Ad hominem arguments substitute commentary on a person or persons for an evaluation of the claims they make and the evidence they put forward for accepting those claims. Personal commentary is not always fallacious, but very often it is. Ad hominem arguments come in a variety of forms, but we can divide them into roughly two types; those that present criticism of the target (i.e. the subject of the personal commentary) and those authority arguments that misuse the personal credibility of a source for a claim.
Critical Ad Hominems
In each of these cases, personal commentary about the target of the ad hominem has been used as a reason to reject a claim she has made even when that claim is not directly related to the personal information presented about her.
Ad Hominem (Abusive): This is perhaps the most common form of an ad hominem argument. It consists of an attack on the person in question for the purpose of dismissing something she has to say. One could for example suggest that a professor of history doesn’t know her topic because she dresses funny, behaves rudely, or speaks with an odd accent. In either case the issue at hand is completely irrelevant to the truth value of any claims that professor has made on the subject of history. Although at least one of these possibilities (that of being rude to students) could be a very serious issue in its own right, it is not a reason to doubt the credibility of a professor on the subject of her expertise.
This type of ad hominem certainly includes outright insults. In politics, for example, it used to be common to call adversaries to the left ‘Commies’ and those on the right ‘fascists’. This abusive vocabulary has shifted a bit of late (and so we sometimes hear folks accused of being both communists and fascists at the same time). Those skeptical of religion may refer to the faithful as ‘superstitious’ while believers sometimes refer to their critics as ‘nihilists’ who believe in nothing at all. An art critic may be called a ‘philistine’. A magazine may be dismissed as a ‘rag’. We can find similar themes for a broad range of subject areas. While it is by no means clear that all references using derogatory language are fallacious, they are certainly fallacious when they are used to dismiss a point of view without providing a direct critique of it.
Because an ad hominem is not merely irrelevant, it is also rude, people often fall into the habit of confusing the issues (rudeness and lack of logical relevance). Internet discussion forums will for example admonish users to refrain from committing ad hominems as if the irrelevance of insults were the real concern to forum administrators whereas such rules are almost always about preserving the peace and reducing personal conflict. That is a question of behavior, not one of relevance. It is important to remember that what makes an ad hominem a fallacy is not the rudeness of the commentary, or even the way it makes another person feel; it is the irrelevance of those comments to the conclusion of the argument.
It should be added that the presence of disrespectful or outright abusive language in an argument does not ensure that the full argument can be dismissed as an ad hominem fallacy. If an argument contains an otherwise compelling line of reasoning, it does not become fallacious because its author has added an insult or three. In such cases the abusive commentary neither adds to nor detracts from the argument in question, though it may raise ethical issues in its own right. An argument becomes an ad hominem (abusive) when the insults themselves are treated as a reason for dismissing the target of the criticism.
Ad Hominem (Circumstantial): The ad hominem (circumstantial) argument is tricky. Some sources will consider this a valid (and hence non-fallacious) form of argumentation, and some varieties of critical theory seem to use it on a regular basis. This form of ad hominem occurs whenever someone calls into question the motives and/or biases of another party, thus undermining her claims by commenting directly on her personal reasons for making those claims.
In politics for example, it has become common to suggest that liberals favor social welfare programs in order to keep minorities dependent on the Democratic Party. Conversely, libertarians are often said to be tools of large corporations. Christian apologists will sometimes suggest that atheists are simply seeking to avoid responsibility for their sins, and non-believers often suggest that religious beliefs are thinly-veiled mechanisms of social control. In each of these examples, questions about the motives or cognitive limitations of those targeted have been substituted for a critique of their point of view.
The ad hominem (circumstantial) needn’t involve an assertion of vested interests. It might also address cognitive biases and emotional states as in the case where religious beliefs are dismissed as products of brain-washing (or at least as the mere accident of upbringing …”you only chose catholicism because you were brought up in a Catholic home.”). Conversely, in his book, The Faith of the Fatherless, Paul Vitz famously advanced the theory that there is a link between atheism and fatherlessness (or at least problematic to relationships with father figures). Whatever the (de-)merits if this psychological theory in itself, whenever Vitz’s theories have been used to discredit atheist views, that use is an example of the ad hominem (Circumstantial).
Problem: What makes this particular form of ad hominem tempting is the very real possibility that the judgment of the other individual has been swayed by personal circumstances. Since the cirumstances raised in the argument are in fact related to the claims in question, the relevance of the argument is trickier than the simple ad hominem (abusive). It is important to address large-scale biases in politics and other social institutions. So there are times when a well-documented account of such biases would be worth considering. This kind of argument remains problematic however insofar as it places an unreasonable burden on the party criticized.
