Pragmatics3:4.367 -385 InternationalPragmatics Association

LEWIS CARROLL: SUBVERSIVEPRAGMATICIST

Robin Tolmach Lakoff

The field of pragmaticsis broad and diverse;one of its problems,as well as its glories, is the difficulty of defining it so as to cover everythingand satisff everybody.The followingremarks can be consideredone smallcontribution toward making any rigorous definition lessattractive. I like to define pragmaticsas "the interestingstutf about language"- the reason many of us were attracted to linguistics.We wanted to know how languagedid the things it did, to us and for us; why some people used it to get their needs met, and othersto get into variouskinds of trouble;why usinglanguage was sometimesfun, and sometimesfrightening; and so on. As I was reachinglinguistic majority, there was no "linguisticpragmatics": those questionshad to be put aside,as we learned to ask only serious questions: questions that assumed that form alone counted; with the presupposition,I am afraid, that if a question(or its potentialanswer) was likely to be communicableto a non-specialistor not likely to put such a person immediately to sleep,it was non-linguistics,uninteresting, or too easy. Today I think just the oppositeis true. I think pragmaticsis, and ought to be, the area of study that tells us what we realfy want to know: everythingyou ever wanted to know about languagebut were afraidto ask.Not, that is,everything about language:there is still, of course,phonology, syntax,and the rest of "central"or "core"linguistics. But no responsibleperson is afraid to ask questionsabout those fields. Many of us, however (and even more so, our students)are in somesense or other afraid to articulatethe questionsthey most deeply desirethe answersto, becausethey havegotten the idea that thesequestions aren't sufflcientlyhard-nosed, scientizable, or respectable.I feel,on the contrary,that they are the most respectablequestions anyone can ask,with potentiallythe most important answersanyone can give or get. You can take that, if you like, as an apologyin the Socraticsense for this paper, a justificationfor its explorationof questionsthat somewould define as "non-linguistic" or frivolous.They are as you will seecompletely linguistic, in that they are concerned with the ways in which languagecan be usedto achievesignificant etfects on reality; and they are preciselyas frivolousas is our need to understandthe persuasivepower of language,its use as a political instrument.If this be frivolity, then make the most of it. Consider lrwis Carroll (Charles Lutwidge Dodgson). Consider his principal contributions to literature and language, the two books 's Adventures in Wonderlandand Throttghthe Lookitrg-Glass.lnmost universitydepartments in which 368 Robin Tolntnch Lakoff significantEnglish literature is studied,these books occupy at besta marginalplace, as "children'sliterature," outside anyone's Canon. At the sametime, a remarkablenumber of important writers of English(not to mentionother languages)have either written explicitly about these books, or have in one way or another incorporatedthem into their own works of fiction or poetry.A glanceat the table of contentsof probably the most significantcollection of writings on the Alices,Aspects of Alice (1971) reveals contributionsby (among many others):W. H. Auden, Virginia Woolf, Robert Graves, Harry Levin, Edmund Wilson,Allen Tate, JohnCiardi, J. B. Priestley,William Empson, Kenneth Burke, and (oh yes) Grace Slick.So it is necessaryto accountfor two things: the relegationof the Alices by seriousscholars to marginality;and the recognitionby thosewho createthe cultureof the centralityof thesetexts. I want to suggestthat the paradox is resolvableif we understandthe Alices correctly,as perhapsthe most subversivebooks ever written.As a result,they have beendefined as "children'sbooks" - not to be takenseriously, rendered harmless by the appellationof "nonsense."No lessan Americanicon than Walt Disneyhas undertaken to sanitizethe texts,make them safeand sunlit.But in their originalforms, theAlices are works more dangerouslysubversive than any other writings. And the topic is relevant to pragmaticsbecause (again unlike other subversiveworks) the subversive content of.theAlices specifically undermines our culture'scomfortable view of language: as an orderly,value-free, cognitive and socialphenomenon. It attacksour assumptions that communicativebehavior is normallyrational and fair; it castsdoubt on our belief that human beingsare sensibleand sentient.Other subversiveworks undermineour faith in the validity or rationalityof systemsbuilt on communicativebehaviors and capacities,e.g. politics and religion(Swift's Gulliver's Travels comes to mind).But Alice goesbeneath and beyondGulliver in questioningthe bedrocklevel of thosecapacities that uniquelymake us human:our rationaland socialsystems, and the languagethat makesthem possible.In this respect,Carroll's works are very deeplysubversive: they overturn our assumptionsabout human communicativeinteraction - our pragmatic capacities. They do this by raisingand exploringa numberof assumptionshuman beings - - ordinaryspeakers and, too often,even specialists - make about how communication works.To list a few: (1) Our rules of pragmaticinteraction are the only/bestpossible such rules. (2) Those who have power have the right to it becausethey think and speak better and more rationallythan the powerless. (3) Our rulesmake sensein their own right (ratherthan as linguisticand social; descriptiveand prescriptiveconventionalizaticlns). Carroll leadsus to theseconclusions by constructingalternate universes which, on first sight (by his "naive"protagonist, Alice), appearto operateby the oppositeof our normal pragmaticrules and (therefore)to be "nonsensical."But whenthey are seen on their own terms,they turn out to makejust as much "sense"as our familiarsystem does- and therefore,either all are "sense,"and thereforeany choiceis arbitary;or all systemsrely on clurwillingness to abideby rulesthat are "nonsensical"- and so we are all, to invoke a tavoriteterm of Carroll's,"mad," since mad persons(or creatures)are : subversive pragmaticist 369 thosewho do not accommodateto "rational"systems. Seen in this light, the Alices readily transcendtheir traditional categorizationas children'sfiction, being neither. Ultimately, they deconstructour fondest and deepest beliefsof all: in truth, certainty,and predictability.By so doing they revealtheir author as a man over a centuryahead of his time: not only as a pragmaticistbefore there was pragmatics(and an inspirationto Wittgensteinamong others), but asa prematurepost- modernist,whose suspicionof the validity and rationalityof our socialand intellectual justifications for ourselves leads inexorably to the decomposition of our entire trustworthyuniverse. I will give examples in the remainder of this paper of Carroll's skeptical examinationof each of the above.

