Informal Logic X.2, Spring 1988 Buttercups, GNP's and Quarks: Are Fallacies Theoretical Entities?1 JOHN WOODS University of Lethbridge and University of Amsterdam 1. Researchers who have been follow- ing the development of the pragma- dialectical approach to argumentation theory2 have been attracted by the promise that it shows for transacting fallacy theory in an innovative manner designed to take things well beyond the enfeebled scholasticism of what Hamblin has called the Standard Treatment. As I have tried to show elsewhere,3 the extent to which it is possible for pragma-dialectics to furnish a new unified account of the fallacies turns out to be a function of the extent to which it is justifiable to free the fallacies from the burden of identities forged for them by the old traditional accounts.4 And so, for ex- ample, the pragma-dialectician is interested in being able to disarm, in a quite general way, objections in the form, "But that's not what the ad baculum really is!" Part of the methodological strategy which enables such a burden to be lightened, if not removed altogether, is reflected in the pragma- dialectician's espousal of what elsewhere I have called "Thesis II". Van Eemeren and Grootendorst make the point that fallacies are not buttercups. These authors are dubious of any approach in which fallacies lead a more or less autonomous ex- istence. like birds, trees or flowers .... in which the ... most important job concern- ing fallacies seems to be to detect them and then classify them correctly. A 'theory' of fallacies seems to function like a flora in which the species are described: 'Is this a buttercup or a dandelion?' 5 Thus Thesis II: Fallacies do not lead a life of their own independently of a properly articulated theory. Something is a fallacy only within a theory of fallacies. 6 This is the crux of their approach to the fallacies, and I do not wish to do it the violence of a careless or fanciful interpreta- tion. But an interpretation is needed all the same. The difficulty is this. If one came upon Thesis II equipped with a broadly Duhemian-Quinean (naturalized) epistem- ology of science, it is likely that it would be taken as trivial, something along the lines of "Buttercups do not (any more than but- tercup stages) lead a life of their own, but only within a well-articulated physical theory. " Since on a Quinean view of theories and things, every object is a theoretical entity?, Thesis II could not be taken as anything particularly interesting or novel. So we see at once that van Eemeren and Grootendorst have a stake in preserv- ing this central claim from the charge of obviousness. In fact, it is clear that van Eemeren and Grootendorst are indeed entertaining a quite different notion of theory-dependency, because in forwarding the case of theory- dependency for fallacies they do intend to be attributing to them something striking and distinctive. Theirs is the idea, roughly, that (A) buttercups are not theoretical en- tities in the sense that they intend (for surely they have, also in the sense that they intend, a life of their own). (B) Gross National Pro- ducts are to some degree theoretical entities, for, though the aggregate of goods and ser- vices that a country produces in a year is somehow pre-theoretically "out there," in actual macroeconomic practice, the GNP numbers, in order to be useful to economists, are adjusted in Departments of 68 John Woods Finance with the help of mathematical and statistical heuristics which have no natural macroeconomic counterparts. (C) Some of the entities of physics, for example, the per- manently confined quarks of Quantum Field Theory, are theoretical entities par ex- cellence; they have no life of their own in- dependent of theory, and nothing is a per- manently confined quark, except within QFT. Not only is the view oftheory dependen- cy that van Eemeren and Grootendorst have in mind something rather like this degree- of-theoreticity approach, there is also reason for interpreting their Thesis II about the theory-dependency of fallacies rather ex- tremely, that is, in the sense of example (C). Such is anyhow a sustainable interpretation initially, though it may have to be recon- sidered in due course. I shall return to this point shortly, but for now I wish to con- centrate on the theoretical gains that can be got from their account if we accept this ex- treme interpretation. 