Is it not, then, better to be ridiculous and friendly than clever and hostile?
—Socrates (Plato, Phaedrus 260c)
Euphemistic genital deletion (Generative Semantics rule name)
—Paul Postal (cited in Borkin 1972:28)
All through the war, the Generative Semanticists were having fun—fun that spilled into and all over their work, which had everything to do with the times. Generative Semantics came on the scene as the narrow lapels and narrower ties, close-cropped hair, and corporate ethos of the 1950s, out of which Transformational Grammar emerged, were giving way to the headbands, Nehru jackets, bell-bottom pants, shaggy hair, and stoner ethos of the 1960s, when the cultivated alienation of the hippies had a pervasive influence. The Generative Semanticists—“a bunch of people who got together at conferences to make puns and play Fictionary and smoke funny cigarettes” (R. Lakoff 1989:972, 2006)—are as representative of that period as beads or sit-ins or bongs. Their work teems with themes of drugs, music, casual sex, and, of course, anti-establishment politics.
The stylistic proclivities of Generative Semantics, like those of the period generally, were not just a veneer of cultural references and allusions glued onto an otherwise conventional form. They run deep. The hippies were as seriously at odds with the preceding generation as any group could be, perhaps excepting the Khmer Rouge. The early term for their ethos was youth culture, which makes the generational dividing line clear. It was replaced by counterculture, which makes the scope of the rejection equally clear. The final, even more transient, label was Woodstock Nation, which was coined in a formal declaration of independence from, and state of war with, “the Pig Empire” (Hoffman 1971). Nothing captures the spirit of the period more fully than Charles A. Reich’s The Greening of America, a massively successful Jeremiad against the prevailing economic, political, military, educational, and social values gathered under the ominously Kafkaesque term, The Establishment. Here’s a sample:
From all of this, there emerges the great revelation about the executive suite—the place from which power-hungry men seem to rule our society. The truth is far worse. In the executive suite, there may be a Leger or Braque on the wall, or a collection of African masks, there may be a vast glass-and-metal desk, but there is no one there. No one at all is in the executive suite. What looks like a man is only a representation of a man who does what the organization requires. He (or it) does not run the machine; he tends it. (101)
At the major risk of trivialization and other distortions, at both ends of the analogy, it is impossible to miss the parallels with Generative Semantics, which ended, for many, in a total rejection of the transformational home that nurtured them and then turned them out; George Lakoff, the chief spokesman of this rejection, sounds like Jerry Rubin or Abbie Hoffman on the vacuity of the older generation when he decries Chomsky’s work as “the epitome of emptiness” (1977a:284).
Just as the hippies defined themselves in pained relief against The Establishment, a central tenet of Generative Semantics for the Lakoffs, for Ross, for many of the second-generation Generative Semanticists, was (quoting the male Lakoff) “the idea of not being like Chomsky.” The Generative Semanticists saw Chomsky as dishonest in his handling of data—reworking it to serve his temporary purposes, discarding what he couldn’t rework, ignoring vast regions altogether—and, in very sharp contrast, they jubilantly celebrated all kinds of data, not only data that gave the Interpretivists trouble, but data that gave their own theories the fits as well.
A similar pattern holds for theoretical machinery: where Chomsky’s style is to redefine and retain modifications of his flawed proposals, usually in terms which point to their new and improved abilities, not their former inadequacies, Generative Semanticists tended to renounce theoretical innovations very publicly, to proclaim their errors from the rooftops. And a related pattern, at least in the eyes of the Generative Semanticists, holds for the social extensions of the two models: they found Chomsky exclusionary, even cultish, in his disciplinary politics, defining other linguists in a concentric Dantesque vision. “There were the inner circle” of MIT loyalists, as Robin Lakoff characterizes it, “the various outer circles, Limbo, and Bad Guys” (1989:972).
But Generative Semanticists wanted everyone to join their party. Well, almost everyone. They, too, had a Limbo and a gang of Bad Guys; Chomsky, as the sheriff-gone-wrong, riding at the head of the gang. Generative Semanticists encouraged diversity, welcoming any and all forays into uncharted data, so their Limbo wasn’t populated with hopelessly misguided researchers, but with people working on things Generative Semanticists just hadn’t got round to yet. Where Chomsky works on a rather narrow, almost individual, set of linguistic problems—always using images of isolation to describe his own relation to the rest of the field—Generative Semanticists eagerly looked for problems in other areas of linguistics, in psychology, in philosophy, in logic, in literature, even in the medicine cabinet. The Oedipal reflex has rarely been carried to such extremes in science; because Chomsky was regarded with almost religious fervor when the schism began, he was promptly demonized. He was recast, recalling R. Lakoff’s characterization, “as satanic, the Enemy” (1989:970). Demons are negative exemplars, we know, so we should not miss that the rejection of (a certain picture of) Chomsky was self-consciously ethical: inclusion over exclusion, honesty over casuistry, congeniality over hostility.
The political elements of the Generative Semantics style, however, throw something of a monkey wrench into the simple anti-Chomskyan account of their ethos. The Generative Semanticists largely opposed the war in Vietnam, and Chomsky was one of its most outspoken critics. We will cross this bridge when we get to it, a few pages along.
It may seem digressive, even diversionary, to identify a Generative Semantics style at all, let alone to spend a chapter examining it. Chomsky would concur. He is fond of dismissing the movement as “an irrational cult” and “a fad,” just another symptom of the early 1970s collapse of values, and the kind of attention we’re giving it here can play into such dismissals. But facts are facts: all the major figures partook in an identifiable style, in very identifiable ways. Everyone who talks of the period both remembers the style and has an opinion about it, even Chomsky. For better or for worse (in fact, for both), a loose-jointed, absurdist, politically direct style suffused much of Generative Semantics.1
That style—to take the dry, reductionist, analytic approach that is death to any style, but at least gives us some postmortem results—had three principal manifestations: humor, politics, and data-worship. We begin the autopsy with humor.
