Jonah's thesis at Princeton is entitled "What makes jokes funny?"
Life is divided into the horrible and the miserable.
2.A FORK IN THE GARDEN PATH
The only time a woman really succeeds in changing a man is when he's a baby.
I phoned my Dad to tell him I had stopped smoking. He called me a quitter.
Jesus was a Jew, yes, but only on his mother's side.
5.LOGICALLY VALID BUT SILLY INFERENCE, OR A CONJUNCTION WITH NONSEQUITUR
Organized crime in America takes in over 40 billion dollars a year, and spends very little on office supplies.
6.VARIATIONS ON FAMOUS LAST WORDS
I'm not young enough to know everything.
Life is extinct on other planets because their scientists were more advanced than ours.
Art, like morality, consists of drawing the line somewhere.
This sentence no verb.
It has been my experience that people who have no vices, have very few virtues.
11.COULD YOU REPHRASE THAT
Why is there so much month left at the end of the money?
I am sitting in the smallest room in the house. Your review is in front of me. Soon it will be behind me.
13.RUNNING OFF AT THE MOUTH
Some luck lies in not getting what you thought you wanted but getting what you have, which once you have got it you may be smart enough to see is what you would have wanted had you known.
I don't consider myself bald. I'm simply taller than my hair.
We had a quicksand box in our backyard. I was an only child, eventually.
The production and comprehension of humor must depend upon knowledge of what is funny. This knowledge need not be explicit in the minds of the participants, however it must be represented at some level in order for individuals to recognize jokes and evaluate their funniness. The comic uses this judgment in the creation of the joke, to separate the wheat from the chaff, while the listener uses this judgment to respond to the joke. Of course, humor is a multifaceted phenomenon, and no one theory has managed to encapsulate and explain it entirely (although some have claimed to). However, it is worthwhile to review the major historical theories, since between them all most instances of humor can be understood.
Why is something funny? The various approaches to this question have focused on many different aspects of humor. Some theorize emphasize the semantic content of the joke, while others rely on its syntactical form. In practice, these two elements are usually combined to produce the humorous effect, and it is difficult to isolate these components. Many of these theories are complementary, as a psychoanalytic or social theory of humor may operate in tandem with a cognitive theory. A single joke may be humorous for a multitude of reasons, and it might be too much to expect that a single theory of humor might account for all of levels that a joke simultaneously engages.
Humorous History (or a the History of Theories of Humor)
The derision theory of humor extends beyond ancient times, and many attribute the modern formulation of this theory to Thomas Hobbes who wrote that "The passion of laughter is nothing else but sudden glory arising from the sudden conception of some eminency in ourselves, by comparison with the infirmity of others, or with our won formerly." Georg Hegel also expressed the similar view that laughter is little more than "an expression of self satisfied shrewdness."
Darwin observed that laughter is primarily an instinctual expression of joy and happiness, which may sound obvious to some, and leaves us with the question of what causes joy and happiness. This is where Sigmund Freud's theories become relevant, as Freud devoted tremendous energy to explaining the causes of mental pleasure and pain. In his work on humor Freud concluded:
According to Freud the pleasure derived from humor stems from two sources: the economy of effort in the joke, and the playful regression to old childhood pleasures. Freud also noted the parallel between dreams and jokes, which both seemed to employ the same techniques. These techniques include condensation, application of the same material, and double meanings. The transition from Darwin's theory to the various stages of Freud's theory illustrates an important principle in the study of humor. The theory must strive to explain humor as deeply as possible, without simply grouping all of the instances of humor under one heading. Darwin claimed that the source of all humor is pleasure, but then failed to give a satisfactory account of pleasure. Freud not only attempted such a definition, but also tried to show how the various mechanisms of humor satisfy this account. His theory provides a model for how to link the specific instances of humor with their underlying structure, although he certainly did not complete the task of bridging the gap between the theory and the data. The release theories do not necessarily contradict the superiority theories since some jokes may evoke pleasure through superiority, although most release theorists did not believe that this is the only way for something to be funny.