2. As is now well-known, van Eemeren and Grootendorst characterize as fallacies any violation whatever of their pragma-dialectical rules of procedure for ra- tional conflict-resolution. This is their Thesis I. We can now confidently ask, "Does the van Eemeren and Grootendorst approach give at least the beginnings of a well-articulated theory of fallacies?" It does. A fallacy now is any appropriately disturbing procedural infelicity at any junc- ture of a conflict -resolution discussion. The point is clear and it is also well-stocked with specifications of the procedural rules that van Eemeren and Grootendorst have work- ed out in their several writings and which they have amplified with illustrations of ways in which they can be violated. And this would appear partially to fulfill an im- portant adequacy condition on any suitably comprehensive theory of such things. Clearly, the provisions of Thesis I and Thesis II do seem at first blush jointly to disarm any criticism in the form "But this is not what the fallacies really are". If the fallacies are nothing except within theory, then there exists considerable latitude to construct a theory, in which they cease to be nothing and genuinely become something, pretty much as one pleases. And it seems perfectly open to van Eemeren and Grootendorst to say, "No, they are not that, they are this, i.e. any infelicity contra the procedural canons of rational conflict- resolution. " In doing so, not only do van Eemeren and Grootendorst attempt to simplify fallacy theory and to give it a generality that could hardly have been dreamt of previously, they also appear to have calmed the methodological waters. Endless wrangling about what precisely is the structure of the ad hominem or the dialectical character of complex questions now lapses. It is unnecessary effort, as unrewarding as it is inconclusive. Moreover, since the new theory pro- vides that the fallacies (the infelicities of regulative procedure) are copiously more numerous than the traditional gang of eigh- teen or so, we now have (part of) an answer to the old boundary question: why these? The answer: they are not these. They are as many as there are types of ways of of- fending against the dialectical canon. It also emerges that though not decidable tout court, the van Eemeren and Grooten- dorst fallacies appear to be decidable in a more limited sense. Their fallacies are a recursive class for discussants A ,B, ... in discussions D under the conditions: If 1. A,B ... recognize that R is a bona fide rule of D and 2. one or more of A,B ... holds that move F of D violates R, then F is a fallacy. 3. What to make of it all? A masterly manoeuvre, one might think. As I have tried to say with some emphasis, none of this works without the linch-pin that fallacies are nothing except within theory. Were it other- wise, it would invite and deserve the complaint that what van Eemeren and Grootendorst say the fallacies are they manifestly aren't, and what they say they aren't they are. But iffallacies have no con- ceptual purchase beyond their role in theory then, again, the charge is stilled. We are now at the crux of the business of this paper. It is necessary to ask whether there are good reasons to think that fallacies are, in this way, wholly dependent on theory. If we take theoretical entity hood as coming in degrees, a matter of the distance, so to speak, from sensory stimulation, then "There are some cows over there" give us bovine theoretica of fairly low assay. The GNP of the Netherlands would command a loftier theoretical perch. The entities of QFT would be altogether out of sight. Negative-energy seas of electrons, per- manently confined quarks and spontaneous symmetry breakings would stand free of sensory signatures no matter how diffuse the linkage. Bereftness of sensory engage- ment would seem to vary proportionately with empirical ad hocness. Theoretical en- tities are posited for convenience, or for the further deep rationalization of theory, or for its filling in or smoothing out. This is a large point. There is disagree- ment galore about the linkages of theory with the surface irritations of the theorist. Are theoretical sentences deficient in or empty of empirical content one by one, or is it theories entire that the data underdeter- mine, leaving individual sentences in- capacitated for sensory kinship? It is not necessary to pursue these matter here, for- tunately. But one admonition is called for, I think. Theoretical entityhood is central in the approach of van Eemeren and Grooten- dorst, and as I have said before, "it is a thorny bramble, perhaps no less prickly than the old idea of fallacy which their account tends to place into involuntary retirement."8 In the end, metatheory will have to judge the trade-off. I leave the question to the tender mercies of the ongoing research pro- gramme. However, I must say that it does not seem to me that the English word "fallacy" stands in matters of sensory pro- Buttercups, GNP's and Quarks 69 vocability more to "permanently confined quarks" than to "GNP". "Fallacy" is cer- tainly somewhat sensitive to patterns of as- sent and dissent and dispositions thereto in the linguistic behaviour of human reasoners. But try as one might, one searches for like behavioural moorage for "Lo, a permanetly confined quark!" It may be that the single best symptom of a word's having a purely theoretical denotation is that the word never before existed and had to be invented and conscripted for special baptismal duty. Whatever we make of it, "fallacy" would fail this test hands down. Mind you, it is always possible to appropriate a word that