In the introduction (p Critical Notice REASON RECLAIMED. ESSAYS IN HONOR OF J. ANTHONY BLAIR AND RALPH H. JOHNSON. Edited by HANS V. HANSEN AND ROBERT C. PINTO. Newport News, VA: Vale Press, 2007. Pp. xiv, 1-313. ISBN:0-916475-29-8, US$ 34.95 paper. Reviewing a Festschrift is a rare and precious honor, allowing the reviewer to add his voice, however small, to the chorus of intellectual praise that the volume is meant to convey. This is certainly the case with this collection of nineteen cutting-edge contributions on several aspects of argumentation theory, dedicated to the career and achievements of J. Anthony Blair and Ralph H. Johnson, among the pioneers of informal logic. Aside from sheer scientific excellence, the chief virtue of this volume lies in the variety of topics investigated by its contributors, who also happen to represent almost every existing approach to the study of argument. Such diversity is, of course, also the main limitation of this collection. Any reader looking for well-focused treatment of a single topic or extensive coverage of any specific theory will be better off looking elsewhere. On the other hand, anyone interested in braving the richly diverse landscape of argumentation theories will find here an excellent vantage point. Besides, the open-ended structure of the book reflects the scope of interests addressed in their long career by the scholars to whom the volume is dedicated. The fact that some contributions are only loosely related to the works of Blair and Johnson just adds to the variety, without subtracting from the quality—which is, without exception, excellent. Indeed, the sole person who might complain for this kaleidoscopic abundance is the reviewer of this book, who faces the arduous task of concisely presenting so many diverse contributions, with little hope of finding any systematic connection among them. However, let us try, following the thematic partition suggested by the volume editors, Hansen and Pinto. 1. The first six papers cover the dialectical dimension of argumentation theories, again from a variety of angles. Walton and Godden focus on “the historical development of Blair and Johnson’s views on dialectics,” as we are told in the editors’ introduction (p. xii). This description, although factually true, is a bit of an understatement, since the main concern of Walton and Godden in their article is in fact to contrast Blair and Johnson’s view of dialectics with Walton’s dialogical approach to argumentation. After reviewing several different positions expressed by Blair and Johnson over the last thirty years, it is concluded that their © Fabio Paglieri. Informal Logic, Vol. 28, No. 2 (2008), pp. 170-192. Hansen & Pinto: Reason Reclaimed 171 approach seems to remain consistently dialectical, without turning dialogical. This distinction is then paralleled with the difference between argument-as-product and argument-as process. Even if Walton and Godden concede that a focus on both aspects is legitimate, they express a marked preference for the latter view—a preference, they suggest, that is not completely shared by Blair and Johnson. The ultimate conclusions of this comparison are quite ecumenical, to the point that they cannot really be disagreed upon: “Sometimes a product-based approach is our best bet, or our only bet, given the information we have about the situation of a particular argument. We do not deny this, and agree that several different perspectives can be taken in the study of argument. Yet, we emphasize that our understanding of the nature, purpose, workings and success of argument is deeply enriched by adopting a dialogical perspective whenever possible” (p. 17). Fair enough. More interesting, however, are two specific claims that Walton and Godden systematically use in their analysis: the idea that argument identification and classification presupposes reference to a dialogical context, so that we cannot even label a speech act as an argument without reference to the dialogue where it occurs, and the thesis according to which dialogical approaches are intrinsically richer than dialectical ones. Arguments are provided in support of these claims, and both are used to cross-examine Blair and Johnson’s view of argumentation. While to my mind the outcome of this debate is far from being settled, Walton and Godden’s analysis certainly provides a useful contribution to it, and helps emphasize important differences between various conceptions of dialectics and dialogue. In his richly textured essay, Freeman addresses the question whether or not Johnson’s analysis of the dialectical tier is consistent with heuristics. Ultimately, Freeman is interested to defend Johnson’s position against two criticisms raised by Govier—namely, that Johnson’s dialectical tier implies an infinite regress, and that we lack any practical method to distinguish what Johnson calls ‘standard objections’ from non- standard ones. As for the first objection, the problem is that the dialectical tier of an argument consists of those objections that need to be answered for that argument to be satisfactory. According to Govier, this implies supplementing the original argument with sub-arguments designed to rebut such objections, and these arguments in turn will have their own dialectical tier, requiring further sub-sub-arguments with a further dialectical tier, and so on ad infinitum. As for the second criticism, known also as the discrimination problem, Govier poses the question of what alternatives and objections against a given argument need to be addressed by the proponent, in order to satisfy Johnson’s dialectical criteria for validity. The dilemma here is that, on the one hand, demanding all possible alternatives and objections to be met would impose an exceedingly stringent burden over the arguer, while, on the other hand, discriminating among them poses the difficult challenge of defining suitable criteria for dialectically relevant alternatives and objections. After summarizing this debate, Freeman focuses on one of the Fabio Paglieri 172 solutions proposed by Govier to the infinite regress problem, the so called ‘Benign Interpretation’ of the dialectical tier. This view admits that an argument is strengthened by facing its main known objections and that this process can be reiterated if needed, but does not demand such a reiteration as a precondition for the validity of the original argument. In other words, according to the Benign Interpretation, argumentation is in principle indefinitely extendable, but infinite regress is not needed to attest the prima facie validity of an argument. However, as Freeman observes, Govier was not very sanguine about her own solution, that she saw as either endorsing ad hoc restrictions to the criteria for dialectical validity, or leaving open the threat of infinite regress. On the contrary, Freeman sets to show that the “Benign Interpretation both is closer to solving the regress problem than she gives it credit and provides a framework for solving the discrimination problem” (p. 22). In particular, Freeman makes use of Rescher’s analysis of formal disputation to provide an interpretation of Johnson’s dialectical tier that solves the discrimination problem, at the same time avoiding the pitfall of infinite regress. Without going into details on Freeman’s analysis, I want to draw attention on a certain Toulminian theme which somehow characterizes the whole debate. Famously, Toulmin mentioned the problem of recursion while describing his layout of an argument. His solution to this difficulty was, in its extreme simplicity, quite instructive: “Some warrants must be accepted provisionally without further challenge, if argument is to open to us in the field in question: we should not even know what sort of data were of the slightest relevance to a conclusion, if we had not at least a provisional idea of the warrants acceptable in the situation confronting us. The existence of considerations such as would establish the acceptability of the most reliable warrants is something we are entitled to take for granted” (1958/2003, pp. 98-99). To my mind, Freeman’s article provides a useful formal specification of the basic insight outlined by Toulmin. In their contribution, van Eemeren and Houtlosser provide further developments to the pragma-dialectical analysis of strategic maneuvering, understood as the arguer’s attempt at reconciling the dialectical aims of critical discussion, i.e. settling differences of opinions in a reasonable way, with their personal rhetorical goals, i.e. having things their own way. Insofar as the arguer’s rhetorical purposes change across dialectical stages and across different types of communicative activity, also the nature and aims of strategic maneuvering are sensible to these two dialectical dimensions. As for dialectical stages, van Eemeren and Houtlosser refer to the standard partition of pragma-dialectics: confrontation stage, opening stage, argumentation stage, and concluding stage. As for activity types, the authors first outline the general feature of argumentative discourse, and then proceed to characterize three more specific sub-cases: adjudication (attempting