can you do in text citation?

Sentences in the Language Prison: The Rhetorical Structuring of an American Language

Policy Debate

Author(s): Kathryn A. Woolard

Source: American Ethnologist, Vol. 16, No. 2 (May, 1989), pp. 268-278

Published by: Wiley on behalf of the American Anthropological Association

Stable URL: https://www.jstor.org/stable/645002

Accessed: 24-09-2018 16:28 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide

range of content in a trusted digital archive. We use information technology and tools to increase productivity and

facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at

https://about.jstor.org/terms

American Anthropological Association, Wiley are collaborating with JSTOR to digitize,

preserve and extend access to American Ethnologist

This content downloaded from 140.211.127.19 on Mon, 24 Sep 2018 16:28:52 UTC All use subject to https://about.jstor.org/terms

sentences in the language prison: the rhetorical structuring of an American language policy debate

KATHRYN A. WOOLARD- University of Wisconsin-Madison

A movement against bilingualism and for "English only" in the public sphere has been grow- ing in the United States for the past several years.' Manifestations include the organization "U.S. English" and campaigns it has led against multilingual elections and in support of the establishment of English as the official language at state and federal levels. From 1983 to 1988, the number of states designating English as the sole official language multiplied from three to seventeen, and bills to make English official came before 19 state legislatures as well as the U.S.

Congress in 1988. California is one state where such efforts have been markedly successful. In the elections of

November 1986, a striking 73 percent of the voters approved Proposition 63, making English the official language of the state and ordering the legislature to "make no law which diminishes or ignores the role of English." Proposition 63 was not California's first experience with an English-only campaign. In 1984, the state passed Proposition 38, protesting federally mandated multilingual election materials and voting procedures. That in turn was an echo of San Fran- cisco's Proposition O, a successful 1983 ballot initiative to abolish multilingual elections that inaugurated the current campaign against official uses of languages other than English, influ- encing the course of events in other states as well as California.

These English-only activities respond to a trend toward governmental support and use of non-

English languages that began in the 1960s and peaked in the 1970s, when the Lau versus Ni- chols Supreme Court decision was interpreted as virtually mandating bilingual education, and federal civil rights legislation established bi- or multilingual election practices in many areas of

the country. A number of states authorized even more extensive public uses of minority lan- guages. Such acts were viewed by some as "part of the ethos and political attitudes of contem- porary America. The 'melting pot' notion has given way to a greater tolerance of ethnic and linguistic diversity" (O'Barr 1981:387).

As a popular reaction to a changing political ethos, the English-only campaigns seem to be a textbook case of status politics, a political movement in which a once-dominant social group,

perceiving its cultural values as dishonored by social change and rejected by other groups,

The anti-bilingual movement in the United States is a status movement, but a sim- ple "status politics" analysis does not fully capture its dynamics. Such movements are neither homogeneous nor merely reflective of social structure, but express in- ternal ambivalence and organize sociopolitical change through rhetorical pro- cesses. I analyze texts produced in San Francisco's 1983 campaign against bilin- gual ballots in order to account for the initiative's broad appeal. Rhetorical tropes and recurring themes shift the target of the campaign to aspects of political process of which a wide spectrum of Americans are suspicious, projecting fears onto out- siders and constructing a familiarly benevolent meaning for the initiative. [Amer- ican politics, language policy, status movements, discourse analysis, ideology]

268 american ethnologist

This content downloaded from 140.211.127.19 on Mon, 24 Sep 2018 16:28:52 UTC All use subject to https://about.jstor.org/terms

seeks to reaffirm symbolically its declining prestige (Gusfield 1986; see also Hofstadter 1955). Governmental action is seen as having the power to confirm or degrade the group's status in society. Making an issue of a cultural practice that summarizes their life style and distinguishes

them from the others who reject it, partisans struggle for institutional endorsement of their val-

ues. Whatever the practical efficacy of resulting legislation, status gains come in the symbolic public affirmation of the group's mores.

From this analytical perspective, English-only legislation can be seen as a means-in and of itself, not simply as an instrument of further action-of elevating Anglo-Americans and English

speakers, and of publicly, ritually degrading immigrants and carriers of minority cultures who

are believed to reject and thus threaten the Anglo-American way of life. Such an interpretation

makes particular sense in light of the often-repeated complaint that "my immigrant grandfather

learned English and made it without any special help. Why can't they?" Perceived retention of native languages by newer immigrants is even more of a reproach to the assimilated than to the autochthonous English speaker.

However, as Gusfield has recently suggested, the interpretation of such political movements as symbolic dramatizations of status conflicts can be overly simplistic, missing important as- pects of political and cultural process. Gusfield criticizes his own early work on the American Temperance movement for presenting cultural groups as given, unchanging and monolithic, and political conflict over status as "two football teams rushing toward each other at the scrim-

mage line" (Gusfield 1986:191). These concerns are echoed in Sally Falk Moore's recent com- ment that an event is not necessarily best understood as the reenactment of known drama, or

the exemplification of an extant social order. Multiple voices of contestation and a multiplicity of meanings figure in social process (Moore 1987:729).

