This paper argues for a point about fallacies often ignored in logic texts that gets further distorted in Christopher Johnson's article, ‘Reconsidering the Ad Hominem’.Footnote ¹ Johnson has argued that the traditional view of the ad hominem fallacy is wrong. He claims that appeal to ad hominem considerations may be justifiably invoked in a number of narrowly defined cases and for a number of reasons. Cases would include those in which the evidence for conclusions on both sides is equally strong, the arguments are equally coherent, and it is unlikely that any new evidence would make either side more compelling. In such cases resorting to ad hominem considerations might justifiably shift the weight of judgment.Footnote ² Another example might be the application of research and analysis in public policy. If you have two or more equally strong analyses, it could be highly significant what methodology was used to achieve the results. Since the use of a certain methodology is often tied to personal characteristics of the researcher, those ad hominem characteristics might be relevant in choosing which policy to utilize.Footnote ³

Johnson uses an earlier analysis by Lawrence Hinman to support his case and then lists several of the common objections to Hinman.Footnote ⁴ The objections primarily argue that the answer to the above dilemmas is to look further into the evidence or the premises, not look at the personal characteristics of the individuals which have nothing to do with the strength or weakness of the argument itself. Moreover, against the claim that existentialism or Marxism makes regular use of ad hominem issues, the objection is that the fact that people use bad reasoning at times does not justify it. The objections are the standard ones given in logic texts tailored here to fit the case being made by Hinman and in further detail by Johnson.

My response to Johnson's argument is quite different. I agree that ad hominem considerations are relevant. In fact they are always relevant to some degree or another because they are a form of evidence and they are not just important in certain limited cases. Whether that evidence is good or bad depends on the same criteria as any other evidence, i.e. a string of verifiable and dependable results. The power of Johnson's article lies in equivocating ‘ad hominem fallacy’ with ‘ad hominem considerations’. Based on the textbook assumption that the ad hominem is a standard fallacy, Johnson claims then that this so-called ‘fallacy’ may not be a fallacy at all in certain narrowly-confined cases. But this is true of all the traditional fallacies. What appears to be a fallacy is not a fallacy because, even though it has the same form, the evidence supports the conclusion. Moreover, what Johnson consistently refers to as the ‘intellectual virtues’ are virtues for him because of how they play out in actions.Footnote ⁵ Intellectual courage and fair-mindedness have to do with previous actions and that is why these virtues are attributed to people. It seems glaringly obvious that if someone has been fair-minded in ten previous projects, there is a highly reasonable inductive likelihood that the person will be fair-minded again. But why is this news? If the ad hominem fallacy were a distinct entity, separate from other forms of reasoning and set clearly in a category called ‘fallacy,’ then Johnson's point would be a startling break from old thinking. But in fact the ad hominem is neither a distinct form of reasoning nor primarily a ‘fallacy.’ Equivocating on ‘ad hominem' lends a profundity to Johnson's case while I would argue that his conclusion is, in a strictly non-ad hominem sense, trivial.

Some traditional fallacies relate primarily to the source of the fallacy while others relate more to the reasoning process itself. For the latter consider hasty generalization and composition. The heart of these fallacies is a leap from available evidence to an unjustified conclusion. For hasty generalization it is the assumption that a small or non-random sample points to a generalization about the whole. Composition assumes without further evidence that the whole thing is like the parts. Now both of these may in certain cases not be fallacious. If a small sample has had all the variables carefully understood, then the leap from a small sample to a conclusion has the signature of the fallacy but in fact is not one. Studies with samples of plutonium are a classic example. The same would be true of composition. A classic example is contrasting ‘if all parts of the machine are light, the machine is light’ with ‘if all parts of the machine are gray, then the machine is gray.’ The obvious difference is that chains of evidence other than the parts-whole equivalence dictate whether what appears to be a fallacy is or is not so.

