32.4 Xie Book Review PUBvers © Yun Xie. Informal Logic, Vol. 32, No. 4 (2012), pp. 440-453. Book Review Giving Reasons: A Linguistic-Pragmatic Approach to Argumentation Theory by Lilian Bermejo-Luque Springer, Argumentation Library, Vol. 20, 2011, Pp. xvi, 1-209. ISBN 978-94-007-1760-2 (Hardcover) €106.95 e-ISBN 978-94-007-1761-9 (eBook) €99.99 Review by YUN XIE It has been years since argumentation theorists started to readdress the relationship among different approaches to argu- mentation, with a hope of fostering some possible integration. Recently, more and more proposals have been developed, at- tempting to bring different approaches together into a compre- hensive framework to build some unified theory. Bermejo- Luque’s newly published book, Giving Reasons: A Linguistic- Pragmatic Approach to Argumentation Theory, is exactly an- other endeavor of this kind. The book shares the ambition of fa- cilitating “the integration of argumentation’s logical, dialectical rhetorical and epistemic dimensions” (p. viii), and tries to realize it by developing a novel linguistic-pragmatic theory of argu- mentation, one that conceives of argumentation as some particu- lar speech-acts, and unpacks its normativity into the linguistic and pragmatic nature of argumentation. The central thesis ar- gued in this book is that argumentation should be characterized as a speech-act complex, whose constitutive goal is showing a target-claim to be correct, and accordingly, that the appraisal of argumentation involves both a semantic evaluation determining the correctness of its target-claim and a pragmatic evaluation determining its goodness in showing that correctness. Chapter 1 gives a general introduction of “Argumentation and Its Study”. Argumentation has a widespread presence in human interaction, since it is not only “closely connected to the specifics of human language and communication” (p. 2), but it A Linguistic-Pragmatic Approach to Argumentation Theory 441 has also been valued as something important and good for “hu- mans as both rational and social beings” (p. 2). However, Ar- gumentation Theory as a discipline has only a late emergence, due to the influence of the developments of western philosophy (pp. 6-8). Here Bermejo-Luque defines “Argumentation Theory as a discipline” to be only “the study of argumentation from a normative perspective”, specifically attempting to address four founding questions: “what is argumentation?” “How should we interpret and analyze argumentative practices?” “What is good argumentation?” and “How can we determine argumentation goodness?”(p. 9). Argumentation theorists, when answering these questions, have taken distinct approaches. They emphasize different aspects of arguing, and differ from each other on the definition, analysis and appraisal of argumentation. However, their integration can be adequately achieved, so Bermejo-Luque claims, by her linguistic-pragmatic approach forthcoming in the remaining chapters (p. 10). If Argumentation Theory is to be understood essentially as normative, then what is an adequate way to conceptualize the idea of argumentative value, and how can we justify the norms or normative models of argumentation developed accordingly in our theories? Considerations about these “crucial meta- theoretical questions” are the content of Chapter 2, Why Do We Need a New Theory of Argumentation. Bermejo-Luque seriously doubts that we already have some satisfactory theory with re- gard to argumentation goodness. The traditional account of ar- gumentative value has endorsed a “deductivist ideal of justifica- tion”, thus proposing a characterization of “argumentation goodness only in terms of the status of its premises and their in- ferential relations to its conclusion, ignoring the pragmatic con- ditions of argumentation as a communicative activity” (p. 21). To counterbalance the traditional “semantic-deductivist” ap- proach, which has severe limitations, contemporary argumenta- tion theory gives rise to a “pragmatic account of argumentative value” (p. 23). But this new account, as criticized by Bermejo- Luque, is essentially an “instrumentalist conception of argumen- tative value”, according to which “good argumentation would be a matter of its ability to achieve the typical ends which those en- gaged in the practice of arguing were aiming at” (p. 23). This implicit instrumentalism fails to distinguish argumentation goodness from argumentation success. It is doomed to result in Yun Xie 442 some unacceptable form of relativism, and makes it senseless for us to talk about good argumentation simpliciter (p. 33). Ac- cording to Bermejo-Luque, “it is a mistake to think that argu- mentation goodness consists in any type of perlocutionary achievement” (p. 33), even if we add to this achievement some additional “qualifications” which “demand more than mere per- locutionary effectiveness”, such as Perelman’s “idealized uni- versal audience”, Johnson’s “rational persuasion” or Pragma- dialecticians’ “reasonable resolution