Building on these observations, I argue in this paper, first, that politico-cultural groupings are

to a great extent created rather than merely reflected in status movements like the English-only

campaign. When such movements turn to electoral politics, coalitions are built out of groups with disparate interests through the linking of ideological themes. Political rhetoric weaves to-

gether diverse concerns into cultural frames that are viewed as legitimate by, and thus help to constitute, electoral coalitions (Hunter 1987:99). Second, I argue that status politics expresses internal ambivalence about norms and institutions as much as external conflict over them. Part

of the appeal of such movements may be their capacity to assuage doubts by projecting sus- pected negative aspects of social life onto outsiders.2

I will develop these two points in relation to the English-only movement through the exam- ination of rhetorical processes in one particular campaign, San Francisco's Proposition O of 1983. In his analysis of "mythical realities" in Hawaii, Marshall Sahlins (1981) has argued that historical change is organized by "structures of significance," but that the use of an existing meaning structure to absorb a new political development in turn introduces innovations in the

structure itself. I will argue that electoral success for the English-only position was enabled by recourse not just to Anglophile and/or xenophobic sentiment, but to traditional structures of

significance in American political culture previously associated with liberal support of minor- ities. Some voters for Proposition O may have been motivated by concerns other than the de-

fense of the mainstream English-speaking way of life, but these were tied successfully to the language issue.3 The use of traditional themes in this context by a new faction altered their meaning at the same time as it organized political change.

The conjoining of cultural themes in ideological schemata that authorize political action is a modern kind of mythmaking, in which a second-order semiological system is created from

existing cultural signs (Barthes 1972:114). The present analysis begins with discourse analytic techniques to identify recurrent themes in the campaign texts that serve as semiotic building blocks (Agar 1983; see Bilmes 1981; Quinn 1986 for related approaches). Largely following Barthes, I examine how unexpressed premises and transformative tropes such as metaphor and metonym link these themes in apparently natural causal relations, providing the ideological

sentences in the language prison 269

This content downloaded from 140.211.127.19 on Mon, 24 Sep 2018 16:28:52 UTC All use subject to https://about.jstor.org/terms

framework for voter alliances. As suggested by Bourdieu in his analysis of orthodoxy, I argue that this rhetorical naturalization of politics "straightens" the image of the world and returns it

to a state of innocence (Bourdieu 1977). The rhetoric of Proposition O projects onto ethnic minorities enduring American fears not just about ethnolinguistic threats but about other dark

corners of political life, as identified by Bellah et al. (1986), and restores to innocence the elec- toral process in which supporters participate.

San Francisco's Proposition 0

San Francisco enjoys a reputation as "not only the most unconventional city in America, but also one of the most tolerant, most diverse, and most democratic" (Fosburgh 1984). For this reason many people are taken by surprise when they learn that the voters of San Francisco were

among the first to approve one of the new generation of anti-bilingual measures. Proposition O directed city officials to urge the federal government to amend the federal Voting Rights Act

so that the city and county of San Francisco would no longer be required to provide election materials in any language other than English. Although 62 percent of the voters supported Prop-

osition 0, the referendum had no immediate practical effect, since federal legislation lies well

beyond the purview of the San Francisco electorate. Nonetheless, its overwhelming approval was a significant act in a city with a more than 40 percent minority population, which is cele- brated and celebrates itself for its cultural diversity.

The events that led to Proposition O began at least as early as 1965. The federal Voting Rights

Act of that year aimed to eliminate abuses at the polls that had served as barriers to black par- ticipation in the voting process for years. To this end, the 1965 act banned the discriminatory use of "tests or devices" such as literacy tests as prerequisites to registration or voting. The law

also specified that citizens educated in another language in U.S. public schools (that is, Spanish in Puerto Rico), could not be denied the right to vote because of lack of proficiency in English (U.S. Public Law 89-110).

The Voting Rights Act was renewed and amended in 1975, extending its protection to "lan-

guage minority groups." The definition of "devices" banned under particular conditions was broadened to include monolingual English ballots for minority-language voters. The new law

specified that wherever 5 percent of the voting-age citizens of a political subdivision were mem- bers of a "single language minority," and where the illiteracy rate of this language minority

group was higher than the national average, election materials were to be provided in the group's language until August, 1985.4 Congress set forth its reasoning on this point in the text of the law, finding that voting discrimination against language minority citizens is pervasive and national in scope, that such minorities have been denied equal educational opportunities by state and local governments, resulting in severe disabilities in English, and that the practice of conducting elections only in English therefore excludes language minority citizens from par-

ticipating in the electoral process (U.S. Public Law 94-73, Section 4[f][1]). In 1976, the Bureau of the Census released its identification of counties subject to the new

requirement. Of a total 495,099 voting-age citizens in San Francisco, 5.1 percent were Chinese, of whom 15.7 percent were considered illiterate, and 8.9 percent were of Spanish heritage, of whom 6.7 percent were illiterate. Trilingual election materials-registration forms, ballots, and Voter Information Pamphlets-were thus mandated for San Francisco at least until

1985.5 Charges of noncompliance filed in 1978 led to a consent decree monitored by the Fed- eral District Court, under which multilingual elections were fully implemented.

