"Belief and the Basis of Humor"
by Hugh LaFollette and Niall Shanks

American Philosophical Quarterly, 1993, 329-39
[pdf version]

Humor is a pervasive feature of human life which crosses racial, cultural, sexual, and class divisions. Yet its nature is elusive. This elusiveness should have piqued the philosophical imagination; its pervasiveness should have demonstrated its philosophical importance. However, it has generated relatively little theoretical interest.(1) We find that surprising. An analysis of humor could pay handsome dividends. Practically, it could inform aspects of the current debate over political correctness. Theoretically, it could illuminate discussion of significant issues in epistemology, the philosophy of mind, and ethics.

Preliminary remarks

When theorists have studied humor, they often assumed that laughter was either a necessary or a sufficient condition of humor.(2) It is neither. Although humorous events usually evoke laughter, they do not do so invariably. Humor may evoke smiles or smirks which fall short of laughter. Thus it is not a necessary condition. Nor is it a sufficient condition. People may laugh because they are uncomfortable (nervous laughter), they may laugh at someone (derisive laughter), they may laugh because they are insane or mentally imbalanced (hysterical laughter), or they may laugh because they are physiologically induced to do so (as when someone tickles them relentlessly).(3) Perhaps these other forms of laughter are philosophically interesting, but they are not forms of humor and so are beyond the reach of this essay.

Other theorists have offered typologies of humor. Typically these identify different motives for or psychological benefits of humor. For instance, some theorists (like Hobbes) emphasized the use of humor to demonstrate our superiority over others. Other theorists (like Freud) emphasized the psychological benefits of humor.(4) Although questions about the motives for humor are interesting, they, too, are beyond the scope of this essay. We offer not a study of laughter nor a categorization of the uses and benefits of humor, but a theoretical analysis of humor.

Our analysis does not specify necessary or sufficient conditions for humor. There probably aren't any. We will, however, identify central features of paradigmatic instances of humor -- features which, although perhaps absent from marginal cases, are vividly present in most cases, and certainly present in those which are of crucial interest to philosophers.

Of the earlier theories of humor, we think incongruity theories have been the most promising. Kant proposes an incongruity theory in The Critique of Judgement. Later theorists elaborated on Kant's account by emphasizing incongruous psychological states (Morreall) or incongruous semantic scripts (Raskin).(5) We will not directly challenge these theories. Rather we will offer an account which, although compatible with their core insight, demonstrates their limitations while explaining why incongruity can lead to humor -- and why it sometimes does not. Finally, our account takes humor from the sidelines of philosophy and places it on the playing field with ideas current in epistemology, the philosophy of mind, and ethics. But before we spell out the details of our theory, we must first describe some examples of humor and identify those features of humor which any adequate theory must explain.

A phenomenology of humor

The forms which humor take vary considerably. But even the casual observer will recognize a close connection between humor and language. Often humor springs directly from language, as in puns and jokes that depend on double entendre. For instance, someone asks: "What comes after advanced calculus?" Answer: "False teeth." Our familiarity with the phrase "advanced calculus" leads us to expect some remark about higher mathematics. What we get instead is a remark about oral prosthetics.

Some people may find this joke humorous.(6) Others will not. It is not humorous (or non-humorous) simpliciter. Humor is context-dependent. It depends, among other things, on the listener's beliefs. This joke is more likely to be humorous to a teacher or a student whose beliefs lead them to expect a remark about math.(7) Were we to ask the same question to a convention of dental hygienists, they might well interpret it as a genuine request for information -- not as a joke at all. Their operative beliefs would likely lead them to think about oral hygiene, not higher mathematics.

Other jokes hinge on semantic ambiguity. Consider the British newspaper headline, "Obscene Performance: Magistrates to Act." The reader may know what is intended, but may nevertheless be struck by the ambiguity. The humor arises when she oscillates between viewing magistrates as performing obscene acts and viewing them as halting such acts.(8)

Other jokes, although still clearly linguistic, are embodied in narrative form. In Gulliver's Travels Jonathan Swift portrays the Lilliputians as preparing for conflict to the death because their neighbors open their eggs from the "wrong end." Swift's narrative of the imaginary diatribe of political leaders transforms the reader's understanding of matters of state. We come to realize that, although countries have probably not gone to war for this particular reason, they have doubtless pursued war for equally silly reasons.(9) And that, of course, was Swift's intention. He used humor not to entertain, but to change the reader's political views. The ability of humor to change (or attempt to change) our beliefs deserves explanation.(10)

