Valid Ad Hominem Arguments in Philosophy: Johnstone's Metaphilosophical Informal Logic MAURICE A. FINOCCHIARO University of Nevada-Las Vegas Abstract: This is a critical examination of Johnstone's thesis that all valid philosophi- cal arguments are ad hominem. I clarify his notions of valid, philosophical, and ad hom- inem. I illustrate the thesis with his refuta- tion ofthe claim that only ordinary language is correct. r discuss his three supporting arguments (historical, theoretical, and inter- mediate). And r criticize the thesis with the objections that if an ad hominem argument is valid, it is really ad rem; that it's unclear how his own theoretical argument can be ad hominem; that if an ad hominem argument is really valid, it would have to be based on the proponent's own assumptions; and that the thesis is not true of philosophical arguments that are constructive rather than critical. Resume: J'examine a fond la these de Johnstone selon laquelle tous les arguments philosophiques valides sont des arguments ad hominem. Je clarifie "vaJide", "philosophique" et "argument ad hom- inem". J'i11ustre sa these en employant sa refutation de I'enonce que seulement Ie langage ordinaire est correct. Je discute des arguments historique, theorique et interrnediaire qu'il avance pour soutenir sa these. Je presente les objections suivantes: 5i un argument ad hominem est valide, c'esl en effet un argument ad rem; ce n 'est pas clair comment ses propres arguments thi:oriques peuvent eire des arguments ad hominem; si un argument ad hominem etait reellement valide, il devrait reposer sur les suppositions de I'individu qui presente I'argument; et sa these ne s'applique pas aux arguments philosophiques qui son! constructifs plutO! que critiques. Keywords: ad hominem, validity, philosophical argument, metaphilosophy, Henry W. Johnstone Jr. 1. Introduction The aim of this essay is a critical examination ofthe thesis that valid philosophical arguments are ad hominem. This thesis was advanced by Henry W. Johnstone, Jr., and constitutes a highly original contribution. a brilliant idea. and a constant theme of his half a century of philosophical effort. I In general, his work was a pioneering effort in the informal logic of philosophical argument and included other related themes, such as metaphilosophy and the role of rhetoric and of for- mal logic in philosophy. In focusing on this thesis, I do so because it is probably his key contribution and is emblematic of both the rest of his work and of the informal logic of philosophy. I shall first discuss several clarifications, then a concrete illustration, then some supporting arguments, and finally several objec- tions. ©In/ormalLogic Vol. 21, No.1 (2001): pp. 11-24. 12 Maurice A. Finocchiaro 2. Clarifications The thesis can be expressed in several ways: that "all valid philosophical argu- ments are ad hominem" (P A81 );2 that in philosophy only ad hominem arguments are valid (PA3; VRS6); that the validity of philosophical arguments lies in the prop- erty of being ad hominem (PAS7-92); that ad hominem argument "is the only valid argument in philosophy" (VR134); and that in order to be valid, philosophical arguments must be ad hominem. To prevent misunderstandings, the most immediate clarification needed is that Johnstone is not taking the term 'ad hominem argument' in the sense of contem- porary logic textbooks, i.e., as the fallacy of concluding that some claim is false or some argument incorrect on the basis of premises attacking the character, mo- tives, interests, or circumstances of the person advancing it. Instead, Johnstone is using the phrase in its traditional historical meaning, which may be found in Galileo, Locke, Thomas Reid, and Richard Whately (see PA73n 12 and Finocchiaro 1974). Thus, Johnstone often quotes Whately's definition that "in the argumentum ad hominem, the conclusion which actually is established, is not the absolute and general one in question, but relative and particular, viz. not that 'such and such is the fact', but that' this man is bound to admit it in conformity to his principles of reasoning, or consistency with his own conduct, situation', &c." (Whately 1838, 196). Johnstone rephrases this by saying that "argumentum ad hominem . .. is precisely the criticism ofa position in terms of its own presuppositions" (VR134), in which he subsumes both propositions and arguments under the label of "posi- tion." Elsewhere he states that (in philosophy) an ad hominem argument is "an argument against a philosophical thesis [attempting to] exhibit that thesis as incon- sistent with its own assertion or defense, or with principles that must necessarily be accepted by anyone who maintains the thesis" (VR4S). Finally, these formula- tions are meant to be equivalent to a still different one using the notion of a "self- defeating" position, as can be seen from Johnstone's following definition: "an argument that [purportedly] shows that a statement or argument defeats its own purpose is, to my way of thinking, precisely an argumentum ad hominem" (PA82). It is equally important that by 'validity' Johnstone does not mean formal (or deductive) validity. A key reason for this is that the latter is independent of the truth of the premises, whereas he