Consideration of these examples makes it fairly clear that the fuck of (12a)–(20a) (henceforth fuck1) and the fuck of (2) (henceforth fuck2) are two distinct homophonous lexical items. These two lexical items have totally different selectional restrictions, as is shown by the examples:
(26) Fuck these irregular verbs.
(27) *John fucked these irregular verbs.
(28) Fuck communism.
(29) *John fucked communism.
—Quang Phuc Dong [aka James McCawley] (1992 [1968]:4)
The age of the hippie was not a subtle age. The defining style of the Generative Semanticists was not a subtle style. Their example sentences implicate all three of the most visible elements of the counterculture, elements which (like much of the spirit of the 1960s) rapidly degenerated in the following decade, in this case, into one of the most mindless and hedonistic slogans of all time: “Sex! Drugs! And rock ‘n’ roll!” (respectively illustrated by sentences 1a–c, 2a–c, and 3a–c):
There was also plenty of room for other prominent counterculture characteristics— politics, scatology, general absurdity:
Humor was a point of pride with many Generative Semanticists, in large part because it was a such a clear break from Chomsky, a declaration of the spiritual, in addition to the intellectual, schism with the Ur-Father. Chomsky’s prose has all the humor of Aristotle. (One of the late 1960s joke templates was title-of-the-world’s-shortest-book, like Theories of Racial Harmony, by George Wallace, the racist governor of Alabama, and Problems of the Obese, by Twiggy, the iconically underweight fashion model, and a prime candidate for this title circulating linguistics at the time was The Bawdy Humor of Noam Chomsky.) “We felt Chomsky took himself too seriously,” George Lakoff remembers. “We thought it was extremely important that people be able to laugh at themselves and what they were doing, as well as have fun.” It didn’t hurt that the youthful academic audiences at the cusp of the 1960s/1970s were well disposed to appreciate absurdity, boundary-pushing, and celebrations of their own cultural artifacts.
Was it childish? Well, when my daughter was six years old, she danced around the house delightedly singing out “Ass-kicking peanuts!” with particular stress on ass. She had discovered the use/mention distinction. She couldn’t get busted for saying ass because it was the brand-name of some peanuts she had just been given. She wasn’t using the word. She was quoting it. It’s hard to find any motive more sophisticated than that behind a sentence like:
Like other stylistic fondnesses of Generative Semantics, the role of humor built slowly, and has its roots in the first transformational forays into semantics. Katz includes the occasional stab at humor, of the typically early-1960s patriarchal variety (the example sentence, “Blondes are eyed by oculists,” 1964b:748 and the technical term, G-string, with Postal 1964:57). But it was in the Abstract Syntax period that these traits began to crop up with regularity, the first symptom being a penchant for flamboyantly named protagonists in example sentences, an appropriately contra-Chomskyan symptom. Chomsky’s sentences are notoriously dull (from any but a strictly linguistic perspective), a favored protagonist being the easy-to-please John (a lineal descendant surely of Bloomfield’s “Poor John ran away”).The Abstract Syntacticians, on the other hand, went for Max and Seymour and the ever-popular Floyd; Generative Semanticists, for Knucks McGonagle, Figmeister, and Norbert the Nark. Under the genius of Abstract Syntax—namely, Postal—the sentences also began to populate with gorillas, wombats, penguins, toads. With Generative Semantics, the situations got weirder, the cast of characters expanding to include cultural icons like Willie Mays and Yoko Ono. Patterns and themes recurred. One leitmotif had Richard Nixon loitering in men’s rooms.
Sample sentences were the thin edge of the weird-name wedge. Lakoff called one transformation Flip. Ross named one Slifling; then, following an alliterative bent, proposed Sluicing and Stuffing. Carden proposed Q-magic. These names at least had some reference to the actions performed by the rule (Slifting raised a Sentence node, an S-node, to a higher clause, so its etymology is S-lifting: Flip exchanged noun phrases to relate constructions such as “I like it” and “It pleases me;” Q-magic concerned quantifiers). Later in the game, as respect for the descriptive machinery of Transformational Grammar declined, the names became flagrantly arbitrary; in particular, there was a fad of adopting proper names for rules, like Irving, Ludwig, and Richard. As G. Lakoff explained it, “one way to [remember where you’re fudging] is to use obviously arbitrary names like clyde instead of arbitrary names that sound profound but aren’t, like Determiner” (1971a:iii; see also Ross’s justification for the name Do-Gobbling—1973a:70). Other names went in the opposite extreme, becoming absurdly specific, like Grinder’s Apparel Pronoun Deletion (for the syntactic behavior of certain sentences concerning disrobing) and—the hands-down winner in this category—Postal’s Euphemistic Genital Deletion (for sentences where certain graphic nouns are demurely spirited away, as in Max is playing with himself again).4
Hippies were also eager to outrage older generations, so we should not overlook shock value in data sentences like these. For the routinely tattooed, pierced, and profane citizens of our current century, when the word fuck can be reeled off on television in all its intricate lexical variations with Shakespearean virtuosity, or just boringly iterated with relentless mundanity, a sentence like “Let’s fuck” might seem positively quaint. But to an established linguist, opening a book on politeness in the 1970s and finding those words staring out at him in an article (by a woman!), it would have been as if parasitologist had come into his office, defecated on the desk in front of him, and poked around in his production to point out all the interesting worms.