Based on this remaining fragment of the fifth chapter of Aristotle's Poetics, scholars have attributed the original "frustrated expectation" theory to Aristotle. The idea that humor is founded on a disappointment or unrealized expectation is more fully elaborated by later thinkers, however Aristotle's theory contains the seeds for many later explorations. He goes in a different direction than the derision theory and recognizes the type of humor based on "make believe" that has been put on or set up all in good fun. His theory of humor best applies to absurd, clown like situations where the audience's awareness of the fantasy allows them to find humor in typically morbid themes (such as death, murder, or violence). This brand of humor can be found in many situations ranging from a pie in the face to a poorly made horror movie. The significant feature of this theory is that it focuses on the joke itself rather than on the speaker or hearer of the joke. Later incarnations this theory broaden its scope to incongruity of any sort, but one can not help but wonder what might be contained in the missing sections of Aristotle's theory of humor.
Immanuel Kant developed Aristotle's absurdity theory and postulated that "Laughter is an affectation arising from a strained expectation being suddenly reduced to nothing." Schopenhauer essentially shared this view, and explicitly endorsed an incongruity based theory of humor: "The cause of laughter in every case is simply the sudden perception of the incongruity between the a concept and the real objects which have been thought through in some relation, and the laugh itself is just the expression of this incongruity." Researchers dispute whether incongruity is sufficient, or whether a joke requires incongruity immediately followed by resolution, and we will return to this point later. It is important to note that incongruity-resolutions theories mirror Koestler's "biassociation" theory which refers to "the creative thinking which links two self-consistent but mutually incompatible frames of reference or associative contexts at the same time." Koestler's method for the creation of jokes produces jokes which are funny according to the incongruity theory of humor. Raskin's formulation of the necessary and sufficient conditions for a text to be funny is founded on incongruity, and we will be exploring his bold theory in depth soon.
Most theorists believe that none of these three theories of humor are complete, but they are not necessarily incompatible with one another. For example, Frank MacHovec has recently proposed a theory of humor which attempts to synthesize the features of all of these theories. He speaks of humor's dependence on the convergence of three "cluster-like forces", polarity (direct feeling tone), power (force or valence of the feeling tone), and process (mechanics, details of technique; how polarity and power are focused). This theory demonstrates that most of the various theories can be successfully integrated with one another without excluding or favoring one theory. Where the theories conflict, it is possible that the two versions simply produce jokes of varying funniness. MacHovec has argued that the problem with the incongruity theory on its own is that it is too focused on micro mechanistic, nuts and bolts of the joke, and sometimes misses the larger context in which a joke is placed. Raskin, on the other hand claims to found a script based semantic theory of humor which provides all of the requirements for a joke. It will be important to keep these two viewpoints in mind as we explore the possibility of a complete, unifying theory of humor.
Before continuing to explain the details of these theories, it is worthwhile to point out where they all fall short. In an important sense, none these theories provide us with an answer to the question "why is X funny". Rather, they simply attempt to point out which things are funny without giving us an reason why. It is unclear what type of answer we are seeking to this question since the nature of humor is as mysterious and uniquely human as love. In fact, an answer to a question of this type could lie outside of the range of any theory of humor, as there may not be any reason why we find the things that we do funny. However, we should not allow the possibility of failure to stop us from speculating, after all, we have only suggested that such a theory might not exist, which is a far cry from proving that it doesn't. The following question will help guide this investigation and give it a form which can be grappled with more readily: Are the things that we find funny necessarily funny, or could we conceive of another culture/species that found, for example congruity rather than incongruity funny?
One of the most important strategies we will employ in answering this question is a comparison of humor to other well studied psychological phenomena. In the process of searching for things which are similar to humor, we will find that no perfect analogy exists. Of course, if two items were anaogous in every possible way, they would be identical (or at least isomorphic), so all of our analogies to humor must be disanalogous in some ways. In an analogy between humor and colors, nobody intends to suggest that funny things are colors, only that they are similar in some important ways. It is important to pay attention to the ways in which things are analogous as well as disanalogous to obtain a more complete conception of the concept under investigation. Humor does not fit perfectly into the catogory of an emotion, a peice of sense data, or a conditioned reflex, but it is like them all in certain important respects. Each of the analogies that will be drawn, serve to illucidate a particular set of humor's features through an analogy with something in which those features are salient. This technique is a powerful method of brainstorming, since everything is analogous to something in some respect, and it gives us a chance to compare humor with other strange experiences that life has to offer.