takes validity to refer not only to the proper relationship between premises and conclusion but also to the truth of the premises; that is, by validity he means something analogous to what is usually called "sound- ness." Here I speak of analogy rather than identity because Johnstone avoids speak- ing of soundness or truth of premises. Instead, one term he uses is cogency, according to which a cogent argument is one that is formally valid and has premises which are impossible to doubt because they are exactly what the doubter holds (VR26). He contrasts cogent arguments to rigorous arguments, which he defines as arguments that are formally valid and have premises which are impossible to doubt because to doubt them is to miss the whole point of the argument (VR26). Johnstone's Metaphilosophical Informal Logic 13 It follows that "mathematical proofs, then, are rigorous; and some philosophical arguments are cogent" (VR26). Other terms Johnstone uses to clarify his concept of validity are relevance and force. He does not give an explicit definition or elaborate analysis of these two notions but takes them in an intuitive and ordinary sense. However, his discussion is helpful when he compares valid arguments, criticisms, objections, passports, and contracts, and when he suggests a common core meaning: "These two no- tions of relevance and force are, I shall maintain, the root ideas common at least to valid arguments, criticisms, objections, and judgments, even if not to valid pass- ports or contracts" (P A62). Helpful is also his discussion that relevance is a neces- sary but not sufficient condition for force: "It seems clear that no argument lack- ing relevance can have force. On the other hand, an argument could have rel- evance without having force. These two statements summarize all that I have discovered about the relationship between relevance and force" (PA62-63). Johnstone's concept of validity may also be clarified by noting that it is for him essentially synonymous with effectiveness or success. These two notions focus on actually accomplishing an aim. Since the aim of argument is to support or establish a conclusion, an effective or successful argument is one that actually supports or establishes its conclusion. When the conclusion is not a categorical statement but the conditional claim that there is an internal inconsistency in the position ad- vanced by an arguer (which is the case for conclusions of ad hominem argu- ments), then an ad hominem argument is effective, successful, or valid insofar as it really shows that there is such an internal inconsistency. In Johnstone's own words, "an argumentum ad hominem, like any other argument, will be valid when it establishes the conclusion it claims to establish, and invalid when it establishes a conclusion independent of this" (PA73). This explicit definition makes it clear that Johnstone is not equating effective- ness or success with persuasiveness, Le., mere persuasiveness pertaining to rhetoric in the pejorative sense of this word. It is indeed true that his view of the impor- tance of rhetoric and its role in philosophy underwent an evolution, from an initial dismissive to a final appreciative position, according to which rhetoric in the good sense of the word does indeed provide an essential feature of philosophical argu- ment and of its validity (VR81-85). However, I have no space in this essay to discuss this aspect of Johnstone's views, and the following point about persua- siveness must suffice. In line with his way of thinking, one could say that persua- siveness has two meanings: a persuasive argument could be one which as a matter of empirical fact persuades people; and it could be one which ought to persuade an appropriately relevant group of people. If we understand persuasiveness in the latter normative sense, rather than in the former descriptive sense, then we could equate an effective, successful, or valid argument with a persuasive one; no harm would follow and this connection would provide an additional helpful clarification (see Finocchiaro 1997, 369-71). 14 Maurice A. Finocchiaro We now come to the third key term in Johnstone's thesis, the term 'philosophi- cal.' Part of what he means can be glimpsed by examining his writings and noting that the concrete historical examples discussed most often are the following: the egalitarian argument that all men are created equal and the teleological argument for the existence of God (PA25-39); Plato's simplicity argument for the immortal- ity of the soul (PA58-59); Aristotle's criticism of Eudoxus's argument that pleas- ure is the chief good (PA64-67); Berkeley's criticism of the materialist argument that external bodies provide the causal explanation of our ideas (PA67-69); a self- reference objection to naturalist epistemology (PA69-75); Mill's "proof' of the principle of utility (PA77-79); eight of Hume's arguments concerning causal ne- cessity (PA93-1 04); various realist, anti-realist, functionalist, and anti-functional- ist arguments in the philosophy of logic (VR45-52); a self-reference objection to Norman Malcolm's claim that ordinary language is the only correct language (VR53- 56); Berkeley's argument that to be is to be perceived (Johnstone 1989,8-10); and