of a difference of opinion” (pp. 26-30). Basically, all those instrumentalist conceptions of argumentative value provided by contemporary argumentation theorists confuse “intrinsic argumentative value” with “the in- strumental value that a piece of argumentation may have in rela- tion to some further (extrinsic) end” (p. 34). Therefore, a new theory of Argumentation is needed, which should rightly take argumentation’s intrinsic value as its central concern, and avoid characterizing it in instrumental terms. But what exactly is this intrinsic value of argumentation? And how can we unpack it in some non-instrumental way? Bermejo-Luque gives her answers by defending a “constitutivist” conception of argumentative value, which draws a distinction between the “constitutive goal” and the “additional goals” pur- sued in our acts of arguing. In particular, “the activity of arguing is, constitutively, an attempt at justifying a target-claim”, so “justification is the intrinsic argumentative value”, and “arguing well is justifying” (pp. 38-39). More importantly, “this account of argumentative value in terms of justification is meant to be completely empty” (p. 39), thus it could avoid any possibility of falling into some form of instrumentalism. By taking justifica- tion as being both the constitutive goal and the intrinsic value of argumentation, Bermejo-Luque claims that we have not only captured our pre-theoretical concepts of argumentation and of argumentation value, but also found the solution to the justifica- tion problem of normative models (pp. 44-45). So, in the end, one final relevant question remains: “What does justification, so understood, consist in?” and the answer, according to Bermejo- Luque, turns out to be dependent on “the very characterization of argumentation that we endorse” (p. 39). Naturally, Chapter 3, Acts of Arguing, is devoted to a char- acterization of argumentation, proposed as the theoretical object A Linguistic-Pragmatic Approach to Argumentation Theory 443 for the new linguistic-pragmatic theory to be developed in this book. Bermejo-Luque firstly identifies justification, the constitu- tive goal of argumentation, as showing the correctness of the target-claim by giving reasons. Then she characterizes argumen- tation as a linguistic-pragmatic object. On the one hand, an act of arguing is indeed a “speech-act complex” which is “com- posed of two further speech-acts, namely, the speech-act of ad- ducing and the speech-act of concluding” (p. 60). On the other hand, these speech-act complexes are “second order” in nature, because “they can only be performed by means of a first order speech-act—namely, constative speech-acts” (p. 60). Specifi- cally, a speech-act will be interpreted as “second order” when it is possible to point out “a certain relationship with another speech-act” which turns this speech-act into another kind of illo- cution (p. 61). Regarding argumentation, it consists of two con- statives, R and C, which become respectively the second order acts of “adducing (a reason)” and “concluding (a target-claim)”, and “this occurs because of their relationship to each other by means of an implicit inference-claim whose propositional con- tent is ‘if r, then c’” (p. 60). Obviously, in order to interpret cer- tain speech-acts as acts of arguing in this way, as Bermejo- Luque claims, “we will have to make a presumption concerning the relationship between R and C”, and this amounts to “attribut- ing to the speaker an implicit assertion I, whose content is ‘if r (the content of R), then c (the content of C)’” (p. 61). This im- plicit inference-claim is also “constitutive of any act of arguing” (p. 62), and its propositional content is simply a “particular in- dicative conditional” (p.62) of which “a truth-functional inter- pretation” is favored (p. 64). Bermejo-Luque then adopts Bach and Harnish’s Speech-act Schema (SAS) to characterize the acts of arguing understood as a second order speech-act complex, and illustrates in detail with an example (pp. 64-68). This speech-act complex characterization of argumentation, according to Bermejo-Luque, has two advantages. First, it can provide “a unitary account of argumentation as a justificatory and as a persuasive device” (p. 56). That is, acts of arguing are characterized as speech-acts “whose illocutionary force is that of an attempt at showing a target-claim to be correct” (p. 58), while their perlocutionary effect is inducing reasoning in others, or “an invitation to inference” to the addressee (p. 73), which, as a re- sult, explains the practical achievement of producing beliefs or Yun Xie 444 persuading the others (p. 58). Second, this characterization of argumentation facilitates a possibility to integrate the logical, dialectical and rhetorical dimensions of argumentation (p. 54), since “justifying a claim will be equivalent to showing it to be correct”; and “in order to do that it will have to satisfy not only logical and dialectical conditions, but rhetorical conditions as