The expressed purpose of the ballot initiative that came to be labeled Proposition O was to

urge the abolition of the federally mandated multilingual elections. The petition drive for the initiative was begun in July, 1983, by city supervisor Quentin Kopp, a conservative Democrat with a constituency in the white, upper middle-class areas of the city. The recently established

270 american ethnologist

This content downloaded from 140.211.127.19 on Mon, 24 Sep 2018 16:28:52 UTC All use subject to https://about.jstor.org/terms

West Coast branch of "U.S. English" quickly joined forces with Kopp. Working primarily at shopping centers in white, middle-class areas, the organization garnered 14,400 signatures in less than a month. The Registrar of Voters certified the validity of 12,400 of these signatures, well above the 9679 needed to place the measure on the ballot.

The Committee for Ballots in English, which orchestrated the Prop O campaign, and the co- alition of minority and civil liberties organizations that opposed it each had budgets under $30,000-small by usual campaign standards. Although both sides used direct mailings and phone banks targeted at areas of likely support, the limited budgets led them to rely primarily on the free media and on public meetings to get their messages across.

Only 45.9 percent of registered voters went to the polls. Although the mayor was up for reelection, there were no serious challengers, and other offices were not hotly contested. Prop-

osition O and an antismoking initiative drew the most attention; 92 percent of those attending the polls cast votes on Prop 0, and the measure was approved by 62 percent of the voters.

It was not surprising to local commentators that O carried the city's high income areas and particularly the "West of Twin Peaks" section that constitutes Kopp's territory. Proposition O was rejected in the Chinese neighborhoods and the Latino area of the Mission District, although by a remarkably narrow margin.6 In only one other identifiable neighborhood, Haight-Ashbury,

still the home of many new-left radicals, was there a clear rejection of Prop O (61 percent). Several neighborhoods typically considered liberal and progressive, such as Eureka Valley

and Potrero Hill, did not oppose the measure.7 Election analysts calculated that predominantly gay as well as black areas of the city failed to take a stance against O, with 51 percent of those

voters supporting the measure (Binder 1983). These same neighborhoods had passed another proposition, N, expressing disapproval of Reagan's policy toward El Salvador, with margins ranging from 60 to 76 percent, so it is not simply the case that the progressive voter did not turn out for this election.

These results in traditionally liberal areas were particularly noteworthy because the anti-O campaign has relied heavily on the endorsements of well-known politicians, many of them considered liberal. Among the public opponents of Proposition O were Mayor Dianne Fein- stein, almost of all of the City Supervisors other than Kopp, the Democratic County Central Committee, all the Democratic clubs including the gay and black organizations, and the pow- erful speaker of the California Assembly, black politician Willie Brown. Feinstein was reelected with ease, but she and other leaders neither reflected nor shaped their constituents' perception

of the language issue. Binder (1983:2) reports that "the same areas that gave the Mayor the most votes contributed substantially to the margin of O's victory."

A puzzling question, then, is the meaning of Proposition O's success, given San Francisco's reputation as a liberal and culturally plural city. Although we cannot divine how individual voters construed the language issue, we can see how it was constructed publicly in the cam- paign. By closely examining the rhetoric of printed texts that carried primary messages of the

debate, we may arrive at a plausible interpretation that reconciles the surprising aspects of the election results.

the themes of Proposition 0

A first step in the analysis of the campaign materials is to extract from the corpus of texts (campaign flyers, editorials, news reports, voter information pamphlet) propositions or themes that recur with frequency (Agar 1983).8 Because they recur, we can view these themes as cen- tral to the cultural construction of the issue. The goal is to identify semantic units that are basic

to the overall campaign discourse, rather than to locate effective but possibly idiosyncratic as- pects of persuasive style. Surface syntactic features such as nominalization or passive verbs (Fowler et al. 1979), or creative tropes such as those identified by Lloyd-Jones (1981) in the oratory of the British demagogue Enoch Powell, are not the principal topic of inquiry here.9

sentences in the language prison 271

This content downloaded from 140.211.127.19 on Mon, 24 Sep 2018 16:28:52 UTC All use subject to https://about.jstor.org/terms

Several themes recur in the pro-Proposition 0, anti-bilingual material. Each appeared in at least several texts but is given here in just a single brief example.

Waste: "Bilingual ballots waste scarce tax dollars" (Voter Information Pamphlet, p. 72).10 Logic: "Knowledge of English is already required for citizenship, which is a prerequisite to vote" (V.I.P., p. 73). Unfairness: "Immigrants in the past felt it a duty and a privilege to learn English" (V.I.P., p. 73). National Unity: "A common language is the basis of American nationhood" (S. F. Chronicle 11/1/83). Hinder English Acquisition: "The provisions prolong English illiteracy" (V.I.P., p. 71). Full Life: "The individual who fails to learn English is condemned to semi-citizenship, condemned to low pay, condemned to remain in the ghetto" (Guy Wright, S. F. Examiner and Chronicle 11/6/83). Uninformed Voter: "It is questionable whether a non-English speaking voter can form an opinion and cast an intelligent vote" (Bay Area Reporter 11/3/83). Bossism: "multi-lingual ballots encourage political bossism" (Sun Reporter 11/2/83).

These themes, particularly the first five, sometimes appeared in list fashion, in a graphic il- lustration of the fact that voters might have different reasons for arriving at the same conclusion.

