Leaky registers and eight-hundred-pound gorillas.

Leaky registers and eight-hundred-pound gorillas. How do people talk about the unmentionable--the "I won'tsay what" things? And if people want to communicate about suchthings despite their unmentionability--while still observing the normsthat render some words or topics noxious--what strategies might beavailable? How well can those strategies work? Meanwhile, what mischief,and what unforeseen entailments, can lurk in the observance of theconventions of verbal taboo? It is clear that human communities everywhere have conventionsabout what may or may not be said, by whom, in what circumstances, andwith what degree of praise or opprobrium. To participate in speech atall is to participate in such regimes of value, and indeed those regimesare high on the list of what we study as anthropologists. From among ourown ranks as well as from other disciplines, there is a considerableliterature on politeness, on taboo language, and on euphemism, includingsome works that reach for crosscultural and crosslinguistic universals.Thus we can feel confident that the descriptive details of human sexualreproduction and elimination are likely targets of conventionallinguistic delicacy (or its aggressive disregard), as are assertionslikening particular people to disfavored animals, or statementsunfavorable to the ritual self-regard of one's interlocutor. Thisis not to downplay the cultural and historical specificity thatconditions even which animals are disfavored and which body parts mostproblematic, to say nothing of the larger picture of activities, typesof persons, and social settings that govern what is locally mentionableand what is not. That larger picture also governs when and whyeuphemistic or evasive talk itself may be disvalued. ("Stop beatingabout the bush!") Of more concern for my present purposes, however, are the specificcommunicative means that may be deployed so that the unmentionable isnevertheless referred to somehow, or understood to be the bush aboutwhich people are beating. What are some important communicativestrategies for doing this, and under what circumstances might a strategyfail? When does dirt stick to the speaker, and how hard? Can thesestrategies be subverted? Can a speaker make the dirt stick to someoneelse? Create an odor of toxicity where it might not otherwise have beendetected? I shall first consider various kinds of containment efforts,some of them very familiar, with which speakers try to deal withtoxicity in their own talk: the strategy is to contain the noxiousmaterial in a sort of linguistic insulation--some linguistic cordonsanitaire that insulates a speaker from responsibility for toxiceffects. But this public game of containment-in-the-moment, important asit may be, is not the whole of the matter. Noxiousness does not simplylie in the denotational content of talk; it is more complicated, moreentangled in mediating conventions and processes. So the game ofcontainment represents only the surface, the most overt portion of alarger field of social action and second-order effects. Looking towardthat field demands attention to interlocutors, to discursive contextsand histories, and ultimately to the cultural and ideological settings,the regimes of language, truth-telling, and knowledge in which thosespeakers and their efforts are situated. Lastly, I turn to omissions andabsences, where toxicity, if inferred at all, must lie in the ear of thehearer. Because efforts at containment, ironically, create a kind ofpresent absence--the figurative 800-pound gorilla in the room--somekinds of omissions of toxic material from explicit talk involveprecisely the same issues as linguistic containers do. I limit my discussion to kinds of talk in which the norms ofunmentionability are observed, and not just in the breach. Obviously,this means excluding some fascinating kinds of transgressive talk (orwriting), particularly those that arise from rage or contempt, or from adesire to confront a person or a norm deemed oppressive. Much of suchtalk, however, depends for its effect on people's knowledge of thenorms it breaks, and on the relative markedness, probably alsoinfrequency, of transgressive utterances. When transgression becomes thenorm it is no longer transgression; the unmentionable is unmentionableno longer. I concentrate, therefore, on some ordinary, even conventionalized,strategies for containing toxicity by creating or modeling some distancebetween the problematic materials and the speaker who mentions them. Thestrategies model that distance as it might be seen from the perspectiveof some putative addressee. To glance ahead, we will see thatconventional strategies of containment may succeed, insofar as theyprotect a speaker from some form of retribution an un-insulatedassertion might have risked. Or they may fail; or success may not havebeen the goal in the first place. Yet, for any of these cases, I willsuggest that the containment conventions entail all sorts ofsecond-order indexicalities, often unanticipated by the speaker,sometimes undermining the effort or circumventing the conventions--andcertainly not included in off-the-shelf academic understandings ofverbal taboo. In the connotational penumbra, many kinds of social dramasmay lurk. Containment Efforts Let us begin with the most obviously conventional: the kinds ofcontainers most often discussed under the headings of verbal taboo or,sometimes, etiquette (how to be tactful; how to talk to your in-laws;and so on). Among the means often mobilized for calling disfavoredreferents into mention are special registers and/or footing shifts. Thefirst of these containment strategies, euphemism, works by masking anddistancing: putting a linguistic mask--a circumlocution, or a relativelyabstruse expression, or a codeswitch--between the interlocutors and anydirect engagement with dispreferred, toxic referents. The secondcontainment strategy, a footing shift, establishes a different kind oflinguistic "container": a quotative frame that shiftsresponsibility for the authorship of an utterance onto someone else. Thetoxic referent, in this case, might be mentioned explicitly, butdistance is established between the referent and the current speaker.(Notice that these two strategies are not utterly distinct. A speakercould do both at the same time--quote someone else's use of acode-switch. Or, in some ethnographic cases, the code-switch or thespecial register might be locally--conventionally?--understood asoriginating elsewhere, as if it were always already a kind of quote, away of appropriating someone else's voice.) (1) Simple examples of register-shift insulation would include the use,in English, of vocabulary derived from French or Latin sources, asopposed to Germanic ones. Thus gentlemen dine, perspire, and expectoratewhile peasants eat, sweat, and spit. These are familiar examples, as arethe ways Latinate vocabulary can be piled up in registers such as"bureaucratese." There, the polysyllabic lexicon is oftencombined with syntactic precautions (such as agentless passiveconstructions), as in the text of a memo I once received from auniversity buildings maintenance department. "In view of the factthat electric space heaters are conducive to adverse safetyconditions," the memo advised, "personnel from the PlantOperations Department will commence retrieving them. In the event thatsomeone's personal heater is collected, the owner need onlyidentify it at [...] and he/she can then take it home. Your cooperationin maintaining a safe working environment for the Campus Community willbe appreciated in this effort." (In fact, all the heaters must havebeen personal property; the university had never supplied such things.)I imagine that actually ordering faculty members to get rid of theirspace heaters might have seemed, to the writer, a potentially faultableassertion of authority that might also have called undue attention tothe buildings' inadequate heating. Also familiar, at least to historians and librarians, is thepractice of learned authors in the past who used to switch into Latinwhen they wrote about "the naughty bits" (thus possiblyencouraging generations of students to improve their competence inclassical languages). For example, in a study of Maasai language andfolklore, Alfred Hollis (1905:257) offers texts and translations ofMaasai riddles, including one translated as follows: "What is thething which hides itself in its bed? The louse which the boys uncover.Ut pulex in ruga cutis celat, sic puella in lecto juvene aggresso."