The fallacies linked to the source are directly relevant to Johnson's point but follow the same pattern as above. ‘Appeal to authority’ is closely analogous to the ad hominem issue. As usually put, an ‘authority’ is an individual or group with a track record on the subject at hand, a record that has been tested by careful work or carefully examined experience over time. If someone does not have this record, e.g. a famous athlete advertising an automobile, then ‘appeal to authority’ is a fallacy. If the track record is there, e.g. a Nobel Laureate making claims about his or her specialty, then not only is the fallacy not present but the fact that this particular individual is making the claim adds to the strength of any argument being made. It is not that the ‘fallacy’ becomes acceptable nor is it somehow mysterious that the personal characteristics of the individual are relevant. Of course they are relevant because there is an evidential link between those characteristics and the conclusion being drawn.

Another example relating to the source briefly alluded to by Johnson is tu quoque.Footnote ⁶ A parent's advice to a child not to smoke is logically distinct from the fact that the parent smokes. For the child to say, ‘But you smoke,’ is irrelevant to the logic of the argument; thus the child commits the tu quoque fallacy. But personal characteristics are relevant as evidence and not just in narrowly defined cases. We may question the same Nobel Laureate's claims about forms of government because, even though the person may be very bright, he or she has no track record of action or previous thoughts on that topic that have been examined. For the child in the tu quoque example, as well as for all of us, prior actions over time are a form of evidence, perhaps some of the strongest evidence. If a parent tells a child not to smoke but does it himself, it is perfectly justifiable for the child to assume that, despite what the words say or what the scientific data implies, this person, who has a responsible track record of feeding me, sheltering me, getting on in a blooming buzzing world, knows what he or she is doing. Based on the overall evidence in the parent's life, there must be a deeper reason in support of smoking for such a person as this whose record of ‘getting along’ is, from the child's perspective, evidence of successful surviving. Now all this would completely depend on the situation, the number of evidential chains involving the parent and child, and how the ‘fallacy’ is utilized by the person using it. Nevertheless, the fact that an individual's personal record of experiences and individual characteristics may be relevant in justifying a conclusion is not startling news. It is simply a function of the fact that it is a human being that is presenting the argument and that evidential chains can be found relevant to that argument based on the actions done by that person. With further evidence, whether it be in hasty generalization, composition, appeal to authority or the ad hominem arguments, these ‘fallacies’ simply are not fallacies. This point is generalizable to all the informal fallacies and there is no reason to think the ad hominems are any different.

One more point should be considered. Johnson analogizes the moral virtues and the intellectual virtues and grounds both in what might be called the consistency of character. Moral courage may well imply intellectual courage because both involve facing up to and dealing with fears. He even goes so far as to suggest that, e.g. an interest in fine wine may indicate the intellectual virtue of attention to detail.Footnote ⁷ He admits this is tenuous but possibly relevant. But then he also gives several counterexamples such as loyalty to the integrity of one's discipline having nothing whatsoever to do with loyalty to one's spouse.Footnote ⁸ I think he is right on all fronts here and his discussion is interesting. However, the intellectual virtues he cites are all ones tied to actions, i.e. they are forms of practical wisdom, and it is the actions in the relevant arena that constitute the verifiable evidence relevant to the point at hand. Johnson says, ‘it itself is an intellectual virtue to know when appeal to the ad hominem is useful and justifiable’.Footnote ⁹ What I have objected to in this paper, and what is at the heart of his equivocation, is that one is not ‘appealing’ to anything. The evidence of a person's characteristics in their work or their record on a subject is simply evidence; it may or may not be more important than other kinds of evidence. In the scientific world it is usually less important but still relevant; in the everyday world it may be equally important or more important than other evidence. A loving parent who has nurtured a child for 18–20 years may know only a general outline about the field the child is considering pursuing but, when such a parent suggests that the child at least look elsewhere before making a decision, if nothing else the child should consider the source and that someone else's personal experience may at times be superior evidence to one's own. There is nothing new in taking ad hominem considerations seriously. ‘Reconsidering’ them is most useful in showing how those considerations are in fact an integral part of all evidential considerations.