Microsoft Word - Review of Finocchiaro34.2PubVers2.docx ©    J.  Anthony  Blair.  Informal  Logic,  Vol.  34,  No.  2  (2014),  pp.  219-­‐239.   Critical Review Meta-argumentation, An Approach to Logic and Argumentation Theory By Maurice A. Finocchiaro Studies in Logic, Logic and Argumentation, Vol. 42. London: College Publications, 2013. Pp. vii, 1-279. ISBN 978-1-84890- 097-4. UK£12 US$17.10 CDN$21.121 Review by J. ANTHONY BLAIR Centre for Research in Reasoning, Argumentation and Rhetoric University of Windsor Windsor, ON Canada N9B 3P4 tblair@uwindsor.ca Meta-argumentation sets out, defends and illustrates an ap- proach to the study of meta-arguments. These are arguments about arguments and argumentation. Meta-arguments are to be contrasted with “ground-level” arguments, which are “typically about such things as natural phenomena, historical events, hu- man actions, etc.” (p. 1). But meta-arguments are not just argu- ments about ground-level arguments, for they can be arguments about arguments about ground-level arguments—second-level or higher meta-arguments. An argument, for Finocchiaro, is “a set of statements that attempts to justify a claim by supporting it with reasons, or by defending it against objections, or both” (p. 1). (More about this definition when I get to Chapter 4.) The book is an amalgam of papers most of which have been published or delivered elsewhere. However it is thoroughly and meticulously organized. The Introduction contains a concise outline, and each chapter includes, besides an introduction and interim summaries of component parts, a concluding summary that succinctly describes its contents. Between the Introduction and the Conclusion, the book’s 13 chapters are grouped into three parts. In the first part, three chapters set out Finocchiaro’s “meta-argumentation ap-                                                                                                                 1  Available from Amazon in the UK, US and Canada May 2014 prices.   J.  Anthony  Blair       ©    J.  Anthony  Blair.  Informal  Logic,  Vol.  34,  No.  2  (2014),  pp.  219-­‐239.     220   proach”—his methodology, his basic principles of argument analysis, and a typology of meta-arguments. In the second part, in six chapters he applies this approach in examinations of a va- riety of topics in which meta-arguments figure prominently: dia- lectical definitions of argument, some methods of argument crit- icism, arguments about deep disagreement, conductive argu- ments and self-referential arguments. And in the third part, Fi- nocchiaro presents four case studies of historical meta- arguments, two by J.S. Mill, one by Hume and one by Galileo. Most of this review will be a description of the book’s contents in medium detail, with the interjection of an occasional comment or criticism. The remainder, following, will be a brief appraisal. Part I – The Meta-argumentation Approach Chapter 1. Methodological Considerations: Toulmin’s Applied Logic Logic, Finocchiaro begins, is, or at least should be, theorizing about significant (non-trivial) cases of argumentation and rea- soning. He describes in detail Stephen Toulmin’s method of log- ical theorizing, proposed in The Uses of Argument, which he commends to his readers and which he summarizes as follows: . . . Toulmin’s so-called “applied logic” is a methodolog- ical approach to logical theorizing that has the following features: it is pragmatic, in the sense of taking practice as essentially (i.e., co-equal with theory); epistemological, in the sense of attaching great importance to arguments for knowledge claims; simultaneously normative and de- scriptive; comparative, in the sense of sensitive to the dif- ferences and similarities of different fields; empirical in the sense of opposite of a priori; historical, in the sense of oriented toward the history of thought; and naturalist, in the sense of emphasizing description à la natural histo- ry, or analysis as contrasted to theory. (p. 10) It is this “applied logic” methodology that Finocchiaro thinks we should take from The Uses of Argument, and not the “Toulmin model,” which he regards as just a variation of the standard framework of minor premise, major premise, conclu- sion, support for major premise and rebuttal to objection, con- Review  of  Finocchiaro,  Meta-­‐Argumentation           ©    J.  Anthony  Blair.  Informal  Logic,  Vol.  34,  No.  2  (2014),  pp.  219-­‐239.     221   reason or counter-evidence.2 The latter is a substantive theory about how arguments can be analyzed, and not an example of the method at work; the former is the (undeservedly neglected) method Toulmin proposes for the field of logical theory and the study of argumentation. Finocchiaro contends that logic and argumentation theory need to be grounded “more robustly” on an observational ap- proach to argumentative reality. Formal deductive logic does this when serving as the theory of mathematical argumentation; but it is a prioristic when used to serve as a general theory of argumentation. However, alternative approaches suffer from a lack of a robust observational or pragmatic basis. Witness Pe- relman and Olbrechts-Tyteca, who, while right about many gen- eral points, rely excessively on context-less snippets for exam- ples, and as a result produce questionable analyses (their exam- ple of begging the question being a case in point). In particular, Finocchiaro recommends that the Toulmin method’s emphasis on observation should be conceived as the study of ”observational arguments,” that is inductive arguments broadly conceived: not just inductions by enumeration, but also “abductive arguments, conductive arguments, inferences to the best explanation, hypothetico-deductive confirmation, argu- ments from analogy, statistical syllogisms, and the like” and in addition, “ampliative, probable, plausible, and defeasible argu- ments” (p. 15). Argumentation theorists’ arguments, he notes, will thus be about ground-level arguments and thus be meta- arguments. “Logical theory and argumentation theory are or ought to be instances of meta-argumentation” (p. 15). And at the meta-argumentation level, some of the arguments will be induc- tive (e.g., enumerative inductions or arguments from analogy) and others will be deductive (e.g., about the entailment relations among observations, concepts or theoretical principles). The Toulminian approach to theorizing about argumenta- tion that Finocchiaro is advocating overlaps, he notes, with oth- ers known under different labels: “informal logic, practical log- ic, naturalized logic, logic of real arguments, argument analysis, philosophy of argument, argumentation theory, and immanent dialectical approach” (p. 16). Moreover, it is an approach that                                                                                                                 2  Thus   Finocchiaro   joins   those   who,   erroneously   in   my   opinion,   treat   Toulmin’s  “warrant”  as  a  premise—see  Hitchcock  (2003)  for  an  alterna-­‐ tive.  