(2) While it is not to be supposed that Hollis's Maasai sourceprovided the Latin--the Latin switch pertains strictly to therelationship between Hollis and his reader--in my next example, thecode-switch is attributable to quoted characters. Again, only the readerwho knows the switched-into language has access to the text. Here, thenovelist Dorothy Sayers has her characters, Lord Peter Wimsey andHarriet Vane, switch into French when in bed on their wedding night(Sayers 1937:60): He [Lord Peter] laughed suddenly. <<Enfin, du courage! Embrasse-moi, cherie. Je trouverai quandmeme le moyen de te faire plaisir. Hein? tu veux? dis done!>> << Je veux bien.>> <<Dearest!>> << Oh, Peter!>> In deference to the reader's tender sensibilities, I refrainfrom continuing the quote. In these examples, Latinate or French expressions distance theAnglophone speaker or writer from a potentially toxic referent. When thecontainer is a mask, as in the Latin renderings of the "naughtybits," one effect (often noted in the literature) is to restrictthe audience. Ways of speaking and writing that must be acquired throughlengthy formal education are available only to elites; for theelite-oriented speaker or writer, only those hearers who can be trustedto handle the information in a socially appropriate way can receive itat all. (A similar process is at work in "thieves' cant,"children's secret languages, and the like.) Moreover, talking aboutnaughty bits in this masked manner will protect speakers from any Iesemajeste, any damage to their image in the eyes of the public, whichmight ensue if it were known to all what they were speaking about. Pasdevant les domestiques--or pas devant, for short--as the WASParistocracy used to say (Friend 2009:155), where the servants werenon-initiates in the code. As Erving Goffman (1981:97-98) noted,however, even exclamations--self-talk--if taboo, are subject toself-censorship around inappropriate audiences: "a man who uttersfuck when he stumbles in a foundry is quite likely to avoid thatparticular expletive should he trip in a day-nursery." Yet, no matter what bystanders might be present, when I stumbleover a loose brick in the sidewalk I'm not likely to exclaim,"Oh defecation!" I might suppress the self-talk altogetherrather than render it in this high-register form which fails to servethe self-talk's expressive purpose. For the native speaker ofEnglish, these forms distance the speaker emotionally, in some sense,regardless of audience. Learned later in life than English, and acquiredthrough formal schooling or through special means--at least as regardsLatin and French as languages in themselves, rather thanEnglish-language borrowings--such expressions lack the immediacy ofbasic English words to the English-speaker. Some degree of emotionalintensity, of personal engagement with the messier aspects of humanlife, is missing. Similar processes might be at work in the semantic"bleaching," or "vagueness"--an abstract quality, asopposed to denotational earthiness--often found in the honorificregisters of languages where such registers are lexically based. (3) More is at issue here, however, than distancing and bleaching. Forbleaching, one could manage just as well with circumlocutions andsuper-ordinate (less concretely specific) terms, and indeed theseconstructions are common in lexical honorifics. But Anglophonespeakers' uses of Latin, Latinate euphemisms, and metropolitanFrench also draw on those forms' secondary indexicality--theassociations those words have with elite speakers and formal contexts,lending the user some upgraded aura of elitehood and conferring a bit offormality on the occasion. In the Sayers example, perhaps theelite-marking second-order indexicality of French neutralizes--in herview at any rate--some supposedly downgrading moment-of-copulationdialogue. (4) (More than one reader, however, has found her code-switchinto French merely embarrassing.) Meanwhile, the author of themaintenance memo I mentioned earlier not only evades responsibility formaking demands on faculty, but perhaps also, in deploying Latinatevocabulary and complex constructions, seeks to upgrade the discoursewhen addressing persons perceived as erudite. (5) Alternatively, the memo might be drawing on the extreme formalityof discourse in those official military, legal, or corporatebureaucratic contexts where Latinate forms and circumlocutions prevail.(6) David Crystal (1995:174) offers a list of some thirty expressionsused by businesses that had to "let people go" during the 1991recession: career change opportunity, personnel surplus reduction, andvocational relocation are examples. Most often cited to illustratebureaucratic evasiveness, such forms are not just the duplicitousstrategies of those who would pull the wool over the public's eye,but are instead more pervasive signs of an official--henceimpersonal--world. (See Jane Hill 2000, 2008 for discussion ofrelationships between this kind of "serious public discourse"and various "light" or "tough" or"plain-speaking" styles with which politicians may try toconvey personal leadership qualities, or connect with a popularaudience.) Meanwhile, footing shifts such as reported speech offer anotherpath through which unmentionables can be mentioned. In the followingWolof dialogue from a family gathering in a rural village (see Irvine1993:125), Mati, a girl aged about 11, defends herself against thecharge of having insulted an older woman. Mati repeats the insult, butshe puts it in the mouth of another child:Mati: Xanaa, yaay. Ngente noo Mati (age 11): It was like this, mama.and, dem dagai Mbajaan, di-- We [three girls: Mati, Fatu Ture, anddi fob ay matt--xam ngga ne-- Ndei] were walking together, going to the Bajaans' granary--uh, gathering firewood--you knowBB: Booba xaleu jom!--booba BB (elderly woman, related to thesabaar bi nggai wax. Bajaan family): This shameful child!-- Millet sheaves, you mean. [i.e., stealing them]SG: E, bayyi ko ba mu yeggal rek. SG (middle-aged woman, politicalNggai di naa lan? leader): Hey, just let her finish. You were saying what?Mati: Ndei ne ko, bu la Benn Sarr Mati: Ndei said to her, "if Benn Sarrjappee, di na la noq. Kii ne ko, catches you, she'll screw you [makegaana gi. Fatu Ture ne ko, gaana you sorry]." SHE said to her, "Leper."gi--ak sa tangk gu nepperel. Fatu Ture said to her, "Leper--with your stinking feet."