However, a complex argument structure tied the Uninformed Voter and Bossism themes (and sometimes Full Life) to each other and to the stated goal of Proposition 0, the abolition of mul-

tilingual elections. Example 1, from an editorial in the San Francisco Chronicle, the major daily morning newspaper, represents this argument in the most dispassionate tone of all the texts:

(1) Non-American born citizens should, thus, be able to comprehend a ballot printed in English. If they cannot, they go to the polls unable to even understand what candidates have been saying, a poor basis on which to exercise electoral judgment.

This situation, it seems to us, presents a potential danger which far exceeds the mere irritation of the multilingual ballot. In close elections, a candidate might prevail only because he has assembled the largest number of poorly-informed and incompetent voters IS. F. Chronicle 11/1/83].

Several unstated premises enable the inferential chain that links language proficiency to Bossism: (1) that English is the language of political information, and specifically; (2) that can- didates themselves campaign in English; (3) that print is the medium of such information, or, alternatively; (4) that lack of literate proficiency in English entails lack of aural proficiency.

The last two assumptions are noteworthy because of their questionable validity: the dispro- portionate influence of television is a much-remarked fear in contemporary politics, and there is no natural link of language ability to literacy.11 The first two underlying assumptions are even

more significant, because they are also the goal of Proposition O. In the transforming logic of modern myths, the starting point is the endpoint as well, in Barthes' structuralist view (Barthes

1972). That which is to be (re)produced through rhetoric as a form of political action is that which is assumed. The goal of Prop O is to bar languages other than English from electoral uses, but the argument justifying this action departs from the assumption that only English is used in electoral politics. The discourse re-presents the assumed state as natural, correct, and morally just. Such moral justification is necessary because, with government sanction of mi- nority languages in the 1970s, it no longer goes without saying that the language of politics in America is English. This is the straightening of opinion that Bourdieu (1977:164) notes in or- thodoxy, aimed at restoring a primal state of innocence, English-language political discourse.

On the basis of this set of assumptions, the Chronicle reasons that citizens who need to use non-English ballots are particularly easy prey for unscrupulous candidates capable of "assem- bling voters," quite as if, in the image of the familiar machine metaphor of politics, assembling meant fabricating them.2 The ease with which the argument is put forward may derive from a more deeply and generally held tenet of linguistic chauvinism: that English is a vehicle of in- formation, while other languages are obfuscating cloaks, vehicles of manipulation (Claudia Strauss, personal communication). "Truth" is more likely to come in transparent English, free of the seductive packaging of foreign languages.13

The warrant for believing that a monolingual ballot would end the perceived danger of Boss- ism, an especially crucial link in the argument, is unstated. Opponents of Prop O were infuri- ated by what they saw as an inconsistent argument and unwarranted conclusion. One attempt to point it out is given in example 2:

272 american ethnologist

This content downloaded from 140.211.127.19 on Mon, 24 Sep 2018 16:28:52 UTC All use subject to https://about.jstor.org/terms

(2) Opponents of the use of bilingual voting materials claim non-English speakers can't vote intelligently because they can't understand election debates in the mass media. No better rationale exists for the bilingual voter pamphlet. It provides full information on both sides of the issues-much of which is absent even in the English-language press [J. Avila, Sunday Examiner and Chronicle, 10/16/83].

As understood by Avila and his sympathizers, abolishing bilingual election materials would eliminate a major source of unbiased information and thus contribute to rather than diminish the perils of uninformed voting. In their view, the conclusion of the Uninformed Voter-Bossism

argument rests on two further unstated propositions: (1) that uninformed voters should not be allowed to vote, regardless of their citizenship, and (2) that restoring the monolingual ballot will effectively bar that participation. It would be difficult to find convincing backing for the proposition that "unqualified" voters, effectively organized by bosses, would be prevented from voting by monolingual English ballots. However, the opposition was not free to attack this

weak link since a similar proposition is central to its own argument for bilingual ballots. This

implicit agenda of restricting the right to vote was attributed to Proposition O by opponents (example 3) and denied by supporters (example 4):

(3) We must not let Proposition O deny taxpaying, law abiding Latino and Asian American citizens their right to vote IV.I.P., p. 75].

(4) Private assistance has been traditionally afforded by family members, friends, political parties and various associations. .... would NOT prevent citizens from exercising the right to vote [V.I.P., p. 73].

Opponents' insistence that the O position was either illogical or a violation of civil rights fell on deaf ears. From the anti-O viewpoint, it appears either that 62 percent of the usually fairly

progressive voters of San Francisco advocated depriving some citizens of their right to vote, or that they were unable to spot gaping holes in the English-only argument even when these were pointed out to them. However, other features of the discourse allow an alternative explanation of the appeal of Prop O and the failure of the anti-O campaign to get its message across.

The themes of Uninformed Voter and Bossism are linked to bilingual ballots in these cam-

paign texts through figurative relations of metonymy and metaphor, rather than syllogistic rea-

soning. These tropes depict the proposed political action not as the revocation of citizens' rights, but as an act of liberation.

Metonym is the trope most essential to arguments such as the Chronicle's. Non-English lan- guages are associated with Uninformed Voters and with Bossism, and the focus of concern shifts from language to Bossism. The foreign languages then function as a metonym, substituting

for the ensemble of themes. Metonymically, an assault on Bossism can be carried out through an assault on language. The rhetorical figure is convincing because, derived from association, it seems natural:

what allows the reader to consume myth innocently is that he does not see it as a semiological system but as an inductive one. Where there is only an equivalence, he sees a kind of causal process: the sig- nifier and the signified have, in his eyes, a natural relationship [Barthes 1972:131].