Parmenides's argument about the nature of being and Aristotle's refutation of it (Johnstone 1989, 11-12). Although these arguments are not always explicitly dis- cussed in the context of the question of their validity and ad hominem character, they do convey a flavor of what Johnstone is talking about; furthermore, even when he discusses them in the context of other issues, those other discussions connect indirectly with this question. The general impression is that Johnstone is studying arguments characteristic of certain particular branches of philosophy. Clearly these are such branches as metaphysics, theory of knowledge, ethical theory, and logical theory; collectively considered, these could be labeled first philosophy, systematic philosophy, specu- lative philosophy, or theoretical philosophy. It is equally clear that he is not refer- ring to arguments common in other parts of philosophy; for example, in the historiography of philosophy, scholars often advance arguments that are straight- forwardly historical and inductive, involving questions of factual accuracy, causal connection, genetic origin, empirical consequence, and cultural evolution; and they also engage in philological arguments concerning the linguistic integrity of texts, the correct meaning of passages, and the etymology of words. It would be arbi- trary to disqualify such arguments from being "philosophical," but it would be uncharitable to advance them as counterexamples to Johnstone's generalization. Let us say they do not fall within its scope and thus do not falsify it; he is simply talking about other kinds of arguments. Analogous remarks apply to arguments prevalent in various branches of ap- plied philosophy, e.g., philosophy of science, of religion, and of art. On one occa- sion where Johnstone seems to talk about the nature of science, he is quick to point out that, appearances to the contrary, he does not intend to get involved in questions of the philosophy of science, and that his references to it are merely illustrative and not substantive (PA22). A similar caution regards analytic philoso- phy, for I agree with L. Jonathan Cohen's (1986) thesis that typical arguments in Johnstone's Metaphilosophical Informal Logic 15 analytic philosophy are inductive (see Finocchiaro 1991). These qual ifications are crucial in order to appreciate the strength and weakness of Johnstone's thesis; without such qualifications, one might raise irrelevant objections to it, irrelevant because based on a misunderstanding of his meaning. That is why I regard my remarks above as part of a clarification of his thesis. He is talking about arguments in systematic philosophy. 3. A Concrete Illustration To understand better Johnstone's thesis, it is useful to give an illustration. Con- sider the view that ordinary language is the only correct language in philosophy, which is a presupposition of the school that goes by such labels as ordinary- language philosophy or linguistic analysis. As Johnstone indicates, the view can be found explicitly stated in an essay by Norman Malcolm (1942, 357). Now, sup- pose one were to criticize Malcolm's view by arguing that the history of philoso- phy readily shows that great philosophers were typically using words in ways that deviate from ordinary language; for example, when Plato speaks of eidos, Kant of Ding, Croce of spirito, Whitehead of actual occasion, and Sartre of projet, they are not using these words in their ordinary sense. This objection would be ineffec- tive against Malcolm because his position denies the correctness of these philoso- phers' language, and so these cases do not provide counter-instances consisting of correct language which is non-ordinary; the objection begs the question. After pointing this out, Johnstone asks us to consider the following criticism. In stating his position, Malcolm is not using the phrase "ordinary language" in the ordinary sense. To use the phrase in an ordinary sense would be to use it in a context like this: suppose a reporter from the popular media interviews a Nobel Prize winner in physics to convey to ordinary people a sense of what kind of person he is and what his discoveries amounted to; suppose also that the physicist is able to explain himself clearly and comprehensibly in a down-to-earth manner, without using scientific jargon and uncommon sentence constructions; in this case it would be proper to say that during the interview the physicist used the most ordinary language. However, Malcolm's "ordinary language" is not equivalent to this because for him technical talk among physicists would also be "ordinary," and so would also be technical talk among artists and art critics. Thus, appearances to the contrary, Malcolm's phrase is itself not ordinary language; since this context is philosophical, by his own principle, his language is not correct. In short, as Johnstone puts it, Malcolm's motto "seems to impugn its own correctness" (VR54). Next, Johnstone points out that, unlike the first criticism, this second one is valid, in the sense that it succeeds in establishing that Malcolm's thesis (as origi- nally formulated) must be abandoned. To be sure, the thesis could be "revised," for example by saying that in philosophy only "ordinary language is correct lan- guage, and pronouncements about ordinary language can also be correct" (VR56). But this revision would underscore the fact that the criticism hits the mark and 16 Maurice A. Finocchiaro establishes the critical conclusion that it is not true that in philosophy only ordinary language is correct. The final point to understand in this illustration is that this successful criticism is an ad hominem argument. Applying Whately's definition, the criticism tries to show that Malcolm, in accordance with his principles, is bound to admit that non- ordinary language can be correct. Using Johnstone's definitions, the criticism tries to prove that Malcolm's thesis is inconsistent with its own assertion or defense, that it defeats its own purpose. 