The  computer  scientists  Finocchiaro  cites  in  support  do  so  in  order   to  make  the  model  deductive,  which  suits  their  programming  needs  but   strays  from  Toulmin’s  intent.     J.  Anthony  Blair       ©    J.  Anthony  Blair.  Informal  Logic,  Vol.  34,  No.  2  (2014),  pp.  219-­‐239.     222   Finocchiaro has previously advocated under the rubric “histori- cal-textual approach” and put into practice in his books about Galileo’s ground-level arguments for and against the motion of the earth. What is new about this book, he says, is its focus on meta-argumentation: the theoretical meta-arguments of logicians and argumentation theorists to justify their theories and theoreti- cal claims, and famous historical meta-arguments. Chapter 2. Elementary Principles Chapter 2 introduces the basic details of Finocchiaro’s theory of argument interpretation and evaluation that corresponds to the approach described in Chapter 1. He begins by setting out the basic concepts, distinguishing reasoning from argumentation, and identifying different sorts of objections and counterargu- ments. He states what these things are (without defence), which in my view unnecessarily invites disagreement from those who think they are something else, whereas I think he is in fact stipu- lating how he understands or will use these concepts. “Reasoning,” he says, “is the activity of the human mind that consists of giving reasons for conclusions, reaching conclu- sions on the basis of reasons, or drawing consequences from premises” (p. 18). “Reasoning is linguistically expressed in ar- guments,” conceiving arguments as “series of propositions some of which are [allegedly?] based on others, where the intercon- nections are expressed by means of reasoning indicators” (p. 18). Note that this account without my bracketed insertion does not allow for bad arguments, but that is a slip, since two para- graphs later Finocchiaro refers to “[t]he relationship between reasons and conclusion claimed in a given argument” (p. 19, my italics). It is also slightly different from the definition introduced on the first page of the book and later defended at length in Chapter 4: “a set of statements that attempts to justify a claim by supporting it with reasons, or by defending it against objections, or both” (p. 1). The reader might want to question whether these two definitions are equivalent, but in any case, as we will see, it is the latter definition that Finocchiaro favours. An objection Finocchiaro defines as an argument against a controversial proposition. If the proposition has already been argued for, then an objection to it is a counterargument. Turning to argument structure, Finocchiaro distinguishes serial arguments, in which the reason for the conclusion is itself the conclusion of another argument; linked arguments in which two or more reasons “depend on” one another to support the conclusion; and independent arguments, in which there is no Review  of  Finocchiaro,  Meta-­‐Argumentation           ©    J.  Anthony  Blair.  Informal  Logic,  Vol.  34,  No.  2  (2014),  pp.  219-­‐239.     223   such interdependence among the reasons to provide the support. (He does not expand on what such dependence consists of. Two or more independent reasons can supply support to the same conclusion without being linked.) In this section Finocchiaro also introduces an argument structure diagramming system, which includes a complicated set of numbering conventions, to display and record these structural features of arguments. His conventions are no more complicated than others (such as those used in the Pragma-dialectical system), but like all such conven- tions they take some getting used to—which the reader had bet- ter do, since Finocchiaro uses them extensively elsewhere in the book. He stipulates that the evaluation of an argument “attrib- utes some merit or flaw” to it; it must include “the articulation of some rationale for the evaluative claim” (thus entailing meta- argument); and it will tend to come in degrees. He proceeds to offer the standard catalogue of types of criticism of arguments, with some interesting variations. There are the three basic types: criticisms of the conclusion, criticisms of the reasons, and criti- cisms of the relationship between a reason and the conclusion. He downplays the first as uninteresting, since it merely reiterates the disagreement that gave rise to the argument being criticized (although I would just note that if such a criticism refutes the conclusion, it follows that there must be something wrong with the argument(s) offered in support of it). Arguments criticizing a reason can have the conclusion either that it is not true, or that it is without foundation (i.e., without needed or adequate sup- port)—the “truth” vs. “acceptability” distinction found else- where (e.g., Johnson and Blair 1977). Finocchiaro distinguishes weak from strong arguments that criticize the relationship (of relevance) between a reason and its conclusion. The former al- leges mere unclarity; the latter, that the conclusion does not fol- low. He lists six different ways in which strong criticism of the reason-conclusion relationship of an argument—he dubs them “disconnects”—can be prosecuted. One is to show that the premises can be true while the conclusion is false, demonstrated by a counterexample, or by showing it does not follow neces- sarily, or analytically, or in virtue of the argument’s form, or that it violates the rules of deductive logic. A second, found es- pecially in arguments in which the conclusion explains the rea- sons, is to show that the conclusion is no more likely than a specified alternative also consistent with the reasons. A third is to show by means of a strong argument that a presupposition on which the original argument’s reason-conclusion relationship depends is false. A fourth is to show that what in fact follows J.  Anthony  Blair       ©    J.  Anthony  Blair.  Informal  Logic,  Vol.  34,  No.  2  (2014),  pp.  219-­‐239.     