Figure-Ground Reversal in Linguistic Humour
Tony Veale, School. of Computer Science and Informatics, University College Dublin

1. Introduction
Figure-ground reversal (FGR) is a phenomenon most commonly associated with visual perception, and indeed, few readers will be unfamiliar with the gestalt switches required to process such trick images as the Necker cube (see Fineman, 1996). Nonetheless, FGR is not a phenomenon that is restricted to image processing, and may generally apply to any aspect of cognitive structure where salience can be redistributed from primary to secondary features, that is, from those elements that are highlighted, marked or privileged to those that are not. We thus believe that FGR can play a significant explanatory role in theories of scientific discovery, Gedanken experiments, analogy, metaphor and humour, though prohibitions of space preclude us from considering all but the latter in this abstract. Here we choose to explore the role of FGR in verbal humour, arguing that much of the dynamism, inventiveness and deconstructive playfulness of humour arises from a general cognitive ability to apply FGR at different levels of linguistic and conceptual description.

2. Theories of Humour
The phenomenon of FGR plays a very specific and localized role – too localized, we argue – in the General Theory of Verbal Humour (GTVH) of Attardo and Raskin (see Attardo and Raskin, 1991; Attardo et al. 2002), a theory which currently dominates the field of humour research. The GTVH essentially offers a script-based incongruityresolution view of verbal humour, in which a narrative may be compatible with multiple scripts, one of which may be more contextually primed than others. Like its progenitor, Raskin's (1985) Semantic Script Theory of Humour (or SSTH) , the GTVH suggests that humour occurs when a listener is deceived into applying a highly primed script that ultimately leads to an incongruity (e.g., see Veale, 2004), an impasse of interpretation that can only be resolved by switching to an apparently less salient script. The GTVH views humorous resolution, whether partial or complete, as the work of a particular logical mechanism (LM) that applies at the level of scripts. Understandably, LMs have proved to be the most enigmatic elements of the GTVH, prompting Attardo et al. (2002) to enumerate a taxonomy of 27 different LMs.

sententially (by a pattern of words and lexical scripts). the GTVH notion of script is grounded in the work of Schank and Abelson (1977). More recently. such as narrative structure. logical mechanism. the logical mechanism is the most crucial. may be purchased at a heavy cost. In the SSTH of Raskin. and so on. the GTVH offers the possibility that incongruities may arise from a more nuanced spectrum of conceptual and pragmatic concerns. In contrast. causal ordering on a narrative and which reflect a single top-down interpretation of events based on an abstracted distillation of relevant episodic memories. For instance.g. word choice. As noted earlier. to which mathematical processes like sub-graph isomorphism can be applied. who view scripts as schematic structures that impose a sequential. that of the explanatory power of the GTVH as a theory. and so can more easily gain access to the various crevices of language where humour can take root. We note in passing that a Cognitive-Linguistic alternative to the script-switching account of GTVH is offered by Coulson's (2000) frame-shifting perspective. Attardo et al. as in the . This generalization. though it increases the descriptive power of scripts. it is suggested that an LM called falseanalogy is central to jokes whose humour derives from ill-judged comparisons. and inferentially. the GTVH additionally sees humour production as the culmination of a modular process involving a variety of complementary knowledge resources. (ibid) augment this view with a graph-theoretic account of script representation that views scripts as arbitrarily complex symbolic structures. since the Cognitive Linguistic notion of frame is more flexible than that of the GTVH notion of script.. as a by-product of common-sense reasoning (e. since certain elements in a script will be more salient and foregrounded than others. as when one intuits that a joke is racist and activates a Racism script). based on a fixed catalogue of antonymies like DEATH/LIFE. each determining a different element of a joke. called the lexical handle of the script). provided the notion of script is sufficiently generalized to accommodate phonetic as well as semantic information. we concentrate here on the particular crevices that seem inaccessible to the GTVH but readily accessible to FGR. Coulson’s perspective is arguably the more flexible but least developed of the two. and by far the most controversial. In a further evolution of the SSTH. GTVH scripts can be activated in one of three ways: lexically (by association with a single word. which situates the process of humour production within the more general framework of Conceptual Integration Networks of Fauconnier and Turner (1998). these elements are marked to distinguish them from less salient background elements.2 For the most part. the incongruities that trigger script switching are purely semantic. modular element of the GTVH. This representational shift allows the GTVH to encompass even punning as a script-level operation. Furthermore. Since our current focus is the GTVH.