currently lies about. If we can manage to semantically off-load it completely, this is tantamount to having a pristine name to do with as we theoretically please. If the off- loading is only partial, then we can try to make do with the ensuing ambiguity, the new tendril of which we invite to bear the new semantic weight. As things are, it takes time for semantic adjustments to take hold, if ever they will. For the time being, it is open to us and necessary to say that there isn't at present sufficient reason to think that fallacies stand more to quarks than to GNPs, and so no sufficient reason to accede to the most central and most audacious feature of the van Eemeren and Grootendorst account on this, the extreme interpretation of it. Of course, while waiting for the seman- tic facts to come to their senses one could bridge the process by stipulation or lexical invitation. Fallacy theory, old-style, is get- ting us nowhere, a pragma-dialectician might say. So why not realign some of the elderly and frail semantic connections and reconstrue the word "fallacy" in this new way in quest of greater theoretical yield? This is fine, as far as it goes. But stipula- tions aren't authoritative except to those who freely adopt them, and even then only on pragmatic sufferance. For those who don't, and also those who do, the prudent course remains to wait and see how the en- suing costs and benefits shake out. I don't say these things polemically. Far from it; 70 John Woods I am prepared to wait and interested to see. 4. I tum now to a clarification and then a response to an anticipated objection from the pragma-dialectical side. First the clarification. Fallacies, so I have been say- ing, have no pre-theoretical purchase accor- ding to van Eemeren and Grootendorst. They are given a footing only in theory, and the place they occupy in theory can be in principle as distant from or near to sensory irritation as the theorist can arrange within the loose constraints of scientific method. Now I certainly have not meant to say that the fallacies, in the place in theory that van Eemeren and Grootendorst have presently given them, must exhibit the same level of theoretical rarification that QFT currently reserves for permanently confined quarks. Plainly they do not. I mean only that fallacies and quarks have the same pre- theoretical standing; namely none. That fallacies don't stand more to quarks than to GNPs is a fact, if fact it be, about pre- theoretical standing; and the fact is, or so I believe, that fallacies and quarks do not share the same pre-theoretical standing. It should also be mentioned that nothing in the van Eemeren and Grootendorst ac- count, as I am here construing it, requires them to ignore the existence of other theoretical attempts upon the (old) fallacies. They are not saying that the (old) fallacies have never had a theoretical home, no mat- ter how humble or ramshackle-a hovel, as it were-and neither are they saying that un- til now fallacies have had no conceptual pur- chase whatever. All that they need say (and do) is [AJ that all the conceptual purchase that the (old) fallacies have ever and could ever have had is intra-theoretic; and [BJ that those theories have been sufficiently unim- pressive to justify their abandonment and the semantic realignments that I have here been discussing. As for the anticipated objection, perhaps, contrary to my interpretation, van Eemeren and Grootendorst don't quite mean what I say they mean, and so don't think that fallacies have just the same pre- theoretical vacuity as quarks. Fallacies pro- bably don't, in fact, have the same standing as quarks, but admitting it is a nuisance for their account. The greater the degree of pre- theoretical autonomy that the (old) fallacies possess, that is, the more buttercup-like they are, the less they are dismissible out of hand. The more they are somehow there with recognizable (I don't say transparent) buttercuppy structures to tell stories about, the more their abandonment is questionable. That they languish like eunuchs in the pragma-dialectical theory, unanalyzed and innocent of structural engagement with the fallacies that do occur there is a bit awkward. 9 It suggests a certain caprice. It remains open, as I say, to stipulate. "Well, I don't want my word 'fallacy' to be about those things; I have no interest in those things. Let's try something else." But now the stipulation is doubly encumbered: (I) It puts pressure on the new approach to generate good and impressive results, the sooner the better; and (II) It requires that there be some systematic analysis of the old fallacies, if only that despairing of them is not seen as arbitrary. And so, in the end, you find yourself doing fallacy theory the old way, for the nonce if not longer. As it happens, the requirement to attend to the (old) fallacies in something like their traditional identities, is expressly acknowledged by van Eemeren and Grootendorst. The acknowledgement is conveyed by their Thesis III which pro- claims the necessity of construing pragma- dialectically those errors known as the tradi- tional fallacies. to This is striking and im- portant. Thesis III straightaway restricts the latitude of the nouvelle vague theorist to make of the fallacies what he pretty well pleases, and thus diminishes the chances of the new account of evading debate about whether what it says the fallacies are the fallacies really are. Whereupon, the radical ness of the departure announced in Thesis I and authorized by Thesis II, on an extreme interpretation of theory- dependency, is crimped. The theses tug in different directions, Thesis I in the direc- tion of radical innovation, Thesis II in the direction of traditional connotations. 