Of course not all humor appears in linguistic garb. One may find a cartoon or a silent film humorous. In Modern Times Chaplin provides a series of vivid images of a worker tightening bolts on an assembly line. However, even when -- as here -- the humor is primarily visual, some aspects of the humor may be relayed to others via language so that it still elicits guffaws. Even when that verbal redescription fails to capture the nuances carried by the visual images -- as it frequently does -- our ability to understand the humor still depends on our mastery of certain linguistic categories. Were we bereft of the linguistic categories necessary for understanding visual humor, then we could not find it humorous. Someone who knew nothing about machines, factories, and industrial workers would not understand Modern Times and hence, would likely not find it humorous -- at least not for the same reasons that we do.

The previous examples may suggest that humor is always manufactured for consumption; that humor is as humorists do. Not so. We may find experiences humorous as well. One of us has a son who in anger once kicked his older sister in the stomach. His mother chided him. She demanded to know why he kicked his sister. "I didn't mean to kick her in the stomach," he said midst his tears. "I was trying to kick her in the head but she moved." The exchange was humorous. However, our son (and our daughter) failed to understand the humor. This highlights the fact that people with inadequately developed cognitive (or linguistic) abilities cannot appreciate some humor. Prior to the acquisition of any beliefs, children cannot understand humor at all. Once they have acquired a (suitably rich) belief system, they are capable of humor, though the range of humor they can appreciate will be limited by the range and complexity of their beliefs.(11)

Additionally, some experiences may not be humorous when they occur although they may, in retrospect, be a source of great humor. One of the authors of this essay is 6'3" tall. He lived briefly in a house with a six foot entrance into the bathroom. Thus, it is not surprising that on more than one occasion his expectations of clear passage were thwarted. At the time it was not humorous.

Not then. But it is now. While the victim of midnight lobotomies this author was "too close" to see any humor in the event. The throbbing pain prevented any other perception. Now he can "see" the event differently. It vividly exemplifies his lack of coordination and his inability to navigate an ordinary doorway. That's why the event is humorous in retrospect but was not in prospect. He can maintain the appropriate psychic distance -- a matter we shall discuss in some detail later.

This particular case points to an extremely significant phenomenon associated with humor. Sometimes an event which under normal circumstances would be seen as humorous will not be so interpreted because of an intervening physiological or psychological process or state. The previous example illustrated how pain can thwart humor. Moods can do likewise. Even people with a "good sense of humor" will occasionally be immune to humor because they are in a bad mood. On other occasions our moods will alter the character of the humor to which we are susceptible.

More generally we need to explain why some people are characteristically immune to humor. A dullard may lack the intelligence to understand humor; the unimaginative person may be incapable of appreciating humor. A dour person may be disinclined to see humor. And an individual blindly committed to his or her views may be unable to comprehend or acknowledge anything humorous about situations or events related to the focus of commitment. Each of these demands explanation.

Finally, any adequate account of humor must also explain why we may appreciate some jokes only once yet appreciate others on numerous occasions. Extant theories leave this mysterious. For instance, some theories of humor claim surprise is a necessary condition of humor. Were that so then, like Heraclitus's river, we could never step into the same joke twice. Additionally, if, as some theories suggest, mere incongruity were the essence of humor, then, since all jokes depend on incongruity, they should be equally humorous during multiple retellings.

The theory

a) The abilities of creatures capable of humor

Now that we have identified crucial features of humor which any adequate theory must explain, we must offer our theory. This we will do in two stages. First, we shall delineate what we take to be necessary abilities of creatures capable of humor. Then, in the following section, we show how these abilities spawn humor.

Humor is possible only for agents whose belief systems manifest hierarchical cognitive richness. That belief is essential for humor should not be surprising. We recognize that the dullest normal human can see humor which even the most talented bullfrog would miss. The human not only has more beliefs than the bullfrog (if the bullfrog has any beliefs at all); the nature and complexity of those beliefs differ.