to terminate a dispute by an authoritative third party, instead of settling the difference of opinions among the disputants), mediation (attempting to solve a disagreement Hansen & Pinto: Reason Reclaimed 173 with the help of a third party, that acts as a facilitator but has no authority to settle the dispute independently from the disputants), and negotiation (attempting to solve a conflict of interests, rather than a difference of opinions, by reaching a compromise that is considered acceptable by all parties). Integrating these two criteria of dialectical variation produces a four-by-four taxonomy (p. 43), showing how dialectical considerations determine the pre-conditions for strategic maneuvering under different circumstances. Here van Eemeren and Houtlosser build on their previous work with Snoeck Henkemans to show how these different dialectical predicaments can be further specified, allowing fine-grained understanding of strategic maneuvering in argumentative practice. They use the notion of dialectical profile, defined as “a configuration of sequential patterns of moves that the participants in a critical discussion are obliged or entitled to make in order to realize a specific dialectical aim in a particular stage or sub-stage of the resolution process” (p. 44). Some examples are then provided, showing how this notion effectively isolates the space of possibilities where strategic maneuvering is to take place. This is meant to demonstrate how, in argumentative practice, rhetorical moves do not happen in a vacuum, but rather within the well- defined boundaries of specific dialectical rules—a fact that certainly casts a better light on the frequently misconceived opposition of dialectics and rhetoric. In his contribution, Krabbe investigates the role of objections in dialogue. He starts considering two opposite fallacies discussed in the literature: the fallacy of objections, i.e. sabotaging an argument by perpetually objecting to it, and the converse fallacy of objections, i.e. showing an unjustified insensitivity to criticisms raised against one’s own position. With reference to Johnson’s theory of argumentation, Krabbe observes that which one of these two fallacies is given greater prominence depends on what aspect of an argument one intends to stress: if one emphasizes the illative core, then the fallacy of objections constitutes a more urgent concern; if, on the contrary, priority is given to the dialectical tier, then the converse fallacy of objections acquires special importance. However, Krabbe’s aim is “to see how a reasonable and balanced system of dialogue rules may wipe out both extremes” (p. 59—thus catching two fallacious birds with one dialectical stone. In order to do so, Krabbe provides a taxonomy of seven different types of critical reactions to an argument, and then argues that only two of them can be properly conceived as, respectively, mild and strong objections. The rest of his analysis is focused on the latter, since they are considered the most interesting moves for a dialectical understanding of objections. Krabbe sees strong objections as exposures of flaws in the counterpart’s argument, without necessary implying full-blown rejection of it, charge of fallacy, or personal attack. The anatomy of strong objections is described as including a verdict (the negative opinion expressed against the original argument), a substance (the reasons for dissatisfaction that produced the verdict), an evaluative argument (connecting the substance Fabio Paglieri 174 to the verdict), and a substantive argument (showing how specific features of the original argument justify the substance). A dialectical procedure is triggered by raising an objection, in which proponent and opponent (in Krabbe’s terms, Protagonist and Antagonist) reverse their roles. In the objection discussion, the Antagonist will defend the criticism raised against the original argument, while the Protagonist will be entitled to cross-examine it. At the end of this critical discussion, either the Antagonist retracts her objection, or she keeps it, possibly with some modifications – in which case, the Protagonist will know that his argument does not carry presumptive weight with the Antagonist, hence something different has to be tried. In light of this dialectical analysis, Krabbe concludes that the two aforementioned fallacies both result from faulty handling of the objection discussion: the fallacy of objections can be imputed whenever the Antagonist raises an objection without being prepared to defend it in subsequent dialogue, while the Protagonist incurs in the converse fallacy of objections when he rejects an objection without testing it adequately through critical discussion. Notwithstanding some open problems acknowledged by Krabbe (pp. 62-63), his work makes a significant contribution towards an integrated dialectical account of objections and of the fallacies connected with their use in dialogue. Kauffeld carefully analyzes the checkered history and various applications of the notion of burden of proof, in order to question its usefulness for contemporary theories of argumentation. The central issue is whether the concept applies to the macro-level, i.e. with reference to the general obligations incurred by arguers in the context of dialogue, given their respective standpoints and roles, or to the micro-level, i.e. concerning probatory obligations generated by making specific dialogical moves and using certain argumentation schemes. Kauffeld identifies four different positions on this issue, tracing their origins in the history of legal theory: (1) confining the burden of proof at the macro-level, (2) reserving the notion for the micro-level, (3) assuming that the concept applies at both levels in analogous ways, and finally (4) acknowledging that burden of proof plays a role at both levels, but with different degrees of importance and distinct dynamics. In relation to everyday argumentation, Kauffeld strongly supports the latter view, and he uses the public debate on the ratification of the U.S. Constitution to demonstrate the sophisticated interplay of probatory obligations at (and across) both levels. Since similar ramifications are often obscured, rather than revealed, by unchecked use of the generic label “burden of proof”, Kauffeld concludes that we might be better off abandoning the expression, in favor of a more precise terminology – in his own apt metaphor, we need start using a vast array of scalpels, rather than resorting to the old-fashioned meat ax. Kauffeld’s insistence on the nuances of argumentative obligations is echoed also by Pinto’s analysis of the burden of rejoinder, i.e. the obligation to criticize an argument incurred by speakers when they propose using a previously disputed proposition as a premise, or when Hansen & Pinto: Reason Reclaimed 175 they object to taking a previously accepted claim as a premise. Pinto aims to understand when and how a burden of rejoinder arises in similar cases, and according to what standards it should be assessed – that is, what argumentative process would successfully discharge the obligation. His analysis further develops some of his previous work, and it is very much in the vein of the design-theoretic approach to argumentation favored by Goodwin and Kauffeld. According to Pinto, acquiring a full-blown burden of rejoinder is a two-tiered process. First, the speaker endorses an obligation to have valid reasons for supporting use of a disputed claim or for objecting against use of an accepted proposition as premise, by the very fact that these proposals are advanced in the context of a pre- existing debate: thus the speaker is entangling herself in controversy or dispute, which involves the public consideration of reasons. But the fact that the speaker is pragmatically committed to having reasons for her claim or objection does not necessarily entitle the counterpart to demand such reasons to be produced and made explicit—something more is needed for a proper burden of rejoinder to emerge. Namely, the speaker must freely assume such an obligation upon herself, and the only reasonable motives for doing so are strategic. In particular, Pinto argues that the speaker will accept the obligation to provide “justifications on demand” in order to ensure full consideration for the argumentative move she is proposing or objecting to. In other words, the burden of rejoinder is used as currency to make sure that the counterpart takes the speaker’s position seriously. Pinto then distinguishes four different forms in which a rejoinder can be produced: as a straightforward rebuttal of the controversial argument, as offering reasons for discounting that argument, as a means to neutralize it, or just as a way of bracketing it (for details, see pp. 86-87). The type of rejoinder needed to discharge the speaker’s burden in turn depends on the argumentative move that originated it: if the speaker is just objecting to using a previously accepted proposition as premise, it will be enough to neutralize or bracket the argument supporting it; if, on the other hand, the speaker is trying to use a previously disputed claim as premise, a stronger rejoinder is needed, to rebut or at least to discount the counter-arguments against such a claim. 