4. Justification Let us now examine Johnstone's justification of his thesis. In an important pas- sage (P A81-82), he suggests there are three main arguments in its favor: a histori- cal empirical argument involving cases of famous arguments from the great phi- losophers; an abstract theoretical argument involving his conceptions of argu- ment, validity, ad hominem, and philosophy; and an intermediate argument involv- ing a classification of philosophical criticisms. Johnstone's empirical argument is essentially an induction by enumeration in which several typical instances of philosophical argument are examined and each is shown to be both valid and ad hominem. In the context where he explicitly elaborates this argument (PAS7-80), he considers Aristotle's argument against Eudoxus's conclusion that pleasure is the chief good; Berkeley's argument against the materialist claim that material bodies are the likely causes of our ideas; and a self-reference objection to naturalism. However, as indicated earlier, various as- pects of all the arguments in that earlier longer list are used to amplify this set of three. For my purpose here, I shall focus on Berkeley's anti-materialist argument, found in paragraph 19 of his Principles of Human Knowledge: But, though we might possibly have all our sensations without them, yet perhaps it may be thought easier to conceive and explain the manner of their production, by supposing extemal bodies in their likeness rather than other- wise; and so it might be at least probable there are such things as bodies that excite their ideas in our minds. But neither can this be said. For, though we give the materialists their external bodies, they by their own confession are never the nearer knowing how our ideas are produced; since they own them- selves unable to comprehend in what manner body can act on spirit, or how it is possible it should imprint any idea in the mind. Hence it is evident the production of ideas or sensations in our minds, can be no reason why we should suppose Matter or corporeal substances; since that is acknowledged to remain equally inexplicable with or without this supposition. If therefore it were possible for bodies to exist without the mind, yet to hold they do so must needs be a very precarious opinion; since it is to suppose, without any reason at all, that God has created innumerable beings that are entirely use- less, and serve to no manner of purpose. [Berkeley 1929, 134; see PA67] Johnstone's Metaphilosophical Informal Logic 17 My own analysisl of this passage is that Berkeley is trying to show that it is not even likely that material bodies exists (having earlier argued that it is not necessary that they do). His argument is that there is no good reason for this likelihood, while there is one against it. The reason against it is the theological and teleological claim that material bodies would be useless creations. He supports his claim that there is no good reason in favor of the likely existence of material bodies by arguing that the only reason is provided by the following "materialist" argument, and this argu- ment is inconclusive. The materialists argue that it is likely that material bodies exist because their existence would provide the simplest explanation of our ideas and sensations. Berkeley objects that this argument is inconclusive because those who try to explain our ideas on the basis of material bodies also believe that it is incomprehensible how matter acts on mind, and so their explanation does not really succeed. Let us focus, as Johnstone does, on only part of Berkeley's overall argument, namely on what I have called his objection to the materialist explanation of our ideas (the last sentence of the preceding paragraph). The argument is ad hominem insofar as it shows, not that there is no explanation of our ideas in terms of exter- nal bodies, but that the materialists can provide no explanation (given that in their position it is a mystery how matter acts on spirit). That is, Berkeley is criticizing materialism in terms of its own presuppositions; or again, he is trying to show that its thesis of the probable existence of material bodies is inconsistent with its other assertion about how matter can act on spirit. We may also agree with Johnstone that Berkeley's criticism is valid since it aims to show the incoherence of the materialist position, and this coherence is indeed established. As stated before, for this and many other arguments Johnstone's historical