For humans even simple beliefs are best understood not as isolated entities but as part of a pattern or network of beliefs. As Donald Davidson puts it: "a belief is identified by its location in a pattern of beliefs; it is this pattern that determines the subject matter of the belief, what the belief is about."(12) Or, as he says elsewhere:

Beliefs are identified and described only within a dense patterns of beliefs. I can believe a cloud is passing before the sun, but only because I believe there is a sun, that clouds are made of water vapor, that water can exist in liquid or gaseous form; and so on without end. No particular list of further beliefs is required to give substance to my belief that a cloud is passing before the sun; but some appropriate set of related beliefs must be there.(13)

If a belief is constant over time, there will be some (possibly small) subset of relatively stable beliefs which persist. However, the patterns and networks of beliefs within which that belief (or subset of beliefs) "resides" may be in relative flux. A person may, for instance, believe that "Most politicians are crooked," but the exact character of that belief will vary from time to time, depending on the pattern of beliefs with which it is currently associated. At a specific time the character of the belief will be dependent on the other beliefs to which she is currently attending. What determines the particular pattern to which we attend? On some occasions it is merely the circumstances. For example, a parent grieving the recent death of a child is unlikely to have patterns of beliefs selected by members of a comedy club audience (although on some later occasion, she may have those very beliefs). At other times the environment will partly explain the focus of attention, even if it does not determine it. This is as it should be: Otherwise we could not adjust our behavior to fit the context -- and that would ill serve our survival.

For cognitively sophisticated creatures, however, in most case the belief pattern will not be uniquely fixed by the current stimulus. The patterns will also vary depending on other beliefs we have, in particular, what Quine calls our "higher-order beliefs"(14) Humans have beliefs about the world.(15) They believe that the earth is round, that the sun will rise tomorrow, and that clouds are made of water vapor. However, we also have beliefs about our first-order beliefs. These higher-order beliefs largely constitute the hierarchical cognitive richness we think essential for an adequate account of humor. Philosophers have always been especially interested in higher order beliefs concerning the reliability of our first order beliefs. Here, however, we wish to focus on other roles of higher-order beliefs -- roles more central to an understanding of humor.

Higher-order beliefs enable us to move beyond our immediate conditions to predict what will happen and what might happen. They permit us to predict not only what might happen in real life, but also in imaginary lives. Thus, our inductive beliefs not only empower us to predict what will happen if the car in which we are riding stalls on the freeway, they likewise empower us to predict what will happen in a story we hear about someone's car stalling on the freeway. Higher-order beliefs also determine what is relevant in particular circumstances. They empower us to identify patterns which might be relevant in these circumstances and to determine which of the alternative patterns is more likely.

Put differently, these higher-order beliefs structure our first-order beliefs bringing some of them to the focus attention while relegating others to the periphery. This power of higher-order beliefs to shift patterns of beliefs in and out of focus is essential for healthy functioning. The grieving parent may eventually move the memories of the child from the focus of her attention to its periphery and ultimately into the remote background. But that does not mean she did not care about her child. It merely means that life must go on and that her grief-related beliefs should ultimately shift into the background.(16)

Think of beliefs as points in an "epistemic space" which have complex arrays of connections with other points in that space. The various patterns of belief are interconnected collections of such points. These patterns provide us with a wide array of perspectives from which to view and interpret events of interest. Within that space there is no privileged "absolute" perspective, but a multiplicity of relative perspectives. Our second order beliefs structure, rank, and evaluate these patterns and perspectives in the various contexts in which we find ourselves.

This is the fertile ground from which humor springs. Humor is inherently relational -- no event, person or thing is intrinsically humorous. It is context dependent. It depends upon the circumstances, the teller (if there is one), the current beliefs of the listeners (or viewers), and the relationship (if any) between the teller and the listener. For the moment, though, we wish to focus on the beliefs of the listener. For unless listeners have the ability to view a subject matter from multiple perspectives, then they cannot experience humor. But this ability is not sufficient for humor. They must also have the appropriate psychic distance. What we mean by "psychic distance" is not some mysterious force, but merely the workings of our higher order beliefs to determine, in a given context, to which patterns of our first order beliefs we currently attend -- and which other patterns might be relevant in that context.

While standing at the appropriate psychic distance from an event, we have a perspective that we cannot have while standing "close" to it. From the distant perspective we can "see" contrasting belief sets unnoticeable if we are too close. However, it is not merely that we can see these different sets or even that we have the appropriate psychic distance. We must also be able to move the focus of our attention rapidly back and forth between some subset of these alternate patterns.

b) Putting it all together

This "flickering" in the focus of attention -- this active oscillating between these different but related belief sets -- is humor. Humor is not something passively witnessed. Like thinking, it is something in which the subject participates. Thus, to have a sense of humor on a given occasion is to be disposed to engage in the activity of flickering between different patterns of belief. Our second order beliefs determine which subjects (i.e., patterns of belief) are (or are not) candidates for humor (i.e., patterns between which we can flicker). Psychic distance provides a space within which to flicker.