well” (p. 58). The integration of the three dimensions, regarding both the analysis and evaluation of argumentation, is the main theme of the following four chapters. Chapters 4-6 attempt to develop these three dimensions respectively, and Chapter 7 aims to ex- plain their integration in more detail. Bermejo-Luque deals with “The Logical Dimension of Argumentation” first in Chapter 4. The basic position argued in this chapter is that Logic should be understood as a “non-formal normative theory of inference” whose goal is to characterize a “non-formal concept of validity” (p. 96). Particularly, the logical dimension of argumentation fo- cuses on its “semantic properties”, and thus plays an important role in argumentation evaluation by providing criteria for deter- mining the goodness of inferences that supervene in our acts of arguing (p. 88). To defend such a particular conception of Logic, Bermejo-Luque relies heavily on Toulmin’s ideas of logic in The Uses of Argument (1958). She argues that Toulmin has al- ready proposed this promising characterization of logic (p. 81), but he fails to see a crucial distinction between “the real objects on which inferences supervene—namely, acts of arguing and reasoning—and their semantic representations” (p. 82). So Ber- mejo-Luque distinguishes strictly the communicative acts of ar- guing, or argumentation, from “arguments”, which are con- ceived of as “just theoretical reconstructions”, namely, each be- ing “a representation of the semantic and syntactic properties of an act of arguing” (pp. 56-57). The logical study of argumenta- tion takes argument, or more specifically, the inferential struc- ture represented in argument, as its object, developing normative theories with regard to its goodness. In line with the criticisms made by Toulmin against the formal approach to Logic, Ber- mejo-Luque has proposed a broader concept of “non-formal va- lidity” that is “not analytic but pragmatic” (p. 96), according to which “valid arguments are arguments whose warrants represent correct claims, in the sense that they entitle us to put forward our conclusion with the qualifier with which we have actually put it A Linguistic-Pragmatic Approach to Argumentation Theory 445 forward” (p. 98). As a result, logical evaluation becomes a task to “determine which qualifiers really correspond to the premises and warrants that represent the reason and the inference-claim of the corresponding act of reasoning and arguing” (p. 100), and on that basis, to judge “whether the correctness of premise and the warrant actually make the conclusion, as qualified in the act of arguing, correct” (p. 175). Accordingly, in the logical dimension “the working concept for argument appraisal will be the concept of qualifier” (p. 168). Bermejo-Luque further distinguishes two kinds of qualifier: ontological and epistemic. She believes that this distinction is expressed in two forms of our uses of modal terms. When we say “p is true”, “p is necessary” or “p is probable”…“we are saying something about its representativeness respecting the world” (p. 170). This type of modal term is called an “ontologi- cal qualifier”, by means of which we express “the type of prag- matic force with which we put forward the corresponding con- tents in claiming” (p. 170). By contrast, when we say “it is likely that p”, “it is necessary that p”, or “probably p”…“we are saying something about the status of this claim as knowledge, about the confidence we should put in this claim or our entitlement to it” (p. 170). This type of modal term is called an “epistemic quali- fier”; with it we make “an illocutionary act of concluding” (p. 171). Moreover, by imposing the ascription of qualifiers to all the elements of argument (i.e., premise, conclusion, warrant and backing, rebuttal), Bermejo-Luque puts forward an extended Toulmin model of argumentation (p. 115), in which the qualifier of the conclusion is epistemic, while all the others are ontologi- cal. Thusly, the proposed “non-formal concept of validity” is now better clarified: “For an argument to be valid its conclusion has to be qualified with an epistemic qualifier which corre- sponds to the ontological qualifier that correctly qualifies its warrant” (p. 176). However, as we can imagine, sometimes in reality our ac- tivity of arguing involves not just a neat complex of claiming the reason (and the inference-claim) and concluding the target-claim, but also some more efforts to justify or defend our reasons and inference-claims, especially when they are controversial or chal- lenged. That is to say, our acts of arguing actually need to ex- pand in some way in order to succeed in achieving their com- municative goal. Bermejo-Luque regards this aspect of argu- Yun Xie 446 mentation as its “dialectical dimension”, and proposes to explain it in terms of “the recursive nature of acts of arguing” (p. 120). Chapter 5, The Dialectical Dimension of Argumentation, is focused on analyzing the dialectical properties of acts of argu- ing. It