88 Bickenback & Davies, Good Reasons for Better Arguments Despite these flaws that I have described, this remains a very important and good book, and well worth the read. There is too much that has not been said about trust for this not to be an important book. Its most important virtue is the implicit way in which it recognizes the complexity of trust and distrust, and the complex ways in which society itself requires trust. By taking on so many of the facets of social trust, Social Trust and Human Communities continually reminds us that wherever there is society, cooperation, and human flourishing, there is also trust. Note lIn her well-known article on trust, "Trust and Antitrust," Baier discusses a hypothetical scenario in which, trusting strangers in a library to simply leave her alone, she is instead saved from a falling brick by being pushed out of the way. In a case like this, Baier concludes that the heroic stranger, who did not simply leave her alone, would have done more than she was trusted to do, not less, and that Baier would, in such a scenario, have reason to be grateful to the stranger. Govier, however, misinterprets Baier as being "dis- turbed" rather than "pleased" by the stranger's actions. See page 115 of the book under discussion. Celeste Friend, Department of Philosophy Miami University. Oxford OH 45056 U.S.A-3644 friendcm@muohio.edu Good Reasons for Better Arguments by Jerome E. Bickenbach and Jacqueline M. Davies Peterborough: Broadview Press, 1997. ISBN 1-55111-059-8. Paper. CDN$26.95, US$21.95 Reviewed by Don S. Levi This new textbook is intriguing because of its reliance on the ideas of Habermas. Although many of the lessons, including those on Venn diagrams and truth tables, Mill's Methods, fallacies, and how sampling or analogy can go wrong, are not new, there are several topics that are not usually found in other texts, including a section on game theory; whether fictional analogues are problematic; the paradigm caselcounter-example technique for definition; and such fallacies as appeal to vanity, two sidedfairness (where it is assumed that one must be either for or against),just world hypothesis (where we favor © In/oTmal Logic Vol. 20, No.1 (2000) Book Reviews/Comptes rendus 89 explanations that make things come out right) and objectionable cause (where we want something we find objectionable to be responsible for something deplorable). However, the real selling point of the text is its reliance on Habermas for their view of argumentation as a means of "pursuing agreement through rational discourse" (p. 11). That selling point would have been stronger if Bickenbach and Davies had developed a Habermasian pedagogy to go along with their Habermasian theory. To illustrate effectively how argumentation can be a means of achieving agree- ment, the different sides on a controversial issue should be shown to be par- ticipating in a dialectical exchange where their views are refined or even trans- formed by participating in the exchange. Although the authors repeatedly refer to argumentation as a process, there is insufficient evidence in their teaching practice that they actually think of it that way. Their neglect of the importance of dialectic shows in their focus on "trans- parent persuasion," which "makes the rationality of agreeing with the conclu- sion transparently clear" (p. 29). The agreement is to be reached by an ideal- ized audience, which, seemingly, is going to be influenced only by the sound- ness of the premises of an argument, or the sufficiency and relevance of those premises for the conclusion, and not by how the argument is stated. This emphasis on an idealized audience may explain why the authors so often cite as examples arguments that they themselves devised, arguments where there does not seem to be an arguer, let alone an audience. Habermas may not be responsible for their neglect of rhetoric, but he is responsible for their key concepts of instrumental and communicative ration- ality, which are only two of a number of technical concepts, including 'trans- parent persuasion', that the authors teach. They cite Aristotle as the authority for their key principle that the standards for good argument are relative to the aims of the arguer. When the aims are clear and uncontroversial, and only the means for achieving them are the subject of reasoning or argument, then 'instrumental rationality' applies. When the ends are unclear or there is a con- flict over what they should be, then it is a matter of 'communicative rational- ity'. The authors consider this second kind of rationality to be of especial importance because its goal is the same as the fundamental goal of communi- cation, namely, that of "bringing us closer to the best understanding of our- selves and our world that it is possible for humans to have" (p. 10). This distinction between the different kinds of rationality may be difficult to apply to particular examples of argumentation. But the real use to which the authors put the distinction is in classifying the three types of reasoning that they discuss: deduction and induction have an instrumental rationality; 'prac- tical reasoning' has a communicative rationality. Although deduction and in- duction have different standards of rationality, reflecting the difference in how much certainty each aims at achieving, the standards involve "established procedures and rules, formulae, algorithms that can-like tools-be manipu- 90 Bickenback & Davies, Good Reasons for Better Arguments lated to achieve a variety of different goals" (p. 9). "Practical reasoning" is