Some have argued that there is no separate cognitive step which evaluates whether or not a sentence is funny after its meaning is apprehended. There is no way to understand the sentence without "getting" the joke, just as there is no way to get the joke without understanding the sentence. As evidence for this assertion they point to the fact that spies have often been exposed by telling them a joke in their native tongue, resulting in a fit of uncontrollable laughter. This does not prove that humor is a reflex, but it does demonstrate a fact we all know well; sometimes it feels impossible to prevent oneself from laughing, and laughter often slips out before we have had a chance to fully process the joke consciously.
Unlike the evaluation of a scientific theory or even a work of art, humor is not the sort of thing that we evaluate and then decide what we think of it. It usually either strikes us as funny, or not to some degree or another. Furthermore, when somebody thinks that something is funny, he leaves little doubt about his opinion to those around him. It is hard to tell if somebody thinks that a painting is beautiful, but a comedians know instantly if a joke is floplaughter betrays the soul.
Becoming Emotionally Attached
Although emotions are primarily non-symbolic it is obvious that they still involve some inferential reasoning, even if much of it unconscious and not very complicated. The evaluation that a particular peice of sense data is threatening, which in turn elicits the emotional responce of fear, needs to be flexible enough to adjust to chaning environments where vaslty different types of things are dangerous. This evaluation also must be made very quickly, but the organism can not sacrifice the inferential power absent in a reflex. This is where emotions entered the evolutionary stage.
Emotions have the ability to effect perceptions, as emotionally charged words are harder to recognize than emotionally neutral ones, and as everybody knows, yoru mood effects that nature of your interpretation of reality. Emotional responces are also tempered by social forces, and this is born out in our everyday emotions such as disgust and shame. There are numerous examples of foods (inscets, pets, live animals) that are considered beyond repulsive in some parts of the World, while they are delicasies in another. Similiarly, different cultures place different emphases on what is veiwed as shameful (lying, nakedness, etc.) and what an appropriate responce to shame should be. (ranging from nothing to repentance or ritual suicide). These examples demonstrate that the roles of cognition and emotion are interwoven, and many systems exist which invlove the interaction between unconscious inferences and physiological responces, uniting the domians of reflex and consious decision making.
Returning to the comparison between humor an beauty, the question of what makes an object or an idea beautiful has puzzled philosophers for ages. While someone might propose a criterion such as symmetry or asymmetry, this type of answer is lacking in the same way that the theories of humor are. The follow up question, 'But what makes symmetry beautiful?'' leaves us with virtually the same predicament with which we started. Perhaps at this point in the line of questioning we must postulate a bottoming out foundation point and claim that the perception of humor is much like the perception of color. Red does not look red for any reason, it just looks red because that is how we experience it. The perception of color does not rely upon rules that tell us which sense data correspond to which colors. We discriminate between colors by being able to tell that a thing looks like the experience that we have learned to label 'red' or 'green'.
On the other hand, it is not all that difficult to imagine possible reasons for the evolution of humor. The expression of anger and hostility via superiority humor is obviously a vast improvement for society over actual physical violence, since although a biting remark might embarrass someone, very few people have suffered direct physical harm from humiliation. To this day, humor is used within social communities to maintain power relationships and dominance hierarchies. One need only ovbserve a social group interacting to watch this principle in aciton, as different people make the same types of cracks, which are responded to very diferently, simultaneously reflecting the relative social positions of both the joker and the respodants.
Superiority humor also provides an effective method of social control, as people have a hard time accepting and responding to direct criticism, but seem to handle the criticism far better if it is expressed as a joke or a parable. It is important to note that many discussion of political reform, subversion, and criticism takes the form of jokes, which thinly mask the very serious sentiments beneath their punchlines. This is especially true under repressive regimes, such as the former Soviet Union, where the joke served as a form of underground subterfuge, but is also true in many other contexts such as William Shakespeare's fool bearing the message of the truth to Henry IV.
The role of comedy in social reform and criticism is crucial even in an "unrepressed" context such as modern day America. Jay Leno's and David Letterman's nightly bashing of politicians and the media are poignant examples of the power of humor, and even seemingly innocuous cartoons like the Simpsons often carry with them messages that would be far more difficult to convey without the medium of humor to sugar coat their critical observations. The humor in these jokes also allows people who are unwilling or incapable of listening to the underlying message to dismiss it as nothing more than superficial entertainment. In some cases, the laughter response may have simultaneously developed as a defense mechanism for some to deny the seriousness of a potentially disturbing information, and to provided a deceptive disguise for unacceptable topics of communication.