2. The second group of essays is gathered under the heading of “The nature and scope of argument”, which gives the section a rather broad latitude— to put it mildly. Indeed, this is just a useful umbrella to cover a variety of contributions on several key issues in argumentation theories. Once again, such a thematic diversity does not hinder their scientific quality at all. Weinstein’s chapter provides a perspective on truth as an emergent concept, inspired by inquiry in physical chemistry rather than mathematics, and intended to reconcile the enduring tension in Fabio Paglieri 176 argumentation theories between acceptability and truth. Weinstein shows that most of the current views of argumentation, including those of Blair and Johnson, remain poised between these two poles. This in turn leaves them exposed to some classical concerns on the adequacy of either truth or acceptability as the sole ideal of argumentation: in a nutshell, the problem is that acceptability of a claim will never be able to ensure its factual truth regardless of how tight we make the dialectical constraints on rationality, whereas the truth of a proposition will still leave it impotent for argumentation, unless its truth is known, believed or accepted by the parties. Then Weinstein raises some deeper problems on the notion of truth itself, confronting three powerful intuitions that have shaped its conceptual history: truth as coherence, implying that allegedly true claims must be able of withstanding rigorous logical scrutiny and avoiding inconsistency; truth as correspondence, positing a close relation between our beliefs and the shape of things in the real world as made manifest by our experience, or with reference to the elements of some well defined model (as in Tarskian semantics); truth as pragmatic adequacy, requiring that our progressive process of inquiry makes our representations of the world well attuned to our practical purposes and actions. Weinstein focuses on the first two concepts, essentially to argue that both of them presuppose the availability of a well defined and coherent model or frame of reference prior to assessing the truth of our utterances. This highly idealistic requirement is given the lie by the history of our most magnificent intellectual achievements: if there is a constant element across different fields of scientific inquiry, it is the continuous evolution of the models against which the truth of theories is assessed. This being the case, “the problem of correspondence seems intractable, since, short of instrumentalism, science must await upon its own discoveries to have a picture of the world against which its claims are to be ultimately judged. This calls for a radical revision of the metamathematics of truth” (p. 96). Coherently with this program, Weinstein then outlines his own suggestion on how to define a model of emergent truth. Leaving aside the details of Weinstein’s proposal (interested readers will find them outlined on pp. 102-106, as well as in some of his previous works), I think two points deserve special emphasis: one is a criticism aimed at the view of truth endorsed in many contemporary argumentation theories, the other is a positive consideration that might well inspire future research in this area. As for the criticism, Weinstein observes that “most informal logicians focused on truth in the most obvious and ordinary sense. That is, truth as conformity with the facts of the matter manifest in ordinary experience and analyzed in standard ways. The plausibility of such a common sense stand has been rarely argued for, but rather has been taken as obvious” (p. 99). Aside from the inadequacy of this simple view of truth to account for scientific inquiry, there is a mild irony here, insofar as this is precisely the notion of truth commonly endorsed by modern logic—the very tradition from which many argumentation theorists and informal Hansen & Pinto: Reason Reclaimed 177 logicians would like to distance themselves. Weinstein suggests that the argumentative enterprise would be better off working with a different and more sophisticated understanding of truth. This leads us to the positive contribution: the take-home message that I personally extract from Weinstein’s careful analysis is that, when arguing, we are not only trying to probe the truth of factual claims in view of some shared frame of reference on reality, but also negotiating what might be the most adequate frame of reference in the first place. Relevant models for assessing the so called “truth of the matter” might shift during discussion, and this is part of the dialectical process as much as the evaluation of any given claim, premise, or argument. This insight might give us a slight sense of vertigo, but I think it captures a significant layer of complexity intrinsic to any argumentative process. Moreover, even if mathematics is not needed to express this emergentist perspective on truth, it shows that “these notions can be given precise content and so are not to be scorned on the grounds of vacuity” (p. 104). If so, then perhaps significant attempts to integrate Weinstein’s truth (no pun intended) in contemporary theories of argument would be advisable. Bailin and Battersby focus their contribution on reason appreciation, that is, the inclination to fulfill the normative demands of reason in one’s own conduct, given “a respect for reasoning based on an understanding of its nature, role and significance, and a recognition of its subtleties and aesthetic aspects” (p. 107). The authors argue that reason appreciation is a crucial feature both for understanding argumentation and for teaching it effectively, insofar as mere detached knowledge of the inner workings of reasons and arguments is not enough to guarantee their application in practice. Bailin and Battersby explain why dispositional accounts, as those outlined by Siegel and Ennis, are relevant but not sufficient for the analysis of reason appreciation, since dispositions remain silent on why reasoners behave (or restrain from behaving) in certain ways, therefore providing little insight for pedagogical purposes. With the aim of suggesting an alternative account, the authors conceive of reason appreciation as involving both an attitude of respect towards the practice of reasoning as a whole, and aesthetic sensitiveness to the subtleties of its concrete applications. Respect for reason is based on both epistemological grounds (basically, reason commands respect as the chief means by which we foster our quest for truth) and moral considerations (as emphasized also by Siegel, reason plays a crucial role in making us truly autonomous, as opposed to being stimulus-driven, impulsive, or easily manipulated agents). As for what they call “the aesthetics of reason” (p. 118), the authors insist that it is not to be confused with the rhetorical dimension of arguments, although it partially overlaps with it. More generally, aesthetic considerations on reason involves appreciating the elegance and inventiveness of a given argument, beyond and besides any rhetorical value it might possess, and they are linked with an affective response to it, usually associated with a feeling of enjoyment. Finally, Bailin and Battersby conclude their interesting analysis with Fabio Paglieri 178 some tentative suggestions on how to foster reason appreciation as part of the teaching of argumentation. While I am sympathetic with their emphasis on the somehow neglected topic of reason appreciation, I find debatable the suggestion that greater reason appreciation would also result in “civilizing the discourse” in social interaction, to use their expression (p. 120). I do not see how this follows at all: being an excruciating sophist and a practiced liar does not necessarily prevents me from appreciating reason, both epistemologically and aesthetically – it just so it happens that I do not care for extending the benefits of reasonableness to my fellow arguers, given my own practical goals. The point is that, while the noble quest for truth is indeed a necessary end of my own reasoning as a rational being, it is not necessarily my goal while arguing with others. Sometimes I may actually intend to lead us both to the truth of the matter, but just as often I may just want to achieve persuasion, in order to shift your conviction and/or your course of action to accommodate my own agenda, whether or not this implies any concern for truth. Sure enough, such an attitude might well disqualify me as a rational arguer, depending on the theory of argumentation one endorses. However, this banishment from the community of rational arguers would not be due to any lack of reason appreciation on my part, but rather to my failure at complying with the social constraints projected by conventions of fair play in argumentation. Incidentally, the need to distinguish reason appreciation from appreciation of dialogical rules descends from the need to avoid oversimplified identification between reasoning and arguing. Although these two practices share many similarities and