5. It seems, then, that the best way to understand the pragma-dialectical approach concerning the theoretical entityhood of fallacies is as one which provides for their transition from one theoretical environment to another, from the old traditional theories to the new pragma-dialetical theory. In describing the transition, Thesis III flatly disallows the supposition that the old fallacies have been wholly denatured, for at a minimum they can be taken as held over by the new theory for the time being and on sufferance, while possessing precisely the theoretical purchase they they've had all along in the old theories. This can happen in two ways. In the first way, the fallacies of old are absorbed into the new theory with their natures more or less intact, and are given their more or less traditional place there along with the other things that the new theory decides to recognize as fallacies. This can be likened to the passage of the Ptolemic heavenly bodies to the Copernican theory, in which their Ptolemic natures are by no means ex- tinguished (though their behaviour is dif- ferently described). The second mode of transition is one in which a new theory borrows from an old one some of its basic ontic commitments. The borrowed entities are accommodated in the new theory on sufferance, and there undergo an ever-increasing suppression of their natures, as genuinely new entities (often of the same name) emerge in the new theory and ultimately bring sufference to an end. Here we might think of the transitional borrowings of the particles of atomic physics by subatomic physics. Experiment decreed that sufferance be short and it was. The transition was for the old particles a substantial change, as Aristotle would say. Of course, old particles still have a home in atomic theory, but they have none in Buttercups, GNP's and Quarks 71 quantum mechanics. If you believe that Thesis I dominates over Thesis III, then it is possible to hold that the transition of the fallacies from the old theories to the new is like this second case; and thus that the fallacies do not sur- vive the transition, which visits upon them a substantial change-a change that ex- tinguishes their identities in the new theory. But if you give to Theses I and III, approx- imately the same weight, and so if the tran- sition is of the former, more ecumenical kind, then as I have already indicated, the old fallacies persist and call out for an ar- ticulation of their natures, whatever else goes on in the new theory. To their credit, the spirit of wait -and-see is well appreciated by our pragma-dialecticians: The dialectic approach that van Eemeren and I have worked out for fallacies does not, of course, place the final seal on the sub- ject. Within the framework that has been developed there are still a lot of things that need to be worked out, fIlled in and probably also put right. II SO now we have it. I do not think that fallacies are theoretical in the same extreme way that quarks are. And contrary to my earlier conjecture, van Eemeren and Grootendorst don't think this either. If fallacies aren't buttercups, that is not to say that they can't be like GNP's, that is, like things having some degree of pre-theoretical conceptual identity which a new theory must not trifle with altogether. The task of the new theory is to characterize the fallacies in innovative and even startling ways, but not in ways that extinguish their traditional identities, or in ways that do so all at once. The decision to accord to the fallacies a level of theoriticity that preserves non- trivially their pre-theoretical or traditional connotation is a significant decision. It en- cumbers pragma-dialectics with an old albatross, and it remains yet to be seen how it will be disposed of. 6. I say again that it is a virtue of some recent writings on the fallacies that they are 72 John Woods discussed in relative freedom from their traditional connotations, fixed by what is said to be an enfeebled scholastic tradition known, since Hamblin, as the Standard Treatment. The allure of the liberation is not hard to appreciate. One wants to be able to re-examine the fallacies with some pro- spect of revealing novel features of them which may contradict the tradition, yet without running into objections in the form, "But that's just not what we