As far as incongruous humor goes, it is clear that our minds, optimized to process rational, consistent data (or so we would like to believe) needed to evolve a mechanism for dealing with inconsistent, ungrammatical, or irrational input. As the Star Trek character Spock teaches us there are many ways that we might have dealt with such data. Upon hearing an incongruous or ambiguous sentence, Spock responds with a blank look of incomprehension, the humanoid equivalent a the "Syntax Error" messages that computers report when confronted with incongruity or ambiguity. Humans needed to develop some method for registering and dealing with the fact the not all input is sensible, and it would be awfully inconvenient if we "core dumped" every time we encountered something of this sort. Perhaps laughter is indication that we should stop trying to make rational sense of something, and helps us recognize the futility in trying.
Is Humor in the Belly of the Laugher?
In his book Color for Philosophers, Hardin relentlessly demonstrates that the psychology of color is far more complex than philosophers had ever previously imagined. He shows that the perception of color is not simply a function of the light that is reflected off the surface of an object, and in fact seems to involve the active imposition of a perception by the agent.Hardin, Color for Philosophers, p. 2. Hardin cites Naussau's work (Naussau 1983, 23) which distinguishes between at least fifteen different causes, grouped in five categories, for "standard" color perception. Colors are not simple qualities of an object that exist "out there" independent of our perception. Rather they arise out of a complicated interaction between the external world and out internal world which together produce the perception of color.
Whats in a Definition?
If "x is 'y' iff x is D" is a definition of 'y', then 'y' and D are synonyms (intersubstitutable).
Moore made this argument for [y = good], and concluded that aside from a few uninformative examples such as "x is 'good' iff x has intrinsic value." this question (Q) is always meaningful, so, there are no definitions of good. It is important to note that good is not the only concept that could be substituted for y, and that we can apply his argument to the case where [y = funny].
Moore asserts that apart from trivialities such as 'x is good iff x has intrinsic value' and 'x is good iff x ought to exist for its own sake' there are no definitions of x that make (Q) vacuous. One problem with this argument is that Moore provides no procedure for determining if there exists some non-trivial D such that (Q) is self-answering. Of course, Moore can insist that his argument shifts the burden of proof to the proponents of rigorous definitions for 'y', who must now produce such a D to defeat Moore's claim, however, it is still possible that such a D does exist, it's just that nobody has ever thought of it. In other words, the Open Question argument describes one of the properties of an atomic concept, namely that it is always meaningful to ask "Is x 'y'?" when y is atomic. However, Moore's argument does not provide a any procedure, aside from intuition, which determines whether or not such a y exists. In this respect the argument is fundamentally inconclusiveit can tell you with certainty that a concept is definable, but it will always leave a lingering doubt as to a concept's indefinability.
The applicability of this argument to the concept of humor is not as clear cut as Moore's argument against the definablitiy of good, since in the case of [y = good] Moore expresses a claim which seems correct, and with which many philosophers agree. However, it remains to be seen whether we can find a non-trivial D such that the question "Granted x is D, is x funny?" is always self answering. The existence of such a D would provide us with a definition of funny that would specify the precise conditions under which x is funny. As Moore's argument demonstrates, it is not at all obvious that such a definition needs to exist for all concepts. The question is whether 'funny' is the type of concept that is indefinable, or if it can actually be specified by some synonymous criterion.
If there were no D that provided us with a Moorian definition of humor, it would not be time to wave the white flag. Our inability to define a term in Moore's sense of definability should not threaten our conviction of the power, significance, and meaning of that term. We would hard pressed to formulate a Moorian definition of many concepts, such as identity, game, heap, or lamppost, and yet most of us have little difficulty using them meaningfully in a sentence. Moore's standards of definition may be too stringent for all but a few concepts, especially as a scientific theory strives to capture a working definition of a phenomena that may change over time. An alternative to Moore's stlye of definition is the prototype model of a concept advocated by Ludwig Wittgenstien. Under this schema, the determination of a new judgment is based upon the comparison between known instances. This procedure is a common cognitive technique that is employed in many different tasks and may play a central role in determining the funniness of particular situation. Something may be judged funny if it is similar in a crucial way to another previously known funny instances. Peter Strawson suggests this approach in his analysis of personal responsibility.