Creative humour is surely resistant to analysis by scripts if. one cannot help but feel that the designation of FGR as a particular logical mechanism. cannot be said to exist in advance. it seems almost paradoxical to assume that a theory of creativity can be constructed on a relatively static substrate of script structures. For instance. Here a false analogy is created between the sun and a light-bulb. thereby possessing a granularity of application that transcends the level of a simple conceit or formulaic Ur-skeleton. by definition. A switch from the first script to the second thus results in a figure-ground reversal. reducing the phenomenon to a conceit that can be found in any number of related jokes that each instantiate the same template or Ur-joke (see Hofstadter and Gabora. This is particularly jarring if. 1989). Certainly. The latter. as a novel construct. understates the importance of FGR in humour. in the joke “How many X’s does it take to change a light bulb? 100 – one to hold the bulb and ninety-nine to spin the room around”. Consider the following witticism from the noted gold-digger. Attardo et al. Zsa Zsa Gabor: . suggesting that when the sun is not shining it is not "turned on". much less a script for stealing wheelbarrows from a factory by filling them with dirt and pushing them though the front-gate. the primed script ("twist the light-bulb") and the unprimed script ("twist the room") make complementary identifications of figure and ground. not hot. since it is highly unlikely that one would already possess a script for turning a room to screw in a bulb. FGR can operate at many different levels of conceptual description. truly novel events cannot be anticipated. such a mechanism does exist in the guise of figure-ground reversal. In these jokes. as we shall argue. The GTVH similarly views figure-ground reversal as a specific logical mechanism or LM that constructs a scenario in which two scripts offer inverted interpretations of the same narrative event. Figure-Ground Reversal as a Pervasive Mechanism of Humour Indeed. Nonetheless.3 old joke where a mad scientist builds a rocket to the sun but plans to embark at night to avoid being cremated. there is not so much a switch between two different scripts but a switch between a standard script and a creative variation of this script. (ibid) enumerate six different reversal-based LMs (of 27 in total). no matter how baroque the taxonomy of combination mechanisms. FGR is here seen by the GTVH as a specific logical mechanism that is literally realized within the world of the joke: the X's in the above actually perform a physical FGR by twisting the room (the ground) rather than the light-bulb (the figure). and reversal in general as a family of LMs. and the GTVH offers no mechanism for its construction. 3. one needs more flexible notion of script than the top-down formulation offered by Schank and Abelson. and serial divorcee. role exchange and vacuous reversal. However. such as chiasmus. and hence.

the one-liner “If God wanted us to be vegetarians he wouldn’t have made animals out of meat”. Meat is a profiled figure element of the base concept Animal. in language of cognitive grammar. Meat is a derived concept whose existence requires the existence of animals. since the comparison of scripts can only follow the dynamic creation of the new “script”. which are those that depend on a logically prior concept (or set of concepts) for their existence. (1998). Consider. How often do we use complex words like "seminar". FGR easily explains these morphological jokes. Though one can represent this re-conceptualized meaning as a graph structure. Furthermore. In contrast. another FGR is required to recombine these parts and return the modified whole to the foreground. actually I am an excellent housekeeper. the re-conceptualized meaning cannot be constructed by an LM. The effect of FGR on the definition of a derived concept like Meat . indexed by its lexical handle “housekeeper”. Indeed. linguistic humour stretches the notion of script to breaking point with its deconstructive ability to manipulate even the sub-structure of words. In the blending vocabulary of Fauconnier and Turner. Whenever I leave a man. be used to invert the definitional basis of derived concepts. This deconstructive ability of FGR can. a figureground reversal must take place. Consider humorous word formations like "manunkind". but where do these scripts originate? Clearly. Once alteration of a particular component is performed.4 “Darlink. since a complex word is a salient figure that is constructed from a backdrop of smaller lexical units. even if that LM is GTVH’s concept of figure-ground reversal. As an animal product. "disseminate" and "seminal" without attaching any prominence to the backgrounded words that give rise to them? Whenever a speaker decomposes a complex word into its component units. since decomposition gives greater salience to the parts than to the whole. as in "right" to "left" to highlight a more liberal attitude to copying. one can expect to find a housekeeping script in the lexicon. it is hardly a script of the GTVH mould. I keep the house!” The GTVH entreats us to view this joke as a juxtaposition of scripts. "copyleft" and "McJob". for the instance. but where does the juxtaposed meaning “a taker and keeper of houses” reside? The most obvious answer is that it is constructed dynamically by the hearer. which would require the GTVH to posit some form of "morphology script" before it could gain any sort of explanatory traction. or. more generally. using FGR to deconstruct the conventional meaning “housekeeper” into “a keeper of houses” and to reconstruct a new meaning on the fly. FGR here performs an "unpacking" of a complex word-blend to reconstruct (via the "web-principle") the input spaces that originally gave rise to it. or "kind" to "unkind" to highlight the essential cruelty of humanity.