(pause) (pause)SG: Fatu Ture mi! Lammenam SG: That Fatu Ture! Her tongue is likedefe mel ne jaasi. a sword. The other child, Fatu Ture, eventually got spanked for the offense,while Mati got off with a lecture that focused on a later moment in thereported, fault-filled episode. Mati's containment effort, framingthe insult within a quote, evidently worked, inasmuch as primaryresponsibility for dirt was displaced onto someone else. It is obvious enough to a Wolof-speaker--and it requires littleexplanation to an outsider--that calling someone a "leper withstinking feet" is insulting. (7) More subtle in its disparagementis the activity of a Barcelona talk-show host, as reported by SusanFrekko (this volume). In Barcelona, Catalan is "we-code" forethnic Catalans and Catalan nationalists, while Castilian (Spanish) is"they-code," in John Gumperz's (1982) terminology. TheCatalan talk-show host is bilingual, and she entertains calls fromCastilian-speakers as well as Catalan-speakers. But the host seldom saysanything original in Castilian. Instead, she echoes the utterances of aCastilian caller, or she quotes cliche phrases conventionally associatedwith Castilian-speakers. In this way, conspicuously quoting real orstereotyped Castilian others, the talk-show host distances herself fromthe Castilian source even while uttering Castilian words. What sheutters in Castilian is always framed as someone else's voice. Ineffect, she denies responsibility for Castilian speech and allowsCastilian voices to be disparaged. Thus in a socialscene--Barcelona--where ideologies of social group distinctness andsegregation are salient, the talk-show host's particular manner ofcode-switching creates noxiousness as an attribute of Castilian languagethat need not have been implied, had the pattern of bilingual usage beendifferent. Like a writer's use of scare quotes, the speaker'sdistancing herself from Castilian is what implies that it is"other" and defiling. Notice too that in this case it is theCastilian words, not their referents, that are problematic for theCatalan nationalist because of the Castilian social category they index. Yet, reported speech, code-switches, and euphemistic language asmeans of avoiding responsibility for talking about something problematicdon't always work. Dirt can stick; registers (and footing shifts)can leak, spilling some portion of their potentially toxic contents ontothe speaker. In the case of the Wolof child, Mati, the pause after herutterance and before SG exonerates her may hint at that possibility.Mati escapes the main responsibility for the utterance, but maybe shedoes not quite escape all of it. A clearer example is offered by laneHill, citing a Mexicano woman who explains that one good reason to speakMexicano is to recognize when she is being insulted: (8) Porque qemanian amo nicmatiz de ome, hasta nechtehuicaltiIiz in nonantzin, pero in amo nictenderoa tlen nechilia, hasta nechiliz, "Chinga tu madre," con perdon de Dios, quen nicmatiz tlen nechilia? Because if I didn't speak two languages, someone might say something about my mother, but when I don't understand what he is saying to me, even if he says to me, "Fuck your mother," begging God's pardon, how would I know what he is saying to me? As Hill (Hill and Irvine 1993:13) comments, "Here, even thoughthe obscenity is assigned to a third party, the speaker has stilluttered it in her own voice ... clearly she does not see herself asmorally neutral in this role; she must ask God's forgiveness forthe obscenity, even though it is not 'hers.'" Somethingtoxic has stuck to her, at least in God's eyes. Just to permitdisfavored words to come out of one's mouth can leave a stain. Wemay be reminded of how children were once punished for uttering"dirty" words by having their mouths washed out with soap. The trouble with relying on a footing shift, then, is that onestill utters the toxic words. Their indexical relationship with thespeaker's mouth may command too much attention for the footingstrategy, that social distancing through which responsibility for thewords is attributed to someone else, to overcome. In quoting,responsibility for authorship of noxious material is contained;responsibility for animating it, and thus allowing toxic material toenter the ongoing talk, is not. In the next example, although the tone is jocular, the indexicalconnection with a disdained speaker is what makes otherwise innocuouswords noxious. As in the Catalan case, it is words themselves ratherthan their referents that are problematic. The example comes from theletters of the poet John Keats. Writing to a close friend, Keats (1899[1820]:431) remarked: You must improve in your penmanship ... I would endeavor to give you a facsimile of your word Thistlewood if I were not minded on the instant that Lord Chesterfield has done some such thing to his son. Now I would not bathe in the same River as Lord C. though I had the upper hand of the stream. I am grieved that in writing and speaking it is necessary to make use of the same particles as he did. (Keats to Charles Wentworth Dilke, March 4, 1820) Here, the indexical connection has been described as if physicallyconcrete. The humor rests on the innocuousness of particles; presumablythose linguistic forms would have a very low rank on the five-pointRevoltingness Ratings scale devised by Keith Allan and Kate Burridge asbackground to their study of euphemism and dysphemism (Allan andBurridge 1991:52, and see their "Appendix R"). Since the scalepurports to apply to referents rather than words, it is unsurprisingthat no "particles" are to be found in Appendix R. (9) Theparticles' only noxiousness is a product of their contiguity withLord C. Ex post malefacto: Replies and responses. This review ofconventional strategies for mentioning the unmentionable throughcontaining it in register shifts, code-switches, periphrasis, andfooting shifts has been meant only as a sketch. I have not intended acomprehensive view, but instead only to glance at the window throughwhich it is seen. There are large literatures on these processes and onthe many varieties of disvalued referent or linguistic taboo they seekto mask or evade; see, for example, Hughes (2006) Encyclopedia ofSwearing, or the similarly encyclopedic works of Allan and Burridge(1991, 2006). There is even a journal, Maledicta. Some of these worksmake it clear that what is toxic at one place and time may not be toxicat another. Moreover, what is mentionable in the presence of oneaddressee or bystander may be unmentionable in the presence of another,even if it is uttered only in self-talk--as we have already seen inGoffman's example of the stumbler-over-loose-tiles. Similarly,whether ethnic jokes, even slurs, are toxic fighting words or signals ofintimacy depends, at least to some degree, on whether the speaker andaudience are composed of co-ethnic insiders or not. What these last examples suggest, as some inventories ofmalediction and Revoltingness do not (at least, not as such), is theimportance of attending to the interactional setting of usage ratherthan only the inherent toxicity of the linguistic expressionsthemselves, much less their referents. It is not just that expressionssuch as "bloody" are disfavored in Britain far more than inAmerica, or that American attitudes toward blasphemy have shifted duringthe past century or so. Important as those facts are, it is alsoimportant to note that toxicity can be created interactionally, out ofthe most ostensibly innocuous ingredients--out of anything (e.g., anyutterance in Castilian; English particles). Another example from Goffman's writings will furtherillustrate the point. In a discussion of the ways in which the meaningof an utterance is dependent not just on its own contents, and not juston whatever the speaker might have intended, but on the ways it is takenup by an interlocutor-the meaning projected back onto it, so tospeak--Goffman (1981:68-70) offers a long list of ways person A'sutterance, "Do you have the time?" might be responded to byperson B. The responses include such possibilities as "Say, are youtrying to pick me up?" or even "How much for the wholenight?" Casting the original utterance as a pickup line or even asolicitation, these responses reveal the vast potential for an aura oftoxicity to be created in interaction. In the "Do you have the time?" example, Goffman does notdiscuss whether there is a way for person A to deflect the imputationcast upon him or her, an imputation that is conspicuous even though itis not mentioned explicitly. Containment after the fact, or in responseto someone else, may require its own set of strategies. Sometimes,however, these response strategies can resemble the register shifts andcode-switches we have already considered. For example, in a recent showof the radio program "Moths," the actress Sarah Jones, who isadept at creating a wide social range of characters, told of beingstopped by police in Los Angeles shortly after her arrival in that city.