mean by the [e.g.] ad baculum." The freedom is desirable because it is the freedom to in- novate theoretically in ways that improve upon our former effort. As we have seen, the methodological in- strument of this liberation is to be found in claims for the theory-dependency of the fallacies. In moderate versions, the theory- dependency of fallacies l2 restrains the degree of pre-theoretical conceptual pur- chase which the fallacies need be assumed to have and thus leaves to theory at least some of the business of determining their character. This averts a procedural vexa- tion which for ease of exposition I shall overstate to the point of parody: "no dif- ference from the Standard Treatment is allowable, because such difference would falsify the fallacies as they really are (in logical space, so to speak)." Thus the com- plaint against the Standard Treatment is for misunderstanding its own approach to the fallacies. The mistake it makes, on this parody, is that of supposing that in its own account of them, the fallacies are not theoretical entities, and thus that what it says about them realistically describes their true and antecendently fixed natures. Such is the confusion of theoretical postulation with identity conditions. The virtue, then, of some of the new ap- proaches is the prospect of proceeding with the investigation of the fallacious scien- tifically, that is, within the flexible limits of the scientific method under the loose en- couragement of nerve-end hits. Doing fallacy theory scientifically stands in con- trast to doing it analytically, that is, by way of the careful description of intuitions which are thought uniquely to determine the con- ditions of adquacy of the very enterprise itself. It is the way of conceptual analysis as opposed to the way of theoretical con- struction. To some extent, therefore, the doctrine of theory-dependency favours or- do cognescendi over ordo essendi and thus leaves certain features of the fallacies to be determined by conditions on theories, such as simplicity and comprehensiveness, that exceed the data whatever they are. Recall van Eemeren and Grootendorst, in a char- ming phrase: "Fallacies aren't buttercups." It is possible of course to identify more or less extreme versions of theory- dependency, for a domain of enquiry D, by way of the interplay between the mode of scientific postulation and the mode of con- ceptual analysis. A. Extreme theory dependency reserves the option that any and all putative truths in D may be overturned by postulation. B. Moderate theory dependency holds to a minimal core of truths in D as fix- ed by conceptual analysis, and permits postulational deviations from that core to oc- cur only with great modesty-in dribbles, as it were-and only over large-ish intervals of our intellectual history. C. Minimal theory-dependency reverses the dependencies of extreme theory-dependency, and then some. For the radical minimalist, nothing seriously said in D is subject to or secured by postulation. All D truths are held to the standards of con- ceptual realism. If, as in my parody of a few paragraphs back, the Standard Treatment is taken as a radical minimalist position, 13 then the danger exists that it will enforce the at times silly things it says about a fallacy as part of its very concept-that is, it threatens to enshrine error in identity conditions-in consequence of which corrections and im- provements risk dismissal a priori. The virtue of moderate theory- dependency is twofold: first, it allows for theoretical innovations that genuinely im- prove fallacy theory without at the same time losing sight of the things that are generally recognizable as fallacies, as the same things, more or less, that our theoretical forebears were wrestling with. 7. A virtue, of course, is a good thing. Like any good thing, there can be too much of it. Extreme theory-dependency, for ex- ample. I have in mind what I think is the most extreme contemporary version of the no-buttercup theses, a view espoused by Maurice Finocchiaro. Finocchiaro is celebrated for his notorious dismissal of the fallacies as existing only in the mind of the interpreter, and for the kindred view that fallacies aren't self-subsisting entities that have their own independent existence. 14 A fallacy, on this view, is not seen as an ob- jectively verifiable fact but rather as the in- tersubjective interaction of the persons engaged in dialogue. 