justification tries to show they are both valid and ad hominem. To be fully convinc- ing Johnstone would also have to show that these arguments are valid because they are ad hominem, and/or that they are valid insofar as they are ad hominem and invalid insofar at they are not ad hominem. But I do not wish to criticize his historical argument on this basis because to attempt to show such claims would introduce theoretical considerations, which would turn his historical argument into the theoretical justification; and although he does not discuss these considera- tions in the context of the historical argument, he does discuss them elsewhere, as a separate justification of his thesis. So let us go on and discuss his theoretical argument. Johnstone first argues that, unlike the situation in natural science where truth and falsity are independent of the supporting evidence, the truth or falsity of a philosophical statement is relative to the argument that proves or disproves it. By this he means that a philosophical statement is one such that "it is impossible to think of the statement as true without at the same time thinking of an argument in its favor, and it is impossible to think of it as false without at the same time thinking of an argument against it" (PA23). The essential reason for this metaphilosophical 18 Maurice A. Finocchiaro claim is that "the argument for a philosophical statement is always a part of its meaning. Furthermore, ... the argument against a philosophical statement is always a part of its meaning" (P A32). But, if the truth or falsity of any philosophical statement is relative to the argu- ment that establishes or disestablishes it, then, unlike the truth or falsity of a scientific statement, it is not relative to objective facts. Hence there is no argumentum ad rem to establish or disestablish any philosophical state- ment. This leaves open only the possibility of an argumentum ad hominem. But any valid argumentum ad hominem will be found to have the same characteristics as each of my examples has been found to have. It will exhibit the self-defeating nature of an argument or statement that it attacks. It will be directly relevant to this argument or statement. It will borrow its force from the energy with which what it attacks is asserted. [PA76] One could object here that Johnstone's initial metaphilosophical premise is not true, by focusing on a paradigm example of a philosophical statement, namely the existence of God. But I believe this objection would distract us from the main thread I want to develop in this essay. So let me note a sense in which Johnstone's argument has some plausibility. That is, let us apply these ideas to Johnstone's own thesis that valid philosophical arguments are ad hominem. Earlier I clarified the meaning of this thesis by explaining the notions of validity, philosophy, and ad hominem. However, it should also be noted that once one has explained what Johnstone means by these terms, one has gone a long way toward establishing the correctness of the thesis. One could say that the thesis is almost analytically true, given the meaning of the terms involved. Of course, this is not the whole story because the thesis also has applications to historical reality and normative implica- tions regarding philosophy. On the other hand, the analytic aspect of Johnstone's thesis is part of the story. So his position does have a considerable amount of coherence and self-consistency. As regards Johnstone's third argument (PA8l-92), there is no space to elabo- rate it here, but it is so original and suggestive that it deserves a few comments. This argument is intermediate between abstract and empirical. On the one hand, it analyzes the notion of a statement or argument "defeating its own purpose" and identifies several ways in which this can happen. On the other hand, it articulates a classification of philosophical criticisms and several subtypes are distinguished; these are all arguments that charge some other argument or statement with the following flaws: unintelligibility (e.g., tautological emptiness, occultness, ambigu- ity, or inconsistency); dogmatism; tu quoque; "throwing out the baby with the bathwater"; denial of one's own presuppositions; and self-contradiction. Then Johnstone tries to show that there is a one-to-one correspondence between such philosophical charges and the various types of "defeating its own purpose." Now, recall that an ad hominem argument is one claiming that some statement or other argument defeats its own purpose; then it is easy to see that this intermediate justification amounts to an attempt to show the equivalence between types of Johnstone's MetaphilosophicallnJormal Logic 19 critical arguments and subtypes of ad hominem arguments. To the extent that Johnstone's classification of philosophical criticism and his subdivision of ad hom- inem argument are exhaustive, he may be taken to have shown that all philosophi- cal criticism is ad hominem, and consequently that all valid philosophical criticism is ad hominem. Whether this is equivalent to showing that all philosophical argu- ments are ad hominem depends on whether all philosophical arguments are critical. And this brings us to a major criticism of Johnstone's thesis, by contrast to some of the minor ones already mentioned, which were not stressed but were rather regarded as suggestions for clarifying the thesis. 