(10) Evidently assuming that she was a streetwalker, the policemenchallenged her about what she was doing and where she thought she wasgoing. She was indeed walking on the street, chatting with a Latinafemale companion. What was the reason for the police challenge? Was itthat respectable people in Los Angeles don't walk, so the policecould reasonably suppose the two women were displaying poor character?Or was it that Jones was "walking while black"? At any rate,she promptly launched into a stream of high-register RP British English,haughtily protesting the outrageousness of the policemen'streatment. The policemen retired in confusion. In this case, what wascontained by Jones' register shift is a toxic implication createdby others, and it was contained after the fact, through her response. Summary of containment efforts: As we have seen, for linguistic orother semiotic material to work as a container it must offer indices ofpositive value--whether a positively-valued self, or addressee, orreferent--to contain, and insulate the speaker from, something disvaluedor dangerous that has also been semiotically indexed. Whether thecontainment effort seems to succeed, or whether it fails, or isdeliberately sabotaged, the effort must depend on a regime of semioticvalue: a regime of conceptions about how language (and other semioticform) represents; about what's toxic, and what's valuable;about what insulates, and what does not. Bear in mind, however, that although registers leak, not allleakage is necessarily undesirable. The speaker who quotes the Biblemay, in some circles, be socially elevated by the use of sacred words,even though it is clear to all that he or she is not their author.Similarly, the person who, as spirit medium, acts as conduit for someimportant supernatural being may be revered for it. A sacred voice canleak onto its animator. In a different, and non-sacred, culturalcontext, the Wolof griot who "transmits" (jottali)--speaksfor--the highest-ranking patrons ranks higher than those griots whospeak for less exalted people. There is a calculus of effects. Betweenthe container and the contained, what is upgrading (in indexical value)may counter, or supersede, or fail to insulate, the downgrading semioticforms. Finally, one should also take note of the possibility that any ofthese containment efforts might serve only to heighten the sense oftoxicity, of "ostentatious taboo," as A. W. Read (1964) calledit--a point to which I shall return. Theories of Causation, Regimes of Truth and Knowledge Not only do the English "particles" of which Keats wrotelack intrinsic toxicity, they are deployed within a culture (andideology) of language that conceives of words as "mere" form,an apparatus for conveying ideas, divorced from a world of materiality.It is not to be supposed that Keats believed uttering a particle wouldturn him into Lord C., or affect him other than through the mediation ofhis (Keats') own ideas and disgust at Lord C. Even though, as apoet, he manipulated words self-consciously, paying attention to theirphysical sounds and constructing text-artifacts with them, we need notimagine that he believed words could materially influence persons orobjects without the mediation of ideas and emotions they might call upin someone's mind. Poetry may be magical, but its effect, in hiscultural tradition at least, is not taken to be quite the same as thepractical magic of spells. In contrast, among Korowai of West Papua (Indonesia), as describedby Rupert Stasch (2008 and this issue), some words are taken to bematerially damaging. A special register, xoxulop, exists "forwrecking," and the use of a xoxulop expression in itsreferent's presence can cause that referent physical harm. So, tocall a string bag (ainop) mbam-manu-lefu ('placenta') willcause it to come apart. To call a dog "death adder" mightcause it to be bitten by one of those snakes. The damage effect istransitive: so, if someone utters the word yanop-mamunga ('humanblood') in the presence of people eating red pandanus sauce, the"wrecking" expression will damage the sauce and thereby causethe people eating it to become sick (Stasch 2008:6). Stasch notes thatit is not clear exactly what the mechanism might be through which damageoccurs, although there is some connection with cosmogonic forces,witches, emotional ambivalences and the destructiveness of the uncanny,of unheimlichkeit. As Stasch points out, the register is the inverse of a euphemisticone. Xoxulop is well suited to angry expletives and malicious talk,and--perhaps a little like the bonding effect, the insiderness, thatexpletives and taboo language can create in some Euro-American contextsamong people who use them reciprocally--xoxulop expressions cancontribute to forming social ties among persons who control them. We aretold of cases where someone helpfully "wrecks" an object thathas offended someone else, as for example when Stasch himself bumpedinto a house's protruding sticks, and a bystander sympatheticallyswore "death-platform" (the xoxulop expression for"house"). However, controlling the register also requiresknowing exactly which words to avoid, and when. Since many"wrecking" expressions are themselves ordinary words not veryhigh in any Revoltingness Rating, there is always a danger of accidents.The danger is all the greater since the connection between a xoxulopexpression and its referent is not obvious. The xoxulop expression isthe hidden name of its referent. It must be learned, if it is to beavoided in its referent's presence. Someone who does not know thehidden name, or does not realize that the referent is present--blithelytalking about death adders without realizing a dog is lying in thebackground--or does not know a particular person's hidden name andutters it in that person's presence--will cause harm. Meanwhile,those who do know the hidden names establish a special bond by jointlyavoiding them. Notice that for Korowai these hidden names are the "true"names; it is the containing expression, the means of avoidance, which is"false." The locus of truth (what is taken to be truth) inthis case differs from that in cases like the Catalan bilingual radiohost, where the speaker's "true" ethnic identity isindexed in the container--the Catalan utterances--not in the Castilianutterances contained as reported speech. The cultural comparisons thesecases might evoke are complicated, however. One case involvestruth-in-denotational-reference; the other involvestruth-in-social-indexicality (authentic social identity of speaker assignaled by code usage). Moreover, identifying the locus of"truth" does not in itself indicate whether uttering baldtruths, or "unmentionable" truths, in this or that context,can ever be valuable. Those who reject