With Bossism rather than the threat to English as a primary target of Proposition 0, the ini-

tiative was likely to find wide support. Bellah et al. (1986) point out in their recent study of American political culture that one conviction that unites ordinary Americans-liberal, con- servative, progressive-is the negative evaluation of political brokers and special interest pol- itics. To the extent that Americans think of politics as what Bellah et al. call "the politics of interest," they see it as "not entirely legitimate morally" (1986:200).

This metonymic structuring of the ballot issue was sometimes taken to the extreme of asso-

ciating foreign languages with illegal aliens and fraudulent balloting, thus giving the languages even greater symbolic freight. Equally powerful is a recurring metaphor of imprisonment in the

presentation of Bossism and its relation to language. The metonym, the metaphor, and the log- ical gap that they bridge are all displayed in example 5, an excerpt from a taped interview I had with the head of the Committee for Ballots in English (7/25/84):

(5) KAW: "And you see them [multilingual election materials] primarily as disincentives to learning En- glish?"

sentences in the language prison 273

This content downloaded from 140.211.127.19 on Mon, 24 Sep 2018 16:28:52 UTC All use subject to https://about.jstor.org/terms

R: "There are two things: disincentives to learning English and ... certain Hispanic leadership who strongly support bilingual ballots in Spanish. They want the core constituency, for their own political ambition, Spanish speaking. There're a number of them who're very happy to have the Hispanics stay in the barrio, keep them there, their only language is Spanish. 'We will tell you how to vote.' Now this can and is becoming an increasingly serious problem in California. And we have enormous numbers of the Hispanics, in the millions, we don't even know, if we count the illegals. Here's another thing we're very much afraid of, and I may get into this, at least officially, telling the U.S. Attorney and telling the Attorney General, this is unconfirmed, I can't verify this, but the city of Oxnard, until the last election, 50 percent of that vote were illegals, undocumented-50 percent."

KAW: "How would that have happened?"

R: "Uh, this is really pretty simple, you get the leader, there are many who are in and around on the farms, they're Hispanic, so the Hispanic leader, he gets them all in the house or in the barn and he says, 'I'm going to pass out some pieces of paper, you sign them and put your address in, and I'll put them in for you.' These are absentee ballots, then they go to the registrars or the county clerks of whatever county they are in, and the ballots come back in time of election to whomever these people are, get them back in the same barn or the same room in the house and mark all their ballots, send them back in. Easy. (uh hmmm) Nothing to it, in California. (hmm) First to get, first to get an absentee ballot, and you make it out."

KAW: "If that is a problem ... do you think that the fact that the materials are bilingual makes any difference to that process? In other words could you stop that process or make a dent in it by having monolingual ballots?"

R: "Probably not. . ."

Discussion of the Spanish language here conjures up images of illegal aliens, and the restric- tion of Spanish is metonymically a means of controlling those aliens. The metaphor of impris- onment also contributes to this striking exposition. Hispanics are "kept in the barrio" by the

leaders. The imprisonment image becomes quite literal in the hypothetical sequence with il- legal aliens: in order to politically exploit workers, leaders corral them physically, "get them all in the house or in the barn" and then "get them back in the same barn . . ." again.

The imprisonment metaphor, one of the most consistently exploited rhetorical figures in the

campaign, is fully explicit in example 6, taken from a statement that appeared under the name

of Quentin Kopp, the Supervisor who introduced the initiative:

(6) Without an impetus to learn English, it is far too easy to become sequestered in a language prison, a prison that many politicians attempt to perpetuate, manipulate and control.

Without the ability to speak our language, an individual is forced to follow the tenets of self-appointed leaders of that minority. There is no choice. Information from other sources cannot penetrate the lan- guage barrier [Bay Area Reporter, Political Supplement, 11/3/831.

A final example of this important image occurs in conjunction with the Full Life theme men- tioned earlier:

(7) the individual who fails to learn English is condemned to semi-citizenship, condemned to low pay, condemned to remain in the ghetto [G. Wright, S. F. Examiner and Chronicle, 11/6/83].

In this rhetoric, minority group members are presented not as active agents but as acted-upon

prisoners. A relation of exclusion and opposition between ethnic minorities and ethnic leaders is set up. We find "politicians," "self-appointed leaders," "ward-type manipulators," and "po- litical scoundrels" "commanding," "manipulating," "imprisoning" the "ethnic groups," "mi- norities," "inarticulate voters," and "citizens" (all terms taken from actual texts).

Proposition O becomes, then, not an attack on minority rights, but a crusade to liberate mi- norities and protect their real rights and interests. Language is the prison, or the bars on the prison, and minority leaders the jailers. To release the imprisoned, the language must be re- moved. Without prison walls to guard, the jailers will be out of a job. Symbolically, removing the offending languages from the ballot is an act toward freeing the minority-language citizens.