For many Generative Semanticists, this style leaked out of the example sentences and rule names to permeate the prose. Anecdotes, inside allusions, and schoolyard expressions like “one swell foop” showed up regularly, most notably in the second-generation papers. Jerry Morgan, for instance, offers this account of some missing arguments in his article:
[My earlier paper] contained the ultimate solution to the problems of pronominalization, reference, identity, as well as an item of overwhelming and irrefutable empirical evidence against the lexicalist position. Unfortunately, it was handwritten on a package of Puritan Hog Chow, and was eaten by a hungry Chicago policeman who tore it from me during a tear-gas attack on three jaywalkers, thereby being lost to mankind. (1976 [1968 or 1969]:340n1)
But, lest we get carried away, we need to keep in mind that most of the humor, while far from subtle, was not exactly center stage. There are few belly laughs in Generative Semantic papers. Aside from some of the wilder underground papers (many collected in Studies out in Left Field: Defamatory Essays Presented to James D. McCawley—Zwicky et al. 1992 [1971]), the papers don’t exist to tell jokes; serious work goes on in and around the humor.5 Even the Quang papers have significant linguistic points. The jokes participate in the papers—sometimes propelling them along, sometimes offering a commentary on the author’s confidence in the analysis—but mostly just contributing a general tone of goofiness to the work, in deliberate counterpoint to Chomsky. To use a slightly effete word for the practice, it has more whimsy about it than humor. It depends on odd situations, insider references to other linguists or to cultural figures, and a general sense of kookiness, not on punch lines or on farcically sustained imagery. There is also a very clear gradation of the proportion of whimsy as a function of audience. The rabid-hog quotation above, for instance, comes from an underground paper circulated amongst core Generative Semanticists, of a piece with “Camelot, 1968.” To stay with the same author, Morgan’s CLS contributions are still loose and eccentric, with subheads like “Pickles and Strawberries” and the extended participation of Ernie Banks, but they are markedly less informal. His more mainstream publications, as in Language (1977), are positively tame (though certainly not Chomskyan).
The presence of second-generation Generative Semanticists here raises another factor in the levity of Generative Semantics papers, a factor Hagège contemptuously dismisses as the “whiff of amateurism” (1981 [1976]:21): many of the more outrageous Generative Semanticist offerings did in fact come from amateurs; that is, grad students. The chief Generative Semantics schools (Chicago, Champaign-Urbana, and Michigan) all followed the practice that Chomsky and Halle had begun in MIT, of urging students out into the fray as soon as possible. But where inexperience and exuberance tended to manifest among MIT students as discourteous polemics, in Generative Semanticists the trend was toward loopy humor, political asides, and declarations of awe at the glorious complexity of the data.
And of course style is a personal trait, like speech patterns; there were several very distinctive people at the helm of Generative Semantics. Newmeyer’s term rambunctious captures much of their collective spirit (1986a:137), but there was individual spirit at play as well, and not all the humor had the same motivations. With George Lakoff, humor may well have been in some part the “concealed intellectual aggressiveness” that Arnold Zwicky recalls of the movement generally. Lakoff is an aggressive guy, given to some rather grating rhetorical tactics, and it would be difficult to deny that this aggressiveness had vent in his publications, one of which, I’m sure you recall, ends with “Nyaah, nyaah!” (1973b:290). His example sentences also took occasional pot shots at the other side, such as juxtaposing lexicalists and whores in parallel sentences (1971c:333), and offering up this gem of agonism, “Chomsky is the de Gaulle of linguistics” (de Gaulle being the authoritarian president of France for the entire decade of the 1960s, notorious for his strong-arm response to student protests; 1973d [1972]:235).
For McCawley, who has a very widespread reputation for gentleness, even when directly challenging someone’s argument, the humor was probably more celebratory. Ross falls somewhere in the middle; Robin Lakoff is a little closer to McCawley; and Postal’s humor, a juxtaposition of deadpan scholarly prose and example sentences teeming with wombats and gorillas and anaphorically orphaned blondes, is more difficult to pin down. Many of the second-generation Generative Semanticists also had their own cluster of stylistic character traits—some leaning more toward politics, others toward absurdity—each one indicating a slightly different configuration of motives, and a different hierarchy of role models. But there is one defining factor that shines through all of these tonal variations—aggressive, gentle, itinerantly goofy—that linguistics was great fun.
This History is humbly dedicated to those Valiant young Knights who, in quest of the Holy Grail, shed their Blood under the onslaught of Savage wild boars in the Forests of Lincoln and Grant, and in the Stone Valley of Michigan, in the Duchy of Czechago, Summer, 1968. Richard the Leatherbuttocked, Lord Mayor.
—Sir Lancelot of Benwick Morgan le Fay and the Green Knight (1976 [1968]:249)
We come now to the point at which a simple not-being-Chomsky account of the Generative Semantic ethos founders somewhat. Most of the Generative Semanticists and most of the Interpretive Semanticists opposed the war in Indochina, and Chomsky was a leader in what was called the Anti-Vietnam movement; among other things, it was this opposition that put him on Firing Line, sitting next to William F. Buckley.
One explanation for this situation might be a mild twist in the Oedipal story that puts the political explicitness of many Generative Semanticists into the trying-to-out-Chomsky-Chomsky file, much along the plotline that many people assumed in the Generative Semantics origin story—emerging from Abstract Syntax as Lakoff and Ross went deeper into Transformational Grammar than its leader dared to go, having more courage in their convictions than he did. It’s also a trajectory many people associate with Lakoff’s career especially, as he moved into the Chomskyan influenced domains of psychology, philosophy, mathematics, and politics.