In asserting that another party’s point of view is the product of personal bias and/or ulterior motives, one can bypass direct consideration of any claims made by the target of criticism. An argument using the ad hominem (circumstantial) may thus be equally effective whether it is directed against a well-reasoned position or one that is deeply flawed. An additional problem lies in the fact that some things which might appear as biases to some could as easily be viewed as sources of insight to others. (If a traumatic incident could be viewed as the sole reason for adopting a religious belief and hence rejected as a mere bias, it could as easily be described as a gift through which someone came to an unusual understanding of the world.)
So what should we do?
One way to approach this would be to inquire into the degree to which the personal credibility of the target is at issue in the first place. Where the target of an ad hominem (circumstantial) has presented objective reasons for their position, it would be inappropriate to use a critique of their probable biases against them. This leaves open the possibility of using the ad hominem (circumstantial) as a valid means of criticizing those whose personal credibility has been offered as grounds for accepting a claim. Simply put, if someone gives you an allegedly objective reason to believe something, it is best to set the ad hominem (circumstantial aside) and consider their argument. If they are content to rely on their personal credibility for validation of their views, then the ad hominem (circumstantial) goes directly to the issue of that credibility. It is then quite relevant.
An additional approach to this issue might be to develop a two-pronged response to those one suspects of having biases, presenting both a direct critique of their views and an argument on their biases at the same time. The first serves to show that they are wrong. The second helps to show why, and perhaps why it matters.
Ad Hominem (Tu QuoQue): This variation of an ad hominem argument occurs when someone uses a perceived inconsistency on the part of the target as an excuse to dismiss something she has said. This would normally occur in the context of discussing some moral proposition, concrete recommendation, or otherwise value laden topic. It happens when people talk about what folks ‘ought’ (or ‘ought not’) to do.
A certain degree of inconsistency is a pretty common feature of human behavior, but then again so is outrage at such inconsistencies. The prospect that someone is holding others to moral principles they don’t live up to themselves can be very frustrating. So, whenever people advance an ought without living up to it, we should not be surprised to see someone respond with a tu quoque fallacy.
The central problem with a tu quoque fallacy lies in the fact that any degree of inconsistency on the part of someone advancing a value does nothing to show that the value itself is not a good one. Even a person who has done nothing whatsoever to live up to a value may be quite right to insist on its importance. One might suggest to such people that they really ought to take their own sermons to heart, but not that their sins disproved the messages of those sermons (so to speak).
As I write this, for example, President Obama is pursuing negotiations with the state of Iran over a prospective deal regarding nuclear power in that country. Republican senator Tom Cotton penned an open letter to Iranian officials calling the deal into question, and 47 Republican Senators signed the document. Many Democrats have argued that this is a severe breach of protocol insofar as it is not appropriate for members of congress to interfere directly in diplomatic negotiations with a foreign country. In response to this, some Republicans have pointed out that Democratic senators have done similar things in the past, thus underscoring the hypocrisy of Democrat’s criticizing Cotton and his cosigners. But of course the alleged inconsistency on the part of Democrats here does not prove that they are wrong about Senator Cotton’s actions. It could as easily prove that they were wrong themselves when any of their number did likewise.
This example illustrates an additional feature of the tu quoque fallacy, that it is often reflexive, i.e. that the accusation of inconsistent behavior may facilitate the same. In the previous example, those accusing the Democrats of shifting positions on the propriety of direct messages to foreign ministers in opposition to the President may effectively shift their own position on the subject. In effect, some Republicans may support Cotton (by accusing Democrats of hypocrisy), even if they condemned past examples of Democratic interference in Diplomatic negotiations by Republican Presidents. The Republcans thus get to flip-flop on the issue while advancing the criticism that their opponents have done the same. I like to call this the meta-hypocrisy shuffle, and it is particularly common in politics.
Yes, the reverse occurs as well.
Qualifiers (Non-Fallacious Variants of Ad Hominem): It should be noted that ad hominem fallacies are not always considered irrelevant. When a person (and/or her character) is the topic of discussion to begin with, it certainly is not fallacious to speak about them or their behavior. So, for example, if someone has asked which American President is the worst in history, it would not be fallacious to say bad things about any one of them. far from irrelevant, that would actually be the point of the discussion. Likewise, it is not fallacious to subject authority claims in support of a conclusion to personal scrutiny. Neither would it be fallacious to refuse to discuss an issue with someone on account of personal criticism. It is one thing to say that someone is wrong because you don’t like them and quite another to say that you don’t wish to discuss their views at all. Finally, if the point of an insult is simply to insult another person without trying to prove them wrong on a particular claim, then that may be mean, but it is not a fallacy. What distinguishes an ad hominem fallacy from the many other varieties of personal criticism and abuse is the use of that personal criticism to discredit a point of view.