224   from the reason is a conclusion that is inconsistent with the pur- ported conclusion. A fifth is to show that the argument relies on equivocation: a term used in the premises is true in one sense, but false in the sense needed for the conclusion to follow. A sixth consists of showing that the argument begs the question because the conclusion is the same as a latent premise. The rest of Chapter 2 contains a demonstration of the above theoretical apparatus by the interpretation of the structure, and the evaluation, of two examples of arguments by Galileo. One is Galileo’s critique of the so-called observational argument for heavenly unchangeability; the other is Galileo’s discussion of the anti-Copernican argument from vertical fall. Chapter 3. Basic Types of Meta-Argumentation Finocchiaro here distinguishes the following three types of me- ta-reasoning / meta-argumentation. Argument analysis is the reasoned interpretation or evaluation (or both) of an argument, be it a ground-level argument or another meta-argument. Self- reflective reasoning and argumentation is argument analysis of an argument one constructs oneself. The construction of argu- ments that support premises or premise-conclusion relations, or of arguments against objections, is evidence of self-reflective meta-reasoning on the part of someone constructing a ground- level argument. As an example, Finocchiaro analyses an argu- ment of Galileo that illustrates self-reflective reasoning (thereby also exemplifying argument analysis). Argumentation theory is “the formulation, testing, systematization, clarification, and ap- plication of concepts and principles for the interpretation, evalu- ation and practice of argument or reasoning” (p. 39). Meta- arguments theorizing about argument and reasoning, Finocchia- ro says, tend to be more general, more systematic and more con- ceptually explicit than argument analysis or self-reflective rea- soning; the differences are a matter of degree. Finocchiaro closes the chapter with some comments about the relationship between argumentation theory and formal de- ductive logic—thereby expanding on points made in Chapter 1. Despite a large overlap, formal logic tends to be less empirical and more a priori; to be more focused on a special class of ar- guments (mathematical proofs), on the micro-structure of such arguments, and on their evaluation just in terms of deductive validity; and to be more systematic than argumentation theory. Chapter 3 ends the methodological section of the book. Review  of  Finocchiaro,  Meta-­‐Argumentation           ©    J.  Anthony  Blair.  Informal  Logic,  Vol.  34,  No.  2  (2014),  pp.  219-­‐239.     225   Part II – Theoretical Meta-arguments The six chapters of Part II, 4–9, consist of a series of meta- argumentative analyses of a variety of issues in argumentation theory, enabling Finocchiaro both to defend his own views on these topics and also to illustrate argumentation-theoretic meta- argument. Chapter 4. Dialectical Definitions of Argument In Chapter 4 Finocchiaro first introduces some useful, sharp dis- tinctions among various concepts of argument. He distinguishes between three versions of what he terms the “traditional” defini- tion and three versions of what he terms the “dialectical” defini- tion. One form of the traditional concept is exemplified by Co- pi’s “illative” definition: “a group of propositions of which one is claimed to follow from the others, which are regarded as providing support or grounds for the truth of that one” (p. 45), which specifies a relation but no structure, plus an intention of the arguer to justify, and thus a purpose. It is to be contrasted with the more abstract, non-teleological variant (e.g., Skyrms, Kalish and Montague) according to which only a set of sentenc- es or propositions with a premise-conclusion relation defines an argument; and also to be contrasted with the rhetorical variant, according to which the function of attempting to persuade is built into the definition (e.g., Scriven). Finocchiaro dubs the three variants of the dialectical con- ceptions of argument that he distinguishes: “moderate” (e.g., Goldman, Finocchiaro himself, p. 1), “strong” (e.g., Johnson) and “hyper” (e.g., the Pragma-dialectical school, e.g., van Rees). The moderately dialectical conception defines an argument as a claim supported by reasons or a claim defended against objec- tions or a combination of the two. The presence of at least one of the first two disjuncts is necessary, and of any of the three is sufficient to identify discourse as argument. The strong dialecti- cal conception requires both reasons supporting a claim and also a defense against objections to the claim as necessary (and suffi- cient) conditions of an argument. And the hyper dialectical con- ception takes only a defense against objections as both a neces- sary and sufficient condition of an argument. (Presumably de- fense against objections to the truth or acceptability of the rea- sons—as distinct from defense against objections to the claim itself—and defense against objections to the probative force of the reasons, would be meta-argument, not ground-level argu- J.  Anthony  Blair       ©    J.  Anthony  Blair.  Informal  Logic,  Vol.  34,  No.  2  (2014),  pp.  219-­‐239.     226   ment, although Finocchiaro does not make those distinctions here.) These distinctions make it clear that different argumenta- tion theorists are actually working with importantly different conceptions of what constitutes an argument. The remainder, and bulk, of Chapter 4 consists of fine- grained analyses (reconstruction, and evaluation) of what Finoc- chiaro plausibly takes to be Goldman’s case for the moderate dialectical concept of argument and Johnson’s case for a strong dialectical concept. Although criticizing Goldman on some points, Finocchiaro accepts parts of his argument. In his detailed assessment of Johnson’s arguments for the “dialectical tier” view, he finds much to criticize, and he concludes that what is defensible in those arguments supports only the moderate dia- lectical concept of argument. Finocchiaro closes the chapter with the consideration of three objections to this conception, and he argues that they fail to make their case. Chapter 5. The Hyper Dialectical Definition of Argument Finocchiaro begins Chapter 5 by making a convincing case that the Pragma-dialectical school is committed to the hyper dialecti- cal definition of argument, citing passages from van Eemeren and Grootendorst, Snoeck Henkemans and van Rees. He then argues that members of this school apparently fail to argue for their definition, although he proposes on a closer examination that what he calls an empirical “argument sketch” can be found—in his words, the defence that “if we apply the various principles of the pragma-dialectical theory, we can show that all argumentative discourse consists of attempts to overcome criti- cal doubt; namely that, appearances to the contrary, all actual arguments can be shown to conform to the hyper dialectical conception” (p. 68). And the research program of the Pragma- dialectical school, he avers, appears to support this claim. How- ever, he notes a problem with this defence. For to make their case the evidence must show not only that enlightening analyses of actual argumentation can be made using the hyper dialectical conception, but also that these analyses are better than those made using alternative conceptions of argument: the strongly dialectical, the moderately dialectical or the illative. In other words, to be consistent with their own concept of argument, their evidence must be presented in such a way as to address and remove the doubts, criticisms or objections of opponents. This they do not even set out to do, and so in terms of the hyper dia- lectical definition of argument, the empirical support they pro- Review  of  Finocchiaro,  Meta-­‐Argumentation           ©    J.  