are deeply connected with each other, considering argumentation as just dialogical reasoning would fail to do justice to the multiple and frequently conflicting aims of the arguers. In his contribution, Finocchiaro compares some of Blair and Johnson’s views of argumentation with the dialectical approach to freedom of argument developed by John Stuart Mill in the second chapter of On Liberty. With reference to the distinction between illative and dialectical components (or tiers), Finocchiaro presents three versions of dialectical accounts of argumentation: his own modest version, according to which either the illative or the dialectical tier is sufficient to produce an argument, but neither of them is necessary; the strong version, famously championed by Johnson, where both the illative core and the dialectical tier are necessary to have an argument, although neither of them is sufficient without the other; and the hyper version, that, according to Finocchiaro, is represented by the pragma-dialectical school, which makes the dialectical tier both necessary and sufficient. Against this background, Finocchiaro embarks in the analysis of Mill’s argument on the liberty of thought and discussion: this constitutes a masterful analytical exercise (pp. 123-132), and a valuable asset of Finocchiaro’s contribution in its own right. Reasons of space prevent me from giving the specifics of his reconstruction, so let us keep to the essential core. Mill aims to show that freedom of argument is desirable, Hansen & Pinto: Reason Reclaimed 179 and he provides three main reasons for this claim (each supported by complex sub-arguments): “first, it enables us to determine whether an opinion is true, whereas its denial amounts to an assumption of infallibility; second, it improves our understanding and appreciation of the supporting reasons of true opinions, and our understanding and appreciation of their practical and emotional meaning; third, it enables us to understand and appreciate every side of the truth, given that opinions tend to be partly true and partly false and people tend to be one-sided” (pp. 132-133). Incidentally, in the course of the analysis Mill appears to be enlisted among the supporters of a moderate dialectical approach, insofar as Finocchiaro is correct in interpreting Mill’s definition of argument: “an argument just is an attempt to justify a conclusion by giving reasons in support of it or defending it from objections” (p. 128, my emphasis). Finally, aside from its historical value and scholarly refinement, Finocchiaro’s reconstruction of Mill’s argument manages to touch upon important concerns of contemporary argumentation theories—among others, “it raises in a vivid and striking manner the issue of the viability of an epistemic approach to argumentation theory, but also suggests the reverse issue of the possibility of an argument- theoretical approach to epistemology” (p. 133). In his contribution, Groarke takes issue with the skepticism voiced by Blair and Johnson on the existence and relevance of visual arguments. After summarizing similarities and differences between Blair’s and Johnson’s view on this topic, Groarke turns to describe what he considers two dogmas of informal logic that have “prevented informal logicians from fully embracing the visual” (p. 138), and that he imputes to both Blair and Johnson: namely, verbalism and reductionism. In his own words: “Verbalism is the view that arguments are made up of propositions (sentences or their equivalents), and that the sentences that correspond to them are the best way to present or summarize an argument. (…) Reductionism maintains that the important argumentative elements of any visual argument are verbal, or can be reduced to verbal equivalents” (p. 139). In contrast with both these claims, Groarke argues in favor of the value and autonomy of visual arguments with respect to their verbal counterparts, both on rhetorical and, much more to the point, logical grounds. This leads him to consider the dogmas of verbalism and reductionism unjustified, hence concluding that “informal logic and argumentation theory (…) need to recognize visual arguments, and understand them in visual terms” (p. 152). Personally, I am inclined to agree with the first part of this sentence, whereas I am not fully sure about the latter: more to the point, my perplexity rests on what I see as a potential weakness in Groarke’s line of argument against Blair and Johnson. The point is that much (not all) of Groarke’s case against verbalism and reductionism hinges on the fact that images can convey evidence in a very powerful and effective way, and that they can precisely express information that would be hard if not impossible to articulate by verbal means (maps and diagrams are good examples of this Fabio Paglieri 180 fact). However, here lies a confusion, I fear: while Groarke’s arguments are successful in asserting the evidential value and informational perspicuity of images, this is not at all conclusive for proving that images, even those that provide us with a wealth of evidence and/or information, constitute arguments per se—which is precisely the point under dispute. To my mind, Groarke is right in considering excessive some of the reservations expressed by Blair and Johnson on visual arguments. But we should also beware of the converse exaggeration— that is, magnifying the argumentative nature of images beyond the boundaries of common sense. Groarke is well aware that not all images are visual arguments: indeed, he suggests that images “can be understood as arguments so long as they attempt to forward evidence for some conclusion” (p. 150, my emphasis). This suggests that the premise- conclusion mechanism, characteristic of the illative core of any argument, visual or otherwise, must be somehow intrinsic to an image, for it to count as a visual argument. If this constraint is relaxed, the notion of visual argument immediately becomes meaningless, since any image at all is of course liable of being used in the context of argumentation to provide evidence or to facilitate reasoning. So the fact that an image can be used to forward evidence for some conclusion is not enough: the image must have in itself some indication of the relevant premise-conclusion link that it is meant to support. Unfortunately, by this standard only two of the illustrations presented by Groarke in his article actually count as valid examples of visual arguments: namely, the cartoons on p. 137 and p. 141. All his other illustrations are just images, as opposed to visual arguments, that can be used (in a variety of ways, I hasten to add) as effective props to foster a process of argumentation. By way of example, let us consider the infamous pictures from the Abu Ghraib Prison in Iraq, depicting US soldiers abusing Iraqi prisoners. As Groarke emphasizes, these pictures have been used most effectively to provide evidence of the misbehavior of parts of the American force in Iraq. Nevertheless, they cannot be said to be, in and by themselves, visual arguments to that conclusion – witness the fact that the very same pictures could be used as evidence to demonstrate torture techniques to sadistic jailers, to prove some idiotic macho mystique to rookies in the force, or to ridicule the alleged superiority of so called Western civilization. Again, here the point is that these pictures convey information on what happened in Abu Ghraib in a highly reliable (not to say emotionally moving) way: as such, they can be used as good evidence in the context of argumentation for a variety of purposes. But if this was enough to turn them into visual arguments, we would then be forced to conclude that every image is a visual argument, inasmuch as it can be used to provide evidence or convey information (and of course any image can do that). So it seems to me that lumping together evidential value, informational perspicuity, and logical force ends up hindering, rather than fostering, Groarke’s rightful attempt of promoting a better understanding of visual arguments. Hansen & Pinto: Reason Reclaimed 181 3. The third section of the volume is devoted to fallacies, again covering a rather broad array of topics. Tindale proposes a critique of Blair and Johnson’s account of fallacies: after framing their work in the context of previous analyses of fallacious arguments, as those offered by Aristotle, Whately, Copi, and Kahane, Tindale focuses on what Blair named the “injury” view of fallacy—that is, the idea that fallacies do reparable damage to arguments, rather than fatally destroying them. In particular, Tindale takes a closer look at Blair and Johnson’s proposal of categorizing fallacies according to what standard of logical cogency they happen to violate. Notoriously, Blair and Johnson identified three of such standards – relevance, sufficiency, and acceptability—and associated each of them to a general category of fallacies—respectively, Irrelevant Reason, Hasty Conclusion, and Problematic Premise. Tindale questions the perspicuity of each of these categories in turn. As for Irrelevant Reason, the main problem lies in the positive definition (or lack thereof) of what