Flickering, however, should not be confused with mere shifting. We are frequently aware of multiple perspectives we do not find humorous. Philosophy classes often discuss divergent views, yet are rarely ocassions for laughter. Nonetheless, philophers do find students are customarily amused by the logical paradoxes. That is, the students not only "see" the different perspectives, they have the appropriate second order beliefs so they can oscillate back and forth between them.

The tendency to confuse flickering with shifting is understandable. It doubtless stems from talk of "a sense of humor" which sounds suspiciously like "a sense of sight". Certainly humor does involve our ability to see alternative perspectives. But it has an dynamical component which transcends the mere recognition of perspectival multiplicity. Someone with a sense of humor has a disposition to a certain type of cognitive behavior -- the flickering -- which constitutes the humorous response to appropriate stimuli.

That humor consists in a flickering or oscillation between different but related belief sets is exemplified by a technique frequently used by playwrights and script writers for TV sitcoms. An event occurs, and then the event is described by different witnesses. Each redescription reflects the differing alternate perspectives -- and hence patterns of belief -- of the witnesses. The humor arises from the viewer's flickering between the various descriptions of the event. The viewer, however, does not merely passively consider each alternative pattern. Rather she rapidly and actively oscillates between them. This speedy and participatory flickering is the humor.

The classic TV sitcom, "All in the Family" provides a vivid illustration. Two african americans arrive at the Bunker household to repair a defective refrigerator. Various family members then recount their view of the visit. Archie Bunker, the father and all-american bigot, depicts the repairmen as dangerous, menacing "toughs" unconcerned about the quality of their work or the cost to the client. Michael, the son-in-law and a parody of the bleeding-heart white liberal, depicts the repairmen as two submissive blacks, eager to please. The humor comes in layers. Each new description of the event provides further perspectives between which the viewer can flicker.(17)

This flickering can likewise be seen as lying at the base of all the examples of humor we mentioned. The humor in "Obscene Performances: Magistrates to Act" emerges from our rapidly moving back and forth between the image of dignified British magistrates, draped in long black robes dispensing justice, and that of less-than-dignified magistrates wielding long black whips dispensing with justice. Or the child's proclamation: "I didn't mean to kick her in the stomach. I meant to kick her in the head but she moved" is humorous because we oscillate between seeing a child sorry for hurting his sister and a child sorry for missing his target. Without going through the hoary details, the reader can easily see this feature at the center of each case of humor we have described -- as well as others which the reader can personally recall.

c) How this analysis explains significant related phenomenon

The flickering essential for humor can only occur if the listener (or reader or viewer) believes there is some point to the alternate beliefs patterns. That is, they must think that those patterns contain or imply some insight or "truth" about the persons, things or events in question.(18) Consider, for example, the following joke. "What is the difference between men and government bonds?" Answer: "Bonds mature." This remark will be humorous only to those who, in a given context, think the implied description identifies an important/relevant feature of (at least many) men. Someone who was deeply and actively committed to the view that men are paragons of maturity -- and thus could not even contemplate the alternative patterns of beliefs -- would be perplexed and perhaps irritated by the joke.

Or suppose that we defined a `Quayle' as "a highly desirable opponent on the game show Jeopardy." Even someone who supported Dan Quayle -- someone who thought he was unjustifiably characterized as stupid -- would nonetheless realize the point and plausibility of the definition. If, however, someone offered the same definition of a `Carter,' the utterance would be perplexing. Even those who thought Carter was a rotten excuse for President, recognize that stupidity was not among his flaws. In short, if the sets of beliefs elicited by humor were demonstrably false -- not in the least plausible and relevant, even metaphorically -- then the statement would not be thought to be the least bit humorous.

This once again demonstrates our claim that humor is context dependent. In the previous cases humor failed because the listener did not see, understand, or appreciate its point. But there are other forces which may thwart humor, and any adequate theory should explain these. Why, for example, should one event be humorous at one time and not another? Why should some people fail to find an event humorous which others find hilarious?

Since humor is a flickering between different but related belief sets, then any state which prohibits us from seeing those sets or from flickering between them will block humor. Humor may be obstructed in any number of ways -- by non-cognitive states (e.g., pain), emotional states (e.g., moods) or cognitive states (e.g., firmly held convictions), including, but not restricted to religious or political beliefs. We shall discuss each of these in turn.

