Introduction Slang and the Struggle over Meaning

Introduction
Slang and the Struggle over Meaning

The Language that Came Down the Hill

One day after we had been living in Tuiuti for some time, we were called into a
little bar by people we did not know though they later became our best friends
in Tuiuti. Almost immediately the conversation turned to giria (‘slang’). “O
senhor conhece nossa giria?” (‘Do you know our slang?’) No, I did not.
There followed a half-hour of instruction in favela slang terms. I asked if they
used these words la embaixo. Maria Antonia said no. I asked why. She said,
“Por vexame!” (‘For shame’). Her meaning was not that they would feel
embarrassed, so much as that the language has no place there and they, too,
therefore, would have no place “down there.” The language of the favela slang
is an impropriety to the stiff and stolid middle class which defines its users as
brutos, assassinos, ladroes (‘thieves’), maconheiros (‘pot-smokers’),
malandros (‘no-goods’).

On the hill, they are free to use this language, rich, funny, ironic, allusive, and
largely incomprehensible to outsiders. With it, they can mock the system that
presents so many encrencas (‘monkey-wrenches in the works’) and hardships.
… (Leeds and Leeds 1970: 261)


I wish I could so poignantly remember my very first introduction to Brazilian slang. I
returned from my first visit to Rio de Janeiro in 1995 just beginning to understand the salience of
slang to Brazilians and questioning whether one could really write a whole dissertation on the
topic of slang. Through future research trips, I became more attuned to the social and spatial
dynamics of slang speaking. Just as Leeds and Leeds briefly noted 25 years before my arrival,
slang continued to be strongly associated with favelas (‘shantytowns’),1 hills, poverty, and
crime. While speaking slang was clearly attributed to particular spaces and people, I found that
Brazilians of all socio-economic classes and neighborhoods talked about slang often – both
spontaneously and at my request. Bolstered by popular interest and sufficient material, I decided
to make slang my primary focus. I have since sat through countless hours of slang “instruction,”

1 Because ‘slum’ and ‘shantytown’ are poor translations for the Brazilian word favela, and due to the
centrality of the concept to this dissertation, I will follow the convention of other Brazilian researchers and
Jennifer Roth-Gordon
Slang and the Struggle over Meaning
1

---

learned hundreds of new words, and discussed the topic of slang with anyone who even
casually inquired about my presence in Brazil. As far as I know, this is the first dissertation to
present an ethnographic account of slang speaking. It is definitely the first ethnographic account
of the language that came down the hill in Rio de Janeiro, Brazil.

While the social tensions Leeds and Leeds document, between favela and asphalt, poor
and rich, marginal and citizen, all hold true today, quite a few things have changed. To begin
with, Brazil experienced unprecedented economic growth in the 1970’s under military rule (the
so-called “miracle years”), and Brazil’s gross domestic product is now among the ten highest in
the world. Yet the Brazilian elite’s two-part plan to “grow first and divide the cake later”
(Caldeira 2000: 42) never returned to the second part, and Brazil now ranks among the world’s
most unequal in terms of distribution of wealth (Caldeira 2000: 47). While Brazil has always had
more than its share of poverty, the divide between rich and poor now reaches unprecedented
levels. Brazil has experienced recent political change which has also increased the social divide
between “marginals” and “citizens.” Though democratization began in the mid-1980’s, working-
class Brazilians are routinely denied their civil rights, and human rights abuses (by a police force
that was never de-militarized) disproportionately affect poor Brazilians of color (Mitchell and
Wood 1999). As in the U.S., the recent rise in drug trafficking, the consolidation of organized
crime, and the escalation of violence have dramatically impacted the condition of daily life in the
favela and in the city as a whole (Holston and Caldeira 1998). Though it has always been the
site of hardship, marginalization, and poverty, the Brazilian favela today offers far less protection
to its residents than it did 30 years ago.
Of equal importance to what has structurally changed over the past three decades is
what has changed in the way researchers observe and talk about Brazil. It comes as little
surprise that the Leeds fail to notice the overwhelming blackness of the Brazilian favela. Their
description of the favela was published in 1970, at a time when socioeconomic class was
thought to supercede (and address) questions of racial inequality. The observation that “race
matters” in Brazil is now rarely contested among scholars. Though for Brazilians, slang is

treat favela as an English word (without italics). Please see Chapters 2 and 6 for a more detailed description
of the social, economic, and spatial conditions (and stigma) associated with the Brazilian favela.
Jennifer Roth-Gordon
Slang and the Struggle over Meaning
2