would make a reason relevant in the first place: Tindale remarks that “irrelevance is the absence of a relationship between statements, without really specifying what the correct relation should be” (p. 162). More generally, Blair and Johnson’s classification of subtypes of Irrelevant Reason (e.g. straw person, ad hominem, etc.) reveals for Tindale a mixture of internal and external constraints on what determines the alleged irrelevance, making things “more murky than one would like when seeking out a clear set of identified, and identifiable, fallacies” (p. 164)—a situation not redeemed by Blair and Johnson’s distinction between local and global relevance. The problem with the category of Hasty Conclusion, according to Tindale, is that it has the undesired and slightly paradoxical effect of classifying most of our argumentation as fallacious. This is because Hasty Conclusion is charged “whenever an arguer fails to anticipate and accommodate objections, standard or otherwise. A glance at any collection of ‘everyday’ arguments—in the media, in student papers, in advertising—would show how regular[ly] this failure occurs” (p. 168). Finally, the definition of Problematic Premise hinges on the idea of a premise introduced in an argument without support, but which needs instead to be supported for being considered acceptable. Here the problem is of course how to establish what premises are in need of support—a problem solved by Blair and Johnson with the notion of a community of ideal interlocutors, to be used as an imaginative frame of reference for the arguer to assess whether or not a given premise is acceptable without further support. However, as Tindale suggests, this seems to reintroduce in the definition of fallacy a requirement that Johnson at least wanted to expunge from it: the so called appearance condition, according to which a fallacy is not just an invalid argument, but more specifically an invalid argument that appears to the audience as valid. Tindale favors including similar contextual considerations into a theory of fallacy, but he claims that doing so is at Fabio Paglieri 182 odds with Blair and Johnson’s intended aims, and that their current account is not well equipped, as it is, for dealing with these issues. Reforming their approach would involve recognizing that, whenever fallacies are conceived as deviations from standards of validity, they inherit the double nature of such standards: some of them are better described as failures of reason, involving violation of logical laws, while others are rather flaws of conduct, resulting from infringement of dialectical obligations incurred in the context of dialogue. This distinction can also be mirrored in terms of internal vs. external factors determining the fallacious nature of a given argument—or its virtuous character, for that matter. This leads Tindale to conclude that “some shortfalls, errors or flaws, or however we wish to describe them, are internal to the argument product as it is produced, involving a problem between the components of the argument. While other shortfalls, etc., lie outside of the argument per se and involve something of its external relations with features of the context, like the audience or the issues or previous argumentation” (p. 170). Of course, the question remains on whether a broad understanding of the notion of “fallacy,” as the one suggested by Tindale, would increase or rather diminish the usefulness of the concept. Indeed, one may well concede that many problems in argumentation are to be imputed to the contextual use of an argument rather than to its structure, and yet insists that we should reserve the term “fallacy” only to the latter cases, regarding instead the former as instances of flawed dialectical practice. As a case in point, Govier is a prominent advocate of a “deflationist” approach to fallacies, that she endorses also in her contribution to this volume, devoted to discuss the problematic uses of questions in argumentation. She focuses on two kinds of questions frequently associated with flawed arguments or fallacious moves: complex questions and rhetorical questions. A standard example of the former, also known as compound, heavy, double-bind, or many questions, is the following: “Have you stopped cheating on your taxes?” However, as Govier reminds us, also flattering questions fall within this category, as in “Which of your three daughters is the highly talented violinist?” The characteristic of complex questions is that they are worded in such a way as to require acceptance of a statement that is not argued for, in order to answer the question—since any answer provided to such questions would imply acceptance of their unsupported premise. The only way to avoid this is by questioning the question, either asking for further clarification or directly rejecting the unsupported premise behind it. Rhetorical questions, like “Don’t we have the right to defend ourselves from terrorists?” are partially different, since they do not carry a real request, but rather express a statement “because of the way in which they anticipate and invite one particular kind of answer from the audience” (p. 181). What they do have in common with complex questions is their function of insinuating a given claim, without subjecting it to critical scrutiny. However, Govier correctly observes that Hansen & Pinto: Reason Reclaimed 183 neither all complex questions nor all rhetorical questions need to be automatically regarded as problematic for argumentation, since not all statements need to be subjected to explicit critical scrutiny. For instance, if the unsupported premise of a complex question is a well known fact to both parties, or if a rhetorical question is used to state a conclusion rather than introduce a premise, then their use is not dialectically objectionable. Moreover, even when questions are argumentatively problematic, what makes them so is partially independent from their complex or rhetorical nature. To use a simile, to have a hidden partition in the trunk of your car is perfectly legitimate (indeed, that is the place where the spare wheel and the emergency kit is often stored); what is not legitimate is to stuff it with illicit goods and smuggle them across the border. Similarly, whether or not a complex and/or rhetorical question is problematic depends on what argumentative material the arguer is putting in its trunk. However, Govier devotes the last part of her article to discuss why insinuating claims, as opposed to explicitly debating them, is in general an objectionable policy for argumentation. She considers four different approach to reasoning and argumentation (Adler’s belief ethics, pragma- dialectics, Walton’s dialogue types, and Habermas’ dialogue ethics), and she concludes that in all of them, for partially different reasons, insinuation is regarded as a condemnable practice—with the exception of extreme situations like quarreling and adversarial negotiation, as in Walton’s taxonomy of dialogue types. Finally, Govier reflects also on the intimidating character of complex and rhetorical questions, speculating on how it contributes to their problematic role in the context of dialogue. In his contribution, Hitchcock argues against the existence of an ad hominem fallacy, starting from Blair and Johnson understanding of it as a fallacy of diversion, and then offering a nuanced reconstruction of the adventurous history of this notion. He disentangles four different conceptions of the ad hominem argument: as an argument ex concessis, as a case of tu quoque, as an abusive personal attack, and as a circumstantial personal attack. The first is the original meaning intended by Aristotle, in which an argument is ad hominem in the sense of the proponent taking some premise accepted by the opponent (and possibly not shared by the proponent) to argue for a conclusion that the opponent is unwilling to accept. The tu quoque interpretation refers to arguments where an inconsistency between words and deeds is used to put pressure on the proponent of a thesis or criticism that does not appear consistent with her actions—on the general principle that “you should put your money where your mouth is”. The abusive ad hominem is closer to the widespread interpretation in contemporary writings, and indicates cases where the proponent of a given position is attacked on the grounds of personal qualities or past history, aiming at general discredit and with no relevant connection to the position under debate. Finally, the circumstantial ad hominem is a specification of the previous case, in which the personal attack against the proponent is aimed at suggesting some bias in her assessment of the issue under consideration. Given these Fabio Paglieri 184 different notions, the question is whether any of them should be regarded as constituting a fallacy—a question that Hitchcock answers in the negative. He refers to Govier’s definition of fallacy, that requires three necessary conditions: a fallacy is defined as (1) a mistake of reasoning that (2) occurs with a certain frequency in real arguments, and (3) has the characteristic of being deceptive, i.e. it disguises itself as