15 My purpose in this final section is to determine the theoretical uses to which Finocchiaro puts this methodological in- sight, and to contrast its employment with that of van Eemeren and Grootendorst. Finocchiaro has little time for the substantive' 'fallacy", which he disparages for the semantic convention that obliges us to acknowledge the fallacies in terms that he has repudiated just lines above. A freer semantic rein is given to "fallacious" and its substantive cognate "fallaciousness"; and he is prepared to shape his analyses with these as the intended targets. He cannot, of course, permit himself the commonplace that something is fallacious just in case it is or is involved in a fallacy. Unlike van Eemeren and Grootendorst, Finocchiaro's fallacies are not allowed to appear in theory. Nothing is a fallacy, period. And so, in ex- tremis, the fallacies vanish. This radical derangement of the customary cognateness of "fallacy" and "fallacious" is open to regret, if not disap- Buttercups, GNP's and Quarks 73 pro val. But it is more interesting, I think, to concede Finocchiaro the point, if only provisionally, and to attend to what he goes on to do with it. With fallacies out of the picture, what are we to make of fallaciousness? As it happens, Finocchiaro locates fallaciousness in hostile territory bristling with tough immigration laws. For one thing, the account offallaciousness must now honour the harsh provisions of logical consequence, or its converse, entailment. For fallaciousness strikes, says Finocchiaro, when and only when conclusions don't follow from premisses. Notice that since fallacies are nothing whatever without theory, and since Finoc- chiaro finds no place for them in his theory, fallacies don't exist for Finocchiaro. Period. It is different with fallaciousness. It is ac- commodated within Finocchiaro's theory, but in ways that make it unclear whether Finocchiaro is employing exteme or moderate strictures on theory dependency. The one thing that is clear is that Finoc- chiaro's theoretical accommodation of it holds fallaciousness to conditions which (themselves extremely or moderately theory dependent, it doesn't matter) govern the semantics of following from. Straightaway this proves troublesome. I cite three examples: (l) If something is fallacious if and on- ly if it is an inference whose conclusion is not entailed by its premises, then "Goldbach's conjecture is sound, since Goldbach's conjecture is sound" is not fallacious. And this seems unpromising for any account of fallaciousness seeking to deal with circularity. True, if Finocchiaro is an extreme theory dependentist, he could claim the latitude to hold that circular arguments aren't fallacious. But plainly such is not his intention. In fact, Finocchiaro sees the present ob- jection coming. He pleads that the conclu- sion in such cases cannot follow from a premise, since it is not a premise. But this won't do, it simply makes for more trou- ble. If we say that A does not follow from 74 John Woods A, then we have it that it is possible that A and yet that not-A, that is, that there is a possible valuation making A and not-A true. Since this holds for arbitrary A, we're up to our ears in omega-inconsistency, for everything now is the case. 16 (2) "A second reason why one might be entitled to say that the conclusion does not follow from the premises is that it may not follow" [with greater likelihood] "than some other conclusion." Well, apart from the simple semantic point that' 'nor more than" doesn't imply "not", (2) ransacks all valid deductive in- ferences, in none of which does the con- clusion follow with any greater likelihood than in any other. By the test of (2), all valid deductions are fallacious, which, though a novel way to beget Mill's howler, is not a congenial result. In his mention of equi- likelihood, perhaps Finocchiaro is thinking of "following from" in probabilistic terms. But this too runs into similar difficulty. It follows with equi-probability from the premise that this is a true coin, that it will come up heads on the next fair toss, and that it won't. To deprive probability theory of this basic rule is all but to put it out of business. (3) "The fifth type of fallaciousness may be called semantical, and is meant to take care of equivocations. This is the case when the conclusion does not follow because the premises contain a term which has two meanings such that, if it is used in one sense, one of the premises is false (though they would imply the conclusion), whereas if the term is used in the other sense, the premises do not imply the conclusion (though the [ambiguous] premise becomes true); in short, in the context the conclu- sions cannot follow from true premises. " Well, let A = "McDuff is having a pic- nic at the bank" and B= "McDuff is hav- ing a picnic". Suppose now that in A 'bank' means a financial institution. Unless the bank is in California or McDuff is its chief executive officer, the semantic assignment to "bank" is enough to falsify A, or let us anyhow suppose. So, by (3), "A, therefore B" is all right, presumably because the premise is false (a confusion of the strict and material conditional, but never mind, it is all right anyway). Imagine now that by 'bank' we mean lip of a river, and that for this interpretation A is true. Thus, by (3), "A, therefore B" is invalid. But surely it is not. What has gone wrong? The point is not simply that counter-examples can be found to some of Finocchiaro's principal claims; that is not hard to do in argumentation theory, given its present and quite general immaturity. It is rather that Finocchiaro