But this line won’t take us very far. It is, first, extremely difficult to out-Chomsky Chomsky on political issues. He does not use sentences like “Shit on Lyndon Johnson” or “In a real sense, Nixon is a murderer” to illustrate linguistic principles, as the Generative Semanticists did (from Quang 1992a [1968] and G. Lakoff 1972a:210). But he says things in long, detailed, closely reasoned political articles like “President Nixon has described Cambodia as ‘The Nixon doctrine in its purest form.’ . . . For once, he may have spoken the exact truth” (1970b:192), after describing that doctrine as the use of
cluster bombs and napalm and bribes for the rich . . . to destroy the political and social fabric of the rural society . . . to convert [American] imperial aggression . . . into regional conflict . . . [through a program] carried out in secrecy, with mercenary troops, CIA subversion, and massive bombardment (1970b:187)
Throughout the period, Chomsky went on record, repeatedly and forcefully, denouncing Johnson, Nixon, Kissinger, and their ilk in the most scathing and closely documented terms, relentlessly framing their actions as war crimes on a level with the atrocities of the Third Reich and Stalinist Russia—sentiments more specific, more caustic, and far more thoroughly prosecuted than anything coming from almost any other intellectual one might name, let alone any other linguist. He was (and remains) a more tireless, dedicated political rhetor than any of the Generative Semanticists, even given George Lakoff’s celebrated later forays (1996, 2004, 2008); in fact, it might not be much of an exaggeration to say he was more tireless and dedicated to political concerns in the period than all the Generative Semanticists combined. Even the very comparison of the respective political efforts of Chomsky and the Generative Semanticists is otiose, if not spurious, like saying Milton was a better poet than most Generative Semanticists, Miles Davis a better musician than most Generative Semanticists, Gretzky a better hockey player than most Generative Semanticists. The difference in scale is orders of magnitude. The Generative Semanticists were more explicit than Chomsky, in linguistic articles, about their political concerns, and certainly more irreverent. But they were not, by even the wildest stretch of the term, more political (the way they might easily be called, for instance, “more semantic” or “more empirical”). Nor, I should think, would any of them dispute this observation.
It is, second, not coincidental that the shared political outlook of Chomsky and most Generative Semanticists (and, indeed, most Interpretive Semanticists) was shared dissent. They were largely united in opposition to the war, and to the political figures behind it. On positive issues (like what should be done to correct the mistakes of American foreign policy, who should do it, how it should be carried out), there would have been far less agreement, maybe none at all.
And, third, the war was unspeakably obscene. It was fortunately impossible for many people of conscience to remain silent in the teeth of both America’s vicious program of terror in Indochina and the parade of ugly, blatant lies coming from the government about that program. I’ll let a more recent author visiting that region in the aftermath of the American invasion lay it out for us:
Once you’ve been to Cambodia, you’ll never stop wanting to beat Henry Kissinger to death with your bare hands. You will never again be able to open a newspaper and read about that treacherous, prevaricating, murderous scumbag sitting down for a nice chat with Charlie Rose or attending some black-tie affair for a new glossy magazine without choking. Witness what [he] did in Cambodia—the fruits of his genius for statesmanship—and you will never understand why he’s not sitting in the dock at The Hague next to Milošević. (Bourdain 2010:162)
That’s Anthony Bourdain, in his Cook’s Tour, doing precisely what distinguished the Generative Semanticists from Chomsky on this count. Bourdain brought his outrage into his day job, writing a culinary travelogue. The Generative Semanticists brought their outrage into their linguistics. Chomsky’s political outrage shows up, with very rare exceptions, only and directly in his dissident activities.
But, of course, the temper of the generation that infused Generative Semantics made for a very different style in the way their political outrage entered their day jobs. Humor and absurdity were principal devices of counterculture rhetoric.
There were certainly earnest dissidents at the time—Dave Dellinger, Cesar Chavez, Fred Hampton, Noam Chomsky—but, for better or worse (actually, for both), the zeitgeist was channelled more fully by less focused, more boisterous people, like Ken Kesey, Jerry Rubin, and Abbie Hoffman. The dominant social spirit was the loose mélange of anger, liberation, and confusion that I have collected under the umbrella hippie, and, even in my broad usage, that umbrella does not cover Dellinger or Chavez or Hampton or Chomsky. The counterculture is Kesey seducing the young and appalling the old with his portable acid tests, Rubin and Hoffman leading phalanxes of outrageously clad longhairs on a mission to levitate the Pentagon. It is the Women’s International Terrorist Conspiracy from Hell (WITCH) casting hexes on the administration during sit-ins at Chicago; the slogan of young Parisian radicals, Je suis marxiste, tendance Groucho; a young linguist in a tie-dyed T-shirt lecturing at Yale about the morphophonological ramifications of words like Kalamafuckingzoo. Much of the energy was political, but much of it was also just the sheer, balmy celebration of life.
Politics and absurdity, in any case, are not very easy to disentangle at the turn of the decade, the 1960s sliding into the 1970s. The political situation was a marriage of Lewis Carroll and Franz Kafka. Martin Luther King Jr., two Kennedys, and Malcolm X had all been shot. George Wallace, who was about to be, stumped around the country campaigning for the presidency by denouncing “left-wing theoreticians, briefcase-totin’ bureaucrats, ivory-tower guideline writers, bearded anarchists, smart-aleck editorial writers and pointy-headed professors” (O’Neill 1971:389), while his running mate advocated a nuclear solution in Vietnam. Little bags of hair from the Chicago Seven, shorn after their arrest, were auctioned off at Republican fundraisers. The U.S. government’s response to increased protests over the Vietnam War was to attack Laos and Cambodia. Governor Ronald Reagan ordered over two thousand National Guardsmen to “quell” a demonstration against the University of California–Berkeley’s expansion into what became People’s Park, resulting in one death, 128 wounded. The Weathermen staged absurd occupations of working-class schools to wake up the proletariat, usually being driven off by hippie-hating toughs. Enraged at the violence, repression, and deafness of the power structure, or possibly just bored, various collectives and individuals bombed over three thousand public structures in 1970 alone, and made over fifty thousand bomb threats. Protesters were killed at Kent State, Jackson State, and Berkeley; maced, clubbed, and arrested all over the United States. Chicago police mugged proudly for photographers after murdering Fred Hampton. Other black dissidents, and some whites, were jailed on weak drug charges or driven into exile. The reduction of American casualties by turning over much of the ground-fighting to the South Vietnamese in order to concentrate more fully on bombing, defoliation, and napalm attacks, actually reduced public outcry. The temper of President Nixon, or, more accurately, his distemper, paled even Garson’s notorious off-Broadway satire of his predecessor, Lyndon Johnson, in MacBird!:
crony: Peace paraders marching.