Anthony  Blair.  Informal  Logic,  Vol.  34,  No.  2  (2014),  pp.  219-­‐239.     227   vide for the Pragma-dialectical conception is at best an incom- plete argument. Finocchiaro notes that there is a reply to this criticism available to the Pragma-dialectical school, namely that what might have appeared to be an illative monological case for their conception of argument “can be easily reconstructed as being dialectical replies to the objections of an antagonist” (p. 69). However, he argues, such a move fails to establish the exclu- siveness or the primacy of the hyper dialectical conception, be- cause a refutation of an objection to a position constitutes an ar- gument for that position. Similarly, an argument for a position constitutes an answer to doubts about it. Finocchiaro sees this as a symmetry of the dialectical and illative tiers. Hence, neither can be said to have primacy over the other—which he takes to be an argument for the moderate or disjunctive conception. Although he is now a proponent of the moderate dialecti- cal definition of argument, it was not always so. Finocchiaro has the grace to quote a 1980 argument of his own in favour of the hyper dialectical definition. “Modesty and potential embarrass- ment” prevent him from evaluating it (p. 71). He closes the chapter with an examination of J.S. Mill’s strategy, in The Subjection of Women—of apparently exclusive- ly replying to objections to equality for women—as a possible instance of someone committed in practice to the hyper dialecti- cal definition of argument. He finds, however, that Mill had a particular reason in that book for focusing on objections; more- over, in parts of the book Mill advances illative arguments too, and so can be seen as a practitioner of the moderate dialectical conception. Chapter 6. Common Methods of Argument Criticism This chapter might more precisely be titled, “Six Ways of Argu- ing for Invalidity” (Finocchiaro doesn’t try to show that they are common). Since they are ways of arguing that the reasons- conclusion relation in an argument is invalid, they are types of arguments criticizing arguments, and thereby, types of critical meta-arguments. Finocchiaro begins by following Krabbe in setting out several ways of proving invalidity. When defining a valid argu- ment as one for which there is no situation in which the premis- es are true and the conclusion is false, if it can be shown of an argument that its premises are true but its conclusion is false, then that argument has been proven to be invalid. (And given that all formally valid arguments are valid, if follows, by modus J.  Anthony  Blair       ©    J.  Anthony  Blair.  Informal  Logic,  Vol.  34,  No.  2  (2014),  pp.  219-­‐239.     228   tollens, also that the argument is formally invalid.) A second way is to show that a situation is possible in which the premises are true yet the conclusion is false. A third way is to produce a paraphrase, expressing the gist of the argument, which consti- tutes an invalid logical form. A fourth way and a fifth way are methods of logical analogy, of which Finocchiaro discusses two versions, one due to Oliver and one due to Govier. These are methods of constructing, by generalizing from the argument in question, a logically parallel argument whose structure is flawed, thereby refuting the original argument. A sixth method (due to Woods and Hudak)—arguing “by parity of reasoning”— is to construct a parallel argument to the one being evaluated, which has a similar deep structure by virtue of which it has the same logical form, so that the two arguments stand or fall to- gether. Chapter 7. Deep Disagreements, Fierce Standoffs, Etc. In this chapter Finocchiaro analyzes the argumentation involved in the dispute about the impossibility of resolving deep disa- greements by means of argument that was initiated by Robert Fogelin’s 1985 Informal Logic article, “The Logic of Deep Dis- agreements” (reprinted in Fogelin 2005). Fogelin’s thesis was, as Finocchiaro interprets it, that “there are important disagree- ments that are not subject to rational resolution, but only to rhe- torical persuasion” (p. 85). Fogelin’s article began a discussion that has included, so far, close to a dozen articles. The argu- ments involved in this dispute are all meta-arguments, and Fi- nocchiaro’s analyses (reconstructions and evaluations) of them would thus seem to be meta-meta-arguments. The chapter is thus a case study of meta-argumentation. As well, Finocchiaro uses the opportunity to employ, and thereby illustrate, his prem- ise-conclusion numbering system and diagramming convention. In detail he analyses Fogelin’s argument and then those of Lugg, Davson-Galle, Feldman, Turner and Wright, Campolo, Frie- mann, Adams, Phillips, Godden and Brenner, Woods, and John- stone. I will not attempt to summarize his reconstructions and closely-argued evaluations here. Suffice it to say that the chapter is a tour de force—a model of historical analysis of a multi- party debate that has wound its way back and forth, with many participants referencing their predecessors, over a quarter- century. No responsible scholar who takes up the issue of the possibility of the rational resolution of deep disagreements can fail to deal with Finocchiaro’s treatment of the meta- argumentation of this dispute. Review  of  Finocchiaro,  Meta-­‐Argumentation           ©    J.  Anthony  Blair.  Informal  Logic,  Vol.  34,  No.  2  (2014),  pp.  219-­‐239.     