---

overwhelmingly linked to the favela, and the favela is strongly associated with blackness, few
studies of Brazilian Portuguese have ever addressed this connection between race and language.
As Silva explains:

[T]here don’t seem to be linguistic traces belonging to groups of blacks or their
next descendents. Besides this fact, in Brazil, also unlike other countries, there is
not a clear distinction of ethnic groups connected to skin color or type of hair.
In Rio, especially, there is a true continuum … (Silva 1995: 160, emphasis in
original, translation my own).


Linguists have often claimed, like Silva, that either the fluidity of racial classification or
the stronger connections between language and class (Guy 1989) preclude such empirical
investigation. Despite recent interest in Brazilian race relations, sociolinguists have yet to reinsert
race as a meaningful category of analysis. In this dissertation, I do not attempt to identify a
“black language” akin to that of African American English, nor do I argue that skin color
predisposes a speaker to a different lexicon. Instead, I focus on the way that certain linguistic
forms and ways of speaking come to be racialized, and I investigate the way that slang becomes
one site for a struggle over meaning. This struggle over meaning is not merely linguistic: what
words come to mean and how they may be employed in discourse are fundamentally struggles
over who is permitted to determine meaning and regulate appropriate ways of speaking. As
such, I argue that slang is the site of linguistic and social struggle in Brazil, and I pay careful
attention to the way that race matters in this struggle.

Towards a New Understanding of Slang

In order to situate my own understanding of slang and its role in Brazilian society, I
begin this study by examining the various definitions of slang that have been proposed by
linguists and lexicographers. It is important to emphasize from the very beginning that defining
slang is an ideological act. There are no linguistic criteria which isolate the lexical features
associated with slang from lexical items associated with jargon or, for that matter, the standard
language. Instead, linguistic resources come to be recognized as slang through an ongoing social
process which relates certain linguistic forms and uses to individual speakers, social groups, and
Jennifer Roth-Gordon
Slang and the Struggle over Meaning
3

---

real-world speaking contexts. In the same way, the linguistic concepts of standard language,
vernacular, dialect, register, jargon, cant, and argot rely on associations with kinds of speakers
and ways of speaking rather than “objective” linguistic facts. Taken together, these terms inform
the ways both linguists and non-linguists think about language. In this section, I briefly discuss
the relation of these terms to each other and to slang.

Of these linguistic classifications, the standard language is perhaps the most difficult to
denaturalize. Often seen as existing as an entity somehow separate from speakers and
conversational contexts, the standard language is carefully guarded by language prescriptivists
and reified by state institutions such as schools. Yet standard language, much like slang, is an
everchanging set of conventions that is recognizable only because speakers and institutions
continue to participate in its recognition. The standard language is imagined as a coherent set of
linguistic features and placed within an imposed linguistic hierarchy. Non-standard language is
proposed as a somewhat less coherent, “catch all” term for anything that occupies a position
below the standard (Bex and Watts 1999). As non-standard speech is often attributed to a lack
of access to the standard (or to an outright rejection of it), slang is generally categorized as non-
standard speech. The vernacular is similarly contrasted to the standard, as an example of local
vs. global or institutional language. Slang often shares with the vernacular a strong connection to
locally based communities.

If slang is linked to local communities and non-standard speech, jargon occupies the
opposite end of the spectrum. Lexicographers have defined jargon as the vocabulary of
technical terms specific to an occupation or profession learned through shared training. In
contrast to slang users, jargon speakers are perceived as having earned entitlement to a
specialized and precise lexicon of their own – and as invested with the power and knowledge
necessary to develop this lexicon. Jargons are thus equated with a kind of socially-sanctioned
knowledge (an expertise), such as medicine or law. The merit and respect often accorded to
jargon and jargon speakers indexes the association between words and powerful speakers.
Jargons, white-collar professionals, and whiteness can be mutually-constituting: From
prestigious social positions and occupations, privileged speakers invoke specialized words
Jennifer Roth-Gordon
Slang and the Struggle over Meaning
4

---

which reify their authority and respect in professional domains. Jargons therefore constitute
linguistic styles that index specific sites and forms of power.