macbird: Stop ‘em!
crony: Beatniks burning draft cards.
macbird: Jail ‘em!
crony: Negroes starting sit-ins.
macbird: Gas ‘em!
crony: Latin rebels rising.
macbird: Shoot ‘em!
crony: Asian peasants are arming.
macbird: Bomb ‘em!
crony: Congressmen complaining.
macbird: Fuck ‘em!
Flush out this filthy scum; destroy dissent.
It’s treason to defy your president. (Garson 1966:73–74)
Linguists were among the concerned protesters. A very beat looking Haj Ross, for instance, can be seen in the foreground of a picture in Time accompanying the story of how a protest over Dow Chemical’s napalm production resulted in Dow’s lab director being barricaded in a Harvard conference room for seven hours (November 3, 1967:57). Jim McCawley was a vocal participant in the 1966 University of Chicago protests against allowing the Selective Service access to class rankings; at a particularly vulnerable time in his career, he snuck into a meeting for tenured-only faculty on the question, took notes, and reported back to a student meeting. Jerry Morgan was very active in the Chicago convention protests. One of the more important collections of linguistic papers from the period is dedicated “To the Children of Vietnam” (Jacobs & Rosenbaum 1970). And, of course, there was Chomsky.
Much of the political activity of linguists may have been indirectly related to the tough, impassioned stand Chomsky took against American imperialism in Indochina. Certainly, a partial motivation for many students who joined the MIT linguistics program at the time (and long afterwards) was a political affinity with Chomsky, just as Chomsky’s principal motivation for studying at the University of Pennsylvania was his political affinity with Zellig Harris. More generally, a partial motivation for many linguists going into the field, and particularly into Transformational Grammar, was Chomsky’s political reputation. He lectured on both at MIT, and many students took both classes. Although he kept his politics and linguistics fairly distinct, his political feelings were very widely known, and widely applauded, in the linguistics community. McCawley ends one of his long letters to Chomsky about the respectively controversy with the postscript, “There is no truth to the nasty rumor going around that the CIA is subsidizing my research in hopes of thereby diverting your energies from the war” (January 18, 1968).
The Generative Semanticists, while no more deeply concerned about the social and political pathologies of the day than the Interpretive Semanticists, were far more overt about those concerns in their linguistics—no doubt a function of the influence of leading figures in each camp. There is some minor leakage here and there, of Chomsky’s politics into his linguistics—the mention that he got one of his example sentences from an activist associate (1972b [1969]:67), for instance, or that given discourse situations concerned his refusal to pay taxes or his anti-war speeches (1975b:61–62), fairly simple illustrations taken from his life, with no overt political meaning in the context. Chomsky is adamant that his two classes of interests, language and politics, are almost entirely unrelated in terms of his roles. An interviewer once asked him what the ‘two Chomskys’ would say to each other, the activist and the linguist, should they bump into each other. Apparently nothing. The Chomsky who was being interviewed here (the political one) didn’t even entertain the question, dismissing it with “There is no connection, apart from some very tenuous relations at an abstract level” (1988c:318). But, will he, nill he, the dissociation is far from absolute. The interview where we get the ‘no connection’ answer appears in a book entitled Language and Politics (1988, 2004), for instance, and while the topics are mostly segregated (an interview or essay or chapter on language, a different one on politics) the two topics appear often between the same covers. The mutual ethotic reinforcement of the two Chomskys, each reputation bolstering the other, is not entirely accidental, and it is undeniable that his political reputation helped contribute to the sense in the 1960s and 1970s especially that linguistics was a field for people with social consciences.6
Jackendoff was only marginally more political in his linguistic work than Chomsky, occasionally incorporating sentences such as The president is insane, one suspects, beyond all hope (1972:97). On the other side of the debate, the picture initially doesn’t look much different. Postal never blended politics and linguistics, Ross and the Lakoffs only occasionally and peripherally, taking a few swipes at Republicans, for instance.
But McCawley was a different story. He was more politically active than the others—Selective Service protests, demonstrating for civil rights in Jackson, Mississippi, burning his draft card, refusing to pay his taxes—but two other factors are considerably more important than his political activities. First, there was his complete lack of reserve about involving his political views in his linguistics—if Postal was the presiding genius of offbeat characters and oddball creatures, McCawley was the genius of what Pullum (1991 [1987]:101) calls “the Fuck Lyndon Johnson school of example construction,” with its obscenities and its political dramatis personae. Second, McCawley was also the most influential Generative Semantics teacher, and it is with the second generation that the FLJ examples really flowered.
Too, there are the matters of space and time; kairos. The official home of Generative Semantics, the site of its yearly CLS festivals, was Chicago—which witnessed the Democratic convention riots, by the police; the subsequent conspiracy trial and its attendant circus; the killing of Fred Hampton; the Weathermen’s Days of Rage; assorted acts of autocratic weirdness by Lord Mayor Richard Daley, and assorted acts of anarchist weirdness by groups like WITCH. And, of course, both Generative Semantics and the counterculture were in ascendance in exactly the same period. There had to be cross-pollination. Interpretive Semantics didn’t take off fully as a movement until the popularity of the counterculture began to recede.