euphemism by proposing to"call a spade a spade" locate a valuable truth in thepotentially offensive (toxic) expression, not in its milquetoastsubstitute. Does it follow that euphemism is "false"? Does it dependon the way the container expression is constructed? And should the"truth" necessarily be uttered? What will uttering it do? In amiddle school science class, as described by Joel Kuipers (this issue),pupils must learn to claim they observed and analyzed some laboratoryphysical process independently of other pupils, even though they areworking together and have been encouraged to work as a team. Eachstudent keeps a lab notebook, which must assert independent analysis andmust never mention copying from another student's book, even ifthat is exactly what took place. In the science class, genre conventionstrump truth-in-reporting, even though scientific "truths" areostensibly at issue. We who compare and analyze these cases must explore the particularlocal regime of truth, the relevant genre constraints, and theethnotheories not only of evaluation but also of causation, that informtoxic talk and govern any efforts to contain the toxicity--or,contrarily, to let it fly. The study of "unmentionables" that opens a window intolocal, cultural regimes of truth also opens another, into regimes ofknowledge. Expressions that are toxic in themselves, and those that aretoxic only in relation to specifiable hearers/bystanders or settings,must be learned, along with their relations to persons and places. Astranger who enters a Korowai household and innocently utters the word"five," not realizing that its homonym (the word for a speciesof turtle) is the hidden name of a member of the household, is notassumed to be acting maliciously, despite the damaging act (Stasch thisissue). Would the act be seen as more damaging if it were clear that theutterer knew what he or she was doing? Cases like these show theimportance of personal biographies and what a speaker's audienceknows of his or her background and experiences. Consider, in this light, the Sarah Jones example in which theactress countered a police challenge in Los Angeles. The containmenteffect of ]ones' register-switch hinges on the problematicattribution of a "true" social identity to a speaker, from thesemiotic material available in an encounter between strangers. It is notclear from the actress's story whether the policemen had heard herspeak to her companion before she shifted to RP British English. Whetherthey had or not, the story as she told it implies that they took RP toindex her "true" respectability, as opposed to whateversemiotic evidence they were previously relying upon. As another example,recall how the Catalan radio talk-show host makes it clear that she is"really" Catalan. She establishes lengthy streams of creativeCatalan talk while limiting the Castilian to conspicuously quotativematerial, thus containing it and containing any hint that she herself isnot authentically Catalan. Containment efforts depend crucially,therefore, on being able to establish a "true,""real" persona that one's interlocutors will take as theone that counts, the one that perdures over longer stretches ofbiographical time. (11) In these two cases (Sarah ]ones and the Catalan talk-show host),the interlocutors were strangers who could not know more about ourprotagonists' biographies than what could be inferred from theirperformance at the time. Among Korowai, too, much rests on aperson's opportunity to acquire knowledge, in their case knowledgeof specific linguistic facts, the words that serve as hidden names forthings other than their more commonly-known referents. Any accidentalusages of hidden names would result from a speaker's ignorance ofthat particular lexical item's esoteric reference. Now, contrastthese cases with those that concern knowledge of wordsthat are unmentionable because they are taboo, not because they arehidden. If we are scolded for using four-letter words in front ofchildren, it is not so much because we might lose status in thechildren's eyes as because children's acquaintance with thosewords should be delayed. Their use conflicts with a cultural image ofchildhood innocence. If a child uses them herself, she not only revealsher willingness to indulge in "vulgar" talk but also betraysher knowledge of its forms--hence acquaintance with the people andsettings that are those words' habitat. ("Where did you learnthose words? That's the way dirty old men talk," I once hearda woman admonish a six- or seven-year-old girl.) In short, a"regime of knowledge" concerns not only what an interlocutorcan infer a speaker knows, but what that speaker ought to know. The traces of such indexical links persist in time, for linguisticusage can reveal knowledge that can only have come from pastcontiguities. No footing shift can protect against the brute fact oftoxic talk emerging from a speaker's mouth and revealing thespeaker's familiarity with it. The same is true if"vulgar" usages emerge from someone's pen. In this regard a literary example may again be convenient: thenovelist who creates dialogue appropriate to the characters in a story.Charles Dickens has often been cited for the wide range ofsociolinguistic variation, and the relative subtlety of itsrepresentation, displayed in the speech of his novels' characters,including speech deemed "uncouth." (12) While this novelisticskill was recognized in Dickens' work early on, it was a skill notwithout its hazards for a writer claiming respectable social status forhimself. For if writers succeeded in representing details of"low" characters' speech or "uncouth" topics,they risked the imputation of vulgarity. How could they representuncouth speech so well, unless they were intimately acquainted with suchpeople and their vulgar talk? This is a problem taken up by an early reviewer of The PickwickPapers. "Anonymous" wrote: ... Mr. Dickens may well afford to disregard the imputation of vulgarity, invariably and indiscriminately leveled by the tawdry affecters of gentility against every man of genius who ventures to take human life, in all its gradations, for his subject-matter.... Mr. Dickens ... moves as naturally and easily amongst his favourite characters, as if there were no such things as conventional proprieties to offend against, and no more dreams of being accused of coarseness for the appropriate and idiomatic language he places in their mouths, than a Wilson or a Gainsborough would dream of incurring such a penalty for placing a pigstye in a landscape. (Anonymous 1837:506-7) "Anonymous" adds that earlier novelists eitherrepresented stage-coach passengers and the like less well, or were moreheavily reproached for their efforts: Fielding has a chapter headed: 'Of many things natural, but low,' and his novels teem with allusions to this class of objectors, well personified in his Slipslops and Graveairs, who shrink instinctively from contact with stage-coach passengers, and have 'low,' and 'low creatures,' eternally in their mouths. Fielding, however, evidently writhed under the reproach; and few modern writers have attempted the class of subjects which exposed him to it, without ever and anon giving the reader to understand that in so doing they are descending from their sphere ... (Anonymous 1837:506-7) What accounts for the difference "Anonymous" has pointedout? Almost a century had passed between the publication ofFielding's novels and The Pickwick Papers. Class relations inBritain had changed, and sociolinguistic norms along with them. Themercantile and professional classes were expanding; literacy and thedistribution of literature were spreading dramatically. It is in thecontext of those social changes that the differing social positions ofFielding (and various other 18th century writers) and Dickens could bebetter assessed. Perhaps there is something relevant in the differencebetween the two writers' social backgrounds--Dickens could belocated in an urban middle class, while Fielding and some others camefrom the landed gentry and wrote at a time when urban professionalclasses were less well established. Probably more important, however,are the novelists' differing ways of representing "low"language and the "higher" varieties with which it contrasts.