Slang and jargon have been related to the less frequently used categories of cant and
argot. Several lexicographers have defined cant and argot as specialized sets of words and
expressions, particular to a segment of the population (Flexner 1975a, Lighter 1994, Sornig
1981). Argot has typically been used to describe the language of professional criminal groups.
Halliday’s (1976) notion of an anti-language is akin to argot, as a professional jargon associated
with the activities of a criminal counterculture. These researchers claim that cant, argot, and anti-
language differ from slang in their cryptic, private nature. While these terms denote a separate
system which is often unfamiliar to outsiders, slang characterizes language that can be readily
accessible to the general public:

Slang is not a special language in that strict sense of delimitation whereby people
and groups of people distinguish themselves definitely and irrevocably from
others by the language they use. Slang is a language variant open to be used by
anybody who might choose it as a specific stylistic variant. (Sornig 1981: 61-
62)


The fact that slang is accessible to a wide range of speakers to index kinds of people
and/or contexts has led Agha (2001) to discuss slang as a form of register variation in speech.
Register has traditionally been defined as a language variety particular to a social situation or
context of use, whereas dialects have been associated with geographic and/or social attributes
of the speaker. Agha expands the connection between context and register, defining a register
as a “discursive formation involving not only lexical repertoires but metapragmatic stereotypes of
personhood, activity, and social relations which, once linked ideologically to lexemes, become
performable through the use of such forms” (Agha 2001). Others have similarly criticized the
artificial division between attention to speakers (dialect) and attention to context (register),
observing that it is impossible to separate the two. Irvine (2001) reminds us that registers draw
heavily on images of the people associated with situations and activities, just as social dialects
may be strongly connected with conversational situations and speech activities. Irvine relates
linguistic practice to the concept of style, a topic I will return to.
Jennifer Roth-Gordon
Slang and the Struggle over Meaning
5

---

The attempt to delineate slang as separate and apart from these other types of linguistic
classification has sometimes resulted in an extreme view of slang as inherently marked – outside
normal language and linguistic competence. McWhorter invokes this type of radical separation
between slang and language when he writes, “Slang to language is like clothes to people—like
fashion, slang changes all the time, roughly with generations of teenagers, but the underlying
language stays the same” (1998: 129). Few would agree that slang is not a part of – nor has any
effect on – some idealized concept of the “underlying language.” Speakers do not always judge
words as slang as they speak, and using slang words may not be a marked linguistic practice
among certain groups of speakers. Eble (1996) reminds us that slang is within the ordinary
competence of language users and should not be treated as a linguistic anomaly. As Eckert has
argued for sociolinguistic variation:

Innovation does not come in through accident and inattention, or through the
chance encounter. It comes in through very much the same mechanism, no
doubt, as the basic acquisition of language – a process of analysis of the relation
between linguistic form and its effect in the world. (Eckert 2000: 215-216)


The fact that slang has been conceptualized as separate and distinct from “normal”
language may help to explain its marginalization from formal linguistic inquiry. Indeed, even
linguists and lexicographers who have turned their attention to slang have tended to idealize or
exaggerate the uniqueness of slang. In their definitions, certain characteristics are emphasized to
create a sense of linguistic difference. Before introducing my own definition of slang, I caution
against an understanding of slang as ephemeral, informal, rebellious, in-group, and pertaining to
youth. As I will argue, it is not that these characterizations of slang are always inaccurate or
uninformative, but rather that their inclusion as the defining elements of slang is overemphasized.
These definitions contribute to the imagining of slang as inherently different from jargons or the
standard lexicon. This imposed separation helps perpetuate slang’s exclusion from serious
linguistic study.