Strawson has argued that "there is no independent notion of responsibility that explains the property of the reactive attitudes. The explanatory priority is the other way around: It is not that we hold people responsible because they are responsible; rather, the idea (our idea) that we are responsible is to be understood by the practice, which itself is not a matter of holding some propositions to be true, but of expressing our concerns and demands about our treatment of on another." By defining responsibility in terms of people's reactive attitudes, Strawson sets us up an account of responsibility that resists rigorous analysis. Following Strasson's lead, we might find that we can make significant headway by definining funiness in terms of what people find funny.
It has been shown experimentally that some of human knowledge representation probably operates according to the prototype model. That is, when asked to identify whether an object B is a bird, we do not employ a propositional conditional that is satisfied just when B is a bird. Rather, we compare B to an prototypical bird (i.e. a canary) and determine the similarity of B to the prototype. This feature of human cognition is easy to detect through introspection in the case of moral evaluations. We often do not first know what our reactive attitude "should" be, and although an attitude is supposed to be an instinctive response, even our intuitions are at a loss in some cases. In these instances we tend to compare the case with other cases whose evaluations we are sure of by extracting the relevant features and determining if the uncertain case fits the mold (much like identifying object B). However, in order for this strategy to be effective we must have available clear cut, prototypical examples with which we can preform this comparison. Later we will examine this avenue of analysis in greater depth to see if it can be applied to the analysis of humor. Meanwhile, it should be kept in mind as an alternative to the necessary and sufficient condition approach that many theorists have attempted to construct.
The possibility of finding the necessary and sufficient conditions for x to be funny is tantalizing, and has led many researchers on relentless quests for these conditions. If such conditions could be adequately spelled out (where x is restricted to verbal humor), it ought to be possible to write a computer program which would be able to evaluate the humorous content of any given sentence. Conversely, the best way to test a theory of humor is by writing such a program, or at least specifying how one could be written. The possibility of designing a program that takes a sentence as input and returns a value corresponding to its funniness (in the simplest case it would return a message of either funny or not-funny) is intriguing, considering that many people consider humor to be the sole province of human intelligence. Furthermore, this thought experiment raises tricky issues concerning the role that understanding plays in "getting" a joke. Granted that some D exists that is a definition of humor, it is still an open question what form D will take. Could there ever be a rigorous definition of humor, for at least some classes of jokes? Or will D always incorporate semantic information and perhaps common world knowledge as well. Work carried out on the comprehension of stories and texts demonstrates that a great deal of background knowledge is required in order to make the necessary inferences implied by the story. Any computer program designed to evaluate jokes would have to have some method of representing such knowledge.
It also seems that any robust definition of humor would have to be powerful enough not only to evaluate humor but to generate it as well. A very crude heuristic would simply have the program generate random grammatical sentences, and then run the humor evaluator on those sentences discarding the ones that are not funny. This is probably not the process that human comedians employ to come up with jokes, however if we had a good enough evaluator function this method would prove effective.
A more challenging project would be to design a program that took as input an idea or a situation, and transformed that idea into a funny formulation. Intuitively, it seems as if some ideas could never be construed in a funny manner, such as a funeral or an operation. However it is hard to find an example of any situation that has not received humorous treatment at some point or another. For example, Tati's Les Vacances de M. Hulot (Mr. Hulot's Holiday) contains a hysterical funeral scence involving a deflated inner tube being mistaken for a wreath, and Levi's is presently running an ad where the patient on the operating table bursts into song to the beat of his own electrocardiogram. It remains an unresolved question whether any idea can be recast in a humorous fashion, but many jokes can be robbed of their humorous value if they are told slightly differently, so it is reasonable to ponder the reversal of this process and to expect a comprehensive theory to provide an account of this transformation. Consider the following:
We're Off See the Wizard.