5. Criticism In fact, my main objection to Johnstone's thesis is going to be that, although it appears to be essentially true of critical arguments in philosophy, it is not really true of arguments that are non-critical and may be labeled constructive. But before we come to that, let us consider other objections, which although they are also major, can be handled more briefly. Johnstone anticipated almost all these objec- tions, and so the issue is whether he answered them satisfactorily. One objection was advanced by Warren J. Hockenos (1968; see VRS6-61). He argued that, by Johnstone's own definitions, a valid philosophical argument is one which establishes the conclusion it claims to establish (PA73); and an ad hominem argument is one which concludes that a given thesis is inconsistent with its own presuppositions (VR4S, 134). Hence, if a philosophical argument is both valid and ad hominem, it establishes that there is an inconsistency between a thesis and its presuppositions. But if this is so, if an inconsistency is really established, the philosophical argument is "ad rem" because such an inconsistency would be an objective fact (albeit a logical one); if the argument is ad rem, it is not ad hominem; therefore, ifvaJid philosophical arguments are ad hominem, they are not ad hom- inem. It follows that it is not true that valid philosophical arguments are ad hom- inem. Johnstone admitted this objection is essentially valid when he confessed that "it is criticisms of the kind that Hockenos makes, whether actually expressed by others or myself, that have caused me, over the years, gradually to modify my conception of the nature and purpose of philosophical argumentation" (VRS6). Thus, Johnstone partly undertook a rethinking of his distinction between ad hom- inem and ad rem arguments, reconsidering whether these two classes are jointly exhaustive and mutually exclusive (VRS7-58), and wondering whether to admit that some valid arguments are ad rem. He also tried to show that the arguments mentioned by Hockenos are ad hominem after all; for the consistency proof would depend on whether the criticized argument and the critic shared the same concept of inconsistency; if they did not, the position under criticism would accept an alleged inconsistency only when demonstrated on the basis of its own concept of inconsistency, i.e., only when the inconsistency criticism was ad hominem. 20 Maurice A. Finocchiaro It is important to note that Hockenos's criticism is itself an ad hominem argu- ment. This feature may account for its at least partial effectiveness. In a paradoxi- cal sort of way, the criticism may thus reinforce Johnstone's thesis. This raise an issue from which a second objection can made to Johnstone's thesis. The question is whether his own argument is ad hominem. Since, as we have seen, he has three supporting arguments, this question is threefold. However, it is obvious that the really crucial case involves the theoretical argument. His historical argument is not even a candidate, being instead a good example of the inductive arguments typical of the historiography of philosophy, which we noted earlier fall outside the scope of his thesis. Regarding his argument from the classi- fication of philosophical criticisms, we have not said enough about it to fruitfully pursue the question. This second objection could be articulated as follows. Johnstone's (theoreti- cal) argument should be ad hominem, if we apply the thesis to itself, and there seems to be no reason why we should not. But if his argument is ad hominem, then two difficulties follow, an evaluative and an analytical one. The evaluative difficulty is that, despite the fact that by now it should be obvious that the term ad hominem, far from being pejorative, is actually favorable (in Johnstone's scheme), still the implication is that it is not "an objective" fact that valid philosophical arguments are ad hominem, but rather that it is true only in his own system. The analytical difficulty is that he needs to explain how his theoretical argument does have the property of being ad hominem; how its conclusion is critical of some other position and how its premises involve the presuppositions of that other posi- tion. Johnstone was aware of this possible criticism (VR135, 139). Although he did not address the evaluative difficulty, he did respond to the analytical difficulty. But he did not respond in a sustained manner, and the only relevant passage is insuffi- cient: My argument does not, at least in any obvious way, miss the point of anyone who might contend that philosophical statements can be true or false inde- pendently of the arguments used to establish or disestablish them. It ac- quires its force precisely from the force of this contention; for the contention can only take the form of an argument, and this very argument will at once serve as a further illustration of the thesis that I have been advocating. Since it exposes the self-defeating character of what it attacks, my argument to the effect that all valid philosophical arguments are ad homine~learly itself a