is first argued that “argumentation maybe said to be re- cursive in its development and in its very nature as a procedure” (p. 127). On the one hand, “any argumentative discourse may nest additional acts of arguing for any of its claims” (p. 127), no matter whether it is viewed as a justificatory device or as a per- suasive device. On the other hand, “in being a means to show that our claims are correct or to induce beliefs in others, its de- velopment as further argumentation is the way any act of argu- ing is able to warrant its own cogency as a justificatory de- vice…or its own legitimacy as a persuasive device” (p. 127). This “twofold recursive nature of argumentation”, according to Bermejo-Luque, is at the base of some “second order inter- subjectivity of argumentative communication”, a term that refers to an assumption of, or an appeal to, “a shared theoretical ra- tionality” and the arguers’ “ability to acknowledge the justifica- tory power of good reasons” (pp. 123-128). Because of its recur- sivity, argumentation inevitably needs to expand dialectically, by taking additional moves and by going through some specific procedures. Following Rescher (1977), Bermejo-Luque contin- ues to characterize dialectics “as the kind of activity that con- sists of certain basic discursive moves that can be combined in different ways” (p. 128). On that basis, a set of basic dialectical moves is proposed which is also a “set of the constitutive moves of acts of arguing” (pp. 131-132), and some dialectical proce- dures are distinguished and discussed, such as “weak” and “strong” opposition dialectical procedure (pp. 133-135). Moreo- ver, Bermejo-Luque also takes pains to argue that “the practice of arguing is, at a minimum, an als ob activity regarding objec- tivity” (p. 137), with a hope to “solve a common difficulty found in current dialectical approaches to argumentative normativity, namely, the tendency to miss the grip of objectivity as the raison d’être of the activity of giving and asking for reasons” (p. 120). Chapter 6, as expected, deals with the role of Rhetoric in Argumentation Theory, and “The Rhetorical Dimension of Ar- gumentation”. Unlike those theorists who are reluctant to grant the value of Rhetoric in their (normative) argumentation studies, Bermejo-Luque insists that “argumentation always has a rhetori- A Linguistic-Pragmatic Approach to Argumentation Theory 447 cal dimension” (p. 139), and aims to defend the important role “that Rhetoric is to play in developing normative models shap- ing the concept of argumentative value” (p. 139). In general, she argues in this chapter that there are three roles that Rhetoric can play in argumentation studies: “(1) to facilitate its interpretation; (2) to make possible the appraisal of its rhetorical value, i.e. its value as a persuasive device; and (3) to make possible the ap- praisal of its argumentative value, i.e. its value as a justificatory device” (p. 140). To unpack these ideas, Bermejo-Luque pro- poses to characterize “the rhetorical dimension of argumentation as, in general, independent of speakers’ intentions” (p. 148), and to value the contemporary view of Rhetoric as providing tools for interpreting communication as a means of influence. As she has argued, the rhetorical properties of a piece of communica- tion depend on some “causal power to influence individuals”, which is to be “understood in terms of what would be a normal response to such piece of communication” (p. 150). That is to say, we should take some elements or features of the perform- ance as causes that would normally produce the corresponding effects. Consequently, the rhetorical analysis of a performance, i.e., interpreting it as a rhetorical device, is a matter of “discov- ering what we may call its rhetorical import” (p. 151), defined as “the sort of rhetorical effects that it is likely to produce in the addressee, given the circumstances, if the addressee responds in a ‘normal’ way” (p. 153). When an act of arguing is analyzed as a rhetorical act, ac- cording to Bermejo-Luque, it “would motivate our inferring; that is, it would exercise a causal influence on us” (p. 156). In the appropriate circumstances, it will normally result in our coming to believe its target-claim in an inferential way, which she calls “indirectly judging” (p. 156). Accordingly, indirect judgments of the form “target-claim since reason” constitute “the rhetori- cal meaning of the act of arguing as a rhetorical act”, and induc- ing such a kind of indirect judgments is exactly the arguers’ rhe- torical intention (pp. 156-157). Here traditional Rhetoric can be of help in our evaluation of arguers’ ability to satisfy this rhe- torical intention, i.e., determining argumentation’s value as a rhetorical device. However, since “arguing well would depend both on the actual correctness of the target-claim and on the goodness of the argumentative act as a means for showing this” (p. 159), therefore, when evaluating argumentation’s value as a Yun Xie 448 