(13) The sociolinguistic, historical, and literary situation of thesenovelists is more complicated than I can address in a short space.Nevertheless "Anonymous's" comments indicate that in 1837a writer still ran the risk that his characters'"vulgarity" would leak onto him, but that Dickens managed toescape damage. Eight-Hundred-Pound Gorillas and Other Objects of Avoidance If registers are prone to leakage, and if masking and framing arerisky strategies for containing noxious linguistic material regardlessof the particular nature of the noxiousness, what else isleft?--Avoiding the unmentionable altogether. But when avoidance isconspicuous, the unmentionable is implicated. The way we tiptoe aroundthe 800-pound gorilla in the room can be precisely the way weacknowledge, and reveal, its presence. (14) Allen Read's (1964) paper on "ostentatious taboo"points out some ways in which thin veiling of a topic, or evading it bysubstituting "you know what" (or "who" or"where"), can serve the mischievous purpose of ostentatiouslydrawing attention to the very topic that has been evaded. "Suchostentatious taboo," he remarks, "is usually accompanied bywell-known paralinguistic features--smirking, the arched eyebrow, aslyness of manner" (Read 1964:162). For Read, who tells us he hadcollected examples from newspapers and magazines for twenty years,ostentatious taboo generally arises in contexts of cultural andemotional ambivalence. Sex and scatology are obvious evasion-inducingtopics, but politics and other social issues offer examples as well.From Cold War tensions with Russia comes satirical avoidance of wordslike red and revolution: thus, Read reports, a journalist has told us(New Republic, 4 August 1947, p. 2) that "laws are being drafted'to protect Americans from exposure to unavoidableinfra-you-know-what-rays.' Another journalist (Joseph Barry, NewYork Post, 27 June 1960, 24/1) made reference to 'the DAR(Daughters of You-Know-What)'" (Read 1964:162). In a moreextended example, Read quotes Jessica Mitford, writing about herfamily's visit to Constantinople: There we were conducted through the palace, where one of the 'sights' was the last remaining eunuch, a tiny, wizened old man with a face like a dried-up apple and a high squeaky, grumbly voice. That night Muv told me to summon Boud and Debo to the cabin; there was something very serious she wanted to discuss with us. From her stonily solemn face and tone of voice, I could only assume that she had heard some bad news from home--perhaps a death in the family-and my heart pounded with real fear as I went in search of the others. When we were assembled, Muv announced in her very gravest tones: 'Now, children, YOU ARE NOT TO MENTION THAT EUNUCH AT DINNER.' We howled and screamed with laughter, and, although we would not have dared actually to disobey, kept referring all through dinner to the 'you know what' with knowing looks and suppressed mirth. (Mitford 1960:88; cited in Read 1964:166) (I omit consideration of the Mitfords' possible insensitivityto the "last remaining eunuch's" personal situation, andthe possible difficulties of being one of the "sights" fortourists.) Clearly, the expression "you know what" is appropriateinasmuch as the transgressive effect, and the ostentatiousness of theavoidance, rests on knowledge the participants hold in common. Considernow the kinds of knowledge that are relevant to a different kind of800-pound gorilla, the one evaded in an example offered by Paul Grice ina discussion of flouted conversational maxims. To illustrate a floutingof the maxim of Quantity, Grice proposes the following hypotheticalsituation: A is writing a testimonial about a pupil who is a candidate for a philosophy job, and his letter reads as follows: 'Dear Sir, Mr. X's command of English is excellent, and his attendance at tutorials has been regular. Yours, etc.' ... [A] must ... be wishing to impart information that he is reluctant to write down. This supposition is tenable only if he thinks Mr. X is no good at philosophy. ((;rice 1975:33) By violating (flouting) the maxim of Quantity, A creates a specialeffect. Information about A's opinion is conveyed without everbeing explicitly stated. Those of us who have read many a recommendation letter cansupplement Grice's account by pointing to the cultural knowledge onwhich the flouting depends. If A intends to convey, by conspicuousomission, the information that Mr. X lacks crucial qualifications forthe position, he has to assume that the reader shares a sufficient senseof the recommendation-letter genre to know what is missing. And thereader, in turn, has to decide whether A himself is likely to shareknowledge of that piece of academic culture. If I receive arecommendation letter from a Mr. A in Tokyo concerning a job candidatefor a position in my institution in the US, can I assume Mr. A shares myunderstandings of what an effective recommendation letter shouldcontain? If the letter is only half a page long, can I infer that Mr. Ais flouting the maxim of Quantity, as I would be doing if I wrote such aletter? And if the letter is favorable, can I infer that it representsMr. A's "true" feelings about Mr. X, rather than someother motive, or a different sense of the conventions of the genre?Unless I know something about Mr. A and his experience of the Americanacademic world, I cannot. So the possibility of conveying unmentionablenegatives (about Mr. X) by omission, though real, depends on what theinterlocutors know about genre and other cultural conventions, and whatthey know about one another. Knowing what one's interlocutor has knowledge of--prerequisitefor gorilla-avoiding--can be difficult. The following example attests tothe indeterminacies inherent in tiptoeing around the gorilla andwondering whether others realize it is there. Recently I had lunch witha younger woman friend. After a lull in the conversation she suddenlyasked, "How old were you when you got divorced?" A pause--Ihesitated, not from offense but from surprise, since the question seemedto come out of the blue--and said, "Oh, I was thirty-six or so, whydo you ask?" (Many possibilities went through my mind.) Anotherpause. She said something like, "Well, it just seems likethat's about the age, late thirties, early forties, when thesethings happen ... " I agreed that it's common; and the topicchanged. What I didn't mention was that a mutual friend had told mejust a few minutes earlier that she (Mutual Friend) and her husband wereseparating, but that the news was not yet public. Did my lunch companionknow? At the time, I inferred that she probably did--that this was the800-pound gorilla we were both avoiding. Certainly I considered thetopic unmentionable, in case she did not yet know. Our conversationimplied, to me at the time, that she probably knew about our mutualfriend and felt equal reluctance to mention it overtly. Yet it ispossible that she was avoiding a different gorilla altogether. Conclusion Much of the literature on "verbal taboo," from classicworks like Edmund Leach's (1964)"Animal categories and verbalabuse" to