229   Chapter 8. Conductive Arguments, Pro & Con Reasoning, Etc. The terms ‘conduction’ and ‘conductive’ as referring to a partic- ular type reasoning and argument were coined in a monograph on ethical theory by the American philosopher Carl Wellman (Wellman 1971). According to Wellman, conduction is “that sort of reasoning in which 1) a conclusion about some individual case 2) is drawn non-conclusively 3) from one or more premises about the same case 4) without any appeal to other cases” (Wellman, p. 54). There may be one reason or several given for the conclusion, or the conclusion may be drawn after weighing both positive and negative considerations (Wellman, pp. 55-57). Conductive arguments are allegedly neither deductive nor in- ductive. Wellman’s views were brought to the attention of argu- ment theorists by Trudy Govier in (Govier 1980), and she intro- duced her own version of conduction in her textbook (Govier 1985, 7th ed. 2010) and in other places noted by Finocchiaro. David Hitchcock developed a sympathetic but somewhat differ- ent version of conduction from those of Wellman and Govier. Criticisms of one or another aspect of Govier’s approach have been registered by Derek Allen, Robert Ennis, and Frank Zenker, and a broader critique of conduction has been registered by Harald Wohlrapp. In Chapter 8, Finocciaro sets out and eval- uates the respective arguments for these various views. To be sure, all this argumentation is meta-argument, and so is grist for Finocchiaro’s mill. Moreover, as he importantly notes, balance-of-considerations conductive reasoning and ar- gument requires a meta-argument as a component of the argu- ment itself—a meta-argument to show that the pros outweigh the cons (or vice-versa). He also argues that conductive argu- ments are in effect endorsed (without that label) by J.S. Mill, Ralph Johnson and Christian Kock, and that they are to be found in advertisements. Finally, in the interest of establishing that conductive arguments are not uncommon historically or current- ly, Finocchiaro analyzes the argumentation of a piece by New York Times columnist David Brooks (included in an appendix to the chapter) and of Galileo’s Dialogue, as instances of conduc- tive argument. In my opinion, there is a temptation to conflate (a) con- ductive arguments, which involve the weighing of pro and con considerations (both accepted by the arguer) plus a judgement that those on one side are weightier than those on the other, with (b) arguments that entertain pro and con considerations, and contain support for the pro side against criticisms and the refuta- J.  Anthony  Blair       ©    J.  Anthony  Blair.  Informal  Logic,  Vol.  34,  No.  2  (2014),  pp.  219-­‐239.     230   tion of arguments supporting the con side. In the former, the proponent accepts both the pro and the con considerations, and therefore must balance their respective probative force before arriving at a conclusion. In the latter, the proponent accepts (and reinforces) the pro considerations but rejects (and rebuts) the con considerations, thereby passing straight to the conclusion. In some cases it seems—going by Finocchiaro’s account of Gali- leo’s arguments—that Galileo is not accepting contrary argu- ments but rather, while entertaining them seriously, in the end rejects them. In such cases Galileo is considering the pros and cons of his geokinetic and heliocentric theory, but not reasoning or arguing conductively. Chapter 9. Self-referential Arguments By ‘self-referential arguments’ here Finocchiaro means “argu- ments that are either intended to exemplify [their proponent’s] own theoretical principles, or are de facto subject to interpreta- tion or evaluation in light of their own principles” (p. 162). Giv- en this definition’s second disjunct, it strikes me that anyone’s arguments are self-referential in this sense, since all arguments are liable to ex concessis ad hominem review. Be that as it may, in Chapter 9 Finocchiaro analyzes arguments by Blair3 that logic does not exhaust the subject matter of the philosophy of argu- ment, by van Eemeren and Houtlosser that fallacies consist of derailments of strategic maneuvering, and by Johnson about several principles about the role of objections in argumentation and its evaluation. As he reconstructs it, Finocchiaro declares Blair’s argu- ment to be deductively valid, but that it uses too narrow a defini- tion of logic, and violates of some of his own declared princi- ples, such as that full-bodied argument must take the arguer and the audience into account. He argues that Blair: identifies logic with deductive logic, thus excluding inductive logic; admits permitting instructors in the department he chaired, contrary to his own prescription, to teach that deductive logic as all that is required for argument analysis; and addresses his arguments to an audience that already accepted his conclusion, and thus is in no need of argument. Next, Finocchiaro provides an exposition of an article by van Eemeren and Houtlosser on the tu quoque fallacy, one of the places in which they spell out their revision of the Pragma- dialectical concept of fallacy. They replace the previous con-                                                                                                                 3  Full  disclosure:  the  Blair  in  question  is  this  reviewer.   Review  of  Finocchiaro,  Meta-­‐Argumentation           ©    J.  Anthony  Blair.  Informal  Logic,  Vol.  34,  No.  2  (2014),  pp.  219-­‐239.     231   cept, that fallacies are violations of the rules of the ideal model called a “critical discussion,” with the revised concept, that fal- lacies are uses of rhetorical devices with the aim of settling dif- ferences of opinion in one’s own favour (called “strategic ma- neuvering”) that violate these rules. Finocchiaro criticizes the example used to illustrate a case of tu quoque strategic maneu- vering by William of Orange as unclear and not definitive. His principal criticism, however, is that without needed justification of certain features of the theory, which is not provided, the au- thors could be accused of ad hoc alterations to the Pragma- dialectical theory of fallacy, and thus be fairly accused of the same methodological failure (ad hoc-ness) for which they con- stantly berate proponents of alternative accounts of fallacies— and in consequence be non-fallaciously charged with tu quoque themselves. The last part of Chapter 9 is a discussion of an article by Johnson in which (according to Finocchiaro) he asserts three obligations of an arguer to deal with objections (namely to re- spond to strong objections, to objections that are close to one’s arguments, and to objections that are salient) and three ways to deal with such objections (state the objection accurately and faithfully, respond adequately—i.e., defuse the objection, acknowledge that the objection is appropriate). Finocchiaro fol- lows his exposition of Johnson’s article with seven more or less extended comments, some critical, and several constructive de- velopments or refinements. The latter include the fruitful no- tions of pre-empting objections and strengthening objections before dealing with them. Part III – Famous Meta-arguments In Part III, Finocchiaro analyzes four “historically influential” meta-arguments that “have become classics” on “topics that are intrinsically important, universally significant and perennially interesting” (p. 178). Two are by J.S. Mill (on liberty of argu- ment and on women’s liberation), one is by Hume (on intelli- gent design) and one is by Galileo (on the motion of the earth). These are in effect case studies of important thinkers’ arguing about arguments, affording Finocchiaro the opportunity to offer his interpretations and to argue about these thinkers’ meta- argumentation: thus themselves instances of meta-meta- argumentation. J.  