lrt me begin with a brief bibliographicalsketch of the Alices'creator. Charles Lutwidge Dodgson was born in England in 1832. He took first class honors in mathematicsat Christ Church, Oxford, in 1854,and the next year was appointed mathematicallecturer in that college,a positionhe held until 1881.He made a number of contributions,regarded as competent but not groundbreaking, in the fieldsof formal logic and mathematics,and died in 1898.Hisposition required celibacy,and he is not knownto havehad relationshipswith adult women.He was,however, very fond of little girls,and enjoyedtaking them on trips, photographingthem, and telling them stories he had made up. One such story he told to the daughtersof Dean Liddell (co-author of the Liddell and Scott Greek-EnglishLexicon), probably in 1862.In 1865, the first version of Alice's Adventuresin (W) was published under the pseudonym lrwis Carroll. Throughthe Looking Glass (LG) appeared in 1871. We might begin by lookingat the pseudonymitself, especially as concernsabout namesand naming occur frequently in the Alices. It is evident that Dodgson'sgiven namesCharles Lutwidge, inverted and translated,create the pseudonym(Lutwidge : Ludwig = Louis : lrwis; Charles : Carolus : Carroll). So to create the Alices Dodgsoninverted the unique elementsof himself(his givennames) - just as the books invert his, and our, reality. Both booksutilize an unusualheroine/protagonist/interpreter/experiencer, Alice. Her age is given (LG VI) as "sevenyears and six months,"and in W she is probably about six months younger. She is a female child, of course, of upper-middle class, academicbackground. Drawings of her made by Dodgson'schosen illustrator, , show her as having unruly below shoulder length dirty blond hair, always dressedin the Victorian child'sunitorm of stockings,knee length dress,and pinafore, neat but not sparkling.The DisneyfiedAlice, on the other hand, has pure blond hair in a shoulder-length,smooth and perfect page-boy,and bouffant spanking-clean clothing.In her pictorial form alone,Alice hasbeen sanitized and renderedrespectable by Disney. We can reflect on the significanceof Carroll's choiceof intermediarybetween ourselvesin our "safe"world, and the peculiaritiesof Wonderland.On the one hand, we can of courseargue, as is surelycorrect, that Carroll choseAlice as she appearsin the books, becauseAlice Liddell was the inspirationbehind the narratives.On the 370 RobinTolmach Lakoff other, we can extendour investigationto includeperlocutionary effect, that is, why the Alices have had the effect on readersthat they have had for over a century; and when we do so, we can say that Carroll'schoice was serendipitous:it enhancesand deepens the effect of the books on us, the readers. First of all, we note her age:around seven.It could not have escapedDodgson (who as a condition of his employmentwas expectedto studyfor and eventuallytake holy orders) that the Church Fathers had establishedseven as the "age of reason." Theology aside,we know (as Piagethas shown)that at seventhe child is on the verge of making senseof the pragmaticworld, intellectualand moral: is beginningto be held seriouslyresponsible for a knowledgeof politeness,is becomingclear about how and where to be direct and indirect,and, as a consequence,has necessarilyhad to develop and test theories about languageand power: who you can be direct or rude to, who not; when and how people say what they mean, or don't; what to take seriouslyand literally, and what not to. So sevenishis the borderline between the infantile state where anything goes, and pragmaticgeneralizations are nonexistent,incomplete, or unreliable; and the adult state in which we are held responsiblefor those aspectsof communication.Alice's forays into universesin which the rules she has been given in the Real World fail to hold are, then, highlyappropriate: they exempliff the confusion of the child at this borderline age.Her obviousneed for senseand clarity represents the sevenyear old child's urgencyin the task of making senseand determiningwhat makessense. A child of that agetherefore must be especiallysensitive to the rules,and is therefore in a particularlygood positionto decidewhen somethingis "nonsense"- an ideal subversive. Furthermore, Alice is a female child. That statusputs her in a specialposition in her society - and, I might add, ours as well. She is both inferior and other, the quintessentialoutsider, the one who doesnot make the rules,and for whosebenefit the rules are not made, but who more than anyoneelse must abide by them. That means that, in one sense,all rules even in the Real (or logical)World, are "nonsense"to her - - all equally arbitrary in terms of their benefitingher, but equallyrequiring obedience if she is to avoid punishment- a frequentcircumstance in WLG, as much (presumably) as in her RW, and making about as much sense.And Alice in WLG gets explicitly interpreted(and criticicized,and corrected)by everyone,as is alsothe casefor children and women in our logicalworld. Although looked at superficiallythe Alice books may seem similar in their treatment of the rules and conventionsof our real world, W. H. Auden (Phillips 197I 9ff) notesan important distinction:In W, all actionis chaos,all emotion,passion; things happen abruptly and unpredictably.But in LG, everythingis overly rigid: there are fixed and explicit rules that dictate how thingsare to be done and understood- fixed but irrational (by our lights).In this sense,both W and LG representexaggerations of our perceptions of reality as somewhere in between these boundaries: some spontaneity,some structure- just enoughof eachto make sensein context.But the fact that neith W nor LG is a rational place suggeststhat neither total freedom nor total order is "sensible"- and yet, of course,societies or their memberstend to idealizeone or the other - seldom a golden mean.The fact that Alice is youngerin W than in LG Lewis Canoll: subversivepragmatici$ 377 suggeststhat in W Alice (as the dreamer and therefore the creator of the system)is still situatedin the childlike,primary-process system in which chaosis toleratedbecause order has not yet emerged.In LG she is older and somewhatwiser: having discovered that there are rules, like all new converts(and early languagelearners) she expects everythingto work by those rules: she overgeneralizes. Neither total freedomnor total rigor is a stablesystem (although the latter might seemto be): both worlds ultimately disappearinto chaotic self-destruction,precisely when (and because)Alice finally acquiresthe power and expertiseto question their premisesand asserther right to determinemeaning and make explicit interpretations (the right of the powerful).We might,then, see Alice's awakeningsthat terminateboth books as indications of a (somewhat precocious)emergence from childhood into adulthoodwith its powers,but its lossof imaginativecapacity (including the ability to make generalizations,linguistic and otherwise). These are the underpinnings of Carroll's exploration of communicative possibilities.lrt us look more closelyat his use of and commentaryon languageitself. The usesof languagein W/LG have been discussedfrequently, directly or indirectly, especiallyby philosophersand logicians.Because most of these commentarieswere made prior to the incorporationof pragmaticsinto linguistics,and mostly prior to the full appreciationof the contributionsof pragmaticistslike Austin, Searle,and Grice, they havetended to focuson semanticanomaly, rather than pragmatic.In this context I will reversethis focus.I will touch briefly on non-pragmaticissues in W/LG, but speak more fully about the pragmaticanomalies of the communicativesystems in the Alices. In any case, as we would expect, even purportedly extra-pragmaticissues have pragmaticroots, explanations,and conequences.) As is typical of a universein which reality seemsuntrustworthy and chaotic, languageitself in WLG is treated prescriptively,as though by forcing languageto behavelogically or reliably, one can achievecontrol over the chaos elsewhere.One exampleis Alice'sconcern in W as to how to addressa mouse:

(1) So she began: "O Mouse, do you know the way out of this pool? I am very tired of swimming about here, O Mouse!" (Alice thought this must be the right way of speaking 10 a mouse: she had ncver done such a thing before, but she remembered having seen in her brother's latin grammar, "A mouse - of a mouse - to a mouse - a mouse - O mouse!")(W II).

The frequencyand salienceof punsin theAlices has often been remarked upon. Puns,clf course,are the casesin which languageceases to "make sense"with respect to our usual notions of semantics,that is, where the word "makes sense" by correspondingto an extralinguisticreality. In punning,words are meaningful through their linguisticvalue alone (or primarily):their soundinglike somethingelse in the linguistic system; so puns constitute or refer to languageas a closed system, as something meaningful in itself. Of course, if language were totally that w&y, communicationwould truly be nonsense,so it is not surprisingto find puns occupying an honorableplace in nonsenseworlds (in which,of course,they make sense)- indeed, they rnake meaning,and are used for explanationsof "reality" - not, as in our world, 372 Robin Tolnnch Lakoff

mere "fun."

(2) "Mine is a long and a sad tale!" said the Mouse, turning to Alice and sighing. 'lt is a long tail, certainly," said Alice, looking down with wonder at the 'but Mouse's tail; why do you call it sad?' (W III)

In our normal understandingof language,we distinguishbetween names and ordinary nouns:the former can be said to havereference but not meaning,while the latter have both. In W/LG, this relationship is sometimesreversed. Thus everyone'sfavorite semanticist, (a largeegg):

(3) "Must a name mean something?"Alice asked doubtfully. 'My "Of course it must," Humpty Dumpty saidwith a short laugh: name means thc shapc I am - and a good handsome shape it is, too. With a name like yours, you might be any shape,almost.' (LG VI)

And a bit later:

(4) "There's glory for you!" [said Humpty Dumpty] "l don't know what you mean by'glory,"'Alice said. Humpty Dumpty smiled contemptuously."Of courseyou don't - till I tell you. 'There's I meant, a nice knock-down argument for you!" 'glory'doesn't 'nice "But mean a knock-down argument,"'Alice objected. 'it "When / usc a word,nHumpty Dumpty said in rather a scornful tone, means just what I choose it to mean - neithcr more nor less.' (LG VI)

Alice (like many rational individuals)assumes that naming is a purely semantic exercise. But Humpty Dumpty suggeststhat the right to give names and make definitionshas a strong pragmaticcomponent: it is basedon power. This discrepancy is the basisof the current disputeover so-called"political correctness.": ia the right to chooseyour name and make interpretationsinalienable? Alice takes the conservative position:meaning is immanentin languageitself, and it is pointlessor irrationalfor humans to interfere in the process;Humpty Dumpty, for all his arrogant elitism, expressesthe liberal-radicalposition: language is madeby peopleand can be changed by them. The idea of reterencedepends crucially upon objectand personalconstancy. If languageis to refer reliably to reality,so that speakerscan be confident of its ability to transmitmeaning, those objects and personsreferred to mustremain, in somesense, the same.Constancy is implicitin and basicto the socialcontract that enablesus to live as social beings,devising and using linguisticconventions and making senseto ourselvesand clneanother. But in W/LG constancycannot be countedupon. Thus for instanceAlice, especiallyin W, is hauntedon severaloccasions by the problem of whethershe is, in fact,"herself'any more. She determines that sheno longerhas either the appearance(being variously larger or smaller)or the skills(doing sums,reciting poetry) that "Alice" had previouslyhad. So she must be someoneelse (in fact a very specificsomeone else: Mabel) and when she goes"back home,"she must lead that Lewis Canoll: subversive pragmaticist 373 person'slife - quite unsatisfactory,she decides.(W II) Pragmatically,this passageis strikingin its abrogationof the Gricean maxim of Quality: Alice's failure to recognizethe communicativeconvention of metaphor. "f'm not myself," a phrase normally used figuratively to mean "I'm not behaving characteristically,"receives a literaland highlyphysical meaning: "I'm a differententity from the one I used to be." Poetry in WLG violatessemantic and pragmaticrules, yet "makessense." In the Alices, Carroll creates a great deal of poetr/, which tends (unlike poetry for the sophisticated,but not unlike the didactic use of poems to inculcate morality in Victorianchildren) to be usednot asdisembodied "literature" unrelated to the text;but with both function and intendedmeaning - sometimes,to be sure,only to showwhether Alice is or is not functioningas "herself';but often, to glossthe workingsof W/LG,, to providereal-world "meaning" within thosecontexts, as the RW usesprose. (So again, our conventionsare turned around.)Some of the most famous examplesin WLG violatesemantic assumptions:

(5) Jabbcrwocky (LG I)

'Twas brillig, and the slithy toves Did gvre and gimble in the wabc: All mimsy were the trorogoves, And the mome raths outsrabe.