justificatory device, we will also need to bring in the contempo- rary “hermeneutic, non-intentional conception of Rhetoric” (p. 158) to “deal with the pragmatic conditions which determine how good an act of arguing is as a means of showing” (p. 159). With regard to this part of rhetorical argumentation appraisal, Bermejo-Luque develops a proposal by adopting Grice’s Coop- erative Principle as the regulative conditions of acts of arguing, for “the Cooperative Principle is a standard for any talk ex- change aimed at being ‘a maximally effective exchange of in- formation’, that is, aimed at ‘showing something’” (p. 161). The last chapter, Chapter 7, is titled “Argumentation Ap- praisal”, in which an overall framework of the normative model of argumentation is given. Good argumentation, according to this linguistic-pragmatic theory, is argumentation “showing its target-claim to be correct” (p. 165), which involves both a “se- mantic appraisal” and a “pragmatic appraisal”. Semantic condi- tions, provided by the logical dimension, determine the correct- ness of the target-claim of acts of arguing, while pragmatic con- ditions, provided mainly by rhetorical dimension, determine the quality of acts of arguing in showing that correctness. A dialec- tical dimension, however, will be involved in both (p. 166). Se- mantic appraisal takes the concept of “qualifier” as the key ele- ment, whereas the pragmatic appraisal uses the Cooperative Principle as a regulative condition, and regards the distribution of the burden of proof as another essential tool (pp. 192-193). To illustrate some applications of this theory, Bermejo-Luque gives a detailed analysis of enthymemes and incomplete argu- mentation, as well as of some traditional fallacies (pp. 189-192). Particularly, she argues that enthymemes are indeed “not ‘in- complete’ in the sense of lacking something that is constitutive of any act of arguing, but it would only be ‘acts of arguing lack- ing inference-claim backings’” (p. 185). And concerning the pragmatic failures of acts of arguing within this new model, she indicates that some new “kind of argumentation failure” be- comes discernible. This new kind of failure, as opposed to “bad argumentation”, is named “false argumentation”. As Bermejo- Luque attempts to show, it refers to acts “that fall short of being ‘real’ argumentation” (p. 194), i.e., they “take the place of ar- gumentation to pretend to be real argumentation”, but indeed violating the constitutive normativity of acts of arguing (pp. 199-200). A Linguistic-Pragmatic Approach to Argumentation Theory 449 In general, Bermejo-Luque’s Giving Reasons is an impor- tant contribution to the field. The chapters and sections of this book are clearly structured, and many of the arguments are well elaborated, and easy to follow. Most importantly, the discus- sions of many topics and issues are grounded on the author’s extensive reading and good understanding of contemporary ar- gumentation literature, as well as other relevant fields; hence many conclusions reached in this book are quite illuminating. It might be better, I think, if there could be one more chapter in the end summarizing all the basic ideas and contributions, together with a full-fledged application of the proposed new framework to some particular and illustrative example. As also a young scholar in this field, I am truly impressed by the depth and inno- vations Bermejo-Luque has achieved in her theorizing about ar- gumentation in this book. But still, I also find some points she has made remain controversial, and some of her arguments in need of further development. However, here I will restrict my- self to comment only on the following two respects. To begin with, I find that some of Bermejo-Luque’s critical readings of other argumentation theories are wanting. When urg- ing a need for her own new theory, she took a critical attitude towards many prominent and influential theories developed in recent literature: Pragma-Dialectics, Informal Logic, Rhetorical Argumentation theory, Epistemological Approach to argumenta- tion, etc. She aims to show that all of them have fundamental weaknesses, but her treatments of these theories are hasty, and her criticisms turn out to be premature. For one example, she has criticized Ralph Johnson’s theory as being unable to provide a suitable understanding of good argumentation, on the basis of an analysis on the concept of “the rational persuasion of an ad- dressee”, which Johnson has regarded as the primary purpose of arguments. Bermejo-Luque examines the concept of “rational persuasion” from different points of view, arguing that, under- stood in some possible ways, it either cannot characterize argu- mentation goodness appropriately, or would only be reduced to “a matter of the rationality of a claim” (pp. 28-30). It is not very clear to me how and why this analysis goes against Johnson’s theory, since here strangely Bermejo-Luque didn’t specify any- thing about Johnson’s own account of rational persuasion