13. Finocchiaro, op.cit., page 16. 14. The psychologists Nisbett and Ross include a chapter on causal analysis in their recent empirical book about human inference strategies. They do not include written arguments in their data, as they obviously might have. Rather, they report considerable evidence from laboratory situations which causal errors people do commonly make. 15. Finocchiaro, op.cit., page 15. 16. Ibid., page 17. 17. Ibid., page 16. 18. Ibid., page 17. 19. Scriven M., Reasoning (McGraw-Hili, 1976), page 73. 20. Here I invent an argument which resembles arjituments which I have heard orally on several different occasions, but which I have not yet seen in print 21. Compare Ralph Johnson, "Charity Begins at Home" and Trudy Govier, " Uncharitable Thoughts about Charity", in the I nformal Logic Newsletter, 1981.) 22. Cited in the C.AU.T. Bulletin in September 1978. 23. From a letter to the editor which appeared in the Toronto Globe and Mail on January 3, 1979 .• Dr. Trudy Govier, 3207 Canmore Rd. N. W.,Calgary, Alberta T2M 4)8 When Two Wrongs Make A Right Leo Groarke York University One of the argument forms discussed in contemporary textbooks on informal logic is that which constitutes the fallacy "Two Wrongs". The alleged fallacy occurs when an argument encompasses the notion that some wrong act legitimates another act which would otherwise be immoral As Howard Kahane puts it in Logic and Contemporary Rhetoric: Just as it is almost second nature for politicians to attack their opponents by means of ad hominem argument, so also it is natural for them to defend themselves against the charges of others by using the fallacious idea that two wrong. make a right. The erroneous rationale behind this fallacy is that if the" other side" does it, or some other evil, then it's all right if we do it also.1 I n a similar vein, Vincent Bary writes the "The fallacy of two wrongs is an argument that attempts to justify what is considered wrong by appealing to other instances of the same action.',2 Others adopt a similar point of view. 3 Despite their agreement, however, there are good reasons to think that the standard analysis of two wrongs reasoning 10 is mistaken, or at least incomplete. For though there are occasions when a two wrongs defence of an action is not acceptable (and is perhaps despicable), there are other occasions when such a defence is appropriate and does legitimate an act which would otherwise be immoral. It is in these cases that two wrongs dO"make a right", and it is important to distinguish them if we are to avoid dismissing arguments which can justify particular ways of acting. II. Before we turn to those cases where two wrongs rea· soning is legitimate, we may note the kind of situation where it is fallacious. Suppose, for example, that some government accuses another of subjecting dissenters to torture and other abuses that contravene the United Nations charter on human rights. I n response to such charges, one can imagine the government in question re- plying that the nation which has leveled the charges employs similar-or worse-practices in its treatment of dissent Here we have a clear case of two wrongs rea- soning which illustrates why such reasoning is sometimes illegitimate, for the government in question does not deny that it uses tortur~, and does not justify such practices, but simply directs attention to other cases. I n reply, it may be said that even if jts charges could be substantiated, this doesn't make aausive practices acceptable, and does not excuse the acts in question. At most, it shows that both governments are guilty of the wrongs such acts entail. To provide another example of fallacious two wrongs reasoning, we may turn to Kahane's discussion, where he notes that: Senator Robert Dole, then Republican National Com- mittee Chairman, was guilty of two wrongs make a right when he defended President Nixon against charges of impropriety in the ITT case. (The charge was that ITT had received favours in an important anti-trust suit in return for their huge donation to Nixon's 1972 re-election campaign.) Dole's counterattack was to schedule a news conference to disclose" .. .improper activities involving the Democratic National Convention involving vast sums of money im- properly received from big business". Dole hoped his attack would take some heat off Nixon, and it did until the Watergate scandal brought ITT back into the public eye.4 In response to Dole's defense of Nixon, it can simply be said that it is irrelevant, for it does not show that Nixon was not gUilty of impropriety in receiving campaign donations, but at most shows that the democrats were guilty, too. Yet such considerations do not excuse or legitimate wrong- doing on Nixon's part, and it follows that Dole's attemptto excuse Nixon is fallacious. Despite the merit of the standard analysis in such cases, there