more recent offerings like Erin McKean's (2001)Verbatim (see the section entitled "Not in polite company"),and even much of the material in Allan and Burridge's (1991, 2006)books, has focused on the denotational semantics of taboo language andthe "Revoltingness" or insultingness of the things denoted. Inmany of these works, a broad historical or cultural context enters intothe analysis largely as a way to explain why the assemblage ofdisfavored things can differ from one place, or one historical era, toanother. (15) Even Read's (1964) paper on "ostentatioustaboo" devotes more attention to the range of topics tabooed thanto the audience effect of ostentatious usage, although it is the latterconcern that struck me as more original and interesting. While usefulfor some purposes, denotational semantics is a narrow lens through whichto look at verbal taboo. A broader view can be found in some works that focus on politenessor on sarcasm. John Haiman's Talk Is Cheap: Sarcasm, ,Alienation,and the Evolution of Language (1998) opens up the discussion in severaldirections. Phonology and intonation, for example, offer devices forturning ostensibly positive assessments into sarcastic--butdeniable--deprecations. Thus a cartoon of Garfield the cat, reproducedby Haiman (1998:34), shows Garfield standing on a bathroom scalethinking, "I've been dieting." The scale says,"Riiiiight." Garfield: "No, really. Is this a face thatwould lie?" Scale: "How should I know? I've never seenit." (16) A chapter on "Un-plain speaking" explores otherkinds of metamessages, including various forms of euphemism, but ranginginto special registers (such as "gayspeak") and rituallanguage. Haiman's emphasis is on a notion of the divided self andthe ways a speaker can be self-consciously alienated from "theactual referential content of his or her message" (1998:10).Meanwhile, the well-known work on politeness by Penelope Brown andStephen Levinson (1978, 1987) shifts the spotlight onto thespeaker's intention in interaction, that is, the linguistic devicesthat allow speakers to show respect for an interlocutor's"face" or fail to do so. Like Haiman's analysis, however,Brown and Levinson's centers on the linguistic properties of thespeaker's utterance. "Politeness" or its lack is readdirectly off the linguistic form. These are a scant few of the works in a very large literature, andthere are many others that are worthy of discussion were space to permitdoing so. (But, f***, I've got to conclude.) What I have tried todo in this paper is to widen the lens still further. The moral life oflanguage does not reside in the linguistic properties of utterancesalone, nor only in the moment of interaction. The words not spoken, thediscourse contexts, the interactional and societal histories, theresponses by interlocutors, the conventions of genre, the regimes oflanguage, truth, and knowledge that prevail in the interlocutors'social worlds--all these are relevant as well. Multiple contingenteffects are available through entailments, implicatures, interdiscursiverelations, and second-order indexicalities, and while some of theseeffects can be drawn upon for strategic and even mischievous reasons,not all can be foreseen. Registers leak; gorillas lurk; and both the leaking and the lurkingreach far into the social dramas where some matters are unmentionableand people talk about them anyway. ACKNOWLEDGEMENTS Special thanks are due to Michael Lempert for helpful comments onthis paper. The flaws that remain are my responsibility, not his. Thanksalso to Dorothy D. Wills, who provided the recording and transcript fromwhich the Wolof example in this paper is drawn. Dr. Wills, who wasconducting research in a nearby town, visited the community where I wasworking and provided me with copies of her recordings there. REFERENCES Agha, Asif. 1993. "Grammatical and Indexical Convention inHonorific Discourse." Journal of Linguistic Anthropology 3(2):1-33. --. 1994. "Honorification." Annual Review of Anthropology23:277-302. --. 2007. Language and Social Relations. Cambridge: CambridgeUniversity Press. Allan, Keith and Kate Burridge. 1991. Euphemism and Dysphemism:Language Used as Shield and Weapon. New York, Oxford: Oxford UniversityPress. --. 2006. Forbidden Words: Taboo and the Censoring of Language.Cambridge: Cambridge University Press. Anonymous. 1837. [Review of] 1. The Posthumous Papers of thePickwick Club; containing a faithful Record of the Perambulations,Perils, Travels, Adventures, and Sporting Transactions of thecorresponding Members. Edited by 'Boz.' With Illustrations,Nos. I to XVII. 2. Sketches by Boz: illustrative of Every-day Life--andEvery-day People. The Third Edition; in 2 vols. London. 1837. 3.Sketches by Boz, &c. The Second Series. Second Edition, London.1837. Quarterly Review (London) LIX:484-518. Bakhtin, Mikhail. 1981. The Dialogic Imagination. Michael Holquist,ed. Caryl Emerson and Michael Holquist, trans. Austin, TX: University ofTexas Press. Brown, Penelope and Stephen Levinson. 1978. "Universals inlanguage usage: Politeness phenomena." In Esther Goody, ed.Questions and Politeness, 56-289. Cambridge: Cambridge University Press. --. 1987. Politeness: Some Universals in Language Usage. Cambridge:Cambridge University Press. Crystal, David. 1995. The Cambridge Encyclopedia of the EnglishLanguage. Cambridge: Cambridge University Press. Dixon, Robert M. W. 1971. "A method of semanticdescription." In Danny Steinberg and Leon Jakobovits, eds.Semantics, 436-471. Cambridge: Cambridge University Press. --. 1980. The Languages of Australia. Cambridge: CambridgeUniversity Press. Fowler, H. W. 1940 [1926]. A Dictionary of Modern English Usage.London: Oxford University Press. Friend, Tad. 2009. Cheerful Money. Boston: Little Brown. Goffman, Erving. 1981. Forms of Talk. Philadelphia: University ofPennsylvania Press. Gorlach, Manfred. 1999. English in Nineteenth-Century England.Cambridge: Cambridge University Press. Grice, Paul. 1975. "Logic and conversation." In PeterCole and Jerry Morgan, eds. Syntax and Semantics 3: Speech Acts, 22-40.New York: Academic Press. Gumperz, John J. 1982. Discourse Strategies. Cambridge: CambridgeUniversity Press. Haiman, John. 1998. Talk Is Cheap: Sarcasm, Alienation, and theEvolution of Language. New York: Oxford University Press. Hill, Jane. 2000. " 'Read My Article': Ideologicalcomplexity and the overdetermination of promising in AmericanPresidential politics." In Paul Kroskrity, ed. Regimes of Language,259-291. Santa Fe: School of American Research Press. --. 2008. The Everyday Language of White Racism. Chichester, UK:Wiley-Blackwell. Hill, Jane and Judith T. Irvine. 1993. "Introduction." InJ. Hill & J. T. Irvine, eds. Responsibility and Evidence in OralDiscourse, 1-23. Cambridge: Cambridge University Press. Hollis, Alfred C. 1905. The Masai: Their Language and Folklore.Oxford: Clarendon. Hughes, Geoffrey. 2006. An Encyclopedia of Swearing: The SocialHistory of Oaths, Profanity, Foul Language, and Ethnic Slurs in theEnglish-Speaking World. Armonk NY: M. E. Sharpe. Irvine, Judith T. 1993. "Insult and responsibility: Verbalabuse in a Wolof village." In Jane Hill and Judith T. Irvine, eds.Responsibility and Evidence in Oral Discourse, 105-134. Cambridge:Cambridge University Press. --. 1995. "Honorifics." In Jan Blommaert, JefVerschueren, and Jan-Olaf Ostman, ads. Handbook of Pragmatics, 1-22.Amsterdam: John Benjamins. (Reprinted 2009 in Gunter Senft, Jan-OleOstman, Jef Verschueren, eds. Culture and Language Use. Handbook ofPragmatics Highlights 2. Amsterdam: John Benjamins.) --. 