different because its objectives may come to be questioned in the process of determining how best to realize them. Moreover, any given argument with a certain over-all aim, may contain sub-arguments with different aims, and so the same argument may conform to the standards of both instrumental and communicative rationality. There are problems with what the authors have to say about each of the three types of reasoning that they discuss. Their chapter on deduction not only is too brief to be of use for the teaching of deductive techniques, but also is of limited value when it comes to explaining why these techniques are of interest, in part because the examples that the authors provide are so con- trived. They say that deductive standards apply to mathematics because the aim of proof or calculation is certainty. But they do not show how they apply. They also claim that formal deduction is "feasible only if the statements used as premises or conclusion are either entirely true or entirely false" but that the way we "ordinarily speak" is "not bivalent" (p. 263). This is because words like 'stylish' or 'trustworthy' apply as a 'matter of degree' and are applied with differing degrees of confidence. However, the authors do not explain why it follows from the fact that there are degrees of being trustworthy that an application of the predicate is true or false only as a matter of degree. What is said about induction and causation is also troubling. "In general, if someone believes that two events are causally connected, she believes that one typically precedes (or follows) the other" (p. 288). This claim seems obviously wrong if it applies to specific events rather than to events of a certain type. Someone may know the cause of a crash because she saw one car run a red light and broadside another car in the intersection, and not be- cause she has any beliefs about what typically precedes a collision. Moreover, the fact is that often a causal explanation provides an understanding of how something works or why something happened, and is not just a claim that certain correlations are more than merely coincidental. "Practical reasoning" is the most distinctive and important topic in this book. The authors refer to it as 'practical' because it is concerned with choosing between different alternatives, and they think of it as beingjudged in terms of its communicative rationality because the alternatives, at least in the case of conductive reasoning, "cannot, on the face of it, be measured or compared (p. 322)." The different considerations cited, for example, in opposition to assisted suicide or capital punishment, seem hard to quantitatively compare, and any standards used to do so are open-ended in the sense that the standards themselves may be questioned. This conception of practical reasoning explains why they confine their lessons on it to game theory, argument from analogy and conductive argu- ment. The first lesson is an obvious choice, though it has limited practical applicability. The other two are less obvious. The lesson on argument from Book ReviewslComptes rendus 91 analogy is included because the authors say that its value depends on the weight given to the similarities and differences between analogue and original, even though its value seems to depend more on the light the analogue shines on the original. The other 'practical reasoning' they discuss is conductive reasoning, which is defined as reasoning where several different considerations are together supposed to argue for a certain conclusion. The authors focus on legal rea- soning, which was the subject of a textbook by Bickenbach. The weight given to any piece of evidence, for example, in a criminal trial, will depend on how difficult it is for it to be explained away. So, the standards used in determining how much weight to assign are open-ended in the sense that the dialectical process may subject the standards themselves to criticism. But why suppose that there are any standards? Their answer is transcen- dental: we must have standards because otherwise communicative rationality is not possible. This claim should be challenged. We need to be able to justify this or that decision or claim in the face of questions or challenges to it; but it is not necessary to invoke some very general rule or principle to do so. The real problem is that because of their focus on standards for the weighing of different options or considerations, they neglect the importance of the kinds of things that a dialectical exchange would reveal: why the argument is being given, what concerns it is revealing, who or what it is addressing, and what is at issue in the argument. If they had focussed more on these rhetorical con- siderations, then they might have reconsidered their classification of all rea- soning as deductive, inductive or practical. Good Reasons for Good Arguments introduces readers to a very different approach to the teaching of critical thinking.This review has suggested that there are some problems with its theory, which is more advanced than its pedagogy. However, if that pedagogy better reflected its theoretical orienta- tion, then perhaps there would be fewer problems with that orientation. In particular, if more of their illustrations were as interesting or thought provok- ing as the argument of the book itself, then it would be possible to better appreciate how to think about or resolve some of the problems with that argument. Don S. Levi. Department a/Philosophy 1295 University a/Oregon. Eugene OR 97403-1295 US.A. dslevi@oregon.uoregon.edu