are problems with the notion of a two wrongs fallacy, for there are other instances where one wrong does legitimate an action which would otherwise be morally unacceptable. To see the problems that arise in this regard, we might first consider Johnson and Blair's claim that those arguments which appeal to the two wrongs line of reasoning are fallacious even if they amount to no more than attempts to mitigate blame or criticism. According to their analysis: Two Wrong. is a fallacy because, at least in the paradign versions, it is an attempt to do the impossible-to prove that a wrong act is not wrong. A wrong act, multiplied no matter how many times, cannot become right The vari- ations on the basic fallacy don't go quite as far. They are attempts to excuse, or to mitigate blame, or to block cri- l ticism. They don't try to justify the wrong, but their appeal to an earlier wrong remains an irrelevant line of defense. A wrong, repeated, becomes no more plausibl~ blame is no less in order, criticism is just as appropriate.!) To see why this claim is problematic, consider a situation where a bully acts in a malicious way toward a group of classmates. I magine that he taunts them and throws stones at them while they are playing in a schoolyard. Such acts are clearly wrong, and we can imagine him being scolded and punished afterwards. But now imagine the same acts proceeded by a long period where the group of classmates mercilessly provoke the boy in question. Suppose, for example, that they continually tease him because of his weight or colour, and throw stones at him when circumstances ensure that they will not be caught Given such a context, it does seem that the boy is not as much to blame as in the first instance. This is not to say that his rock throwing is acceptable (though in certain cir- cumstances it might be), yet it does seem that the provocations he has suffered (the wrongs committed by his classmates) do mitigate our grounds for blame. In particular, there are less grounds than in a situation where no provocation occurs, and it follows that a previous wrong can make a"wrong" act right As Socrates points out in the Republic, it is not wrong to lie if someone intent on murder asks where his intended victim is. On the contrary, this is a case where the existence of a wrong (the decision to murder) does entail circumstances in which an act which would otherwise be wrong (a lie) is not immoral. I ndeed, it would in this case be morally wrong not to lie, and this shows how much the existence of a previous wrong can change our judgment of subsequent acts that mayor may not occur. Other cases where a wrong creates circumstances which legitimate otherwise unacceptable behaviour are easily come by. In cases of self-defense, for example, the injuring of an individual may be excusable because it is the only feasible response to similar actions on the part of an assailant And in cases where the punishment of an individual is justified, it is their violation of others' liberties which makes it permissable to deprive them of their own liberty, though the acts which punishment involves are not acceptable in other circumstances. I n all these cases, a moral argument can legitimately follow a two wrongs line of reasoning. To see how serious the problems are with the usual account of the two wrongs fallacy, it should be noted that there are many other cases where two wrongs reasoning can also justify an action. To take an important example, arguments in defense of civil disobedience always en- compass the notion that some past or present wrong demands some protest which would otherwise be morally unacceptable (because it breaks the law). ~hus, ~andhi claimed that the British were wrong to stay In I ndla, and concluded that this justified civil disobedience on a massive scale. In a similar way, those in the American civil rights movement, the protests against the Vietnam war, and the more recent protests against nuclear arms, have sometimes engaged in civil disobedience, and always on the grounds that some other wrong warranted suc.h action. The following argument is, for example, the baSIC reasoning used by those anti-Vietnam organizations which arranged the illegal occupation of buildings and other government property: 11 Premise: The government has (wrongly) sent Americans to Vietnam. Conclusion: It is not wrong to send protestors to occupy government property. It would be a mistake to dismiss all arguments along these lines simply because they contain a two wrongs line of reasoning. Of course, such reasoning may be erroneous in a particular instance, but there are cases where civil disobedience is warranted, and in such cases one wrong does legitimate an action which is otherwise unacceptable. Another context in which two wrongs reasoning is sometimes appropriate is international politics. When a nation declares war on