"Bcware the Jabberwock,my son! (The jaws that bite, the claws that catch!) Bewarc the , and shun The frumious Bandcrsnatchl"

Hc took his in hand: Long time the manxome foe he sought - Then rested he by thc Tumtum trec, And stood awhile in thought.

And, as in uffish thought he stood, The Jabberwock, with eyes of flame, Came whiffling through the tulgey wood, And burbled as it came!

Onc, two! One, two! And through and through The vorpal blade went snicker-snack! He left it dead, and with its head He went galumphing back.

"And hast thou slain the Jabberwock? Come to my arms, my beamish boy! Oh, frabjous day! Callooh! Callay!" He chortled in his joy.

'Twas brillig, and the slithy toves 374 Robin Tolmach Lakoff

Did gyre and gimble in the wabe: All mimsy were the borogoves, And the mome raths outgrabe.

The problemsin ""lie purely in the realm of lexicalsemantics. And indeed,our ability to understandthe poem perfectlywell on its own terms (exceptfor the first stanza,which is helpfully glossedby Humpty Dumpty in LG VI) showsthat context is crucial in understandinglanguage. Since only content-words(nouns, verbs, adjectives,and adverbs)and only someof those,are representedas "nonsensewords," we can use the rest of the languageto clarifu the puzzling instances;and once we identify the genre of the poem as heroic,our understandingof the genre allows us to fill in still more. Pragmatics(an understandingof context,from sentencestructure to discoursegenre) explicatessemantics. Another caseis more complicatd(W XII):

(6) "They told me you had been to her And mentioned me to him: She gave me a good character, But said I could not swim.

He sent them word I had not gone (We know it to be true): If he should push the matter on, What would become of you?

I gave her one, they gave him two, You gave us three or more; They all returned from him to you, Though they were mine before.

If I or he should chance to be Involved in this affair, He trusts to you to set them free, Exactly as we were.

My notion was that you had becn (Bcfcrreshe had this fit) An obstacle that came between Him, and ourselves,and it.

Don't let him know she liked them best. For this must ever be A secret, kept from all the rest, Between yourself and me."

Here the problem is in text semantics,and more specificallydeixis. Since deixis involvesboth semanticand pragmaticcriteria, pragmaticsalone cannot "make sense" of this poem. It remainsat one levelperfectly transparent; at another,perfectly opaque. The languageis extraordinarilylucid; but becausethe pronouns are not deictically anchored,within the poem or outside of it, the apparent lucidity yields only greater Lewis Canoll: subversive pragmaticist 375 confusion:we feel we ought to understand,but can't. Even those personal pronouns whose reference is normally obvious (first and second personsj ure of uncertain referentialstatus to the reader (and Alice), sincethe poem appearsin the form of a letter that is read in a courtroom,the identityof its writer and intenctedrecipient to be determinedin the judicial process.It is worth noting that this poem functionsas major incriminatingevidence in the trial (of the Knave of Hearts).ttris is interestingbecause thejob. of the reader of the poem is analogousto the task of the judging body in a trial: to match the identity of the perpetrator with that of the defendant:ls the "he" who committedthe crime the same as the "he" sitting in the dock? As in a real trial, the evidenceis frequently ambiguousor vague,and yet "meaning"must be found. Alice (who has been growing bolder as she gets larger) declaresthat she doesn't "believe there'san atom of meaningin it," which is literallytrue - yet the poem functionsas evidence.If nonsensemakes sense, what is sense,and what is nonsense? A poem like Haddock's Eyes (LG VIII) also moves between semantic and pragmaticanomaly.

(7).l'll rellrhee everything I can: There's little to relate. I saw an aged aged man A-sitting on a gate. 'Who are you, aged man?' I said, 'And how is it you live?' And his answer trickled through my head, Like water through a sieve.

'l "He said, look for butterflies That sleep among the wheat: I make them into mutton-pies, Ald sell them in the street. I sell them unto men,' he said, 'Who sail on stormy seas; And thar's rhe way I get my bread - A trifle, if you please.'

"But I was thinking of a plan To dye one's whiskers green, And always use so large a fan That they could not be seen. So having no reply to give To what the old man said, 'Come I cried, tell me how you live!' And thumped him on the head.

"His accentsmild took up the tale: 'I He said, go my ways, And when I find a mountain-rill, I set it in a blaze; And thence they make a stuff they calt Rowland's MacassarOil - 376 Robin Tolntach Lakoff

Yet twopencc-halfpennyis all They give me for my toil.'

"But I was thinking of a plan To feed oneself on batter, And so go on from day to day Getting a little fatter. I shook him well from side to side. Until his face was blue: 'Comc tell me how you live,' I cried, And what it is you do!'

'I "He said, hunt for haddocks'eyes Among the heather bright, And work them into waistcoat-buttons In the silent night. And thesc I do not sell for gold Or coin of silvery shine, But for a copper halfpenny, And that will purchasenine.

"'l sometimes dig for buttered rolls, Or set limed twigs for crabs: I sometimes search the grassyknolls For whecls of Hansom-cabs. And that's the way' (he gave a wink) 'By which I get my wealth - And very gladly will I drink Your Honour's noble health.'

"l heard him then, for I had just Completed mv design To keep the Menai bridgc from rust By boiling it in wine. I thanked him much for telling me The way hc got his wealth, But chiefly for his wish that he Might drink my noble health.