a form of valid reasoning. By this definition, Hitchcock argues that the ad hominem argument is not a fallacy, since, depending on its interpretation, either it represents a perfectly legitimate argumentative move, or it is not an argument to begin with, so it cannot be a fallacy (incidentally, here we encounter again a deflationist approach to fallacies, contra Tindale’s broader conception). The details of Hitchcock’s case are nicely summarized as follows: “In its original meaning, an argumentum ad hominem is a perfectly legitimate dialectical argument from the concessions or commitments of an opponent that one needs not to share. The tu quoque historically emerged from this sense as an appeal to commitments implicit in the behavior of one’s critic; it legitimately challenges the critic to explain away an apparent inconsistency. The purely abusive ad hominem can be a legitimate attack on an opponent’s ethos, a response long sanctioned in the Western rhetorical tradition. Otherwise, it is an objectionable diversionary tactic, but not a kind of reasoning, and so not a fallacy. The circumstantial ad hominem, in the sense of an abusive ad hominem which attributes the position of one’s opponent to self-interest or to a dogmatic bias, raises legitimate suspicion about the credibility of the opponent’s statements and arguments” (pp. 198-199). Hitchcock’s reconstruction certainly puts significant pressure on accounts of the ad hominem as fallacious, like the one endorsed by Blair and Johnson. Besides, and this is no marginal benefit, it contributes to the historical and conceptual clarification of what exactly falls under this rather ambiguous label. Adler’s chapter is devoted to presuppositional blindness, that is, our tendency to accept propositions that are presupposed, rather then being explicitly focused, in a given argument or piece of discourse. The complex questions discussed by Govier (see above) are a clear instance of this tendency, and Adler opens his article with the following anecdotic example of the same phenomenon: “Charles II is said to have invited members of the Royal Society to explain to him why a fish weighs more dead than alive. Various explanations were proposed. Afterwards, he observed that it does not” (p. 201). Adler remarks that “what is troubling about these—our—lapses is that the erroneous assumptions are hidden in plain sight” (p. 208). Indeed, to presuppose is not to keep silent: rather, it is a case of manipulation of the attentional focus, one in which the audience plays the role of a willing accomplice of the arguer. Imagine Charles II as having asked: “Why do fish weigh more dead than alive?”—a complex question, again. Here the (presupposed) claim that fish are heavier after their death is plainly stated, and yet our attention is diverted away from it: we tend to take it for granted, and move on Hansen & Pinto: Reason Reclaimed 185 speculating about possible reasons for this bizarre phenomenon, complying with the focalization proposed by the questioner. Moreover, the very obviousness of what is being presupposed often contributes to distract our attention from it, as something unworthy of further consideration. Adler sees this tendency as dependent upon the conversational rules that tacitly regulate communication, very much in a Gricean vein. This implies that presuppositional blindness can be mitigated, but not eradicated – because such a drastic cure would be far worse than the occasional illness it is meant to overcome. Possible strategies for correcting presuppositional blindness include explicitness and stating the obvious: both can be warranted for critically probing vicious instances of presupposition (as a case in point, Adler extensively discusses Michael Drosnin’s pseudo-scientific rant on the so called “Bible Code”; see pp. 202-208), but both would be highly disruptive for the regular course of conversation, if applied indiscriminately. Quite simply (or obviously, if you like), “what is obvious is not worth saying typically because it is uninformative” (p. 210). The gist of Adler’s reflections is that conversational rules create a ”weak spot” in our argumentative defenses, and we must learn to live with it. This, in turn, should inspire us a measure of intellectual humility, even stronger than the celebrated Socratic avowal of ignorance. Indeed, “we need to be humble and restrained, even on some matters where we do know and take ourselves to be competent. For we may still miss crucial implications, even when they are right before us and, in part, just because they are right in front of us” (p. 212). 4. The fourth and last section of the volume is entitled “Other dimensions of the informal logic program,” which is tantamount to saying “everything else.” The editors will be certainly excused for this small act of thematic surrender, because by now it should be clear that this collection is meant to pile wonder upon wonder, with little care for any specific focus—and this, I hasten to add, provides a wonderful festive flavor to these Festschrifts, conveying the genuine enthusiasm and deep affect surrounding the intellectual figures being celebrated. Especially because the contributions in this last session do not lower at all the standard of excellence uphold by the volume as a whole, but rather add their own finishing touch to the overall composition. Siegel’s contribution deals with the role of cultural variation in argument evaluation, but he gives the topic a curious twist, by ending up discussing the merits and limits of an argument on multiculturalism. However, this analysis is effective to exemplify Siegel’s thesis, according to which any argument, in order to be meaningfully assessed, needs to be evaluated in light of universal principles that are independent from cultural variation—which implies that multiculturalism can have no Fabio Paglieri 186 authority over considerations of argument validity. Siegel refers to his previous work for extensive arguments in favor of such a thesis, and here confines himself to showing how this is made apparent by dissecting a specific argument against the tenability of multiculturalism. Very briefly, the argument that catches Siegel’s attention is one originally provided by Stanley Fish, to claim that any form of multiculturalism serious enough to be worthy of the name is untenable. Fish begins by opposing ‘boutique multiculturalism’ to ‘strong multiculturalism’. The first he dismisses as uninteresting in view of its superficiality: boutique multiculturalists, as defined, are avid fans of ethnic restaurants and shops, and make a show of their open-mindedness towards other cultures in academic circles and high-profile gatherings, but they remain unwilling to seriously mingle with the ‘other’ in any serious matter—children’s education, religious practice, sexual habits, moral norms, and the like. In contrast, strong multiculturalists are depicted as terminal believers in the ideal of tolerance towards other cultures—and it is this very ideal, according to Fish, that leads them into an untenable position. The dilemma generates when a strong multiculturalist is faced with an intolerant culture, i.e. one that does not endorse multiculturalism (and we can certainly think of many that do not). What should dictate tolerance in these cases? If the multiculturalist embraces such a culture, valuing its character, supporting its development, and fostering its traditions, she is thereby reneging on the basic tenets of multiculturalism by approving of a culture that denies them. If, on the other hand, the multiculturalist refuses to extend tolerance to the intolerant culture, she is renouncing her credo in an even more direct way, proving that her open-mindedness was conditional since the very beginning, being subjected to some hidden supracultural considerations. Either way, Fish concludes, strong multiculturalism is untenable. Siegel agrees with Fish that strong multiculturalism is indeed flawed, but insists (and rightly so) that tertium datur, in this case—that is, it is possible to formulate a version of multiculturalism that is both meaningful and consistent, thereby avoiding the downfall of both its cosmetic cousin and its fanatic uncle. According to Siegel, the correct interpretation of multiculturalism dictates that “all cultures should be valued and regarded as worthy only if they extend that value and regard to other cultures” (p. 221). The supracultural constraint presented in this definition is not, as Siegel emphasizes, an expression of some hidden moral agenda, as feared by Fish, but rather a strictly epistemological demand for coherence in expressing the multiculturalist position—the very same principle of rationality that makes Fish’s argument, and arguments in general, cogent. The irony of this rather convoluted story is that argumentation, to be able to rescue multiculturalism from the vice of Fish’s criticism, needs to be understood and assessed independently from any specific cultural perspective. This, Siegel seems to suggest, is in itself a good reason to keep argument validity out of the melting pot. Against this claim, and in favor of throwing back cultural considerations