Pain prevents psychic distance and thus makes it impossible to flicker between alternate belief sets. As the earlier example noted, when the author's head was throbbing with pain, he was simply incapable of seeing any humor in the event. The pain narrowed his focus of attention. At that time there was no perspective from which alternative epistemic vistas could be contemplated, let alone flickered between. Later though, after the pain subsided, he found the event extremely humorous -- then he could intellectually engage a variety of belief patterns. Therein lies the humor.(19)

The ability of moods to dampen humor requires a slightly different explanation. To understand how moods alter our susceptibility to humor we must understand the cognitive and dispositional aspects of moods. Moods operate by disposing us to attend to some belief sets rather than others. Thus, the salience of beliefs may vary with time even though the content of the beliefs may be (fairly) constant. A person who knows she is dying from cancer knows this all of the time. Some of the time she attends to this belief. But especially if she is to avoid constant misery, at other times her attention will be focused elsewhere. Her moods will vary depending on which beliefs sets she currently attends. Her stock of beliefs may be more or less constant; her attention to particular patterns of belief will vary. Her second order beliefs can determine, at least some of the time, when she attends and when she does not.

Likewise for all moods. Moods alter our receptivity to humor by shifting various groups of first-order beliefs in and out of focus. When we are in a bad mood we focus on some feature(s) of our lives which is (are), from the current perspective, negative. While focusing on these negative features, we are less capable of even seeing alternate belief patterns. But even we see them, bad moods often stop us from flickering between them. It is not enough to merely "see" different belief patterns. To experience humor we must participate in the movement -- we must flicker -- between those patterns. Thus, bad moods may thwart humor directly or indirectly. They may stop us from seeing the point of the joke or they may dispose us not to flicker between alternative belief patterns. Or, even if we are capable of flickering, a bad mood (or a purely physiological condition like exhaustion) may dampen the intensity of flickering, that is, it may dampen the humor.

Moods can thus inhibit humor in the same ways that they can inhibit other cognitive responses. When we are in a bad mood we often find it difficult to read, to listen, to converse, or think. We may read a book, but fail to "engage" with the characters. We may carry on a conversation, but without really "being there." We may think about our problems, but our thinking is unfocused and imprecise. Likewise, we may see something which under normal circumstances we would find humorous, but, because of our mood, fail to participate in it.

Humor may also be limited by cognitive states. We all have higher order beliefs about which belief sets are appropriate in a given context. If we are friends of a grieving parent, then our higher order beliefs about what is appropriate in that context will move some belief sets outside our periphery of attention. Thus, these higher order beliefs will make it unlikely that we will see humor where we would normally see it.

Finally, some people may be so committed to a group of beliefs that they may be unable to achieve the requisite psychic distance, and thus, cannot flicker. For example, a person may be so blindly committed to a particular religious or political perspective that she can never get "far enough away" to see any humor related to it. She cannot see alternative patterns or flicker between them. She may even view attempts to get her to see alternate perspectives, let alone flicker between them, as irreverent at best, and blasphemous at worst.

The range of beliefs to which some individuals are committed may be so broad -- and the related patterns of belief so pervasive, that they are incapable of humor under most circumstances. That is often what happens to people who have "no sense of humor." They cannot achieve the requisite psychic distance.(20) These individuals have second order beliefs which dispose them on virtually all occasions to be wholly focussed on a given pattern of belief to the exclusion of others. Or, they have second order beliefs (e.g., that our beliefs should be treated with utmost seriousness) that make it unlikely that they will flicker between competing belief sets they do see. And they have these second order beliefs relatively permanently. In principle such a person might eventually achieve the requisite psychic distance. But for the present, at least, they cannot see or understand the point of the humor.

d) Further refinements

Our analysis also helps explain why some jokes get stale after first telling, while others persist through multiple retellings, and still others persevere seemingly forever. If the flickering central to humor is between perspectives that are not strongly divergent (as in obvious or blatant humor) or subtly divergent (as in smart humor), then the hearer might find the joke initially slightly humorous, but they are unlikely to even evince a weak grin upon its retelling. Similarly, if the subject matter of the humor is not particularly relevant to us, we may find it hard to motivate the flickering between different groups of beliefs about the persons, things or events in question. Imagine telling Dan Quayle jokes to an audience 100 years from now when the Bush administration has been completely forgotten by all except a few dedicated historians.(21)

On the other hand, if the movement induced is about some matter which is more interesting, more fascinating, and perhaps open to still more related belief patterns, then it is likely to be a joke which can be retold numerous times. Or if the range of alternate belief sets concern some matter which is fundamental, yet difficult to completely explain or categorize -- that is, if upon retelling it is likely to prompt continuous reflection upon the humor's subject matter -- as with the best of comedic literature -- then the humor will likely survive. That explains why Chaucer's The Miller's Tale and Chaplin's Modern Times continue to be regarded as classics: Their subject matters are relevant and they are compatible with multiple patterns of belief.

