Monkey See,
Human Do
by
Mongo Yalbag
One of the more benighted cogs in the SpecGram frozen ice cream machine of wit outdid himself a decade or so ago, writing, “The curious thing is that so few philosophers of language have taken proper notice of such texts. The usual analysis is that there’s word meanings, and then additional meaning is added or occurs (or whatever metaphysical sense you like) at higher levels of organization, syntactic or whatnot. Yet such passages indicate that higher levels of organization can actually subtract or cancel out the meanings encoded at lower levels, until in extreme cases the overall meaning actually attains negative values, causing you to possess less knowledge than before you read the text.” Fumbling the ball, however, he replied to a query, “I am not as yet sure what is meant by what I wrote, because until I write it up for submission to SpecGram, the exact meaning added in composition is indeterminate. It is conceivable that my theory will apply viciously to itself, in which case I should be given an extra academic degree or something.” In the last sentence at least he shows unwonted self-knowledge and tacitly admits that his academic degrees are quite extraneous.
It is easy, at least for non-linguists, to flesh out the basic idea into a formal model through the use of set theory: Reference may be defined by the sets of entities in the world referred to by an expression, and semantic composition may be modeled by operations on sets, with adjectives, for example, selecting the subset meeting a criterion corresponding to the quality named by the adjective, all easily formalizable and indeed often formalized in that way. The special feature of the view of meaning embodied in this schema may be captured by the changes in the measures of the referent subsets before and after composition. To ensure that an increase in meaning, which corresponds to a smaller subset (more specific reference, or “picking out” of the things of the world) after composition, is assigned a positive value, we may use the logarithm of the subset measure before composition minus the logarithm of the subset measure after composition. All this is easily done, but you won’t find the details here, as it would be (1) casting mother-of-pearl portraits of Euler, Gauss, and Riemann before baby swine with intellectual deficiencies to publish the results here, (2) certain to ensure the eternal neglect of valuable results by banishing them to the stygian darkness of the outer reaches of scholarly discourse were one to do so, and (3) likely to feed the unseemly pretension of the editors of this rag that they actually matter in the grand scholarly scheme of things. Instead, they are currently in press at a better journal. (As a small example of the power of this method, we may note that the meaning of this last statement is quite small: ln(97,463) − ln(97,461) = 2.052×10−5, which is, however, the same order of magnitude as we find when applying this method to measure the meaning of every editorial in SpecGram.)
However, there are some difficulties that arise with this simpler schema in the fact that set-theoretic methods alone do not suffice to exhaust the measurement of meaning, as the statements of the criteria applying in composition require a more subtle touch. Fortunately, the same louche lout who pointed out the approach above somehow took time out from, one assumes, boozing and scratching to indicate in broad outline how these difficulties may be addressed: “Hugo’s little essay in Les Misérables on French culture in 1819, the year during whose summer Tholomyès seduced Fantine, for example, is one of those things that inescapably shapes the reader’s view of that time, and it’s a judicious enough little essay, or tour of the highlights of his milieu as a young man, and his view of Louis-Philippe as so much better than his regime is right on the money too—Hugo and Daumier pretty much put their permanent imprint upon ‘La Poire’ (the pear), as he was called because of his William Howard Taftian physique. I have joked that when I read Hugo (and Stendhal), I planned to specialize in French history, but with Balzac that idea came to an end. (Misinterpreting ‘when’ and ‘with’ in any but the most strictly chronological sense there would be the cum hoc, ergo propter hoc fallacy, which might have inspired Hugo to write an essay on logical fallacies as breeding grounds for humor, and vice versa.)”
Unfortunately, or fortunately (given how badly his missteps would tarnish the end product), he did not follow up on this but instead ruined what remained of his faculties by reading too much (i.e., any) Gide, leaving it to me to fill in the gaps. This required a rich extension of the lambda calculus beyond my former set of such operations as vindalooing, tandooricization, chutnification, and pakorafication to include padthaiization, meekrobification, khaosaiization, and a vast cookbook of others. A subset of these operations, defined in terms of the Hamming distance of minimal moves away from the basis set of the valid syllogistic forms, allows us to apply the general schema for handling processes and logical operations upon hierarchically nested subsets and their unions and intersections to capture the exact changes in meaning that logical fallacies cause in jokes.
As a first test of this model, I used it to classify the logical fallacies underlying every joke in SpecGram (to the extent the jokes could be distinguished from the general obscurantism, stupidity, and vacuousness of the rag), and established that not only does SpecGram manage to use every logical fallacy to a humorous end at nearly the Shannon-entropic theoretical maximum, it is the breeding ground of several logical fallacies otherwise nowhere attested. And so, to my surprise, in the final analysis I must bid a fond adieu to this wretched rag, for by perverting logic and science in unprecedented ways into untold realms of negative meaning and mind-destroying logical fallacies for moderately amusing yucks and giggles, it has despite itself birthed (though never published) the greatest advance in semantics in history.