Perhaps one of the first things noted about slang is its fast-changing and fleeting nature.
This is an idea shared by both linguists and the general public, and it is often seen as a
fundamental facet of slang. In fact, slang researchers have long refuted this hard and fast
Jennifer Roth-Gordon
Slang and the Struggle over Meaning
6

---

characterization of slang. Pointing to the example of ‘out of sight,’ first recorded in 1893 but still
novel in 1970, Dumas and Lighter observe, “novelty in a locution is apparent rather than real
newness” (1978: 7). Along the same lines, Sornig has refuted the idea that slang has a
temporary nature:

Not so much attention has been allotted to the undeniable fact that some slang
terms are of considerable antiquity and stability. Ephemeral though slang may be
in general, there are slangy words that have survived through the ages,
unchanged in both their denotative and their connotative vulgarity. (Sornig 1981:
20)


In her extensive study of college slang, Eble (1996) finds that most slang terms enjoy
only a brief time of popularity, but she reminds us that not all slang is ephemeral and not all
ephemeral words are slang. “New words bombard us every day, and with a few exceptions
they feel as comfortable and nonthreatening as the words that have been around for
generations” (1996: 18). Current examples of neologisms abound in today’s high-tech world.
By and large, these lexical innovations are immediately considered either part of the standard
lexicon (email, fax, cellular) or attributed to the jargon of computer “geeks” (‘QA,’
‘motherboard,’ ‘GUI’). Undoubtedly, some of these words will fall out of the lexicon as either
they or the technology they denote become outdated (‘joystick,’ ‘floppy disk,’ ‘dot matrix’).
Fast and fleeting though these words may be, they will never have been considered slang. Thus
slang, like jargon and the standard lexicon, is made up of new words, temporarily in-vogue
words, and words that outlive their speakers.

While slang researchers have been quick to notice that claims of ephemerality have been
overstated, many continue to define slang as necessarily informal. I would argue that this
assumption is problematically based on 1) an understanding of a formal/informal dichotomy and
2) a naturalizing of a particular kind of speaker and speech. It should be clear that
conversational contexts do not come neatly labeled as formal or informal, reminding speakers
that slang speaking is prohibited or, alternatively, encouraged. Yet slang researchers often write
as if this were the case. Eble comments, “… slang vocabulary is generally not suitable for
serious topics in written texts with serious purposes …” (1998: 37). Here she writes as if her
Jennifer Roth-Gordon
Slang and the Struggle over Meaning
7

---

audience should automatically bring to mind the same “serious topics” with the same “serious
purposes.” Giles has suggested a move away from objective definitions of speech contexts in
terms of formality/informality, to a more subjective description of how speakers themselves
define a situation. He observes, “Indeed, what is a serious, formal topic for some, could be an
utterly trivial irrelevance for others” (Giles 2001: 211). Imagining an “objective” division
between formal and informal situations also implausibly suggests that speakers who use slang on
a regular basis are always being informal (or always find themselves in informal situations) while
speakers who do not generally use slang necessarily interact with a high degree of formality.

Researchers who include informality in their definition of slang attempt to break this
“either-or” scenario by suggesting that speakers use slang to alter the level of formality in a
conversation. As Lighter, a lexicographer, dramatically explains, “The use of slang undermines
the dignity of verbal exchange and charges discourse with an unrefined and often aggressive
informality” (1994: xii). Eble similarly claims that slang words always diminish the formality of
the conversation in which they occur (1996: 20). These explanations serve to preserve the
formality/informality distinction by assuming that the use of slang renders a formal conversation
informal. But more importantly, this view speaks from the perspective of a “non-” slang
speaker. These authors suggest that an unmarked conversation or speech context has no slang.
Once slang is introduced, the conversation becomes marked [+informality]. Eble offers several
examples of the way slang brings down the formality of an utterance. She compares “The
president appeared uninformed about the bill” with “The president appeared clueless about the
bill” (1996: 117). Reversing the marked and unmarked, we could imagine a different scenario
with three teenagers, where a teenager introduces one friend to another. Comparing the
following greetings, “What’s up, man?” vs. “How do you do, sir?,” we might be more inclined
to mark the latter as [+formality]. As de Klerk reminds us, “among certain linguistic subcultures
(e.g. teenagers, or some less privileged social groups), not using slang may in fact be breaking
norms in some registers” (1990: 592).