Our analysis also helps explain the use of humor as a tool for coping with painful experiences. When something untoward happens, we may be distraught or depressed. We may interpret life as cruel, the world as unfriendly or even meaningless. But with the passing of time we often come to laugh at the very things which once seemed so traumatic. What has transpired? This: through the imaginative flickering between alternative patterns of beliefs we reconceptualize the earlier experience so that the awful is transformed into the absurd; the nauseating into the nonsensical. What was earth-shattering at the time becomes, in the cool light of reflection less serious than we originally supposed. We can laugh not only at the original event, but also at our reaction to it. We begin to see that the world was not as bleak as we then supposed. Or, perhaps it is more accurate to say that we now recognize there are multiple ways of seeing the world and some of these alternate ways are not as bleak as the one we attended to at the time.

An event may have been genuinely traumatic -- and there is nothing wrong with the patterns of belief we had at that time. Nonetheless, there are other related patterns of beliefs -- other perspectives -- from which we can "see" our concern, fear, and apprehension we then experienced as only one among many patterns. We can also legitimately come to see the events as absurd, aesthetically odd, silly, or resulting from our ineptness. Having so "seen" these events, they now become humorous rather than horrible: we can flicker between the belief patterns which focus on the event's seriousness and patterns which focus on its triviality. Surely it is psychologically advantageous to be able to reinterpret terrifying events in humorous ways.(22)

Finally, our account explains the use of humor as a tool for influencing the personal, social, political, and ethical views of others. Humor is frequently used to influence opinions by suggesting that events or persons are less serious or important than supposed. Political satire especially evidences this deflationary role. When political humor gets us to flicker between a politician's preferred description and alternate non-standard descriptions, then politicians or political issues come to be perceived in such a way as to make them seem less important than we might standardly suppose. Comedic portrayals of the British royal family or American Presidents are intended to show that all is not as it appears. Such political figures are especially vulnerable because they are associated with various ranges of standard patterns of belief -- and there is an abundance of alternate descriptions to be considered and between which we can flicker.(23)

A practical application: political correctness

We would be remiss if we did not at least speculate about the relevance of our analysis for the current debate over political correctness. Humor is potentially a powerful political tool because, as noted above, it is capable of focusing our attention to particular descriptions of persons, things or events. Like any tool, it can bring destruction or build beautiful edifices, depending on who wields the tool, and for what purposes. The idea that there might be a need for "humor ethics" is by no means absurd, given the propensity for certain forms of humor to transform the ways in which we think about persons and the relations between them.

In the context of the political correctness debate, what is primarily of concern, is the ability of humor to both establish and reinforce racial and sexual stereotypes. Within some political and social contexts, where most individuals have similar belief sets, telling a particular joke can be a political act with moral implications. There is no such thing as a sense of humor simpliciter, such that feminists, for example, would complain less about "fraternity house" sexism if only they had it. What we find humorous depends on what other beliefs we have -- and hence on what alternate belief patters we can contemplate and between which we can flicker -- and these are not easily disentangled from the society and culture in which we ourselves are embedded, nor from the minority groups to which we may belong.

Certain types of racist humor have been widely used to reinforce racial, sexual, and national stereotypes. The initial tellings of the "jokes" develop stereotypic descriptions of the minorities, while the institution of the telling of such jokes sustains those stereotypes. The group belittled by the jokes will depend on the culture and its minorities, but the content of the jokes is often similar. Jokes told about Poles in the US are virtually identical to jokes told about Irish by the English. Jokes told about Jews in the US are similar to jokes told about Scots in England. Each serves to keep the minorities "in their place" and thus to perpetuate, with varying degrees of subtlety, various forms of oppression.