at all, nor did she discuss any of the criteria Johnson has developed to Yun Xie 450 characterize argumentation goodness. It appears as though she is just focusing too much on the concept, while talking past John- son’s theory. Moreover, the basic strategy Bermejo-Luque has taken to criticize all the other contemporary theories is to charge them with the failure to distinguish the constitutive value and instru- mental value of argumentation, and their inability to provide suitable criteria for argumentation goodness. As she has tried to prove, an epistemic account of constitutive value should be pre- ferred, according to which “good argumentation would be ar- gumentation able to justify a target-claim, in the sense of show- ing it to be correct” (p. 23). Hence, all her criticisms will even- tually boil down to the fundamental question of “why should we prefer such an account and thereby use it to judge the other theo- ries as being faulty”? Or, “why should we characterize the acts of arguing, in the way she has proposed, as constitutively at- tempts at justifying?” The only available answer I could find in the book appears to be grounded on her analysis of arguing as an intentional behavior (pp. 37-39), emphasizing that “in principle, we can argue and then aim or not aim at persuading a universal audience or an addressee in a reasonable way or at resolving a difference of opinion. This is why none of these goals is consti- tutive of arguing…. Yet…it would be senseless to say that we are arguing for a claim but not trying to justify it” (pp. 38-39). However, considering the recent disputes on the issue of the primary use of argument, I am reluctant to take this as a con- vincing argument for the supposed preference for “justification”. Those who favor the option of “persuasion” would also observe that every act of arguing in reality is initiated by controversy or disagreement, thus they could also claim, in a similar way, that “it would be senseless to argue if not trying to persuade (the other or herself)”. Even though I am quite sympathetic to Ber- mejo-Luque’s position, I believe here she still needs further support for her position. In general, she is certainly right to propose that our judg- ment whether an act of arguing is senseless or not will rely ex- actly upon our formulation of the constitutive condition(s) of acts of arguing. It’s always easy to bring forward some possible formulation, but due to the complexity of our argumentative practices, it’s quite hard to prove that it is the right one, espe- cially when your proposed formulation involves the intention or A Linguistic-Pragmatic Approach to Argumentation Theory 451 goal of arguing. Furthermore, if we really need to think of acts of arguing constitutively, I think Bermejo-Luque has indeed also hinted at another and better option—giving reasons—which is just the title of this book! The claim that “it would be senseless to say that we are arguing for a claim but not giving reasons” seems to be much more plausible and defensible. But Bermejo- Luque, surprisingly, chooses instead a particular understanding of “justification”, i.e., “showing the target-claim to be correct by giving reasons”. It’s a richer and more complicated proposal, but at same time, it also brings with it a much heavier burden of proof, namely, to justify the additions of “showing” and “to be correct”, about neither of which is there consensus among con- temporary argumentation theorists. For the second aspect to comment about, I find the logi- cal dimension of argumentation developed in this book to be, though subtle, still in need of further improvements. The basic idea underlying this dimension, as I understand it, is that a logi- cally good argument is one whose qualifier of its conclusion rightly matches the strength of support given by its reasons. Along with this idea, Bermejo-Luque has emphasized the key role of “qualifier” in argument. She not only extends the Toul- min Model by imposing the ascription of qualifiers to all ele- ments, but has also advanced a proposal that “an ascription of qualifiers would be constitutive of arguments, and [logical] ar- gument evaluation would be the process of determining the right ascription of qualifiers to each represented claim” (p. 115). I believe many readers would be in the first place very sur- prised to see a model of argument filled with qualifiers, because it’s just so unnatural when compared to our real life arguments. It certainly seems that, when arguing, we only occasionally qualify our conclusions, and more often than not, we just simply put forward conclusions as being right or acceptable, without any explicit qualifier. Moreover, regardless of this practical oddness, it still remains unclear in Bermejo-Luque account why all these qualifiers are needed, and how they could contribute to the logical analysis and evaluation of argument. As Bermejo- Luque has envisaged it, “all we need to evaluate an argument [logically] is to be able to determine what all these