into the argumentative fray, is Gilbert’s contribution, Hansen & Pinto: Reason Reclaimed 187 which is devoted to speculate on the relevance of intersectionality for informal logic. According to Gilbert, “intersectionality is just the awareness of the importance of examining the matrix of personal characteristics that belong or are applied to a group. The suggestion is that once all of the relevant characteristics have been considered, only then can the appropriate understanding, or, in our case, rules and procedures be applied. (…) The call for intersectionality is the call for letting go of a one-size-fits-all approach to the analysis of argument, and a desire to see all of the significant characteristics that color a person’s identity considered as part of the fit” (p. 230-231). In a sense, Gilbert’s contribution is framed more as a manifesto than as a step-by-step demonstration, aiming at suggesting possible directions of research rather than establishing once and for all any given claim. This is not meant as a criticism, by the way: far-sighted manifestos are crucial to the development of the discipline, and therefore very much welcome. In the same vein, however, we should try to distil carefully the implications of Gilbert’s call to arms for “a greater degree of flexibility” in informal logic (p. 240). This enterprise would imply considering contextual factors (including, among others, issues like gender, class, culture, and race of the arguers; see pp. 232-239) prior to the application of argument analysis and evaluation, since these factors are claimed to influence what rules and standards it is reasonable to expect the arguers to comply with. An important point here is whether this alleged cultural influence should be confined to the application of argument rules and standards, or should be conceived as changing their very nature. Judging from the rich variety of examples provided by Gilbert, I think he succeeds in proving the former claim, whereas I do not see much evidence of the latter. The following is one among many of Gilbert’s vivid examples: “Groups that have been subordinated often require argumentative methods that violate some of the canons of Informal Logic. These may involve the appropriate use of anger (…), or other, even more radical means” (p. 240). Sadly, that is true enough – actually, some of the means that subordinated groups can and have a right to use are not argumentative at all, including taking arms and openly revolting against their oppressors. But what does this tell us, concerning the analysis of argumentation applied to public debate among subordinated and oppressing groups? I believe it reminds us that formal validity is not the same as moral value—and, conversely, formal invalidity does not always disqualify an argument as morally repugnant, therefore prohibiting its use on rational grounds. Insofar as you are unjustly oppressing me, I have the right to use any means that your oppression allows me to fight back, including being as fallacious and one-sided in my arguments as current needs demand. Moreover, Gilbert is right in pointing out that informal logicians should not scorn as ‘irrational’ similar flawed instances of argumentation, because their rationality is rescued at a higher level, in terms of their instrumental value for fostering legitimate and urgent goals, e.g. redressing the power balance in that particular society. But here my agreement with Gilbert Fabio Paglieri 188 ends, since I do not see what can be gained by making the stronger claim that those argumentative practices were never flawed to begin with, from the point of view of their formal validity. In other words, I fail to appreciate how similar cases suggest that the standards for argument validity are themselves influenced by cultural and, more generally, contextual considerations—while I am willing to concede that their application must be analyzed in terms of rationality from a broader perspective. In this chapter, Gilbert seems to oscillate between the weaker and the stronger version of his basic claim on the relevance of intersectionality: on the one hand, he is careful in specifying that certain “laws of thought” appear to be fairly universal (e.g., “I do not intend to prove that anyone walks around believing both P and ~P in any robust manner,” p. 232); on the other hand, he repeatedly suggests that the standards for argument analysis and evaluation are themselves liable to cultural influence, and in his conclusion puts a lot of emphasis on the following statement: “If the very rules of argument preclude the free transfer of communications, then they must be changed” (p. 240). I certainly approve of this conditional (who doesn’t?), but I do not see any evidence so far that its premise is true. Ennis articulates an analysis of the meaning of “probable” when used in the conclusion of stand-alone, affirming arguments—that is, when the use of “probable” is not embedded in reported speech, conditional premises, imperatives, or questions. In particular, Ennis argues for a speech act interpretation of the expression, capable of capturing the difference between ‘Probably P’ and ‘It is probable that P’—namely, the fact that the latter formulation implies a distancing of the speaker from the guarded commitment expressed towards P. Ennis’ thesis is that “to affirm ‘It is probable that P’ (where ‘P’ is a declarative sentence) is to affirm that a guarded committing—to the view that P—is justified” (p. 246). In terms of speech acts, the difference with “probably” would be that saying ‘Probably P’ is a guarded committing speech act, whereas saying ‘It is probable that P’ is a verdictive speech act about such a guarded committing speech act – that is, the speaker is saying that such act of commitment appears justified. Ennis offers a detailed argument to support his thesis: first he defends the clarity and plausibility of his claim, indicating also for what other equivalent expressions it is supposed to hold (‘it is likely that’, ‘the probability is that’, ‘that P is probable’); then he considers three alternative explanations (specific numerical probability, substantial numerical probability, and subjective probability), finding them inadequate; consistency with data is also demonstrated, and some standard objections are presented and resolved (see pp. 250-254 for details). Finally, this allows Ennis to conclude that for his thesis “a good supporting argument has been given, though there is much more to be said” (p. 255). The well- ordered structure of Ennis’ argument is laudable, partly because it invites specific comments to be raised. My personal quibble concerns Ennis’ dismissal of subjective probability as an alternative explanation of the Hansen & Pinto: Reason Reclaimed 189 meaning of “probable” – a dismissal that I do not see as fully warranted. Ennis offers three reasons of dissatisfaction: (1) subjective probability implies a numeralicalness that is not part of the dictionary definitions of “probable”; (2) the same numeralicalness was not in the speaker’s mind at the time of utterance, not even subconsciously; (3) the interpretation in terms of subjective probability dissolves the disagreement between parties with different views on what is probable, while in contrast we perceive a clear disagreement in such cases. The first argument I concede, but obviously it is far from being conclusive: dictionary definitions can be, and often are, quite remote from the actual meaning conveyed by our utterances, and the very theory of speech acts have helped proving this point. As for the second criticism, I think here Ennis is misconstruing subjective probability: its core hypothesis is that, by saying ‘It is probable that P’, I express the fact that my degree of belief in P is some number in between 0.5 and 1.0— but there is no need whatsoever for me to have an explicit representation of this number, not even subconsciously. ‘Believing P with degree X’ is utterly different from ‘believing that my degree of belief in P is X’—the former is what subjective probability suggests as the meaning of the utterance, whereas the latter is a kind of meta-belief that does not constitute a necessary condition of that meaning. Since Ennis’ second criticism applies only to the latter, I argue it does not truly affect its intended target, i.e. the subjective probability interpretation. As for the third criticism, here is how Ennis formulates it: “Suppose A says that it is probable that P, and B says that it is improbable that P. They are clearly in disagreement. They cannot both be right. However, the translation according to the subjective alternative destroys the conflict between what they say. In accordance with the subjective probability alternative, A might say, ‘My degree of belief that the tree will survive is 0.85,’ while B concurrently might say, ‘My belief that the tree will survive is 0.15.’ Now (…) both can agree (and be right) that A’s degree of belief is 0.85 and that B’s degree of belief is 0.15” (p. 249). The problem here is that the particular disagreement that Ennis regards as “disappearing” never existed in the first place—because disagreeing on the probability of uncertain outcomes implies agreeing on the possibility of such outcomes. Here A claims it is likely the tree will survive, while B contests this probability: but both of them are willing to concede that it is possible for the tree either to survive (although B would regard this as improbable) or to wither and die (even if A is skeptical of this outcome). Nothing of this is changed by analyzing the situation in terms of subjective probability: what the arguers can (and always could) agree upon is, to use Ennis’ numbers, that A’s degree of belief is 0.85 and that B’s degree of belief is 0.15. But this just shows they still disagree on what is probable (the numbers differ), while agreeing that, since the matter is uncertain, both outcomes are possible (the numbers are neither 0.0 nor 1.0). Considerations of probabilities are also central to Woods’ contribution, which is devoted to discuss the application of Bayes’ Fabio Paglieri 190 theorem to legal reasoning, in particular concerning jurors’ selection of evidence (what to accept or reject, and to what degree) and establishment of verdict (guilt or innocence, given the evidence). Woods considers Bayes’ theorem, and by extension Bayesianism, inadequate to represent these forms of legal reasoning. In reaching a verdict of guilt G, Bayes’ theorem dictates that jurors should assess the conditional probability of G given the evidence E, i.e. Pr(G|E), by making use of a variety of parameters, including what are technically known as likelihoods, i.e. a number indicating how likely it is that the evidence E was produced assuming that, respectively, the accused was guilty or innocent—i.e., L(E|G) and L(E|~G). Much of Woods’ critique revolves around what should we make of likelihoods in the context of legal reasoning. First he observes that likelihoods cannot be used by jurors in evidence selection, because doing so would fail to make the evidence probative according to the standards of law. Indeed, jurors are expected to select evidence and reach a verdict independently from any assumption on either guilt or innocence (since also the presumption of innocence acts as a default rule, not as a positive statement; see p. 261). In contrast, Bayesianism required precisely this kind of assumptions for the conditional probability of a verdict to be established. This is legally untenable, Woods argues: “One does not prove that an accused is guilty by accepting evidence whose reliability depends in any part on the assumption that he is guilty. (…) One does not find that the accused is not guilty by finding that the evidence is, on assumption of lego-factual innocence, not probative” (p. 261). Moreover, Woods shows that likelihoods cannot be rescued by assigning them a different interpretation in legal reasoning, e.g. as indicating the explanatory force, predictivity, or plausibility of different theories (or cases) built by the parties of a trial on the grounds of (possibly diverging sets of) evidence. Similar interpretations of likelihood are equally untenable, for two main reasons: (1) under certain circumstances, they derange Bayes’ theorem, as Woods shows (p. 264) with a very simple example; (2) none of these interpretations is consistent with the basic features of probability—which is tantamount to saying that probability is not a good model of either explanatory force, predictivity, or plausibility. Taking explanatory force as a case in point, Woods observes that it “does not have the ‘conceptual flavor’ of probability. The Kolmogorov axioms require incompatible states of affairs (or statements) not to have the same probability. But it is obvious that incompatible explanations are sometimes of equal force. The probability axioms also require that the probabilities of incompatibilities sum to 1, but the explanatory forces of incompatible explanations can easily exceed 1. The probability axioms require that probabilistic conjunction be multiplicative. But there are lots of cases in which the conjoined force of pairs of explanations is additive” (p. 265). This leads Woods to conclude that “there is nothing Bayesian in the logical structure of verdicts at the criminal bar. Bayes’ Theorem should not be legalized for this purpose” (p. 267). So Woods’ analysis provides a cogent argument to support a Hansen & Pinto: Reason Reclaimed 191 claim that is somehow already suggested by the conception of proof standards in legal contexts, where “probabilistic finding is insufficient unless it is a finding of certainty, and (…) a finding of certainty is not required” (p. 266). This statement makes clear that, whatever is required for probative validity in legal contexts, it is not a matter of Bayesian probability—because the minimal necessary condition refers to something which is neither probable nor certain, strongly suggesting that probability is not the real issue. The book’s final chapter by Rescher is a fascinating speculation on unknowable facts, unanswerable questions, and vagrant predicates— that is, on some intrinsic limitations of the relationship between language and world. Unknowable facts are, as the name implies, facts that cannot possibly be known, either locally, i.e. by some specific person or group (e.g., ‘F is a fact that I do not know’), or globally, i.e. by any kind of cognitive being (e.g., ‘F is a fact that nobody knows’). When used in the context of inquiry, they immediately produce unanswerable questions, like “What is an example of a fact that you do not know?” The question is unanswerable in the sense that any genuine answer would be false, due to the very nature of the question—in the example above, providing any fact to answer the question would show that such a fact is known to you, therefore your claim that it is not is false. Obviously, similar questions can be easily and truthfully answered by saying “I do not know,” but this is an answer only in a pragmatic sense, which is not what interests Rescher here. Unknowable facts also determine the existence of noninstantiable properties, that are expressed by what Rescher calls vagrant predicates: “F is a vagrant predicate iff (∃u)Fu is true nevertheless Fu0 is false for each and every specifically identified u0. Such predicates are ‘vagrant’ in the sense of having no known address or fixed abode” (p. 271). All these curious propositional beasts share a common origin in terms of general principles, that Rescher illustrates using the Musical Chairs Perplex, named after the famous game: “Here there will be no player who is unseatable: individually considered, any player could be seated. But matters stand otherwise collectively. It is not possible—and impossible as a matter of necessity—that every player can be seated. While seatability is universal among the individual involved, the fact of unseated individuals is inescapable” (p. 273). The analogy with the game makes clear that the issue at stake here is one of numerical discrepancy between linguistic tools and ontological facts. The rest of the chapter is devoted to show that, from what we know of language and reality, it is reasonable to assume that there are quantitatively more facts than truths, where ‘truths’ are understood as linguistic/symbolic representations that happen to match the facts to which they refer. As a consequence, “when reality and language play their game of Musical Chairs, some facts are bound to be left in the lurch when the music of language stops” (p. 279). Unknowable facts, unanswerable questions, and vagrant predicates are just the offspring of such a numerical disparity. Rescher, however, does not explicitly link this conclusion with the Fabio Paglieri 192 enterprise of informal logic and argumentation theory, leaving the reader with the (hopefully answerable) question: “So what?” To my mind, it would be mistaken to think of Rescher’s unknowable facts as “setting the boundaries” of argumentative inquiry, since it is hard to put an unknowable fact to any positive use, including drawing boundaries, for the very fact that it is unknowable—that is, a concrete instance of it cannot be produced. Instead, the boundaries itself are likely to be unknowable, as in “the claim nobody will ever think” or “the argument none will ever conceive”, and this may be seen of a reminder that argumentation and inquiry are open-ended enterprises—not just because of our poor argumentative abilities, but rather out of necessity, if Rescher is right. This, I submit, is nothing but jolly good news. 5. Finally, this remarkable collection is completed by a selective bibliography of publications by Blair and Johnson in informal logic. This provides a valuable bird-eye view on their intellectual and scholarly achievements, the very reason that prompted so many of their distinguished colleagues to pay them tribute in this book. A book that, as I hope these notes would have shown by now, is both broad in its scope and deep in its insight, providing a refreshing experience for any scholar interested in the study of argumentation. Reference Toulmin, Stephen (1958/2003). The uses of argument. Cambridge: Cambridge University Press. FABIO PAGLIERI Istituto di Scienze e Tecnologie della Cognizione CNR Roma