There are currently many humorists circulating what they claim are comprehensive theories. Some are slight variations on preexisting theories while some are beefed up syntheses of them. Victor Raskin has make the bold assertion that he has formulated a set of necessary and sufficient conditions for a text to be funny, what we have been reffering to in our terms as D. He claims that his "semantic theory of humor is compatible with all three approaches [social-behavioral, psychoanalytical, cognitive-perceptual] simply because it does not concern itself with why something is funny, which is the focus of psychological theories of humor. It addresses a different question, namely, "How does a text convey humor?" He presents his theory as a full fledged scientific hypothesis, subject to Popper's falsifiability criterion which states that scientific sentences are falsifiable and unscientific sentences are not. The very application of Popper's critereon to a theory of humor has humorous merits on its own, but these are neutral with regards to confirmation of the theory. His main hypothesis consists of two conditions that are expressed in the theory laden language of linguistics:
Scripts were introduced during the early days of artificial intelligence (around twenty five years ago) to cope with the problem of a machine processing natural language. Roger Schank was the major proponent of scripts, which began as a model of knowledge representation and grew into a more general theory of memory and association. The original problem that Schank was confronted with was the occurrence of sentences and situations that were impossible to understand without some piece of background knowledge. For example, the exchange
The classic prototype of a script is the first script that Schank implemented, the restaurant script. This script stores information that is crucial for drawing inferences about restaurant situation such as the normal sequence of ordering, eating and paying. A story or a text about a restaurant scene might mention that someone entered the restaurant and assume that the reader understands the normal course of events and their proper interpretation implicit in that scenario. The question "Waiter, could you bring another round of drinks?" will be interpreted as an imperative request to bring another round of drinks, based upon the information that the restaurant script provides.
Schank identified a deep problem in language comprehension, and early responces to scripts were very receptive. However, in tim even Schank was forced to abandon the primary role of scripts in the face of mounting difficulties with the script paradigm. One major problem was concerned with the relative specificity of a script. Would a person have separate scripts specifying a visit to a podiatrist's office and the orthadondist's, or do they have a single script specifying a visit to a health care professional? "Such a script is beyond [Schanks] intitial conception of what a script was, becuase it was not specific enough. We had always believed that scripts were rooted in actual expereiences rather than in abstractions and generalizations from experiences."
Part of the trouble with the insistence upon scripts being rooted in actual experiences is that in the limit this strategy ends up postulating a separate script for every possible experience that a person might have, resulting in a highly inefficient middle man, whose explanatory role has been reduced to zero. If everybody must recall an individual script for every situation, they may as well remember the actual details of the situation itself, instead of constructing the virtually infinite number of scripts that such an organization would entail. At the same time, it does seem as if an inordinate number of scipts might be necessary, as there could be no single restaruant script, since for example, at fast food restuarants you pay and then eat, at Diners you pay at the register after you eat, and at fancy restaurants the waiter collects the check from you at the table.
These concerns eventually lead Schank to expand his original idea of scripts into a complex recursive hierarchy which he named MOPs (Memory Organizing Packets). Under this revised model Shank retained the used of scripts, but embedded them within a larger aabstract hierarchy. He maintained that "scripts are not data structures that are available in one peice in some part of memory. Rather, script application is a reconstructive process. We build peices of scripts as we need them from our store of knowledge to help us interperet what we hear."
The finer differences between scripts and MOPs does not concern us here, although it is significant to note that the original proponent of scripts acknowledged himself that the model had serious flaws and required a substantive makeover in order to survive. Raskin writing in 1985 may not have had the opportunity to study some of the later criticisms of scripts, and he does not directly address their limitations in his book. He justifies their usage based on the fact that no other system gets the linguistic job done as effectively as scripts do, and he points to their effectiveness in analyzing jokes as proof that they provide are feasable. However, we must be cautious in evaluating Raskin's evaluation of scripts and determine who is actually doing the work in Raskin's theoryscripts or Raskin himself.