This discourse does not simply redefine minority interests, but actually reassigns minority identity itself by rejecting the minority status of community leaders, and denying their authority

to speak for the ethnic group. This is not a mere euphemism, or a reversal of valences, as when one's "terrorists" are the other's "freedom fighters," but a reordering of the minority/majority

274 american ethnologist

This content downloaded from 140.211.127.19 on Mon, 24 Sep 2018 16:28:52 UTC All use subject to https://about.jstor.org/terms

opposition into a three-part struggle. The majority thereby assumes the right to absorb the mi-

nority, now separated from their leaders, in a new, transformed dichotomy.

In its initial design in 1975, the bilingual ballot was established to give a political voice to so-called language minorities. The rhetoric of the Proposition O campaign reconstructs the idea

of the bilingual ballot as a means of robbing minority members of their voice; those who use bilingual ballots have become "inarticulate voters" in the words of a black newspaper:

(8) multi-lingual ballots encourage political bossism, which means bloc voting, and there are many po- litical scoundrels who are seeking large blocs of inarticulate voters who will vote just as the scoundrels command them to vote [Sun Reporter 11/2/83].

In these texts, the pro-O forces appropriate the voice of the minority and speak for their in- terests. Ironically, the Sun Reporter editorial appears directly under the masthead, "We Wish To Plead Our Own Cause. Too Long Have Others Spoken For Us."14

It is possible, then, that the usually progressive voters who supported Proposition O did not intend to strip people who "need" bilingual ballots of the right to vote. Regulating access to the ballot for voters may not have been the heart of the issue for some supporters; from the evidence of the discourse, regulating access to the ballot for candidates and political brokers was.

This structuring of Proposition O proved to be very effective. It gave the initiative a meaning that was not only acceptable in polite public talk but was laudable even by progressive and pluralistic standards. Additionally, it framed the argument in such a way that a response was nearly impossible. Any leader who spoke in favor of bilingual ballots was effectively discred- ited, since the very act of speaking out for bilingual ballots could be interpreted as providing evidence of self-interest. Myth is extremely difficult to contend with, for oppositional forces as

well as analysts, because "wine is objectively good, and at the same time, the goodness of wine is a myth" (Barthes 1972:158; emphasis in original).

conclusion

The support that Proposition O found among traditionally liberal sectors of San Francisco

becomes explicable once rhetorical figures of metonym and metaphor, and the opposition they create between minority leaders and minority masses, have been identified. The events of Prop- osition O were ordered by recourse to existing structures of significance provided by American political culture. In 1975, Congress asserted that language minorities have been barred from full participation through devices manipulated by entrenched powers. The discriminatory "de- vice" then was the monolingual ballot, and the prescription for redress was the bilingual ballot. In 1983, a similar propositional schema was called into play, asserting that minorities are barred from full participation through devices manipulated by entrenched power. Now the mi-

nority languages and the bilingual ballot were the discriminatory "devices," the monolingual ballot the liberating redress, and ethnic rather than white political elites the abusers of power.

A standard schema used earlier by liberals had been appropriated by new forces and given new implications, in a still familiar program of paternalistic compassion for the less competent.1

While ethnocentric notions about language and literacy certainly underpin rationales for Proposition 0, the initiative was successful not just because it played on English speakers' xen- ophobic insecurities about language and lifestyle. Rather, it linked these fears to even more widely shared American concerns about politics and public life, recruiting a broader spectrum of supporters. The concern with Bossism is a selective invocation of an American political in- dividualism at least as old as Tocqueville. Bellah et al. remind us that Americans are genuinely ambivalent about public life, and particularly uncomfortable with interest group politics and professional politicians (1986:200, 250). Prop O may have been particularly appealing be- cause it addresses this ambivalence by projecting the dark side of political process onto out- siders.

sentences in the language prison 275

This content downloaded from 140.211.127.19 on Mon, 24 Sep 2018 16:28:52 UTC All use subject to https://about.jstor.org/terms

A basic anthropological insight is that ways of talking about the "Other" are ways of talking

about ourselves. Many of the themes of the Proposition O campaign at other times are raised about contemporary American politics more generally: politicians are self-interested and un- scrupulous, voters are uninformed, politicians manipulate the information media, media im- ages manipulate voters, voters are apathetic, and the electoral process is without significance. The rhetoric of Proposition O addresses many of these preoccupying possibilities, but makes them Ethnic and Other. If the discourse depicts Them as uninformed, misled and manipulated,

then the story it tells about Us is by implication the opposite: well-informed through Our mas-

tery of English literacy, We rationally evaluate facts and arguments available to Us in objective printed media, and Our exercise of the cherished right to vote results in the selection of the best

candidate or the correct policy. Self-doubts as much as external threats form the background to this manifestation of the English-only movement.

notes

Acknowledgments. The field research on which this paper is based was made possible by a summer fellowship from the National Endowment for the Humanities. My thanks to Professor Joshua Fishman and fellow members of the 1984 NEH seminar on ethnolinguistic minorities at Stanford University for helpful comments in early stages, and to Alan Hudson-Edwards for generous advice. The opinions expressed here are mine and do not reflect those of NEH. Julia Roberts assisted in the thematic analysis of the texts. Earlier versions of this paper were presented at the "Language and Political Economy" session of the 1986 Amer- ican Anthropological Association meetings in Philadelphia and at the Center for Psychosocial Studies in Chicago; this version has benefited from discussions in those settings. My thanks to Susan Gal, Richard Handler, Claudia Strauss, Selma Sonntag, Michael Silverstein, several anonymous reviewers for American Ethnologist, and Shirley Lindenbaum for careful reading and suggestions, many of them extremely insight- ful. Responsibility for failure to make the most of them is mine. To the rhetoricians of the Prop O campaign I owe the intertextuality of the title of this paper and the links to Friedrich Nietzsche, Fredric Jameson, and Bradd Shore.