Consequently, to be a laissez faire liberal about humor and to assume all jokes are created equal is to adopt a non-neutral political stance. This stance can lead us to tolerate certain forms of humor by presupposing an equality which does not exist in our culture. Women and minorities understandably see such humor as perpetuating their inferior treatment and therefore think such humor should be rejected.(24)

Such humor will likely seen especially offensive when told by white males. You must recall that the teller is part of the humor's context. If the teller is a member of the oppressing group, the humor will more likely be seen as a form of oppression. However, the same joke told by a member of an oppressed minority to other members of that minority might well elicit a humorous response.

The preceding analysis helps explain what specifically is objectionable about racist or sexist "humor." On our view what is morally offensive is not the joke per se, but the underlying beliefs and attitudes which such jokes betoken.

Recall that a person's belief that an event, action, or claim is humorous depends upon the higher order beliefs which she has and to which she currently attends. A joke which belittled women, then, could only be humorous to someone who had the appropriate sort of higher-order beliefs, in particular, beliefs that women are mentally or morally inferior to men. Hence, what is disturbing about this humor is not the bare joke, but what that joke indicates about those who find the joke humorous. Since beliefs are in some important measure dispositions to behavior, then we can plausibly infer that those who find such jokes humorous will likely act in ways detrimental to the interests of women.(25)

Put differently, that people find such jokes humorous indicates the ways in which women are still looked down on in our culture. Thus, by tolerating such jokes we may indicate that these misogynist views are socially acceptable.

We recognize there are legitimate concerns about restricting speech -- even speech we loathe. We cannot here determine whether, on balance, we should tolerate sexist and racist humor on grounds of free speech, or suppress it because of its undesirable effects. Such questions are beyond the scope of this project. Nonetheless, our analysis does help us appreciate the ways in which humor reflects the broader social and political contexts of our lives and can maintain or transform the status quo. Humor is most assuredly philosophically interesting and its analysis important.

Conclusion

We have offered an analysis of humor which seems true to the phenomenon. Perhaps more importantly from our perspective, we have rescued humor from the philosophical hinterlands and placed it in the center of mainline philosophical concerns. If this effort has been successful, we expect that a still more careful study of humor might illuminate work in epistemology, philosophy of mind, and ethics. At least that is our aspiration.(26)



NOTES

1. The only recent philosophical work on humor to be published in a widely distributed "general" journal is G. Hatrz and R. Hunt, "Humor: The Beauty and the Beast" in American Philosophical Quarterly, vol. 28, nu. 4 (1991), pp. 299 - 309. Hartz and Hunt endorse a variation of incongruity theory. Although we find such theories headed in the right direction, for reasons explained below, we think they are inadequate.

2. See "Humor" in The Encyclopedia of Philosophy (Macmillan: New York, 1967), vol. 4, pp. 90ff.

3. Karl Pfeifer has made these points on numerous occasions in his attacks on John Morreall's view of laughter. See references in footnote 4.

4. 4. See Raskin's account of these typologies in his Semantic Analysis of Humor (Boston: Reidel, 1985).

5. John Morreall, in "A New Theory of Laughter," Philosophical Studies, vol 42 (1982), pp. 243-254, suggests that laughter "results from a pleasant psychological shift." This theory, although a theory of laughter, shares some kinship with our analysis of humor, in particular, his recognition that laughter involves some type of psychological shift. However, Morreall fails to identify or understand the nature of that shift; thus his account is subject of numerous counter-examples. See, for example, Karl Pfeifer's criticism of Morreall in "Laughing Matters," Dialogue, vol. 22 (1984), pp, 695-697, and in "More on Morreall on Laughter," Dialogue, vol. 26, (1987), pp. 115-118. Hartz and Hunt's analysis (ibid) is likewise subject to similar objections.

6. There is no better way to kill a good joke than to explain it. But, as we shall see, one benefit of our analysis is that it explains why this is so.

Of course you may not think this is a good joke: it does not need to be killed; it is already dead. Perhaps you think this joke humorous, but dislike other humor we use to illustrate our points. That is to be expected. Different people find different jokes humorous. As it turns out, one strength of our analysis is its ability to explain why this is so. It explains how we can see how others may find a joke humorous even if we do not.

7. For this reason, we expect that "formalist" accounts of humor will fail. The formalist believes that jokes, for example, have a shape or form -- and that all you have to do to generate a new joke, given an appropriate form, is to fill in values for the blank variables, without due care for subject-matter. It will emerge below why this is a poor approach to the analysis of humor.

8. Sometimes it is difficult to know just what is intended, as is shown by the British wartime headline, "British Push Bottles Up Germans". Here the ambiguity may not only not be accidental, it may serve the interests of propaganda.