Raskin heavily employs the notion of scripts in his theory of humor and mostly passes the buck to this system in his analysis of jokes. His conditions stipulate that for a text to be funny it must be compatible with two scripts that are opposed in a certain way. Having outlined a working theory of scripts we must now turn to Raskin's condition of opposition. Raskin postulates three basic types of opposition between scripts, corresponding to the "real" and the "unreal" situation expressed by the text: (1) "The first type clearly distinguishes between the actual situation in which the hero of the joke finds himself or, somewhat more generally, in which the joke is actually set, and a non-actual, non-existing situation which is not comppatible with the actual setting of the joke." (2) "The second type introduces the normal, expected state of affairs and opposes it to the abnormal, unexpected state of affairs." (3) "The third type distinguishes between a possible, plausible situation and a fully of partially impossible or much less plausible situation." Raskin enumerates a short list of the most frequently occuring oppositions, and devotes whole chapters to their standard instantiation in the particular categories of sexual, ethnic, and political humor. Raskin blurs his definition of opposition, by allowing for implicit second scripts in more complicated jokes. The best way to explore Raskin's theory is by applying it to an example, and evaluating its performance. For example, consider this simple "joke" which Raskin spends over ten pages analizing:
Raskin's model of comprehension involves the clarification of every word in the text and the selection of the corrrect meaning. In this joke the words "doctor" "patient" and "bronchial" evoke and reevoke the script of a patient visiting a physician (the DOCTOR script). The fact that the doctor's wife is "young" and "pretty" is irrelevant on this first reading of the joke, and only comes into play after the script switch is triggered by an incongruity. When the wife "whispers" her invitation, after it is clear that there is no reason for the patient to stay if he came to see the doctor (she did not say "He's on the way back." or "You can wait if you like.") the search for other compatible scripts immeadiately produces the alternative scripts LOVER (or ADULTERY). As soon as this alternative script is discovered, all the loose ends of the joke are tied up, and it suddenly is clear why the doctor's wife was whispring, and why it was important that she be young and pretty.
According to Raskin, "the joke is created by a partial overlapp of two opposed scripts, tentatively labled DOCTOR and LOVER. The opposition between the real and the unreal situations evoked by the text belongs to the actual/non-actual type. The non-actual situation exists externally as opposed to being conjectured by the hero(es) of the joke. There is a certain distance between the opposed scriptsthey are niether the negations of each other nor compatible conjunctions of one another. They joke contains a discontinuous contradiction trigger."
It is not obvious that all of this rigourous jargon enlightens our understanding of humor. Raskin's abmitious project is haunted by a number of serious shortcomings. His theory is built on the shaky foundations of script theory, and there may be some problems which will percolate up and threaten the integrity of his semantic theory of humor. Furthermore, as we will discuss below, the very possibility of formulating a rigourous definition of humor may be doomed from the start. However, before we concern ourselves with these braoder methodological issues, it is worthwhile to examine Raskin's theory on its own terms, and to test its robustness while temporarily granting Raskin the benifit of the doubt with regards to the broader issues just mentioned. In particular, before we attempt to determine if any conditions would suffice, we ought to give Raskin a fair chance, and confirm the limits of his hypthesis.
What we need to show in order to demonstrate that Raskin's definition of 'joke' is not satisfactory is either one (or both) of two things: (a) That it suffers from the sin of omission Some texts are jokes, but are not included by the conditions stipulated in Raskin's main hypothesis. (b) That it suffers from the sin of commissionSome texts are not jokes, but are included by Raskin.
Are there any texts that are jokes, but are not included by Raskin's conditions? This question is actually very difficult to answer as it deepnds upon the leaway that we grant to the notion of a script. One reading of Raskin's theory provides us with numerous examples of texts that are funny but not included by the conditions. For example, many jokes contain a single script (euphamisms , witty turns of phrase "The difference between literature and journalism is that journalism is unreadable and literature is not read." - O. Wilde
Both of these sayings contain one script (The script of COMPARISON) and they express no ingongruity or ambiguity. The humorous pleasure seems to be derived both from the sentiment expressed and the sounds of the words used to link the thoughts together. Wordplay is not always funny, so it would be difficult to build into the definition of humor, but its role should not be understated., and even some simple noises), or two scripts that are not in opposition to one another (puns or clever sayings which simultneously convey two parrallel, but unopposed meanings). However, scripts are a very slipery concept, and their vaugness may allow a theorist to emply them without them actually doing any real work for the theory. For example, in many of his analysis, Raskin seemed to be explaining each joke individually and on its own terms in a is far more specific than the mere invokation of scripts.
On one reading, Raskin's theory might be vacuous enough that someone dedicated to it could easily finagle an analysis in accordance with the theory. For example, the deadpan liar is often considered very funny in many settings but based solely on the information contained in the utterance there it is only comprised of a single script. If we were feeling generous, we might explain that there is an implicit second script which pits he the script of the real versus the unreal, the liar versus the truth. However, it is unclear how Raskin's conditions identify such implicit scripts, and he makes up for this deficientcy by fudging the analysis to fit his theory by a seemingly mechanical procedure.