"And nnw, if e'er by chance I put My fingers into glue, Or madly squeezea right-hand foot Into a left-hand shoe, Or if I drop upon my toe A very heary weight, I weep, for it reminds me so Of that old man I used to know - Whose look was mild, whose speech was slow, Whose hair was whiter than the snow. Whose face was very like a crow, With cyes, like cinders, all aglow, Who seemed distracted with his woe, Lewis Canoll: subversive prapnraticiil 371

Who rocked his body to and fro, And muttered mumblingly and low As if his mouth were full of dough, Who snorted like a buffalo - That summer evening long ago, A-sitting on a gate.'

. The problems with this poem extend between, and blur the lines between, semanticsand pragmatics.If we felt a need to be splitters,into what pile would we put the agedman's (and the narrator's)job descriptions?They would appearto represent both semanticanomalies (how can butterfliesbecome mutton-pies?) and pragmatic impossibilities(what is the interactiveutility of dyeingyour whiskersgreen, if you then usea fan largeenough to hide them?) The linguisticproblems here begin with the poem's very name. The White Knight,who recitesit, beginsby saying:

'Haddock's (8) "The name of the song is called Eyes."' "Oh, that's the name of the song, is it?" Alice said, trying to feel interested. 'No, you don't understand,"thc Knight said, looking a little vexed. "That's what the 'The name is called. The name really rs Aged Aged Man."' 'That's "Then I ought to have said, what the song is called'?' Alice corrected hcrself. 'Ways "No, you oughtn't: that's quite another thing! Thc sorzg'iscallcd nnd Means': but that's only what it's colled,you know!" "Well, what ls the song,then?" said Alice, who was bv this time completelv bewildered. "I was coming to that," the Knight said. "The song really is'A-Siuing on a Gate."' (LG VIII)

Alice here is forced to reinterpret common,if imprecise,ways of talking: all of the expressionsin the passagecited are commonlyused with similar meaning,but the Knight insistson distinctions.In one sensethis is a semanticproblem, one of definition or extension.In another,it's pragmatic:the Knight is unwillingor unableto play by the normal rules of everydayconversation, including those of conversationalimplicature. The Knight requires his interlocutor to adhere strictly to the maxims, but as his behaviorwithin the poem itself demonstrates,he doesnot hold himselfto the same high standardsof clarity. In the poem rtselfwe encounterrepeated violations of conversationallogic and speechact felicity. It is often remarked by adults that when the Alices were given to them as children,they found the books disturbing,even frightening.One reason may wellbe Carroll'splaying fast and loosewith hallowedpragmatic principles - frightening for a childjust beginningto understandtheir provenancein the real world! If we can no longercount on the felicityof speechacts; on "normal"computations of illocutionary force;on the appropriateuse of GriceanMaxims and conversationalimplicature; on politenessprinciples - then first of all, a great deal of "normal"discourse ceases to makesense - raisingthe specterof psychologicaland socialdanger unto annihilation (indeed,the conclusionsof both Alices end in the annihilationsof the worlds they create).Pragmatic incoherence both includessemantic confusion, but goesbeyond it 378 Robin Tolmach Lakoff in its dehumanizingcapacity; and therefore,the anomaliesin the Alices make it clear that those who have the power to createand enforcepragmatic conventions (whether explicitly or not) have power indeed. Considersome specificcases in the poem. Its first two lines flout Quantity and Quality, especiallysince a long poem follows.The dialogof the poem is framed in a Q- and-A format, suggestingthat the narrator/questionerneeds the information the aged man can provide, and therefore is willing to listen (cf. Austin 1962 and l-akoff 1972). It further assumesthat the aged man's utteranceswill make sense;or if they do not, that will be the reason that valid communicationdoes not take place. These are the implicit rules of consensualdyadic discourse. As the poem unfolds,these expectations are repeatedlyshattered. It might be objectedthat poetry allowsviolations of ordinary discourseconventions; but when poetry representsa conversationaldyad, it is highly unusual (perhaps unique here) to find the rules of both conversationallogic and conversationanalysis blatantly violated. As one example,many of the narrator'sconversational seconds are worse than dispreferred:e.9., after receivinga perfectlyinformative (by its terms) responseto his question, "How is it you live (stanza 1)?" in stanza 3 the narrator indicates no perception of the answerat all, not even of the fact of there havingbeen a reply, and "thumped him on the head"- hardly a permissibleresponse. In terms of conversational logic and indirect illocutionary force, both charactersexpress their conversational agendasearly and often.The narratorwants information; the agedman, money.To this end he utters two indirect and formulaic requests(stanzas 2 and 7, and perhaps still less directly 4). But the narrator ignoresthem at first, and when he finally hears one at all proceedsto take it literally (st. 8). So while the poem superficiallyresembles a cooperativeinterchange, in fact both participantsare so engagedin their own needs and subjectivitiesthat they make no connectionat all. And although readers can understand the poem perfectly well, as a representationof a discourse event, it exemplifiesa threateningform of illogic - conversationalillogic. And while, finally, on its face the non-dialoglooks to us absurd and improbable,has not everyone,in this world of safety and reason,experienced somewhat analogous interchanges, in which each "side"refused to "get"what the other was after, and while appearingcooperative was usingthe presumptiverules of cooperativeconversational interchange deceptively, to exacerbatepolitical imbalances(sts. 2, 5) rather than legitimately,to exchange information or solidi$rsocial connections. Pragmaticsmight alsobe definedas the field concernedwith the communicative assumptionsshared by humansas membersof a societyor a culture. In more "central" linguistics,asterisked examples often are usedto illustrateextreme cases, to which the rules of grammar do not extend,and which therefore illustratethe boundariesof the applicability of the rules. W/LG communicativebehavior constitutes analogously asteriskedexamples: the recognition of bizarre interactive patterns highlights their antitheses.Then the rules we do observecome into sharp focus. But just as starred examplesmay lose their stars in appropriate contexts,so the interactionsof WLG, absurdwhen contextualized(by Alice or by readers)within the normal world, become perfectly intelligibleseen from within the contextsof WLG as reasonablealternative Lewis Catoll: subversive pragmaticist 379 worldswhich utilize alternate pragmaticsystems. Many of the anomaliesof the Alices arise becausethe context is so markedly differentfrom that of the normalworld. For one thing,the latter demandsan essential distinctionbetween "real behavior"and "games."A big attractionof gamesis that they offer an escapefrom reality,where behaviorhas real consequences;where words and actionsare normallysincere;where behavior is logicallyconnected to its context;where peopleare usuallypresumed to behavewith somedegree of spontaneity.In games,on the other hand, rules are made explicit at the start, and typically are rigid and non- negotiable;"winning" and "losing"are only intra-gameand are explicitlynoted, yet not of "real"consequence;people are not "sincere"in their actionsin the sameway. But in both W and LG the action itself constitutesa game (and one with particularlycomplex and unbendingrules: cards in W, chessin LG); andjust as we would seeit from within any of thesegames, to the inhabitantsof WLG their behavior seems"real," sincere, and spontaneous,although to the outsider it does not. In WLG, it is impossibleto distinguishgames from real life: one n the other. So the WLG charactersfail to distinguish(presumably) spontaneous conversational turns from "turns" in a game:

(9) "....However,this conversationis going on a little too fast [said Humpty Dumpty]: Iet's go back to your last remark but one." "I'm afraid I can't quite remember it," Alice said, very politely. nand "ln that casewe start afresh,' said Humpty Dumpty, it's my turn to choose a subject---"("He talks about it just as if it was a game!" thought Alice.) (LG VI)

On the other hand,fixed non-spontaneousforms of discoursethat normally are insertedinto dyadicconversation as set piecesfiokes, puns, riddles) turn up in WLG as if they were fully spontaneousconversational turns:

(10) "You should learn not to make personal remarks,nAlice said with some severity: "it's very rude.' "The Hatter opened his eyes very wide on hearing this, but all he said was, "Why is a raven like a writing desk?" (W VII)

So just as "game" blends with and becomes"reality," so "nonresponsivediscourse" blendswith that which is fully contextuallymeaningful. In the same way, human behaviorsthat are stylized(like games)yet, to us,logical, meaningful, sincere and thus "real"(like trials,wars, or formal banquets)become in W/LG inextricablyconfused with nonsensicalgames: they begin as abruptly,their rules are as arbitary,they are as soon forgotten.Indeed, Carroll can be said to have anticipatedthe Wittgensteiniannotion of communicativeinteractions as games(and vice versa). The meta- and extralinguisticframes we count on to give senseto our linguistic enterprisesare often absentor distortedin WLG, with the consequencethat, to Alice, meaningis hard to ascertain.As an example,a favorite general framing convention appliedto all kinds of situationsis dichotomousor discontinuouscategorization (cf. Lakoff 1990,ch. 10).Things are assignedto categoryA or categoryB, not partially to both, and our interpretations and responsesto them are based on that rigid 380 RobinTolmach Lakoff categorization(which we assumeothers are usingtoo). One persuasiveexample cuts acrossmoral and intellectualperspectives: the tendencyto considerthings, persons, events, etc. as either "good" or "bad," "right" or "wrong."The ability to do so, and count on consensusfor that assignment,makes both communication and interactive life generally much easier, but also contributes to stereotypingand rigidly judgmental behavior.In WLG this comforting if dangerous dichotomizationfrequently proves unusable, much thoughAlice tries to invoke it. Thus, in LG IV, Tweedledeerecites to Alice a long poem, "The Walrus and the Carpenter," in which the named protagonistsentice a group of oystersto take a walk with them, and subsequentlyeat them. "I liked the Walrus best," Alice immediately responds at the conclusion of the recitation, "becausehe was a little sorry for the poor oysters."Alice thus frames the poem as an exercise in relativistic morality: who was the least bad? Tweedledee counters:"He ate more than the Carpenter,though. You see,he held his handkerchief in front, so that the Carpentercouldn't count how many he took." So what Alice took for morality (compassion)is revealedto be a deceptivesort of utilitarianism.Alice is forced to distinguishbetween right expressionand right action,a conflict she tries to resolve."That was mean!" she says."Then I like the Carpenterbest - if he didn't eat so many as the Walrus.""But he ate as many as he could get," saysTweedledum, now moving the moral contlictto one betweenright actionbased on virtue, and on necessity or compulsion.Alice's simpleVictorian preceptsfail her: the Looking Glassworld does not allow the convenient categorizingframes of the real world. (Older and wiser readers might notice here a convergencebetween the worlds, much as conventional morality deniesit.) Another comfortingreal-world distinction is that betweenwriting that is "factual" and that which is "fictional."One of the fruits of postmodernismis the questioningand manipulationof theseformerly distinctcategories; but the fondnessof the sophisticated for playingwith them only reflectseveryone's concern with the possibilityof finding and retaining boundaries.)As Searlenotes (1979),we do not expect literal verisimilitude of descriptionswithin writings defined as "fiction,"but we do expect somethinglike "plausibilty":actions should unfold by the samelaws of physics,probability and social interactionas in the world we know; or the writer hasthe responsibilitiesof explaining why not, locating the fiction within some mutually recognizedalternative genre (".g., sciencefiction), and remaininginternally consistent. Carroll is creatinga new genre,with new rules of cohesionand plausibility,in the Alices. But the break with conventionis deeper than this. One of our prevailing assumptionsis that, at least sometimes,nonfiction writings may prove predictive of future events,and/or explanatoryof current or past events.But fictional ones per se cannot play this role. In the Alices the two merge, and fiction or fantasy achieves predictive and explanatorypower, while Alice's real world knowledge,ordinarily a predictor of future events, often proves useless.Thus nursery rhymes (sometimes known only to Alice) presageevents. While Alice seesa deterministicrelationship between the poems she quotes and their actualizations,the characterswithin W/LG perceive themselves(as we like to see ourselves)as acting in a non-predetermined Lewis Carroll: subversive pragmaticbt 381 fashion,with control over events.Alice is thus a proto-Whorfian:her languagecreates not only her perception of reality,but her own and everyoneelse's actual reality. On the other hand, the laws of physicsshe bringsfrom RW have no predictive value in W/LG. The CheshireCat appearsand disappearsin all kinds of unpredictableways; Alice growsand shrinksalmost randomly; the objectsfor salein the Sheep'sshop rise from their shelveswhen stared at (perhaps a foreshadowingof the Uncertainty Principle,but certainlyat odds with contemporaryVictorian physicaltheory). W/LG ignoreor overriderules of politenessand conversationalimplicature. The creaturesare regularlyintolerably rude to one another and Alice (even as they insist on playingby the rules of etiquette,as in (12)); Alice is rude out of tactlessnesson frequentoccasions, with her violationsgenerally emanating out of her awarenessof a powerdiscrepancy between herself (or her pet) and her addressees,thus moving from merenon-politeness to rudeness(13):