philosophical argument, and one that I am claiming is valid-is itself also ad hominem. [PA8t] To reinforce my criticism that Johnstone's response is insufficient, I shall now articulate a third related objection. One could object that, to be effective, to estab- lish its conclusion, an argument must be formally valid and have premises that are unquestionable; but any philosophical claim is questionable because this is true almost in virtue of the definition of "philosophical"; thus the premises of a philo- Johnstone's Metaphilosophical Informal Logic 21 sophical argument are always questionable. However, philosophical premises are de facto unquestioned in those circumstances where they happen to coincide with what a person accepts; that is, philosophical premises are not questioned by per- sons who happen to accept them. Therefore, for persons who accept the premises, and only for them, a philosophical argument (if it is formally valid) will establish its conclusion, will be effective, will be "valid" (in Johnstone's sense). Now, such an argument is easily shown to be ad hominem in Whately's sense for it proves, not that the conclusion is a fact or that everyone must accept it, but only that those persons are bound to accept it who accept the premises. But does this yield Johnstone's thesis, whose unpacked meaning is that valid philosophical arguments are those whose conclusion criticizes some statement or other argument in terms of the latter's own presuppositions? Because we are dealing with philosophical claims, the conclusion of a valid philosophical argument is also the denial of some alternative philosophical claim, and so it criticizes some alternative philosophical position; so far, so good. But for such philosophical criticism to be ad hominem, the argument's premises must be the presuppositions of the alternative position; yet we have seen that these premises are propositions accepted by the argument's proponent; so we have a valid argument criticizing a thesis on the basis of alterna- tive presuppositions and thus lacking the property of being ad hominem. It foIlows that valid philosophical arguments are not ad hominem. This objection reaches a conclusion critical of Johnstone's position, but does so by utilizing many ideas that are part of that position. So it is essentially an ad hominem argument. Now, I really see no way of evading this criticism. Thus, here we have a second occurrence of the paradox that Johnstone's thesis is criticized by means of a valid philosophical argument that is ad hominem, Le., by means of an instantiation of that thesis. But let us leave the dizzying atmosphere of such paradoxes and come down to a last and down-to-earth criticism of Johnstone's thesis. This objection points out that although his thesis has much plausibility when the arguments in question are critical arguments, it seems to be off the mark when we consider "constructive" arguments, which Johnstone might call "ad rem"; these are arguments that are not about statements or other arguments, but about things different from statements or arguments. An example is Plato's simplicity argument for the immortality of the soul found in the Phaedo, which Johnstone himself discusses (PAS8-59), although in connection with other issues. One could add the other classic arguments for the immortality of the soul; the arguments for the existence of God; and the argu- ments supporting free will. When faced with such alleged counterevidence, Johnstone could respond, al- though he did not stress it, in a manner analogous to Malcolm's reply to alleged instances of non-ordinary language that is correct. That is, Johnstone could reply that constructive arguments are seldom if ever valid; here it is worth stressing that the subject term of his generalization is "valid" philosophical arguments and that 22 Maurice A. Finocchiaro validity to him means success in establishing the conclusion. Few would be pre- pared to hold that the just-mentioned constructive arguments are valid in this sense. However, Johnstone responded primarily by arguing that such constructive philo- sophical arguments are critical and ad hominem after all (PA35-37, 76-80; VR28, 45). Thi's Johnstonian argument is articulated by examining Mill's "proof' of the principle of utility: The only proof capable of being given that an object is visible, is that people actually see it. The only proof that a sound is audible, is that people hear it; and so of the other sources of our experience. In like manner, I apprehend, the sole evidence that it is possible to produce that anything is desirable, is that people do actually desire it. If the end which the utilitarian doctrine proposes to itself were not, in theory and in practice, acknowledged to be an end, nothing could ever convince any person that it was so. No reason can be given why the general happiness is desirable except that each person, so far as he believes it to be attainable, desires his own happiness. [Mill 1965, 221; seePA77-78] The standard criticism of Mill's argument is this. He argues that happiness is desirable because it is act\lally desired by people. This argument assumes that whatever is actually desired is desirable. But this assumption is false because de- sirable means "worthy of being desired" and not "actually desired." Johnstone criticizes this criticism. He begins by stating that Mill's argument should be given the following more sophisticated reconstruction: (a) happiness is desirable because (b) it is actually desired; and (c) whatever is actually desired is desirable because (d) whatever is actually desired is capable of being desired, (e) whatever is capable of being desired is worthy of being desired, and (1) whatever is worthy of being desired is desirable. Moreover, continues Johnstone, instead of dogmatically declaring (e) to be false, one should understand that Mill had ajusti- fication for it, namely that it is true because (g) "capable of being desired" does mean "worthy of being desired," and this is so because (h) otherwise one would h~ve a way of knowing worth independent of capability, and (i) this would be unacceptable apriorism (0 la Kant). In other words, the standard criticism begs the question; whereas Johnstone's reconstruction makes it clear that Mill's argu- ment is directed against Kant and is trying to provide an alternative to it. Unfortunately, and this is my criticism of Johnstone's account, although Mill's argument is directed against Kant, it is not ad hominem against him; for in Johnstone's reconstruction, Mill's argument is based on an alternative to Kant, rather than on Kant's own presuppositions. As Johnstone himself says, Mill's argument is really an argument "to himself." But to say this is to admit that the argument is not ad hominem. The difference between oneself (one's own system) and someone else (an alternative system) is not insignificant, pace Johnstone, despite the argument advanced in the following passage: Johnstone's Metaphilosophical Informal Logic 23 A constructive philosophical argument, when valid, is very much like a valid argumentum ad hominem. The only important difference is that the philoso- pher using a constructive argument considers what he himself is bound to admit, in conformity to his own principles of reasoning or in consistency with his own conduct or situation, rather than considering what someone else is bound to admit. The constructive argument is thus essentially an argumentum ad seipsum. [P A79] 6. Epilogue My conclusion is that, on the strength of Johnstone's arguments, it is probably true that all valid philosophical criticism must be ad hominem (in his sense ofthese terms); and this claim is important, insightful, and suggestive. But, for the objec- tions discussed above, this claim should not be equated either with the thesis that all valid philosophical arguments must be ad hominem, or the thesis that all philo- sophical arguments are ad hominem. Notes I In this essay I shall take no account of the evolution of Johnstone's thinking, for that would complicate and lengthen it beyond acceptable limits. However, this limitation does not under- mine the primary aim of this essay. 2 Because of the many references to Johnstone's two main books, subsequenfreferences will be given in parenthesis in the text, using the abbreviations "PA" for Philosophy and Argument and "VR" for Validity and Rhetoric in Philosophical Argument. J My analysis differs slightly from that given by Johnstone. but I do this in order to better explain in my own words his key point that this argument of Berkeley is both valid and ad hominem. 1 find Johnstone's own account oversimplified to the extent of making it more difficult to see how Berkeley's argument provides evidence in support of Johnstone's thesis. References Berkeley, G. (1929). Essay, Principles, Dialogues. New York: Scribner's. Cohen, LJ. (1986). The Dialogue of Reason. Oxford: Clarendon. Finocchiaro, M.A. (1974). "The Concept of Ad Hominem Argument in Galileo and Locke," The Philosophical Forum 5: 394-404. Finocchiaro, M.A. (1991). "Induction and Intuition in the Normative Study of Reason- ing," in Probability and Rationality, E. Eells and T. Maruszewski (eds.), 81-95. Amsterdam: Rodopi. Finocchiaro, M.A., trans. and ed. (1997). Galileo on the World Systems. Berkeley: University of California Press. Hockenos, W.J. (1968). An Examination of Reductio ad Absurdum and Argumentum ad Hominem Arguments in the Philosophies of Gilbert Ryle and Henry W. Johnstone, Jr. Ph. D. Dissertation, Boston University. 24 Maurice A. Finocchiaro Johnstone, H.W., Jr. (1952). "Philosophy and Argumentum ad Hominem," Journal of Philosophy 49: 489-98. Johnstone, H.W., Jr. (1959). Philosophy and Argument. University Park, PA: Pennsyl- vania State University Press. Johnstone, H.W., Jr. (1978). Validity and Rhetoric in Philosophical Argument. Univer· sity Park, PA: Dialogue Press. Johnstone, H.W., Jr. (1989). "Argumentation and Formal Logic in Philosophy," Argu- mentation 3: 5-15. Johnstone, H.W., Jr. (1997). "A Bibliography, 1948-1997," Philosophy & Rhetoric 31:6- 19. Malcolm, N. (1942). "Moore and Ordinary Language," in The Philosophy ofUE. Moore, P.A. Schilpp (ed.), 343-68. Evanston, IL: Northwestern University Press. Mill, J.S. (1965). Essential Works. New York: Bantam Books. Whately, R. (1838). Elements of Logic. New York: William Jackson. Maurice A. Finocchiaro Department of Philosophy. University of Nevada. Las Vegas Las Vegas. NV 89154-5028. U.S.A. mauricef@nevada.edu