The trouble with a vacuous theory is not that it is not very informative, and if it wraps around everything without really exlaining anything, its not much of a theory. This is one of the reasons that researchers have begun to demand machine based implamentations of theories; to expose the vaguries and abiguities inherent in many such systems. Raskin's theory certainly does evaluate many statements that we condsider funny as funny, but if we should not expect the theory in its current form to be complete enough that an automaton could learn a sence of humor. Considering that Schank only succeeded in implemnting about a dozen scripts, he was never truly able to test his hypothesis on a full corpus of real world data, and it is unclear whether his theory would have passed the test of the limitless possibilities existent outside of the world of toy modelsan actual novel is more than a simple extrapolation of restaruant, doctor, and lover scripts. Unless we grant scripts imaginary powers that they do not pocess, therby passing the buck to a nonexistent reciever, the dream (or perhaps nightmare?) of defining the set of what is funny sems to be in danger.
Raskin claims that his theory can distinguish a joke from a non-joke, but it can not predict how people will react to it. It seems strange that according to Raskin's scheme, a text might be considered a joke even if it is not funny. In a sense, Raskin's theory redefines the term "joke", and we are free to follow his lead as long as we are conscious of this move. Consider the fact that some identical statements are hysterical in some situations, and grounds for excommunication in others. This obeservation does not devastate to a a theroy of humor, since it merley suggests that the theory should bulild into its conditions a stipulation regarding the context of the joke. Humorists such as Neal Norrick have identified this condition as the stipulation that the listener must be in playful or humorous mode/frame for them to be receptive to potential jokes. Jerry Palmer has described the joke telling process as a negotiation, harkening back to our analogy between humor and colors. If humor is a complex negotiated interaction beteen the speaker (sometime the mute ink splotches on a page) and the listener then all what we hope for from a theory is a specification of the nature of the context that is required for a funny statement to be recieved as funny. If we take this approach, we are not forced to redefine humor in a way that is not correlated to what people consider funny.
Raskin is not a lone maverick for thinking that incongruity is intimatly connected to humor, and some have actually tried to quantify the relationship between the two, expressing the funnieness of a joke as a function of the incongruity it expresses. A number of studies have also been designed in an attemp empirically test to test the hypothesis that non-verbal humor is merely incongruity. G. Nerhardt cunducted a series of experiments in which the subjects were asked to lift wieghts and rate their heaviness. The weights were incremented every trial, establishing an expectation for their heaviness. On the last trial the wieght was either sharply reduced or increased, and this last incongruous wight evoked laughter (the probability of laughter increased "as a linear function of the discrepency form expectation").
An important fact which these studies and examples teach us is the futility of searching for the absolute necesary and suffficent conditions which define the concept funny. In his analysis of the concept of games in the Philosophical Investigations, Ludwig Wittgenstein noted that there is no fixed concept of a game itself. Different games (board games, card games, athletic games, for example) have no one thing in common, other than being human activities which are "played" by
In fact, if humor is indeed an emotion there is a preecedence for the fact that we may not be able to define it. The Polish linguist Anna Wierzbicka encountered this problem in tryring to define the basic emotion 'sadness'. She conceded that the structure of feelings prevents them from being expressed in words since basic feelings are signals which have no symbolic structure, and consequently do not correspond to any analytical concepts. Of couse we would like to be able to say more about a concept other than that it is ineffable and cannot be expressed in words. Following Wittgensteins lead, there meay be merit in examining the disjoint prototypical phenomena that comprise the concept of funny in the hopes that we may discover family resembalnces between various members of this set, which in turn may lend themselves more readily to an analysis.
There is good reason to think that humor is a disjoint phenomena of this type. Aside from the exceptionally large variety of things which different cultures find humorous individual's sense of humor vary from person to person, and even within the same person over his lifetime. People learn to appreciate new forms of humor, either through association with forms that they already consider funny, or eventually by actually creating a new category of things which they consider funny. There is a point in the development of children where they begin to appreciate sarcastic comments, (although some people never do find them funny) whereas previously they simply took them at face value. Not only do our sences of humor become increasingly refined over our lifetimes, but they develop as well, as we no longer (at least we don't admit ir) laugh at jokes we consider childish or primitive and we cultivete new modes of comedic expression.
Can you ever have a theory which "explains" the which w/out a why?
Developing sense of humor for new forms...
- Relate Raskin's model to the semantic taxonomy of one-liners that I have been working on (the ones in my last draft). Try to show how the specific frames satisfy the requirements. Could the various forms of humor be exhaustively generated?