Among the more interesting sidelights to the political aspect of Generative Semantics is the fate of Georgia Green’s McCawley-supervised thesis, whose publication was held up for a year because Cambridge University Press feared libel suits (entertainingly documented in Pullum 1991 [1987]:100–111). The process bordered on the ludicrous, as legal advisors cautioned against including sentences like 6, as defamatory of Mary:
But it brings to light another element of the Generative Semanticist style, a subtle step beyond the FLJ examples—the sentence a clef, where the names refer, more or less obliquely, to living people. Mary and John in 6 are very likely arbitrary, but it didn’t take much imagination in the period to find referents for 7.
As Pullum projects the publisher’s concerns, “Could the John be the indicted Attorney General John Mitchell? Could Martha be his wife Martha Mitchell? Was Ted perhaps Senator Edward Kennedy?” (1991 [1987]:107). Probably.
The Generative Semanticists had a track record in allusions of this sort, with a wide cast of characters, including, in addition to the notorious Mitchells and other political figures, cultural icons, and one another:
Even members of the Interpretivist camp dabbled in the sport. Given the animosity between Lakoff and Jackendoff, for instance, it is hard to imagine that this sentence is entirely innocent:
Beyond the more overt political content of much Generative Semantic work, there were two other ways in which the approach more fully reflected the social mood of the times. First, there was the growing influence of sociolinguistics. While Generative Semanticists were not particularly engaged in sociolinguistics, they were unflaggingly sympathetic toward the field. George Lakoff, for instance, included some of Labov’s work on African American English in his most famous policy statement, “On Generative Semantics” (1971b:280f), and Robin Lakoff carved out her own definitive blend of sociolinguistics and ordinary language philosophy. More generally, Generative Semantics’ concern for language use and social context ensured that they benefited from the growth of social conscience at the time:
The political atmosphere of the late 1960s and early 1970s contributed to this feeling of discontent [with Chomsky’s approach] and was a major factor in drawing many serious students of language into adopting the Generative Semantics program, which, by combining work on formal grammar with concern for the use of language in the real world, promised to satisfy their intellectual interests as well as the demands of their social conscience. George Lakoff [in Parret 1974 [1972]:172] was undoubtedly correct when he [said] “Nowadays students are interested in Generative Semantics because it is a way for them to investigate the nature of human thought and social interaction.” (Newmeyer 1986a:228)
The second way that Generative Semantics more fully meshed with the political temper at the turn of the decade was its rhetorical emphasis on dissent. Dissent was a very marketable commodity in the early-to-mid-1970s. The hippies took generational discord to extremes (Roszak 1969:1–41), making any assault on authority attractive almost by definition. “Bring the revolution home: kill your parents,” went the Weathermen credo. “The Oedipal Conflict has replaced Marxian Dialectics,” said Abbie Hoffman (1971). Jerry Rubin, going upper case, screamed “KILL YOUR PARENTS!” (1971:194).
Generative Semanticists were not as extreme as the Weathermen or the Yippies. No one was demanding Chomsky’s head on a pike. But G. Lakoff was not above describing one assault on Chomsky’s framework as “an exercise in anti-establishment thinking,” or calling Transformational Grammar “as much a part of the intellectual establishment as General Motors is a part of the military-industrial establishment” (1971a:ii), or working his way round to Chomsky as an emblem of establishment wrongness in some muted remarks about the political implications of corrosive data:
Teaching linguistics these days is not without some indirect—very indirect—political consequences. In linguistics, as in politics, much of the relevant data to support or refute many claims are available to the average person. In linguistics, it is in your mind and all you have to do is train yourself to recognize it. In politics, it is all around you, in the newspapers and on TV. Again you just have to be trained to recognize it. Just about any beginning linguistics student, with some careful thought, can in an afternoon think up enough crucial examples to show the inadequacy of our most sophisticated current theories. Similarly any citizen of average intelligence can pick out many of the lies that his government tells him. The thought processes are not all that different, though the subject matter is. Any beginning linguistics student will discover with a little thought that men of great stature in the academic establishment, even very bright ones like Chomsky, can be wrong on just about every issue. It makes one wonder about the “experts” who are running our governments. (Parret 1974 [1972]:170)
The assimilation of Chomsky with The Establishment is not over yet. There is nothing “politically revolutionary about the content of [Chomsky’s] transformational grammar,” Lakoff points out; “if anything, it is reactionary today” (Parret 1974 [1972]:170).
Most data-gathering papers—a genre encouraged in Generative Semantics—followed precisely the prescription that Lakoff offers here, of disconfirming the establishment, except that they also often nod favorably in the direction of the preferred theory: here are some facts which any theory of language should address; no current theory comes anywhere near treating these facts naturally; Generative Semantics, however, comes the closest; Interpretive Semantics is completely out to lunch. By the time that genre flourished, however, G. Lakoff was closer to the linguistic mainstream than Chomsky.
Sentence (1) was brought to my attention by Haj Ross (who in turn had heard it from Avery Andrews).
—George Lakoff (1988 [1974]:25)
I am grateful to J. L. Morgan for producing this sentence [Since Nixon was elected, I’ve come to miss LBJ], thereby setting me off on a productive train of thought.
—Alice Davison (1970:199n3)
No thanks is due to John Lawler for calling my attention to this ugly class of facts.