'"English-only" is a short form of reference to some of the political initiatives examined here that is used by proponents themselves, and as such will be used throughout this paper.

2The notion that status politics involves scapegoating is hardly original; Hofstadter wrote about it in relation to populism in 1955. However, Hofstadter's version of the process is different from the one devel- oped here, in that his scapegoat symbolizes a fundamentally external threat.

31 am using the concept of "motivation" here in the sense that C. W. Mills (1940) suggested, not as a prior, private state of an individual, but as a "vocabulary" that coordinates and makes sense of action.

4The neologistic "language minority groups" meant "Spanish heritage," Asian (Japanese, Chinese, Fil- ipino and Korean), American Indian, and Native Alaskan. In other words, only groups already viewed as (racial) minorities were considered "language minorities." Illiteracy was defined as less than five years of schooling, and by this criterion, the national illiteracy rate was 4.6 percent. Illiteracy in English was un- doubtedly intended but never specified by Congress in the law, an interesting sidelight on the linguistic chauvinism discussed later in this article.

5The Voter Information Pamphlet is a free publication distributed to all voters by the Registrar. It contains not only sample ballots, voting instructions, and information about candidates, but also extensive infor- mation about the often numerous initiatives found on California ballots. For each initiative there is a non-

partisan explanation, an evaluation of fiscal impact, and statements prepared by representative supporters and opponents. In addition, in San Francisco any partisan can insert an argument for a fee, and the pam- phlet can run to a great number of pages.

6According to accepted calculations, neighborhoods with a least 30 percent Hispanic population re- jected Prop 0, but not resoundingly (53 percent) and with an abysmal voter turnout (36 percent) even lower than the citywide rate (Binder 1983). There was considerable debate over results in Asian areas. Although Binder calculated a 60 percent approval rate for precincts with at least 30 percent Asian population (Binder 1983), an advocacy organization, Chinese for Affirmative Action, analyzed five core Chinatown precincts and found the rate of rejection of 0 ranged from 62 percent to 79 percent, figures almost identical to the proportion of Chinese surnames registered (Pettit 1984).

7To attempt a rigorous definition of "liberal" (or "progressive," which as used currently in San Francisco generally means to the left of liberal) in American politics would be to start on a task that might never permit us to return to the question at hand. The rhetoric of the 1988 presidential campaign, which began well after this essay was written, illustrates this difficulty. For the purposes of this essay, I am accepting local com- mentators' characterizations of certain neighborhoods and politicians as liberal or progressive. The im- portant characteristic of liberals for this discussion is that they generally support government spending for social welfare and protection of civil rights.

276 american ethnologist

This content downloaded from 140.211.127.19 on Mon, 24 Sep 2018 16:28:52 UTC All use subject to https://about.jstor.org/terms

8The corpus on which the present analysis is based consists of 62 texts of varying lengths, including reportage, editorials, columns, and letters to the editor collected from 12 newspapers covering the period of the campaign, from July to November 1983. This was a nearly exhaustive survey of community news- papers, mostly weekly, as well as the two major dailies. Partisan publicity and campaign materials were also analyzed, including those published in the Voter Information Pamphlet by the city. The texts are sup- plemented by interviews I conducted with the Registrar of Voters and leaders of both the pro- and anti-O campaigns in the summer of 1984.

9Such an analysis of surface structures could be undertaken, and would lead to some interesting obser- vations. For example, in one Kopp text, almost all sentences that illustrate the pro-O position have human beings as both semantic and syntactic subject, represented by proper names and personal pronouns; active verbs predominate. When the text moves to an attack on the bilingual position, there are almost no human subjects, and instead a high incidence of extraposition of "that clauses" and infinitival clauses, with "it" in subject position. Passive verbs are frequent. While these may be effective rhetorical devices, I would argue that the success of Prop O depended less on such surface structures of particular texts and more on shared propositions recurrent in them.

'0Although strictly speaking, multilingual ballots were at issue, arguments often spoke of "bilingual bal- lots." This may be because O leaders dismissed the Chinese component as dwarfed by the demographic weight of Spanish in California. But it is also undoubtedly because "bilingualism" has peculiar associations and highly charged meanings that "multilingualism" doesn't carry in the United States.

"The fact that both sides of this debate have naturalized literacy as a facet and index of language ability reveals an aspect of language ideology in the United States. In 1975, Congress unreflectingly assumed that native language literacy was unproblematic for minority language citizens, even as it recognized that they are educationally disadvantaged. Prop O discourse facilely creates victims who don't understand English from those who can't read it. It is a rare society in which such an equation is thinkable.