an aggressor, for example, it typically argues that the nation in question has acted in an immoral fashion, and that this legitimates retaliatory action. Pacifists might argue that such responses are always unacceptable, yet this is a contentious conclusion, and one would need a good deal of argument to back it It will not suffice to point out that declarations of war depend on two wrongs reasoning. Another political situ- ation which can make for legitimate uses of two wrongs arguments arises when a particular nation argues that specific policies concerning immigration, employment, the environment, and the export of goods are legitimate because other nations adopt similar policies. I n Canada, the impOSition of tariffs, protectionist trade policies, and restrictions on foreign students have all been justified on these grounds. And though such actions are not always justified, it would be a mistake to dismiss all arguments along these lines. Given such examples, it follows that the version of the two wrongs fallacy known as "common practice" is problematic, for it allegedly invalidates any argument that contains" an attempt to justify some wrongdoing on the basis of some practice that has become commonly ac- cepted.,,6 Given such a fallacy, one who does not believe in protectionist trade policies cannot consistently argue that they may be necessary if they are a common practice among other nations. Yet such an argument cannot be easily dismissed. In view of the couterexamples to the claim that two wrongs reasoning is fallacious, it should be clear that it is a mistake to suggest that two wrongs arguments must depend on an "erroneous rationale" (Kahane) or that it is "impossible" to construct a valid two wrongs argument (Johnson and Blair), or that such arguments introduce"a premise that is irrelevant to the conclusion either that we are innocent of wrongdoing or that our wrongdoing is justified" (Barry). On the contrary, there are occasions when a previous wrong makes a wrong act right. III. The problems with the standard account of two wrongs reasoning result from a failure to define the fallacy in a precise enough way. For though the usual definitions suggest that all instances of two wrongs reasoning are fallacious, it should be said that certain kinds of two wrongs arguments are invalid. It follows that the heart of any satisfactory account of the two wrongs fallacy must be a distinction between those cases where such reasoning is valid, and those cases where it is invalid. To see how such an account can be constructed, we may begin with Johnson and Blair's definition ofthe two wrongs fallacy, for it is more amenable to correction than the definitions already cited. According to their account, the two wrongs fallacy occurs when: 1. M's action, X, has come under criticism. 2. N (or M) tries to defend either X or M by citing Y,Z, or W-allegedly similar actions (the wrongness of which is granted or at least not challenged). 3. Y, Z, or W have no relevance to the defence of X or of M for having done X. 7 In defense of this definition, it might be said that it adequately covers the counterexamples I have suggested, for in all these cases clause 3 is violated (i.e. because Y, l, or Ware relevant to the defence of X). It might therefore be said that the misleading nature of the usual account might be rectified by adopting this definition along with the warning that there are (many) cases where clause 3 is violated. This is a move in the right direction. However, more needs to be said, for such an account is not very helpful unless it gives us some indication of the kinds of conditions which entail the violation of clause 3. It is by making these conditions explicit that we can construct a better definition of the two wrongs fallacy. To see the restrictions on clause 3 that are appropriate, we may note two features that characterize all the examples where two wrongs reasoning justifies some action. To begin with, all are cases where the act in question is an explicit response to the previous wrong which is in question-a response which tries to alleviate its unjust consequences. Hence when one lies to a would- be murderer, one lies in the hope that one will prevent a murder, and when one commits violence in self-defense one attempts to prevent the unjust consequences of an unjustified attack. Similarly, when one commits civil disobedience one protests some law or action which is allegedly unjust, and when one punishes an individual one does so to prevent injustices that might otherwise occur (perhaps to provide compensation or to discourage such acts in the future). All the counterexamples we have considered are analogous in this respect. The second feature that characterizes the suggested counterexamples concerns possible alternatives to the act in question. Thus, an individual can legitimately embark on wrong