—Haj Ross (1974b:100n30)
Generative Semantics was born from a concerted attempt to make the Aspects model work on growing spectra of data—including issues like category membership, lexical composition, pronominal relations, conditions on transformations, performative sentences, quantifier scope, the implications of symbolic logic, and considerably more. It was a research project to extend the power of transformational description. But one of its main side effects was increased suspicion over the descriptive soundness of Transformational Grammar. For every generalization, there were pockets of facts that slipped out of reach. For every rule there were lists of exceptions (“the blood of a wounded theory”—Green 1974 [1971]:4). For every theoretical principle, there were areas in which it fell drastically short.
Inevitably this suspicion of transformational machinery came hand-in-glove with an increased reverence for data, particularly when ordinary language philosophy, sociolinguistics, and research by ecumenical scholars like Bolinger came to exert more influence. Chomsky (although many of his detractors deny it) has a healthy respect for linguistic data, but his principal allegiance is to the formal properties of grammar. The stylistic offsprings of these twin notions—theory-suspicion and data-reverence—were also twins: preoccupation with the inadequacies of theory and humility in the face of vastness of natural language.
One of the principal motives claimed for this data preoccupation was pedagogical, a motive best illustrated by a small, peculiar, barely-above-the-ground pamphlet—perhaps the single most illuminating document for the attitudes that came to be pervasive in late Generative Semantics—called Where the Rules Fail: A Student’s Guide. An Unauthorized Appendix to M. K. Burt’s From Deep to Surface Structure (Borkin et al. 1968). Even its genesis is offbeat enough to fit the Generative Semantics narrative like a batik wall-hanging. It started life as a homework assignment that George Lakoff gave a first-year graduate syntax class at Michigan: take a rather typical explication of a number of the conventional arguments for the Aspects model and tear it to shreds. Lakoff chose a workbook by an MIT student, Marina Burt, who was briefly persuaded to appendicize this attack to the second edition of her From Deep to Surface Structure “as a kind of antitext within a text.” However, when she saw the virulence of the assault, and showed it to “several persons for whom I have great respect” (read Noam Chomsky and Morris Halle, possibly some others), she thought better of it. In her declining letter, she defended her decision with the very reasonable argument that “although you present excellent counterexamples, there is no analysis of the facts which replaces what exists even though we both know it’s not adequate. The feelings are that it is not enough to pick apart an existing theory without offering some kind of alternative.” There is little value in shouting no! when you don’t also offer something worth saying yes to.7
The students were incensed at being denied the opportunity to thumb their noses at Burt in the pages of her own book, and at the person for whom she was, to some extent, a proxy, Chomsky. They had the antitext circulated through the Indiana University Linguistics Club, with a foreword by Lakoff and a petulant preface that complained about being treated like children who “are too immature to be confronted with the realities of science.” Lakoff’s foreword is a source of fascination on its own. It is pitched directly at beginning students of the sort who would use Burt’s workbook, and gives a very good picture of his exhortative teaching style in the late 1960s and early 1970s. It is explicit in its appeal to youth, though it adopts the curious language of gloomy promise:
By pointing out some of the failures of the rules in Burt’s book, we hope to give students a feel of linguistics as a living discipline, where most analyses are hopelessly far away from their goals, and where the old goal of actually writing a complete grammar for a language has become at best a hope for the future centuries and at worst a joke. (i–ii)
Finding out that very little works the way most introductory textbook writers would lead you to believe can be a frustrating experience. On the other hand, it can and should be an exhilarating one. After all, the less that is known, the more there is for you to find out. If you want to do something interesting with your life and are contemplating doing work in linguistics, it should be anything but frustrating to find that there is a lot for you to do. (v)
The Establishment old-timers were clearly mistaken about how language works: Burt’s workbook, Lakoff says, “dates from the days when many linguists thought that transformational grammar in its classical form was basically correct.” The old-timers were not just wrong in this, they were arrogant in their wrongness, and Burt is guilty of adopting “the sanctimonious pose of presenting solid results.” Still, Lakoff generously adds, Burt’s book does present a very valuable opportunity for beginning students, since, if used in the right educational spirit, “it is a good vehicle for teaching students that transformational grammar is not all that it is cracked up to be.” Into this rhetorical stew goes some humor, clear countercultural appeals, and the unmistakable theme that the Generative Semantics program is the road to salvation, in part because it embraces the sort of unflinching skepticism the linguistic Establishment can’t abide, and in part because it simply makes more sense than shallower brands of linguistics (“We know much more about the meaning of a sentence than we know about its surface structure”).
All of the pamphlet is informed by the attitude that true science involves embracing the artificiality of models and the muddiness of data. Lakoff calls for linguistic students to be responsible, like “any student of the physical sciences,” and recognize fudges for what they are. The important point is not to blame Burt, however convenient she is as a scapegoat, but to blame the entire transformational enterprise prior to the new Generative Semantics Enlightenment.
Among the many ironies in this approach to linguistics is the distance it puts between Lakoff and his dissertation only a year after its publication. The dissertation, which had been informally circulating for years, was Lakoff’s first major publication in linguistics, and it took the most decisive early steps towards Generative Semantics, and it was data-happy. His thesis project was essentially to snoop out apparent counterexamples to the Aspects theory and tame them, to save irregular phenomena with transformations. As the Generative Semantics program grew, Lakoff continued to work on taming the most difficult data he could uncover, but by the time his dissertation reached formal publication, he had given up on transformations to do the taming. He taught his students to approach linguistics by engaging in anti-establishment thinking and trafficking in counterevidence. Somewhere, however, the imperative to save the phenomena was misplaced, and much Generative Semantics energy was increasingly devoted simply to celebrating anomalies.