12Little reference was made to the actual rate of utilization of the bilingual ballots, and figures were never cited. This is of special interest since the use rates appear to be quite low. I was able to gauge utilization from the number of voters who checked the box on their registration forms requesting material in Chinese or Spanish. The Registrar of Voters keeps only a running count and was unable to provide statistics for 1983 at the time of my research. But by July, 1984, eight months after Prop O passed, there were 3694 requests for Chinese materials and 1907 requests for Spanish (this latter figure representing a recent sudden increase, according to the Registrar). Out of a total 375,799 registered voters in San Francisco, these constituted only 0.9 percent and 0.5 percent of the voting population, respectively. Prop O advocates, interested in pro- moting images of hordes of incompetent voters, chose not to stress the low use of bilingual ballots, even though it contributed to another theme-"waste." Utilization rates were not cited by the opposition, either, since they do not readily support the contention that this is a much-needed policy increasing informed voter participation. In the heated debate over the social impact of this measure, neither side wanted to draw attention to what the impact actually had been.

3Michael Silverstein (1987:8) has identified a related "Goldilocks" principle in popular American lan- guage ideology, a belief that standard English makes not too many distinctions, not too few, but gets it "ju ... st right," a perfect match to the reality "out there."

'4The Sun Reporter's wording reveals yet another glimpse of the chauvinistic language ideology that allows easy equations to be assumed rather than argued. In this phrase, inability to speak English apparently is equated to inability to speak, which appears to be closely related to inability to think for oneself.

S5This appropriation and transformation of the benevolently paternalistic schema was possible because of an inherent paradox. That paradox can be found in other instances, as in the ideological dichotomization of the poor into the "worthy" and the "undeserving." The worthy poor are those deserving of assistance because they are trying hard on their own; they can be recognized by their reluctance to seek assistance. When they seek out assistance, they become by definition unworthy and moves are made to restrict enti- tlement to welfare services (Mark Stern, personal communication). Similarly, the voice of minorities is so- licited because they are marginal; when they speak and their accents can be heard in the mainstream, they are by definition not marginal, and attempts will be made to restrict their special access.

references cited

Agar, Michael 1983 Political Talk: Thematic Analysis of a Policy Argument. Policy Studies Review 2(4):601-614.

Barthes, Roland 1972 Mythologies. New York: Hill and Wang.

Bellah, Robert N., Richard Madsen, William M. Sullivan, Ann Swidler and Steven M. Tipton 1986 Habits of the Heart: Individualism and Commitment in American Life. New York: Harper and

Row. [First edition 1985, University of California Press.] Bilmes, J.

1981 Proposition and Confrontation in Legal Discussion. Semiotica 34:251-275. Binder, David

1983 1983 Election: A Trend Toward Conservatism Must Still Await 1984 for Confirmation. The Pettit Report 4(23-24).

sentences in the language prison 277

This content downloaded from 140.211.127.19 on Mon, 24 Sep 2018 16:28:52 UTC All use subject to https://about.jstor.org/terms

Bourdieu, Pierre 1977 Outline of a Theory of Practice. Cambridge: Cambridge University Press.

Fosburgh, Lacey 1984 San Francisco: Unconventional City for the Democratic Convention. New York Times Magazine,

1 July.

Fowler, R., R. Hodge, G. Kress and T. Trew 1979 Language and Control. London: Routledge and Kegan Paul.

Gusfield, Joseph 1986[1963] Symbolic Crusade: Status Politics and the American Temperance Movement. Chicago:

University of Chicago Press. Hofstadter, Richard

1955 The Age of Reform. New York: Alfred A. Knopf. Hunter, Allen

1987 The Role of Liberal Political Culture in the Construction of Middle America. University of Miami Law Review 42(1):93-126.

Lloyd-Jones, David 1981 The Art of Enoch Powell: The Rhetorical Structure of a Speech on Immigration. In Politically

Speaking. R. Paine, ed. pp. 87-112. Philadelphia: ISHI. Mills, C. Wright

1940 Situated Actions and Vocabularies of Motive. American Sociological Review 5(6):904-913. Moore, Sally Falk

1987 Explaining the Present: Theoretical Dilemmas in Processual Anthropology. American Ethnologist 14(4):727-736.

O'Barr, William M. 1981 The Language of the Law. In Language in the USA. C. A. Ferguson and S. B. Heath, eds. pp. 386-

406. Cambridge: Cambridge University Press. Pettit, Bruce

1984 The Pettit Report. San Francisco Examiner, 4 January. Quinn, Naomi

1986 Convergent Evidence for a Cultural Model of American Marriage. In Cultural Models in Language and Thought. D. Holland and N. Quinn, eds. Cambridge: Cambridge University Press.

Sahlins, Marshall 1981 Historical Metaphors and Mythical Realities. Ann Arbor: University of Michigan Press.

San Francisco Voter Information Pamplet 1983 Municipal Elections, 8 November.

Silverstein, Michael 1987 Monoglot "Standard" in America. Working Papers and Proceedings of the Center for Psycho-

social Studies 13. Chicago: Center for Psychosocial Studies. United States Public Law 89-110

1966 United States Statutes at Large, Vol. 79:437-446. Washington, DC: U.S. Government Printing Office.

United States Public Law 94-73

1977 United States Statutes at Large, Vol. 89:400-406. Washington, DC: U.S. Government Printing Office.

submitted 12 October 1987

revised version submitted 10 August 1988 accepted 27 September 1988

278 american ethnologist

This content downloaded from 140.211.127.19 on Mon, 24 Sep 2018 16:28:52 UTC All use subject to https://about.jstor.org/terms