acts in response to other acts only if there are no morally preferable acts which are equally appropriate. One can, in short, legitimately commit a wrong only to the extent which is required by the circumstances. In cases of self-defense, for example, one is entitled to meet force with force only to the extent which seems necessary to prevent the attack in question. To continue to use violence after the attack has effectively been stopped is morally repugnant, and illegal. I n a similar way, the justification of cases of civil disobedience must be based on the claim that there is no more acceptable way to protest the law or action which is called in question. If one is, for example, in a position to use legal means to quash some unjust law, then morality demands that they be used. It is only when such means are not available (or for some reason not feasible) that civil dis- obedience is warranted. To take one other example, a nation which declares war is not justified if there is a morally more acceptable and equally effective way to resolve the dispute in question. The other counter- examples can be analyzed in a similar way. Given the two constraints on valid two wrongs reasoning, we may incorporate them into our definition of the two wrongs fallacy, and can say that the fallacy occurs when: 12 1. M's action, X, has come under criticism. 2. N (or M) tries to defend either X or M by citing Y, Z, or W-allegedly similar actions (the wrongness of which is granted or at least not challenged). 3. N (or M) fails to show that X is a response to Y, Z, or W that attempts to eliminate or alleviate its unjust con- sequences, or fails to show that there is no morally preferable response. In essence, this definition stipulates that the two wrongs fallacy occurs when the constraints on valid two wrongs reasoning are violated. Given such a definition, it is never- theless clear that valid two wrongs arguments can be constructed in the cases we have considered, for these are cases where it can be shown that an action is an attem pt to alleviate the unjust consequences of other acts, and where there is no preferable alternative. To show that our definition of the two wrongs fallacy; does indict those situations where two wrongs reasoning is fallacious, we may return to earlier examples. To begin with, it is clear that a country which is accused of torturing dissenters cannot justify its actions by appealing to the wrongs of the accusing nation, for its acts can't be said to be a response that tries to alleviate their consequences. In a similar way, Dole's attempt to excuse Nixon from blame is not successful, for he does not show that Nixon's actions are an attempt to rectify the injustices caused by wrong- doings on the part of the Democrats. To take a final example of fallacious two wrongs reasoning (this time from Johnson and Blair), we may con- sider a case which arose in light of the Soviet Premier Alexei Kosygin's visit to Canada in 1974. The Soviet Union was at that time under attack for alleged discrimination against its minorities, and an account of his meeting with members of parliament reports that: When MPs asked Mr. Kosygin about problems with Jews and Ukrainians, he said such problems exist around the world. He mentioned blacks in the United States and the Irish in the United Kingdom.8 The implication is that discrimination in the Soviet Union is morally excusable, given that it is common practice throughout the world. In reply, it may be said that this doesn't excuse such discrimination and Kosygin's appeal to common practice is not convincing, for there is (ob- viously) no sense in which Soviet discrimination can be said to be required as a response to discrimination else- where. Given that the outlined definition of the two wrongs fallacy does adequately deal with legitimate and illegi- timate uses of two wrongs reasoning, one aspect of the definition deserves comment Thus, it should be noted that it leaves open the question what sorts of acts are morally preferable in a particular case, and that this question must ultimately be answered by turning to moral philosophy. Of course, we could build a more definite conception of morality into our definition. We could, for example, adopt a utilitarian perspective and declare that acts are to be ranked according to their abilityto maximize happiness, but such a move begs important philosophical questions. Qua logician, it seems better to leave such questions for elsewhere, and it is in most cases possible (at least for pedagogical purposes) to rely on intuitively clear examples. The vast majority of cases where fallacious two wrongs reasoning does occur (and certainly those cases considered in introductory texts on informal logic) can be handled without a detai led analysis of such central moral issues. IV. In summary, it may be said that the suggested account of the