see Fauconnier and Turner. the author succeeds in redefining Food in terms of Excrement without affecting the underlying causal order (i. an agent can explore a notion of meat that is not derived from animals. In effect. in which logical mechanisms are little more than the optional whisks and cutting blades that can be attached in different contexts to meet different production needs. strengthens the connection between Food and Excrement to uncomfortably suggest that when one is eating the former. is instead an intrinsic part of the appliance. or by Attardo and Raskin via the logical mechanism of false analogy). It is surely more compelling to believe that figure-ground reversal.. Veale and O’Donoghue. whose outcome is instead a form of metonymic loosening. This tightening is the opposite effect to that created by the Animal/Meat reversal. food is just excrement waiting to happen. 2000). But this extreme modularity is the basis for our gravest concerns about the GTVH.g. The FGR strategy employed here is the causal equivalent of topicalization. though we would pointedly disagree with such a diagnosis..e. The effect. In this vein." By playing with tense. to perhaps arrive at a new conceptualization of an idea previously taken for granted. as in the satirical remark “Eating is over-rated.. syntax and semantics. rather than serving a detachable role in humour. 1998. Remember. the joke suggests a new definition of Vegetarian as simply a person whose diet is not derived from animals but who may eat meat from other sources. the GTVH seems to liken the operation of humour to that of a kitchen appliance. As a research framework. 4. as might be argued by Minsky (1980). perhaps even the motor that drives the rotation of all other components. the use of FGR introduces some intellectual wiggle-room between two concepts that were previously so tightly-coupled as to be logically inseparable. Additionally. Freed from this yoke. that excrement is derived from food).5 is to make the profiled figure appear logically independent from its base concept. allowing the satirist to move an oblique concept from the causal background into a foreground position. morphology. much like research in linguistics is advanced by relatively independent inquiries into phonetics. one is simultaneously eating the latter. One might argue that the above joke is humorous only because it employs a line of faulty reasoning (e. the point is moot because one can invert the definition of a derived concept without doing violence to its underlying logic. a form of metonymic tightening (e. we have argued that FGR is a general tool for performing creative .g. Nonetheless. Concluding Remarks The modularity of the GTVH is intended to foster specialized research on different components of humour production independently..

Schank. R. it often leads to humour that is both thought-provoking and insightful. pp. (2002). Scripts. T. Humor: International Journal of Humor Research 2(4). References Attardo. Edinburgh. New York/Cambridge: Cambridge University Press. Scotland. Coulson. or any comparably static form of representation. . (2000). (1996). (1999). Fineman. M. New York: Dover. (2004). M. Jokes and the Logic of the Cognitive Unconscious. Fauconnier. through which humour can take apart and re-assemble objects at almost any level of linguistic description. 25. and Di Maio. When applied directly to the level of concepts. (1977). Semantic Leaps: Frame-shifting and Conceptual Blending in Meaning Construction. In the proc . memo no. A. However. C. Developing the Incongruity-Resolution Theory.417 –440 Minsky. Computation and Blending. Laboratory. changing the way we think about the world and throwing previously tacit connections between concepts into sharp relief. S. and O’Donoghue. C. (1985).118. M. Conceptual Integration Networks. Synopsis of the Workshop on Humor and Cognition.6 deconstruction.of the 1999 AISB Symposium on Creative Language: Stories and Humour. (1998). S. L. Attardo. Script theory revis(it)ed: joke similarity and joke representational model. Cognitive Linguistics. Humor: International Journal of Humor Research 4-3. Hofstadter. G.I. G. Raskin. a Festschrift for Victor Raskin.I. Veale. Hempelmann. and Raskin. and Abelson. T. Cognitive Science. special issue on Conceptual Blending. V. 22(2):133–187. S. R. this is a level of operation at which the GTVH notion of script. V. S. (1980). D. (1991). and Gabora. 11(3-4). Plans. D. Reidel. 293-347. Semantic Mechanisms of Humor. Dordrecht: D. P. 3-46. 603. F.I. and Turner. (2000). Humor: International Journal of Humor Research 15-1. Veale. New York: Wiley. A. seems entirely inappropriate or misplaced.T. Script oppositions and logical mechanisms: Modeling incongruities and their resolutions. Ritchie. Incongruity in Humor: Root-Cause or Epiphenomenon? The International Journal of Humor 17/4. The Nature of Visual Illusion. (1989). Goals and Understanding. M.

Sign up to vote on this title
UsefulNot useful