Most importantly, there are contexts in which personal criticism may be directly relevant to arguments from authority. When told, for example, that a given professor has taken a stand on the wisdom of a given economic policy, it would not necessarily be fallacious to note that her degree was in poetry (rather than economics), to note that her degree (in whatever field it may be) came from a disreputable institution, or to suggest that other foolish claims made by the professor brought her understanding of the topic into question. Such arguments would be relevant precisely to the degree that they answer any assertion that the professor’s expertize was sufficient reason to accept her conclusions on the policy. This would not prove her wrong, per se, but such criticisms could effectively be used to negate any authority she may be presumed to possess.
Litmus Test: In considering whether or not an argument commits the ad hominem fallacy, ask yourself if personal information about someone has been used as a reason to reject something they have said. If so, first double check to see if that personal information might be relevant after all. If not, then this is most likely an ad hominem fallacy. If the personal information amounts to an attack, then this is an ad hominem (abusive). If the personal information calls into question the target’s motives and personal biases for taking a position, then the argument is an ad hominem (circumstantial). If the personal information calls into question the consistency of the individual’s words or deeds, then it is a tu quoque.
Fallacious Appeal to Authority
A fallacious appeal to authority isn’t normally described as an ad hominem, but it too consists of an appeal to personal information when more objective arguments are called for. So, it seems reasonable to consider it a variety of ad hominem, and that is the approach we are taking here. Appeal to authority is not always fallacious insofar as one might reasonably offer the expertise of a learned individual as a good reason to think her claims are true, provided the claims fall within the overall topic of her expertise. Problems arise when folks are cited outside the context of legitimate expertise. This can happen in a variety of different ways.
Anonymous Authority: This may seem absurd, but authority is often cited without identifying a particular source. This occurs for example when people say that they have read a study without naming it or when someone tells you that they have read statistics proving some point without providing any reference to those statistics. It occurs when students cite “the internet” as their source on an assignment, and even when they cite Wikipedia (as this alone does not tell us who has provided the information they have pulled from it). It should perhaps go without saying that when citing authority one should at least know who the authority is, but people (even highly educated people) often make use of short-hand references to authorities existing somewhere out there, so to speak.
Authority by Association: Authority is sometimes invoked by association. This occurs when close friends or relatives of a given authority are cited on a given subject without providing any substantive reason to believe they are themselves experts or that they have any insights into the views of their authoritative friend/relative. Granted, such individuals may know something about the authority’s point of view on the subject (it’s at least possible) but they may also take advantage of their relationship to express views of their own under the authority of their relationship.
Documentary films and television may also conjure the effect of authority by association when mixing interviews with genuine authorities and those of individuals without any real expertise on the subject at hand. If the viewer isn’t careful, she may not notice the difference. Alternatively, she may forget which source provided which piece of information when recalling the film later. It’s a pretty simple mistake, but one that is easily made, and one which is often encouraged by unscrupulous film-makers.
One would do well to ensure that any authorities one is asked to consider are the actual sources of the claims to which their expertise has been applied, and not merely someone associated with an expert on the subject.
Out of Field Authority: It sometimes happens that a legitimate authority makes a claim outside her own field of expertise. There is nothing inherently wrong with this, and great scholars often make real contributions to fields outside those of their own training and research, but it is important to remember the limits of expert credibility. One would not want to mistake expertise in engineering for expertise in economics, nor that of anthropology for physics. The point here is not to dismiss everything a scholar says outside the field of her own training or research, but to refrain from giving extra weight to their views on the sole basis of expertise in a completely unrelated field.
On a related subject, one should be wary of abstract references to expertise. If a scholar who makes much of possessing a PhD without providing any indication as to field in which she obtained (or the institution which provided it), this is a definite red flag. When presenting credible bona fides, an individual will normally detail the specific degree belonging to her and/or university in which she earned it. Better yet, a professor will normally list the department and institution in which they work. This isn’t always the case, but it’s a common enough trick that one ought to watch out for it. This information doesn’t have to be on the first thing they tell you, but it should certainly be available. If someone wants you to know they have a PhD, but doesn’t tell you much more than that, you may well be looking at someone who is writing outside their field and doesn’t want you to realize this. Unless she has some other basis for claiming expertise on the subject, one would do well to regard such bona fides with suspicion.