In their attempt to define slang, Dumas and Lighter look to establish the criteria which
will cause a “linguistically sensitive” audience to react. “This reaction, which cannot be
measured, is the ultimate identifying characteristic of true slang” (1978: 16). Clearly, Dumas and
Jennifer Roth-Gordon
Slang and the Struggle over Meaning
8

---

Lighter presuppose an audience of non-slang speakers – and ignore the fact that excessively
proper language carries the potential to make listeners react as well. In addition, the assumption
that slang is inherently informal brings along other associations. It feeds into the belief that slang
represents a lack of better vocabulary. While it is assumed that jargon speakers must work to
learn the words of their trade, and even standard language requires access to schools and
training, slang and informal speech are often understood as some sort of linguistic default for
marginalized speakers. Formal language carries connotations of learning and attention to speech,
while informal language is incorrectly assumed to suggest the lack thereof. As Eckert has noted,

While formal style certainly involves greater attention to speech, and while
speakers have to pay careful attention when they’re speaking in the most
extremely standard end of their stylistic repertoire, there is every reason to
believe that similar effort is required at the extremely non-standard end of their
repertoire as well. (Eckert 2000: 18)

I discuss the common ideology of slang as a lack of linguistic skill later on in this
chapter. For now, I hope to stress that the assumption that slang is inherently informal
perpetuates a notion of slang as marked speech, occasionally drawn upon by speakers to tone
down the formality of their otherwise unmarked (and formal) utterances. While speakers may
use slang to index informality, this is surely not all they do with slang, and it is inaccurate to
suggest that speakers who use slang as part of their everyday linguistic practice live and interact
in an always informal world.

Some slang researchers have incorporated social and pragmatic functions into their
definitions of slang. The two most commonly discussed are to promote social solidarity or mark
an “in-group” and to rebel against or oppose social norms. The speakers in this study appeal to
slang to achieve both of these important social and interactional goals, and both of these aspects
of slang figure prominently in this dissertation. Yet there are several reasons I oppose their
inclusion in a definition of slang. My primary objection is that the questions of what slang means
to speakers and what it accomplishes within conversation cannot be determined without
attention to its specific cultural, social, and linguistic contexts. This relates to a larger criticism of
studies that have typically analyzed slang out of context, as I discuss in Chapter 3. Additionally,
Jennifer Roth-Gordon
Slang and the Struggle over Meaning
9

---

authors have tended to lean towards one or the other explanation of the primary function of
slang – suggesting that these are aspects of slang which are contextually-bound and not suitable
for inclusion in a general cross-linguistic and cross-cultural definition of slang.
In her study on American college slang, Eble foregrounds the goal of social solidarity.
She defines slang as “an ever changing set of colloquial words and phrases that speakers use to
establish or reinforce social identity or cohesiveness within a group or with a trend or fashion in
society at large” (1996: 11). While I do not doubt that her speakers use slang to identify
themselves with others, much as the speakers in my study do, emphasizing this as a unique
aspect of slang suggests an artificial distinction between slang and the standard language.
Underlying this view is again the idea that slang is somehow inherently marked. It is as if
speakers call upon slang when they want to communicate important distinctions between
themselves and others and draw from the standard lexicon if they only need to get their actual
message across. This is obviously an artificial and impossible separation. Speakers communicate
referential and social information every time they speak. Strictly avoiding slang and appealing to
only the most standard of lexical items is just another way to “establish or reinforce social
identity or cohesiveness within a group or with a trend or fashion in society at large.” I am not
the first to point out that there is no neutral language, no accentless speech, and no lexical items
that communicate only referential meaning. Indeed, it has become clear to me that a study of
slang is important if only to shed light on how “standard” lexical items continue to be reproduced
as unmarked.
Another approach has been to suggest that slang speakers do not intend to mark their
distinction from other speakers. Major makes this claim of African-American slang: “African-
American slang is a kind of ‘home talk’ in the sense that it was not originally meant for listeners
beyond the nest” (1994: xxix). Yet this explanation of slang as inherently in-group is also difficult
to sustain. Major is clearly not proposing that African-American slang has never been used in
front of outsiders. The only way that this slang could have acquired a sense of being “home talk”
is in relation to other language varieties used by other groups of speakers. African-American
slang speakers – like all speakers – embrace linguistic resources such as slang in the process of
doing in-group/out-group work. In sum, I do not dispute that slang is used to make in-group
Jennifer Roth-Gordon
Slang and the Struggle over Meaning
10

---