qualifiers in argument should be” (p. 109). And what she really means by this claim is that we are “to determine what the actual or correct ontological qualifiers that we should ascribe to the premise and Yun Xie 452 the warrant”, and then “to see whether the epistemic qualifier of the conclusion have been qualified accordingly” (p. 177). Moreover, for the concept of “non-formal concept of validity” that she has proposed for the logical value of argumentation, it bears entirely upon the correspondence of qualifiers between conclusion and warrant (p. 176). So we could actually find that there is no reference to the uses of qualifiers other than those ascribed to warrant and conclusion. Then why do we need to complicate the model of argument in the first place by imposing qualifiers to all elements? Here I believe the only way to defend this complicated model is to argue that all these qualifiers are interrelated in their ascriptions. That is, without first considering the right qualifiers ascribed to backing, premise and rebuttal, it is not possible to ascribe qualifiers of warrant and conclusion and then to determine the goodness of argument logically. But interestingly, this just leads us to see a defect of Bermejo- Luque’s theory, namely, its lack of a sufficient account of the interrelationship among those qualifiers of different elements, explaining the mechanism in which their ascriptions would in- teract with, or be influenced by, each other. As we have seen, Bermejo-Luque is concerned solely in this book with the relationship between the qualifiers of the conclu- sion and the warrant (understood by her as an inference claim in a form of ‘if reason, then conclusion’), about which she con- tends that “for an argument to be valid its conclusion has to be qualified with an epistemic qualifier which corresponds to the ontological qualifiers that correctly qualify its warrant” (p. 176, italics mine). However, even if we restrict ourselves to consider only the qualifiers of these two elements, I think this simplified version of a correspondence mechanism also leaves room for doubt or need for further clarification. What exactly are the pos- sible or acceptable ways of correspondence between the onto- logical qualifier of warrant and the epistemic qualifier of con- clusion in Bermejo-Luque’s account? On the one hand, she ap- pears to interpret it in a way that “the epistemic qualifier should be identical (in modal terms) with the ontological qualifier used”. For example, she believes that “if we want to conclude that ‘necessarily p,’ we have to ontologically qualify our infer- ence-claim as necessary; if this inference-claim is true but not necessary, then our act of arguing will be semantically flawed” (p. 176). And she also explains that, “in order to know what on- A Linguistic-Pragmatic Approach to Argumentation Theory 453 tological qualifier the speaker attributes to her inference-claim, we just need to look at the epistemic qualifier of the target-claim” (p. 183). But on the other hand, it appears as though Bermejo- Luque also needs to make this correspondence relationship to be non-identical, in order to enable her to speak of degrees of justi- fication. As she has argued, “we can make sense of the concept of the justificatory power of an argument by considering what the strongest qualifier is that we can ascribe to its conclusion, given the qualifiers that actually correspond to its premise and warrant” (p. 178). Besides, she admits likewise that “there are epistemic qualifiers without a straightforward ontological coun- terpart” (p. 173), and that “in general the criteria for using a par- ticular epistemic qualifier are dependent upon the sort of evi- dence at the speaker’s disposal” (p. 173). As a whole, I am not sure whether Bermejo-Luque has provided us with a coherent and satisfactory account on the correspondence between epis- temic and ontological qualifiers. Nor am I inclined to agree that the possible interactions among qualifiers of any other elements of argument would be clearer or easier to handle by some corre- spondence mechanism. Anyway, even if a good argument really has a qualifier for its conclusion that rightly matches the strength of support given by its reasons, I think its ascription must depend upon the strength of reasons and the force of the supportive link involved, both of which are subject to their particular uses in different contexts. It is not clear whether, and how, the strength of rea- sons and the force of the supportive relation could be simply characterized by some ontological qualifiers, which in Bermejo- Luque account indicate the “representativeness respecting the world” (p. 170). And it is even harder to prove that the right qualifier for a conclusion is easily determined by some straight- forward function of those qualifiers ascribed to other elements in the argument. Yun Xie Institute of Logic and Cognition, Department of Philosophy Sun Yat-sen University Guangzhou, 510275 China xierobert2005@gmail.com