9. Of course at the time of the conflict the participants did not think their reasons for war silly. Nonetheless, the dispassionate observer -- like the individual reading the novel -- is inevitably struck by the bizarre rationale for war.

10. For instance, repeated viewing of "Monty Python and The Holy Grail" has transformed forever one of the author's perspectives on movies about Arthurian legends. Now while viewing movies of this genre, he will laugh uproariously at seemingly serious events.

11. We will explain below what we mean by talk of suitably "rich" beliefs systems.

12. D. Davidson, Inquiries into Truth and Interpretation, (Oxford: Clarendon Press, 1986) p. 168.

13. Ibid, p. 200.

14. W.V. Quine, The Web of Belief (New York: Random House, 1978), p. 14.

15. We are thinking of beliefs as dispositions to various types of behavior -- as does Quine.

16. This ability to have and move between various patterns of beliefs -- this ability to take various perspectives -- is reminiscent of Thomas Nagel's view in A View from Nowhere (Oxford: Oxford University Press, 1986). Building on Quine's work, Nagel asks "Which vantage point is best?" for understanding the world. His response: "None." There is no privileged perspective from which to view the world. There are numerous views from different locations; there is no view from nowhere. In a given context some of these perspectives may be "normal" or usual, while others may be non-standard or "peculiar". None, however, completely reflects the world as it really is.

17. Of course, if your beliefs are such that sitcoms about bigots are totally and absolutely inappropriate, you may see the alternative perspectives yet fail to flicker -- for you the show is not humorous.

18. By "truth" here we do not mean external, everlasting and context-independent truth. Here we merely mean truth as compatibility with the person's other beliefs. Something is capable of being humorous only if the new redescription is at least somewhat compatible with the listener's beliefs -- even if those beliefs are false.

19. The ability to flicker between different groups of beliefs about ourselves and our actions is the key to self-deprecating humor. The realization that

there are descriptions of our actions other than the ones we would prefer -- descriptions in which we appear pompous or stupid, perhaps -- along with an ability to flicker between them, generates the humor.

20. A person who cannot laugh at her- or himself is particularly sad. Such a person fails to appreciate that there are other perspectives from which to see and understand themselves. Not appreciating the different perspectives, they cannot move between them. Thus, they cannot laugh at themselves or understand why anyone else would do so either. Here again the role of imagination -- our capacity to construct and appreciate different patterns of beliefs, plays an essential role in humor.

21. After the first world war, a very popular picture postcard was published which showed a man addressing a young woman sitting under a tree as follows: "I say Felicity, do you like Kipling?" She replies, "Oh you naughty boy, I've never kippled!" As Kipling fades from high school and university curricula, it becomes increasingly difficult to motivate this joke which evidently had (some of) our ancestors chortling volubly.

22. Why did so many of us find Stanley Kubrick's "Dr. Strangelove: Or, How I Learned to Stop Worrying and Love the Bomb," to be funny during the height of the Cold War? Presumably, at least in part, because it helped us cope with our anxiety over the prospects of nuclear annihilation.

23. Part of what makes pomposity the frequent object of humor is the pompous person's firm belief that there is only one proper way to think or act -- and he "knows" what it is. Redescribing this person brings out the absurdity and promotes laughter (perhaps silent when the object of humor is an authority figure). Likewise when the humor is self-deflationary -- when one laughs at oneself. Self-deflationary humor manifests our realization that there are multiple ways of interpreting our own thoughts and actions, ways which differ from our preferred or intended patterns of beliefs concerning ourselves.

24. It is important to distinguish racist or sexist humor from racial and sexual derisive laughter. If a sexist makes a derisive comment about "women drivers" and his friends cackle, this is not humor since it involves no flickering between alternate patterns of beliefs. it is merely a verbal assault on women or minorities.

Racial or sexual jokes may be humorous to some listeners if it involves the relevant flickering. However, it is humor designed to denigrate women or minorities. As such it is morally odious.

25. Although the joke may not be offensive per se, the telling of a joke in a particular context may be. If someone tells a sexist joke with the intent of putting the women who hear the joke "in their place" then the telling is morally objectionable even if the joke itself is not objectionable.

26. Our thanks to Jeffrey Gold, George Graham, John Hardwig, Karl Pfeifer, Russell MacIntyre, James Rachels, and an anonymous reviewer for the journal for helpful comments and suggestions on earlier drafts of this paper.