1998. "Ideologies of honorific language." In BambiSchieffelin, Kathryn Woolard and Paul Kroskrity, ads. LanguageIdeologies: Practice and Theory, 51-67. New York: Oxford UniversityPress. Keats, John. 1899 [1820]. "Letter to Charles Wentworth Dilke,March 4, 1820." In Horace Scudder, ed. Complete Poetical Works ofKeats, 431. Boston: Houghton Mifflin. Leach, Edmund. 1964. "Animal categories and verbalabuse." In Eric Lenneberg, ed. New Directions in the Study ofLanguage, 23-64. Cambridge, MA: MIT Press. McKean, Erin. 2001. Verbatim: From the Bawdy to the Sublime, theBest Writing on Language for Ward Lovers, Grammar Mavens, and ArmchairLinguists. Orlando, FL: Harcourt. Mitford, Jessica. 1960. Hans and Rebels. London: Victor Gollancz. Mugglestone, Lynda. 1995. "Talking Proper:" The Rise ofAccent as Social Symbol. Oxford: Clarendon Press. Perrino, Sabina. 2008. "Code as containment in Italy'snew multiculturalism." Paper presented at the 107th Annual Meetingof the American Anthropological Association, San Francisco, November 20,2008. Read, Allen Walker. 1964. "A type of ostentatious taboo."Language 40:162-166. Sayers, Dorothy L. 1937. Busman's Honeymoon: A Love Story withDetective Interruptions. New York: Harper & Row. Simons, Daniel J. and Christopher Chabris. 1999. "Gorillas inour midst: Sustained inattentional blindness for dynamic events."Perception 28:1059-1074. Stasch, Rupert. 2008. "Referent-wrecking in Korowai: A NewGuinea abuse register as ethnosemiotic protest." Language inSociety 37:1-25. Judith T. Irvine University of Michigan ENDNOTES (1) See discussion later in this paper of a Catalan-speaker'suse of Castilian as always already "other." See also thediscussion in Irvine (1993) of Wolof griots, whose position and role ina social hierarchy identifies them as people who transmit otherpeople's words. In the Wolof community from which the materials inthat paper were gathered, high-ranking people could hire griots todeliver insult messages, masking the insult's origin via aperson-switch rather than a code-switch per se (although special genresalso enter in). (2) My attention was drawn to Hollis as a source for this kind ofpractice by a citation in Allan and Burridge (1991:19), where anotherexample from Hollis's work is mentioned: "The story of the skyand the earth. We understand that the sky once married the earth. Haecverba dicere volunt. Ut maritus supra feminam in coitione iacet, siccoelum supra terram. Ubi lucet sol et cadit imber, terra calorem recipitet humorem: non aliter femina hominis semine fruitur." (Hollis1905:279. Boldface original.) (3) See Dixon 1971, 1980:64; Irvine 1995, 1998."Vagueness" and "bleaching" refer to the fact thatspecial honorific lexemes are often in a one-to-many relationship withnon-honorific counterparts, and so denote a semantic category ratherthan its particular members. However, the construction and deployment ofhonorific forms and their ideological crystallization into registers iscomplex, and their analysis requires attention to more than just acomparison of lists of lexemes. See Agha 1993, 1994, and 2007:chapter 7. (4) In her early novels about Lord Peter, Sayers presents him asvirtually a caricature of the aristocrat, and one cannot take any of herrepresentations of him as devoid of an element of parody. Yet, the laternovels seem to bring him into fuller realization, and in Busman'sHoneymoon, she conveys (to this reader, anyway) an impression ofauthorial identification with the bride, Harriet Vane. (5) Notice that the values at stake in these examples are actuallymore complex than points on a unilinear scale of "raising" and"lowering." Even if some participants may see French/Latin asnecessarily upgrading and copulation-talk downgrading, others do not.Notice that the space-heater memo's Latinate vocabulary only makesit an object of faculty derision, and also, that a symmetrical exchangeof the F-word can mark solidarity among young men (though this practicemight best be considered a counter-norm). (6) Official statements by government and industry routinely scorehigh in the context for the Doublespeak Awards offered by the NationalCouncil of the Teachers of English. In 1984, for example, the USDepartment of State won the award for announcing that its reports on thestatus of human rights in other countries would henceforward replace theword killing with unlawful or arbitrary deprivation of life. (Cited inCrystal 1995:176.) (7) The insult is greater than it might seem to a Western eye,however, since a traditional Wolof cosmology links lepers with witches. (8) Hill and Irvine 1993:13. The passage comes from Hill'sfield materials and the commentary is hers. (9) Of course the Revoltingness questionnaire, distributed(we're told) among staff and students in two Australianuniversities, had to be presented using words. Since some of thebody-part words on the questionnaire have less technical, morevernacular synonyms, it is possible that the Revoltingness of those bodyparts might be differently assessed had those synonyms been used. Whatwe see on the questionnaire, then, in addition to the findings itreveals, is that some words are too taboo to be represented on such adocument. (10) Broadcast on Moths Radio Hour 5 (September 5, 2009). (11) Similarly, Perrino (2008) describes the use of Veneto dialectby medical personnel, to create an authentic, locally grounded socialpersona in contrast with the standard Italian with which the samespeakers index the institutional setting of the state-funded hospital. (12) See, e.g., Lynda Mugglestone (1995:217): "authors such asDickens ... would ... scarcely have been satisfied with anything lessthan a full representation of such 'uncouthness,' using allthe conventions at their disposal." See also Bakhtin (1981:301 ff.)writing on heteroglossia in the novel. (13) As has been noted, e.g., by Gorlach (1999), Dickens'sstrategies for representing working people's speech, and forcontaining it, were not the same as his predecessors' strategies.Dickens represents working-class speech in much more phonetic detail,and in contrast with protagonists and a narrating voice that maintains astandard. Fielding focuses more on "low" characters'illiteracy and ignorance, and represents their speech in less phoneticdetail. Moreover, other authors to whom "Anonymous" alludesmake explicit statements about the vulgarity of their characters, asDickens does not. On this point Fowler's Dictionary of ModernEnglish Usage (1940 [1926]) concurs with Anonymous. Writers who need totell the reader explicitly that they have not been defiled by thevulgarity they have represented--the pitch they have touched--probablyare not so superior to start with (Fowler 1940 [1926]:585). (14) Of course we also provide for the possibility that someparticipant(s) may not have noticed the gorilla at all. Recall the nowfamous experiment in which Daniel Simons and Christopher Chabris foundthat some subjects, given a task in which they were to pay attention toa video of a basketball game, failed to notice a woman wearing afull-body gorilla suit walk slowly into the middle of the frame andpound her chest. (Simons and Chabris 1999). (15) To be fair, Allan & Burridge actually do more than this intheir work, exploring some of the linguistic forms euphemism can take,and including a chapter on "Distinguishing between jargon, style,and X-phemism" in their 1991 book, as well as similar discussionsin their 2006 publication. In 2006 they include some treatment ofeuphemism and dysphemism in interaction and in self-censorship. Bothworks are mainly devoted, however, to exploring the kinds of referentsthat are subject to taboo and the lexical euphemisms that substitute forthem. (16) Garfield, by Jim Davis, [c]Paws, Inc.; distributed byUniversal Press Syndicate.