(12) "But what am 1 to do?" said Alice. "Anything vou like," said the Footman, and beganwhistling. (W VI)

(13) "And who is Dinah, if I might venture to ask the qucstion'J"said the Lory. 'Dinah's Alice replied eagerly, our cat. And she's such a capital one for catching micc, you can't think! And oh, I wish you could see her after the birds! Why, she'lleat a little bird as soon as look at it!" This speechcaused a remarkablescnsation among thc party. Some of the birds hurried off at once....(W III)

Example (14) involves multiple confusion of illocutionary force and conversationalimplicature.

(14) "Now [said the White Qucen] I'll give you something to believe. ['m just one hundred and one, five months and a day." "l can'[ believe /hat!" said Alicc. "Can't you?" said the Queen in a pitying tone. "Try again: draw a long breath, and shut your eves."(LG V)

"Belief''is one of the underlying("essential," in Searle's(1969) sense) conditions on the successfulperformance of assertivespeech acts - it isn't, as the Queen would haveit, a conversationalgambit in itself.Alice's rejoinderwould normally constitutea polite way of sayingthat the Queen'sutterance is untrue, though it is framed literally asa statementof her own incapacityvia conversationalpostulates (Gordon and l-akoff 1971).The Queen,however, takes it literally and further assumesthat what it conveys is regretat that incapacity;so sheofters adviceon how to overcomeAlice's weakness. Yet the adviceitself is aberrant,violating a preparatorycondition on that speechact: you don't correct psychologicalincapacity by physicalinterventions. Carroll forces us to seeour normaluse of implicatureand indirectnessas interactive games we play;the Queenhere refuses to playwhen it is in her interest- aswe mightwish to on occasion, but generallydare not. (But she'sthe Queen.)In this way the denizensof WiLG play havocwith a basichuman assumptionabout how "sense"is made and both cognition 382 Robin Tolnnch Lakoff and social cohesionachieved: through consensusand convention.In WLG these are flouted where useful, adhered to where useful, insistedupon inappropriatelywhere useful- and thereforebecome totally arbitrary.But sinceall participants(except Alice) accept the non-systematicsystem, that meta-systemitself is consensual,so nonsense makessense. That suggeststhat the wholenotion of "nonsense"as distinguishable from "sense" through some explicit and obviousformal distinctionwill not hold. Nonsenseis not a particular way of talking, thinking, or acting.Elizabeth Sewell (1952) offers a partial definition: Nonsenseis "a collection of words or events that do not fit into some recognizedsystem. (25)" The crucialword here is recognized,i.e., "consensually agreed- upon by all participantsin the interaction."So W/LG shouldbe continuallyand totally chaotic and uninterpretable,but in fact neither is - becausetheir rules (or absenceof rules)constitute a systemsubscribed to by all inhabitants.Rationality, like sanitymore generally,arises out of sharingbelief systems- it has a functional,not a formal, definition. Hence Carroll's frequentinvocation, in bothAlices, of madness.Looked at from the outside,the inhabitantsof both are"mad" and thereforefrightening - unpredictable by externalconsensual rules. But functionallyall are "sane,"because all subscribeto the samesystem. In this respecttoo Carroll is a prematurepostmodernist in the senseof Sass( 1992),who seespostmodernism and schizophrenia as more similarthan different: for example,in questioningor disavowingcertainty, consensus, invariant selfhood, and object constancy.Carroll's worlds share the sameassumptions, and so (when viewedby a non-consentingoutsider like Alice.or the reader)are "mad";but when viewedfrom within, are meta-consensual,that is, postmodern. But consensusis more complexthan it seems.In all worlds- Carroll'sand our own - what we see and how we see it are not really free for all to determine as they wish.Rather, those with power (political,physical, or social)have a great deal to say about how to see "reality,"and how to talk about it: by controllinglanguage they attempt to control, and frequently succeedin controlling, everyone'sreality and possibilities.In WLG Alice and the other charactersfrequently come to conflict over who controls language;wlnt languagecan be made to mean; whethercontrol over languageis ipso facto power.What is unusualin WLG, as opposedto the Real World, is that theseconflicts are openlydiscussed. The Red Queenin LG tells Alice how to talk and how to think:

(15) At the next peg the Quccn turned again,and this time she said, "Speakin French when you can't think of the English for a thing - turn out your toes as you walk - and rcmcmber who you arcl" (LG II)

Humpty Dumpty makesthe relationshipbetween language and power famouslyexplicit:

( 16)"The qucstion is," said Alice, "whcther you con make words mean so many different things." "The question is,nsaid Humpty Dumpty, "which is to be master - that's all." (LG VI) Lewis Canoll: subversive pragmaticist 383

In this perspective,one of the oft-commented upon eccentricitiesof the WLG creaturesmakes perfect sense- that they order Alice around continuallyand (to Alice) pointlessly,mostly for the pleasureof doing so. But this is of coursejust what Alice, as the powerlessoutsider, experiences from the powerful - the adults- in the "sane"RW. Authority and languagein RW and WILG reinforceand createeach other. While she is within the systemand acceptingof its rules (or silently accedingto them), Alice cannotinterfere or comment.But once she transcendsthe system,partly because(W) she gets bigger than any of the others, partly because(in both) she begins to be a competentuser of the system(as she is doing in RW), partly becausethings are getting so far out of control that she must transcendthe etiquette of silence,break out of politeness- by makingexplicit the irrationalityshe perceives in the alternateworlds, she causesthem to explodeand vanish.In the Real World, this dangeris circumvented (usually)either by convincingthe powerlessthat they deserveand indeedenjoy their status;or by bringingoutsiders within the systemand awardingthem the privilegesof insiderswhen they get "big" enoughto constitutea threat. But WLG fail to neutralize Alice in this way,so she neutralizesthem.