Anthony  Blair       ©    J.  Anthony  Blair.  Informal  Logic,  Vol.  34,  No.  2  (2014),  pp.  219-­‐239.     232   Chapter 10. Mill on Liberty of Argument Finocchiaro provides an analysis of Mill’s argumentation in the second chapter of On Liberty, where Mill argues for the liberty of thought and discussion. Finocchiaro uses the numbering sys- tem that he introduced in Chapter 2 to indicate the structure of the support relationships in his interpretation of the arguments. The analysis is fine-grained, and would be of interest to anyone considering the details of Mill’s case. Besides this exegesis of a part of On Liberty, in the last section of the chapter Finocchiaro addresses some issues related to the structure, the contents and the epistemology of Mill’s ar- gumentation. How many lines of argument did Mill offer, three or four? (Four main reasons.) Are parts of Mill’s argument con- ductive? (Yes.) Finocchiaro also argues that the freedom of thought Mill is defending amounts to the freedom to argue. Moreover, he contends, Mill holds that believing something without argument consists both of believing it without support- ing reasons and also of believing it without knowing how to de- fend it from objections—and thereby endorses the moderately dialectical conception of argument (which Finocchiaro also es- pouses). It is clear from Fincocchiaro’s analysis of Mill’s case for freedom of discussion that Mill produces both arguments aimed at directly supporting his claims (the illative component) and also arguments that are meant to support them indirectly by refuting objections to those claims (the dialectical component). Finocchiaro also finds in Mill’s argumentation “the thesis that argument is a key method in the search for truth and the acquisi- tion of knowledge” (p. 189). Noting that Mill’s thesis is defend- ed by a range of epistemological principles, he raises for future research the question of the relationship between argumentation theory and the theory of knowledge: whether one is prior to the other or the two are mutually reinforcing. Chapter 11. Mill on Women’s Liberation The subject of this chapter is not Mill’s ground-level arguments for an end to the legal subordination of women, but rather the meta-argumentation of the book’s first chapter. There, according to Finocchiaro, Mill takes on two presumptions against the book’s thesis, namely that the equality of women conflicts with deep feelings and strong emotions, and that it conflicts with al- most universal opinion and usage. Unless these presumptions can be undermined, he says, any ground-level arguments for Review  of  Finocchiaro,  Meta-­‐Argumentation           ©    J.  Anthony  Blair.  Informal  Logic,  Vol.  34,  No.  2  (2014),  pp.  219-­‐239.     233   women’s liberation will be ineffective and counterproductive. Finocchiaro sees the argumentation of Chapter 1 to be aimed at a causal undermining of these negative presumptions along with showing a positive presumption in favour of equality for wom- en. This is not the place to repeat Finocchiaro’s detailed anal- yses, but his general interpretation is worth emphasizing. He is arguing that Mill needs to clear the way for his ground-level ar- guments in the last three chapters of The Subjection of Women by showing that the grounds for the standing bias against wom- en’s equality are without foundation, and that actually the onus could be seen to rest with its deniers. Although Finocchiaro does not say so, this move might be regarded as a rhetorical device aimed at preparing the audience to be receptive (or less unrecep- tive) to the ground-level arguments. What he does say is that Mill’s arguments against these presumptions can be regarded as indirect support for his conclusion, for they constitute dialecti- cal-tier level rebuttals of objections to women’s liberation. I think that this is so, but only indirectly. For the gist of Finocchi- aro’s analysis is that the presumptions Mill objects to are objec- tions to considering women’s liberation—to entertaining argu- ments for it—not objections to women’s liberation or to argu- ments for it. Of course, if one is opposed to considering a policy change, usually (but not always) it is because one is opposed to making the policy change, and Mill’s adversaries were opposed to equality for women. Finocchiaro ends the chapter by revisiting the issue of whether argument is exclusively dialectical—the view he bap- tized as “hyper-dialectical” in Chapter 5. He concedes that ground-level arguments can be conceived as replies to “prime or minimal” objections, and thus as dialectical-tier arguments. But he contends that the refutation of an objection can equally be conceived as a ground-level (illative-tier) argument: the argu- ment that an objection fails counts as support for the standpoint at issue. “There thus seems to be a symmetry between the illa- tive and the dialectical tiers” (p. 198). He goes on to defend a similar symmetry between the meta-argumentation level and the illative and dialectical tiers. Chapter 12. Hume on Intelligent Design This chapter contains a detailed analysis of the design argument for God’s existence discussed in Hume’s Dialogues Concerning Natural Religion. Finocchiaro begins by considering and reject- ing Stephen Barker’s analysis of it as an argument from analo- J.  Anthony  Blair       ©    J.  Anthony  Blair.  Informal  Logic,  Vol.  34,  No.  2  (2014),  pp.  219-­‐239.     234   gy, then produces his own analysis of the argument from design as it was stated by Cleanthes and elaborated by Philo. The statement of the design argument is followed by a detailed anal- ysis of each of the fourteen criticisms of it that Finocchiaro identifies in the Dialogues. All but two of these are arguments about the design argument, and so are meta-arguments; the last two are ground-level arguments against the conclusion of the design argument. Chapter 13. Galileo on the Motion of the Earth In this chapter Finocchiaro subjects to detailed analysis Gali- leo’s argument against Aristotle’s geostatic argument from ver- tical fall. That is the argument that if the earth moved, objects would not fall vertically (or not be seen to fall vertically); but they do (or they are seen to); hence the earth does not move. According to Finocchiaro’s analysis, Galileo distinguishes be- tween two interpretations of Aristotle’s argument, and argues that on either interpretation, it begs the question (though in dif- ferent ways) and that therefore the argument from vertical fall is not a convincing way to show that the earth stands still. I have just stated the bare skeleton on Galileo’s refutation. Finocchia- ro’s analysis is fine-grained