two wrongs fallacy is preferable to the standard account because it recognizes those cases where two wrongs reasoning is legitimate, and those cases where it is not In contrast, the usual account creates the mistaken impression that any two wrongs argument is unacceptable, and perpetuates the related misconception that acts are right or wrong independently of the circumstances that surround them. The proposed definition not only enables the student learning informal logic to recognize those instances of two wrongs reasoning which are fallacious, but also enables them to recognize valid two wrongs arguments. To see how our account allows a deeper analysis of two wrongs arguments, we might consider an example of two wrongs reasoning which Johnson and Blair discuss. Written in defense of a group of militant Indians who occupied a small park in Kenora, the argument in question suggests that: The occupation of a 14-acre park by the native people in the Kenora area is completely justified. After all, what's a mere 14 acres when they have been robbed of 14 million square miles-the entire North and South American con- tinent 9 According to the standard analysis, this argument is invalid because it contains an instance of two wrongs reasoning. In short, because it argues that: The Indians were wronged long ago. The Indians were justified in occupying the 14-acre park in Kenora We have already seen that such an analysis is problematic because it is not true that all cases of two wrongs reasoning is invalid. In contrast, the revised account does not make this mistake, and tells us exactly why the argument represents a case of invalid two wrongs reasoning (for our purposes, we may ignore the problems that arise in light of its problematic premise). As this is a case of civil disobedience, the first restriction on valid two wrongs reasoning is fulfilled (Le. the Indians in question are at- tempting to alleviate some perceived injustice). Yet th~ argument is still guilty of the two wrongs fallacy because It fails to show that there are no more appropriate means by which the I ndians could effectively achieve their end. Of course, it may be that the argument could be rectified, but this would require additional details which are not pro- vided in the original argument Given this analysis of the argument in question, it can be said that it provides a more sensitive account of the argument, and of two wrongs reasoning, than the usual analyses. Given the frequency of such arguments in every- day moral reasoning, the proposed account can provide a worthwhile addition to an account of informal logic and the fallacies which it contains. Notes 1 Howard Kahane, Logic and Contemporary Rhetoric, p. 28; Belmont: Wadsworth, 1976. 2Vincent Barry, Practical Logic, p. 60 (Second Edition), New York: Holt, Rinehart and Winston, 1980. 13 3For other accounts, see Capaldi, The Art of Deception (New York: Donald W. Brown Inc., 1971) and Johnson and Blair, Logic~ Self-Defense (Toronto: McGraw-Hili Ryerson, 1977). I shall discuss Johnson and Blair's account as I proceed. 4Kahane, p. 29. SJohnson and Blair, p. 78. 6Barry, p. 61. 7Johnson and Blair, p. 78. 8Ibid., p. 9S. 9Ibid., p. 74 .• Leo Groarke, Department of Philosophy, York University, Downsview, Ontario M3J 1 P3 Canada Musclebuilding for Strength in Critical Thinking ~ark VVeinstein Hunter College, CUNY In a recent paper1 Richard Paul argues for teaching critical thinking in the "strong sense". The paper offers his analysis of critical thinking in the"strong sense", but most importantly, attempts to motivate critical thinking teachers to give their students a sense of the "global problems in the analysis and evaluation of reasoning" (p.3). Paul argues that we must help students toward a "clearer recognition of the relationship between world views, forms of life (and) human engagements and interests": placing at the focus of our concern" dialectical/dialogical" "argument exchanges" in which "fundamental lines of reasoning are rarely "refuted" by individual charges of fallacy" (p.3). He requires that such a stance be taken "from the outset" in our critical thinking courses and suggests that we use "multicategorial ethical issues" (pp.2,S) as the means of obtaining strength. Paul rejects what he appears to consider the usual procedure in critical thinking courses: the teaching of ".a battery of atomic technical skills independent of egocentric beliefs and committments" (p. 3). Such an attitude towards the teaching of critical thinking, Paul maintains, merely reinforces student tendefl€ies t-o respond to challenging ideas with "sophistry" and "dismissal"; that is, to illicitly defeat or disregard alternatives threaten ing to "a priori egocentric (or sociocentric) belief systems" (p.2). Since I concur with much of what Paul says, and, moreover, believe that the II Spilit" of his Ploposal is more