Decontextualized Authority: I suppose it shouldn’t need to be said that authorities can be cited out of context, but it happens rather often. Ideally one should expect to learn something about the overall subject at issue and the general point of any larger text or utterance in which a quoted statement appeared. Still, it is common to read quotes without any effort even to identify the source materials at all. This practice is sometimes referred to as quote-mining, and it has become particularly common with the proliferation of visual memes on the internet. It’s worth noting that a number of the quotations floating around the net are simply undocumented, and indeed quite a few may be outright fakes. For this reason Both Monticello and Mount Vernon, for example, have web-pages devoted to the spurious quotes attributed respectively to Thomas Jefferson and George Washington. This hasn’t stopped others from passing those fake quotes off on websites all across the net as well as occasional books and speeches.
The problem with quote-mining goes far beyond spurious quotations, however, insofar as any attempt to invoke someone as an authority ought to include a reasonable effort to account for the context in which she has weighed in on the topic at hand. Often an author’s real point will turn out to be something entirely different than one would gather from a single pithy line. You might find that a given quote seems to resonate without knowing such details, but lacking the context in which it occurred, there is little reason to treat any one-liner from any source as an authoritative statement on any given subject.
Non-Authorities: One of the most common ways in which authority arguments go astray lies in citing people as experts when they simply are not experts in anything at all, as when for example celebrities are treated as experts (…“I’m not a doctor, but I play one on TV”). A somewhat subtler version of this problem arises when people come to be treated as experts in an area wherein they have achieved considerable success, thus giving the sense of their authority a degree of plausibility. Success in a given area may be due to any number of factors other than knowledge, so one shouldn’t be too quick to accept this kind of authority claim. it would be a mistake, for example, to assume a box-office megastar knew a great deal about acting or that a wealthy businessman fully understood economics. Such individuals may presumably know something about the subject in question, but their success is not predicated on their breadth of knowledge or upon their willingness to consider different aspects of the business they are in. So, while it might be interesting to hear the views of such individuals, it is probably best to stop short of accepting personal success as a proxy for expertise.
Qualifier (Non-Authorities are people too): Not being an authority is of course no crime, not even when that non-authority presumes to speak on an important subject in public. The rest of us speak on all sorts of topics we don’t know that much about, so why shouldn’t television and movie stars? There is no reason to expect that a celebrity should be silent on political matters, religion, climate, or crime, but one would not want to mistake their celebrity status for expertise.
Problematic Authority: Sometimes a scholar is legitimately speaking within her area of expertise, but that does not necessarily mean that she is making specific claims about which her personal authority ought to be taken as sufficient reason to accept her position. For example, one would not expect a philosopher of religion to include “God exists” as a true/false question on a test. She may have spent her entire career discussing just that very issue, the existence of God, but few would regard it as reasonable to use her authority to present the answer to this question as a settled fact. Clearly, there are some claims about matters that may fall within an individual’s area of expertise without generating sufficient grounds for acceptance of their claims on that basis alone. In most such cases, public disagreement with other scholars in the field maybe taken as a strong counter-indication. When a given question is subject to considerable dispute within a field, it would be a mistake to take the authority of a given expert as sufficient evidence to accept it
General Note: Some folks are inclined to reject all authority-based arguments as fallacious. This is particularly common when a reader is sufficiently versed in a field to evaluate the reasoning of experts with some degree of confidence. Someone with a Master’s in history may for example be less inclined to simply accept an argument based on the conclusions of a professional historian. She may expect to hear the details of that historian’s argument so that she can decide for herself whether or not he has presented a convincing account of the topic. Someone with less knowledge of the topic may be unprepared to address those details, in which case an authority-based argument may be the best one can offer them. Hence, the utility of an argument from authority is at least partly a function of the prior-knowledge of its intended audience.
Litmus Test: In considering whether or not an argument may commit the fallacy of authority, first ask if the argument bases any conclusions on the personal credibility of an individual. If so, then ask whether or not that person actually possesses any authority on the issue at hand. It may also be useful to ask whether or not the claims cited in the argument relate to factually knowable matters and/or if there is significant disagreement within the field on the truth of those claims. If the answer to the first 3 questions is ‘yes’ and the last one is ‘no’, then it may be reasonable to accept authority as the basis for accepting a claim. It might also be reasonable to expect a more substantive account of the expert’s reasons for making the claim, depending on one’s own willingness (and competence) to handle the details of the argument.