and detailed, taking up some 12 pages, and I will not try to give a more detailed account of it here. Finocchiaro closes the chapter with a discussion of some “conclusions” suggested by his analysis—about question- begging arguments, about the relation between logic and rheto- ric, and about doing argumentation theory as meta- argumentation. He notes that Galileo employed two different kinds of begging the question charges in his critiques. One alleges that an argument relies on a premise identical to the conclusion; the other, that an argument relies on a premise that is no less prob- lematic than the conclusion. Finocchiaro takes this to show that “there are two main types of question-begging arguments” (p. 234). However, it’s hard to say which are the “main” types. There is at least a third type, namely the argument in which a premise presupposes the truth of the conclusion. Finocchiaro might have referenced the literature on begging the question in making his claim. Finocchiaro works with broad conceptions of logic and rhetoric. As he conceives them, “Logic ... pertains to the domain of rationality, reasoning, demonstration, and relationships among truths,” while “rhetoric pertains to the realm of persua- Review  of  Finocchiaro,  Meta-­‐Argumentation           ©    J.  Anthony  Blair.  Informal  Logic,  Vol.  34,  No.  2  (2014),  pp.  219-­‐239.     235   sion, acceptance and relationships among beliefs” (p. 234). He notes that by this account, Galileo’s critique of the argument from vertical fall includes several rhetorical choices Galileo made, all consistent with logically impeccable arguments. He argues that Galileo’s charges of question-begging against the argument require the presupposition that the argument be logi- cally valid, for the persuasiveness of the argument (its rhetorical force) trades on the logical validity of the inference from the guilty premises. Finocchiaro notes that Galileo’s argumentation is meta- argumentation of two types. One is arguments about the struc- ture, merits and defects of the ground-level argument from ver- tical fall. The other is arguments about an issue raised by the argumentation about the ground-level argument, namely the ex- tent of Aristotle’s logical authority, and the domains where Aris- totle’s views are authoritative. Finocchiaro notes that a third type of meta-argumentation might have been raised during this stage of the dialogue, namely argumentation about methodolog- ical principles (in particular, the naive empiricist principle that the human senses are reliable sources of knowledge). All meta-argumentation is theorizing, Finocchiaro asserts, though not all theorizing is meta-argumentation. But must all theorizing about arguments be meta-argumentation? Not neces- sarily, he replies. For it is not clear that such theoretical tasks as concept formation, verbal expression, and propositional explicit- ization involve argumentation. Conclusion Argumentation Theory as Meta-argumentation In a five-page Conclusion, Finocchiaro repeats what he takes to be the principal thesis of each of the preceding chapters, and contends that “the main conclusion supported by these thirteen theses is that argumentation theory can and ought to be prac- ticed as meta-argumentation” (p. 244). Commentary This book has many merits. The tools to assist the reader are helpful. There is a de- tailed table of contents, an extensive bibliography, and a useful index. Each chapter begins with a helpful introduction and there are clear summaries at the end of most sections and of each chapter, and at the end of the book. The writing is clear and the explanations for the most part are lucid. J.  Anthony  Blair       ©    J.  Anthony  Blair.  Informal  Logic,  Vol.  34,  No.  2  (2014),  pp.  219-­‐239.     236   The book exhibits a wide range of scholarship, both in contemporary argumentation theory and the history of science and of philosophy. Finocchiaro’s erudition is intimidating. There is extensive critical interaction with recent and contempo- rary argumentation theorists—I count at least 60 names of peo- ple whose work is mentioned or discussed—and he mentions over 50 ancient, modern and contemporary scientists, philoso- phers and logicians. I find the Toulmin-Finocchiaro principles that should gov- ern theorizing about argumentation and arguments persuasively explained and defended. It seems sound advice that analysis should include under- standing the argument as well as interpreting its structure, should including positive no less than negative evaluation, and that both should be backed by arguments. Finocchiaro’s classification of concepts of argument into traditional and dialectical, each with three variations, is original and illuminating. Without it I could not have formulated the amendment to his definition of argument that I suggest below. The argument-structure numbering system and the dia- gramming system help illuminate extended argumentation. It would be nice if they captured functional roles (e.g., criticism of a premise vs. reply to that criticism; argument directly support- ing the conclusion vs. argument indirectly supporting the con- clusion by refuting an argument for the contrary or contradictory of the conclusion), but they don’t. To be fair, I don’t know of a typographical system that does (although some computer- generated systems might). Finally, Finocchiaro’s many analyses of meta-arguments throughout the book are models of exhaustive and meticulous practice and demonstration of the method he preaches. I have just a couple of reservations. Recall that, for Finocchiaro, an argument consists of “a set of statements that attempts to justify a claim by supporting it with reasons, or by defending it from objections, or both” (p. 1). Depending on what counts as an objection to a claim, an argu- ment so defined might include a meta-argument. Here is why. If an “objection” to a claim is simply a request for a justification of it, or an expressed doubt about it or disagreement with it, unac- companied by any reason, then to defend the claim from such objections simply consists of attempting to justify the claim by supporting it with reasons. That’s a ground-level argument. However, if an “objection” is an argument against the claim, then defending the claim consists of attempting to justify reject- ing that argument by supporting the rejection with reasons. But Review  of  Finocchiaro,  Meta-­‐Argumentation           ©    J.  Anthony  Blair.  Informal  Logic,  Vol.  34,  No.  2  (2014),  pp.  219-­‐239.     