Not all Generative Semanticists would agree with one ex-partisan that “we went out of our way to be funny in our papers so that once our ideas were refuted we could get ourselves off the hook by saying. ‘Oh, did you take us seriously? Couldn’t you see that we were just fooling around?’ ” [personal communication]. But most, I suspect, would acknowledge a kernel of truth in it.
—Frederick Newmeyer (1986a:137)
Newmeyer’s comment about Generative Semantics humor is a bone that sticks in the craw of almost every ex-partisan I spoke with, though his are not the only words that question the role of Generative Semantics humor. The few published comments about it at the time rebuked the Generative Semanticists for the “extreme of cultivating the bizarre or the risqué for its own sake” and for their “juvenile political and sexual references” (Percival 1971:184; Sampson 1976:179), and Lees, with typical bluntness, just says, “It was smartass humor.” The Interpretivists were generally unimpressed—Howard Lasnik says “You have to take your field seriously. You can’t convey to the world that it’s like a standup routine in a nightclub”—though Chomsky himself was somewhat neutral in his remarks to me:
Science is like any other human activity. You don’t have to put a straightjacket on it. If people like to give papers with jokes, that’s fine. It’s neither good nor bad. Maybe it’s more fun to listen to their papers, I don’t know. But it shouldn’t [affect one’s evaluation of] their results.
Nobody else seems to share Newmeyer’s (and his informant’s) feeling that there was hedging going on, that Generative Semanticists wanted to be able to excuse themselves later when their analyses fell apart with the claim that it was all a joke. The founding Generative Semanticists regard his observation as flatly stupid, if not malicious; George Lakoff terms it “an utter falsehood,” and calls the informant “a jerk.” Robin Lakoff goes further yet, seeing the whimsy in almost evangelical terms, albeit in ways the practitioners themselves barely realized at the time:
[We were] inventing a new style of academic discourse. There were several reasons for that. First, we had not been well socialized as linguists or as academics: we had entered the field as a refuge from other fields that would not let us do what we wanted to do. Second, the linguistics department at MIT had scant respect for tradition, and so we tended to finish our degrees unusually fast (thereby being spared the long process of socialization that normally takes place during one’s graduate education). Third, it was the 1960s, a period of youthful rebellion, during which young people were trying on new styles of all kinds. And fourth, by the time we started referring to ourselves as generative semanticists, we had been forced to recognize the existence of a schism between us and our original mentor, Noam Chomsky. His was the most scholarly and serious demeanor imaginable; one way to memorialize the growing rift was to develop a professional style as different from his as possible. Most of these considerations were not in our conscious thoughts; if we had been asked why we did what we did, we would simply have answered, “Because it’s fun.” (R. Lakoff 2006)
Haj Ross says much the same thing:
I think the original and fundamental reason for [our] humor was the lack of fun in the field and the vision of a discipline in which work and play could become one verb. Generative semanticists poked fun at everything, most particularly at themselves, as a vaccine against taking themselves too seriously, a disease that they saw terminal cases of all around them. (Ross 2000:183)
Fun, goes the response to Newmeyer, is not the same as just kidding.
Still, it is difficult not to see at least some muted truth behind Newmeyer’s observation—correlational rather than causal—having more to do with the general looseness of presentation, of which humor was an ingredient, than with humor per se. Informality was part of the working-paper mode of much Generative Semantics; its literature after the first few years rarely made any attempt to hide the tentativeness of its proposals. Many Generative Semantics papers, particularly the ones in the CLS volumes, are rife with expressions of hesitancy about their proposals (which, however, they combined with claims of superiority over Interpretivist treatments of the same phenomena), and stylistic looseness reflected this hesitancy, perhaps tacitly requesting some lenience from the audience. It expresses, in effect, what George Lakoff put more explicitly very late in the debates:
I do not have anything even close to an adequate theory. What I do have, by way of formal treatment, are some grossly inadequate but suggestive ideas . . . So far as I can tell they are only slightly better than what is now available in various versions of generative grammar—and you know how bad they are! I hope you will approach [the proposals] with a charitable heart. The ideas are young and need to be cared for. (1977a:259)
Lakoff’s statement is the fullest one on the tentative nature of many proposals, but it is very far from unique. Nor were Generative Semanticists shy about recanting their proposals when clear proof of something better came along—though no one ever used the excuse “Couldn’t you see that I was just fooling around?” in such a recantation.
In short, while there was clear hedging going on, and humor contributed to that hedging, it is unfair to say that humor was a deliberate escape hatch. The proposals were seriously advanced, but they came with a warning label against unqualified acceptance.
Generative Semanticists’ truth-in-advertising was rarely appreciated. Transformational Generative Grammar had been in flux almost from the outset—certainly from as soon as it grew beyond Chomsky’s personal concern. Lees’s 1962 preface to the second printing of his English Nominalizalions, just two years after the first, apologizes that it “has long since become antiquated” (1968: xxvii). But such confessions were rare in the early years and have always been rare in the work most closely associated with Chomsky, which projects confidence and certainty.
Consumers—of theories just as of washing machines—want confidence and certainty. They don’t want to invest their intellectual capital in something that may be out of date in a few years, or a few months, or a few days. Listen to this complaint about the immediate dissemination of partially digested notions:
What is at issue is whether every new speculation, no matter how ephemeral, should be broadcast in the literature when the development of ideas is so continuous and rapid that the latest speculation is already out-of-date by the time it is published. (O’Donnell 1974:79n1)
Things were starting to look bad for Generative Semantics sociologically, and its ethos contributed to its failing reputation. If you think O’Donnell is critical, here is Robert Wall, an insider, writing a satire, but a devastating one:
Look, maybe the thing to do would be to go into historical linguistics and let syntax go for a while—at least until things settle down a little and these assholes stop printing every hare-brained hallucination that afflicts their heads. (Wall 1970:167n15)