Someconclusions are now possible. 1.The leastpowerful and mostisolated members of a consensualor cooperative systemare ofien its sharpestand severestcritics. For them, the rules aren't the way they are becausethey logicallyought to be, becausethey don't see themselvesas benefittingfrom the statusquo. But as with Alice, they often remain etfectivelysilent, evenwhen they speak:they are allowedto remaincritics because they don't have to be listenedto. 2. The most flexibleand ambiguoussystem is the most lastingand the most comprehensive- if the most frustrating.We saw that W representedone end of an authoritariancontinuum, LG the other,and both asa resultare fundamentally unstable, andexplode if examined.This may suggest something to usabout the sort of theoretical modelsand descriptivesystems we want to construct. 3. Nonsenseis not truly illogical- as long as its rules are determinedand adheredto by mutual consent.Madness is definedby isolation,not by any superficial form of behaviorper se. 4. "Nonsense"as a literarygenre has its own rules.In the first place,despite the name,it must be intelligibleby the mutual agreementof writer and reader.But in approachingworks of nonsense(like genressuch as poetry),readers agree to do extra work (evenas writers do by performingacts of extraordinaryyet controlledcreativity). It istrue that all humaninteraction necessarily depends on trust:trust in one'sown and other participants'rationality and cooperation.But this is especiallytrue in the nonsensegenrer the readermust trust that the writer will continueto make sense,that the writer will provide enough cues and clues to let the reader navigate the rocky terrain; the writer must trust the reader to perseverein the face of chaos and uncertainty.Nonsense literature therefore f-lourishes in a homogeneoussociety in which it ispresumed that everyoneis followingthe samecommunicative principles and shares most cultural rules - the same sort of society,that is, in which irony is most apt to 384 Robin Tolmach Lakoff succeed.This may be why nonsensehad such an efflorescencein Victorian England, and seemson the decline today. 5. And so we can see the Alices as: a. a sociopoliticalcritique of existingsystems: not only politicaland social,but communicative,at severallevels. The realizationthat the waywe are is not necessarily a much lessrfte logicalway to behaveas human beings. b. a commentaryon power,its usesand abuses;its capacityfor blindnessand its fondnessfor predictability,true or false. c. a realization(presaging Wittgenstein and Freud) that in our daily liveswe are neither spontaneousnor logical. Freud (1917) argued that one reason people were loath to accept psychoanalysiswas that it representeda third diminution of human importance: Copernicus removed Earth from the center of the universe; Darwin dethroned Man as just another primate; Freud showedthat Man was not even in control of his own psyche.Carroll providesa fourth diminutionby showingthat we are the slavesof our ingeniouscommunicative systems, not their masters;and that the systemswe depend on were not selectedbecause they were particularly sensibleor universal. d. a proto-postmodernquestioning of the very possibilityof certainty:truth, identity,authority, reason, and finally realityitself are revealedin theAlices as no more than convenientconstructs - not eternalverities.

The abovepropositions imply a pragmaticperspective, involving as they do a relationshipbetween language form and function;intention and understanding;the importancefor the understandingof languageof psychologicaland socialcontext. The conclusionsa reader can draw from Carroll's treatment of these issues are pragmaticallydaring and controversial,that is, threateningto our human senseof uniqueness,rationality, and importance;and our assumptionthat our sharedsystems of knowledgeand understanding,are sharedbecause they are rational.I would argue, following Carroll, that the causalityof the last statementought to be reversed.Our communicativesystems are rationaland meaningfulbecause they are sharedand to the extentthat theyare consensual.If we acceptthese premises, our understandingof what makes us cognitivelyand sociallycompetent must be significantlyreclrganized. This is why I call the Alice books highlysubversive - in a way that strikesat the core of human identityand pride.At the sametime thesebooks illustrate and extend our understandingof pragmaticsand its necessarysystematicities. For thesereasons Alice's Adventuresin Wonderlandand Throtryhthe Lookirtg Glass are worthy of deep and respectfulpragmatic investigation. Lewis Carroll: subversivepragmaticist 385

References

'wonder-world" Auden, W.D. (1971) Today's needsAlice. In Phillips, pp.3-12.

Austin, J. L. (1962) How to do things with words. Oxford: Clarendon Press.

Carroll, Lrwis (1960 p862]) Alice's adventuresin wonderland and through the looking g/ass. New York: Signct (New American Library).

Freud, Sigmund (I9I7) A difficulty in the path of psycho-analysis.Standard Edition, vol. XVII. London: Hogarth Press,137-144.

Gordon, David and George l-akoff (1971) Conversationalpostulates. Papers from the Seventh Regional Meeting of the Chicago Linguistic Society,pp. 63-8a. l^akofl Robin (1972) l-anguage in context. Language 48: XJ7-27.

[:koff, Robin (1990) Talking pow,er.New York: Basic Books.

Phillips, Robert (ed.) (1971) Aspectsof Alice. New York: Vintage.

Sass,l,ouis A. (1992) Madness and modernisrr. New York: Basic Books.

Searle,John R. (1969) Speechacls. Cambridge: Cambridge University Press.

Searlc, John R. (1979) Expressionand meaning. Cambridge: Cambridge University Press.

Scwell,Elizabeth (1952) Thefield of nonsense.I-ondon: Chatto and Windus.