has not decided what to do with a very in- teresting and powerful insight, namely, that fallacies are not things that possess indepen- dent conceptual purchase. Not everyone will share the insight, but for those who do, it is essential that it be allowed to influence weightily the accounts that ensue from it. My conjecture about Finocchiaro is this: (a) If Finocchiaro had had a surer strategic appreciation of his insight, he would never have spared fallaciousness his radical repudiation of the fallacies. If fallacies have no independent conceptual purchase, neither, I should have thought, does fallaciousness. (b) In any event, given that there is something from the old lexicon that he wishes to reconstruct, whether 'fallacy' or 'fallacious' or 'fallaciousness', he both recognizes and does not quite recognize that he is on his own. Where he does not recognize that he is on his own, i. e., free to theorize as he pleases within the broad constraints of scientific method, is in try- ing to reconstruct a credible notion of fallaciousness out of something as imperious as consequence-theory. Knowing this, and so now recognizing that he is on his own, Finocchiaro has to take liberties with something, and he does so. But his liberties-bold and imaginative to be sure- are taken under conditions of maximal strategic disadvantage, for they are taken with consequence-theory, and consequence- theory brings him down every time. I say that Finocchiaro is free to take theoretical liberties pretty much as he pleases within the broad limitations of scien- tific method. But scientific method is at its least permissive in the domain of the logic of consequence, which is precisely where Finocchiaro essays his most novel departures. Thus, the principal methodological dif- ference between the van Eemeren- Grootendorst espousal of the theoriticity of fallacies and Finocchiaro's kindred espousal in this: van Eemeran and Grootendorst at- tempt a radical repositioning of the fallacies, and so place them in a comparatively salubrious climate of their own making. Fallacies, thus, are infelicities against their own rules of rational confict-resolution. I myself do not think that they manage their repositioning of the fallacies altogether unambiguously, but I have no doubt that theirs is the right kind of response to the theoricity of the fallacies. For conflict- resolution theory is stil inchoate, unsettled and filled, therefore, with new theoretical possibilities for the fallacies. Finocchiaro on the other hand, in transacting the business of fallaciousness in consequence theory, chooses a much more conservative theore- tical milieu, one which discourages the very innovations which he seeks to make there. Notes 1 This is a detached bit of an earlier ver- sion delivered to the University of Gron- ingenin May, 1988. I am grateful to E.M. Barth and Jeanne Peijnenburg for incisive and constructive criticism. My Amsterdam colleagues Frans van Eemeren, Rob Grootendorst and Tjark Kruijer were exceedingly patient in their efforts to save me from errors and misconceptions. 2 The principal souces are Frans H. van Eemeren and Rob Grootendorst, Speech Acts in Argumentative Discussions, Buttercups, GNP's and Quarks 75 Dordrecht: Foris Publications (1984) and Frans H. van Eemeren, Rob Grooten- dorst and Tjark Kruijer, Handbook of Argumentation Theory, Dordrecht and Providence: Foris Publications (1987). 3 "Pragma-dialectics: A Radical Depar- ture in Fallacy Theory, " in the Interna- tional Society for the Study of Argumen- tation, Newsletter , January, 1989. 4 That said, there is still some question about the extent to which pragma- dialecticians are free to create for the fallacies new identities. 5 Rob Grootendorst, "Some Fallacies about Fallacies," in Argumentation: Across the Lines of Discipline, ed. Frans van Eemeren et al., Dordrecht and Provi- dence: Foris Publications (1987), p. 335. 6 Idem. 7 W.V. Quine, Theories and Things, Cam- bridge, MA: Harvard University Press (1981), p. 70. 8 "Pragma-dialectics: A Radical Depar- ture in Fallacy Theory," op. cit. 9 See "Pragma-dialectics: A Radical Departure in Fallacy Theory," op. cit. 10 van Eemeren and Grootendorst, "Fallacies in Pragma-dialectical Perspec- tive," Argumentation, 1 (1987), ed. by John Woods, p. 284. 11 Grootendorst, "Some Fallacies about Fallacies," p. 340. 12 As in the pragma-dialectical approach of van Eemeren and Grootendorst, discuss- ed above. 13 For doubts about whether it should be so taken, see the author's, "Why are the Fallacies so Important?" to appear. 14 Maurice A. Finocchiaro, "Six Types of Fallaciousness, " Argumentation, 1, (1987), ed. John Woods, p. 263. 15 Ibid., p. 265. 76 John Woods 16 It is sigificant that omega-inconsistency of Finocchiaro's manoeuvre does not hang on the provisions of ex falso quodlibet, and so does not draw us into a debate between the classical and rele- vant (or paraconsistent) logician. Professor John Woods, Department of Philosophy, University of Lethbridge, Lethbridge, Alberta TIK 3M4. D