237   that is an argument about an argument: a meta-argument. Hence defending a claim against argued-for objections is meta- argument. Presumably Finocchiaro would not be happy with this result, for he needs his definition of ‘argument’ to be independ- ent of his definition of ‘meta-argument,’ since the latter relies on the former. I wonder whether the following modification of Finocchi- aro’s definition of argument might help. Define an argument as “a set of statements that attempts to justify a claim by support- ing it with reasons.” Full stop. But add that supporting reasons may attempt to justify a claim directly or indirectly, or both. In indirect support, the contention is that the claim is supported by the fact that there are good reasons for rejecting arguments against the claim. In the case of direct support, the contention is that the reasons support the claim without reference to argu- ments against it. In the case of an argument offering indirect support, the argument against the argument for the objection is a meta-argument. Thus claims may be supported by arguments or by meta-arguments, or both. In Finocchiaro’s terminology, on this definition, the presence of an illative tier is not necessary, but is sufficient, for the presence of an argument. I need to add that I am following Finocchiaro here, who is interested in the word ‘argument’ in the sense of what gives rea- son to adopt a claim, of which all the definitions he considers are variations. There are plenty of other senses of the word ‘ar- gument’ that he simply does not discuss. Recall that Finocchiaro distinguishes three types of meta- argumentation. Arguments about the interpretation or about the assessment of ground-level arguments—in short, arguments about analyses of ground-level arguments—are one type. Argu- ments expressing one’s reasoning involved in drawing critical challenges about one’s own illative arguments—that the premis- es or the inference can or might be questioned or challenged— and arguments whose conclusions are responsive to those criti- cal challenges, comprise a second type. And arguments with conclusions pertaining to argumentation theory are a third type. Argumentation theory, recall, consists of general propositions about argumentation that are systematic and consistent: “the formulation, testing, systematization, clarification and applica- tion of concepts and principles for the interpretation, evaluation and practice of argument and reasoning” (p. 39). If Finocchiaro were to accept my friendly amendment to his definition, then I don’t know whether he would classify meta-arguments indirect- ly supporting a claim as belonging to the second (self- J.  Anthony  Blair       ©    J.  Anthony  Blair.  Informal  Logic,  Vol.  34,  No.  2  (2014),  pp.  219-­‐239.     238   referential) type of meta-argument or whether he would say that they constitute a fourth type. What is the payoff for noting that many arguments are me- ta-arguments? What are the theoretical benefits of this observa- tion? Finocchiaro contends that there is a method of meta- argumentation, by which he means the Toulmin-inspired method of theorizing about arguments and argumentation that he de- scribed in Chapter 2. I applaud that method, and it certainly ap- plies to ground-level arguments and to meta-arguments, but I can’t see the novelty of that observation. It is assuredly accurate to note that arguments about the pros and cons of various defini- tions of ‘argument’ are meta-arguments, but I don’t see how that alters the contents of that argumentation. He notes that common methods of criticizing arguments are meta-arguments, but makes no claim that this observation alters that nature of those types of critique. Finocchiaro contends that going about arguing over deep disagreements with certain meta-argumentative virtues— e.g., open-mindedness, fair-mindedness, from the opponent’s commitments—are jointly necessary and individually helpful, if not jointly sufficient, for resolving such disagreements. Again, however, I do not see the gain of adding the label of “meta- argumentation,” however accurate it is to do so. That there are arguments about arguments in conductive arguments and in self- referential argumentation is true, but what are the implications of that fact? There is no question that Mill is arguing about ar- guments when defending liberty of discussion and legal equality for women against opposing arguments, that Hume does so in critiquing design arguments for God’s existence, and that Gali- leo does so in arguing against Aristotle’s vertical fall argument for the geo-centric, geo-static thesis. And I think there is no question that Finocchiaro’s analyses of these meta-arguments are scholarly and illuminating. However, I am not persuaded, even by his demonstrations, that the concept of meta-argument was doing any heavy lifting in his analyses. What would count as heavy lifting? I would suggest some such features as the following. Meta-arguments would have dif- ferent properties from ground-level arguments—that is, proper- ties in addition to not being ground-level arguments but rather arguments about arguments. They would be subject to different criteria of evaluation. They would have different structures or different functions. Their conclusions would have a different epistemic standing. It could well be that meta-argumentation is distinctive in some of these ways or in other significant ways. I just don’t see that Finocchiaro argues for any such distinctive properties or that his discussions reveal their existence. So I Review  of  Finocchiaro,  Meta-­‐Argumentation           ©    J.  Anthony  Blair.  Informal  Logic,  Vol.  34,  No.  2  (2014),  pp.  219-­‐239.     239   would file the answer to the question about the payoff for notic- ing that certain arguments are about arguments or argumentation under Unfinished Business. On balance, in my opinion, the book’s virtues outweigh any defects. Meta-argumentation belongs in the library of every scholar of argumentation theory.4 References Fogelin, Robert. 2005. “The Logic of Deep Disagreements.” In- formal Logic, 25: 3-11. Govier, Trudy. 1980. “Carl Wellman’s Challenge and Re- sponse.” Informal Logic Newsletter, 2(2): 10-15. Govier, Trudy. 1985. A Practical Study of Argument, 1st ed. Belmont, CA: Wadsworth. Hitchcock, David. 2003. “Toulmin's warrants.” In Frans H. van Eemeren, J. Anthony Blair, Charles A. Willard and A. Fran- cisca Snoeck Henkemans (Eds.), Anyone Who Has a View: Theoretical Contributions to the Study of Argument, pp. 69- 82. Argumentation Library, Volume 8. Dordrecht/Boston/ London: Kluwer Academic Publishers. Johnson, Ralph H. and Blair, J. Anthony. 1977. Logical Self- Defense, 1st ed. Toronto: McGraw-Hill Ryerson. Wellman, Carl. 1971. Challenge and Response: Justification in Ethics. Carbondale: Southern Illinois University Press.                                                                                                                 4  My  thanks  to  Ralph  Johnson  and  Christopher  Tindale  for  comments   that  led  to  corrections  to  and  improvements  of  an  earlier  draft.