analysis arrived at by other, more structural means.
In any case, one may also raise questions whether the rules are even passably adequate. The following passage can illustrate this point. It comes from the first federalist paper, where Hamilton outlines his plan for supporting the ratification in the subsequent essays. At one point he gives the following summary of the arguments to be developed:
My arguments will be open to all and may be judged by all. They shall be at least offered in a spirit which will not disgrace the cause of truth. I propose, in a series of papers, to discuss the following interesting particulars: – [l] The utility of the UNION to your political prosperity – [m] The insufficiency of the present Confederation to preserve that Union – [n] The necessity of a government at least equally energetic with the one proposed, to the attainment of this object – [o] The conformity of the proposed Constitution to the true principles of republican government – [p] Its analogy to your own state constitution – and lastly, [q] The additional security which its adoption will afford to [q1] the preservation of that species of government, to [q2] liberty, and to [q3] property. In the progress of this discussion I shall endeavor to give a satis-factory answer to all the objections which shall have made their appearance, that may seem to have any claim to your attention [Rossiter 1961: 36].

What is the structure of this reasoning?
First let us note that the conclusion is not explicitly stated in this passage, but it is easily formulated; it is that the constitution should be adopted. To make a long story short, I would say that (m) and (n) are coordinatively compound; that (l) and (m) are linked, and so are (l) and (n), that is, each pair is more intimately interdependent than is the case for coordinative compounding; and that there are five independent reasons, namely (l-m-n), (o), (p), (q2), and (q3).
In other words, here we have a case of “multiple argumentation”, where several independent arguments are given to support the ratification of the constitution. Yet the Amsterdam dialectical rules do not apply. It would be incorrect to say that the federalists accept (as valid) any criticism of the reasons given; they rather are aware of such criticism and try to answer it. Several distinct reasons are given not because the federalists think that any of them is invalid, but because none of them is sufficient. Why then, Snoeck Henkemans might ask, not regard the whole passage and the whole case in favor of the constitution as an instance of coordinatively compound, rather than multiple, argumentation?
There are two reasons for this. First, the five distinct arguments seem to me as different from each other as any arguments are which support the same conclusion. Thus, if this is not multiple argumentation, I doubt any would be. Second, even if we regarded the whole argument as a single one, and the various reasons as merely coordinatively compound, then we would need to make distinctions among different kinds of coordinative compounding. One kind would be that illustrated by the relationship among (l-m-n), (o), (p), (q2), and (q3); another would be illustrated by (m) and (n), or to be more precise by (l-m) and (l-n); a third one by (l) and (m) and by (l) and (n). Regardless of the labels used, the three kinds of relationships are different.

6. Conclusion
There seem to be theoretical-conceptual difficulties, as well as practical ones, with the dialectical approach. The theoretical difficulties cluster around such questions as the following. What is the relationship between actual and potential dialogue? Is actual dialogue really necessary for a dialectical approach? Is potential dialogue sufficient? Must we not make a distinction between atomistic dialogue consisting of an exchange of small units of discourse such as sentences or words, and more realistic dialogue consisting of the exchanges of relatively long pieces of structured discourse? If and to the extent that the latter is primary, does not the structuralist alternative to the dialectical approach acquire primacy? What is the role and importance of the resolution of disagreements, as contrasted with their clarification?[iii] What is the role of criticism and evaluation in the dialectical approach? What is the role of evaluation in argumentation? Is argumentation anything more than inference-cum-evaluation? Is an argument anything more that the defense of a claim from actual or potential objections?
The practical difficulties with the dialectical approach are that its application to actual argumentation suffers from many limitations. This appears to be true even when such argumentation occurs in the context of actual debates, dialogues, and controversies. None of this is meant to suggest that the dialectical approach should be abandoned. On the contrary, this criticism is offered in the hope that by taking it into account, the dialectical approach can become better and stronger.

i. In their new work, Fisher and Scriven (1997) elaborate an account of critical thinking which they label the ‘evaluative’ conception. I am inclined to think their work could be utilized to add further support to this hypothesis.
ii. Although Snoeck Henkemans criticizes the account advanced by James Freeman in some of his earlier papers, her own account is more similar to the one advanced in Freeman’s (1991) book on the topic.
iii. This type of issue is similar to that treated by Tannen (1998) under the label of “debate versus dialogue.”

Barth, E.M., and E.C.W. Krabbe (1982). From Axiom to Dialogue. Berlin: Walter de Gruyter.
Blair, J.A., and R.H. Johnson (1987). Argumentation as Dialectical. Argumentation 1, 41-56.
Copi, I.M., and C. Cohen (1990). Introduction to Logic. 8th ed. New York: Macmillan.
Eemeren, F.H. van & R. Grootendorst (1992). Argumentation, Communication, and Fallacies: A Pragma-Dialectical Perspective. Hillsdale, NJ: Lawrence Erlbaum Associates.
Eemeren, F.H. van & R. Grootendorst, eds. (1994). Studies in Pragma-Dialectics. Amsterdam: International Centre for the Study of Argumentation.
Eemeren, F.H. van, R. Grootendorst, & F. Snoeck Henkemans, eds. (1996). Fundamentals of Argumentation Theory: A Handbook of Historical Backgrounds and Contemporary Developments. Mahwah, NJ: Lawrence Erlbaum Associates.
Finocchiaro, M.A. (1995). The dialectical approach to interpretation and evaluation. In: F.H. van Eemeren et al. (Eds.), Perspectives and Approaches: Proceedings of the Third ISSA Conference on Argumentation (vol. 1, pp. 183-95), Amsterdam: International Centre for Study of Argumentation.
Fisher, A., and M. Scriven (1997). Critical Thinking: Its Definition and Assessment. Point Reyes, CA: Edgepress; and Norwich, UK: Centre for Research in Critical Thinking.
Freeman, J.B. (1991). Dialectics and the Macrostructure of Arguments. Berlin: Foris Publications.
Johnson, R.H. (1996). The Rise of Informal Logic. Newport News, VA: Vale Press.
Rossiter, C., ed. (1961). The Federalist Papers. New York: Mentor Books.
Snoeck Henkemans, A.F. (1992). Analysing Complex Argumentation. Amsterdam: SICSAT.
Tannen, D. (1998). The Argument Culture: Moving from Debate to Dialogue. New York: Random House.
Walton, D. (1996). Argument Structure: A Pragmatic Theory. Toronto: University of Toronto Press.