Lily Maya Wang
B.C.L./LL.B. III, Faculty of Law, McGill University
1 October 2019
“What matters is how you look on paper – not how you sound over the phone. Judging you by your race or color instead of your qualification is discrimination.
It’s unfair, it’s painful… and it’s against the law.”
– National Fair Housing Alliance (NFHA)
Images: Advertisements to alert African American and Latino/a populations about the subtle forms of linguistic discrimination, produced by the NFHA in the United States. The NFHA officially recognized that many members of minority groups were simply unaware of the illegality of linguistic profiling.
“Race in America is a visual phenomenon,” writes Jennifer Lynn Stoever in The Sonic Colour Line. Yet, Stoever argues, “Aural and visual signifiers of race are thoroughly enmeshed.” The focus on the look of race has obscured the ways it sounds, allowing a key tool of white supremacy to work in the shadows. Stoever writes that, “Because racism seems to be a ‘discourse of power that thinks with the eyes’ in a culture driven by an ‘overdetermined politics of looking,’ sound has served as a repository of apprehension, oppression, and confrontation, rendered secondary— invisible— by visually driven epistemologies.”
How exactly, is race inscribed in speech such that sound can play a key role in maintaining racial hierarchies? Does human rights law have the tools to address sonic racial discrimination in areas like employment, housing, and the criminal justice system? If the law is going to truly address racial discrimination, it will need to go beyond seeing skin colour as the exclusive site of racialization and begin to hear sonic racial discrimination.
In this probe, I argue that the law may present various hurdles and paradoxes in finding appropriate solutions to address linguistic profiling. It is my hope that by understanding linguistic profiling we will be able to move towards a more open and inclusive society, celebrating linguistic diversity instead of stigmatizing it.
1. Linguistic Profiling – What is at Stake?
“Linguistic profiling” is the auditory equivalent of visual “racial profiling.” It has to do with inferences and judgments made on the basis of a person’s speech. These include a person’s place of origin, gender, intelligence, and race. In “Linguistic Profiling and the Law,” Dawn Smalls argues that linguistic profiling is not based on “racial genetics” or any anatomical differences, but that “linguistic differences are ‘learned speech characteristics’ that connote ‘shared knowledge about society in general, interlocutors’ positions in society, and appropriate discourse norms given the discourse situation.’”
Linguistic profiling also shows the complex entanglement of race and class in discrimination. Smalls argues that linguistic profiling “most acutely affects linguistic communities of colour,” but it also affects “whites speaking with ‘undesirable’ accents.” Linguistic profiling is not simply about excluding “foreign” voices; it does not treat all non-native accents alike either. In their 1993 study on accent discrimination, Quinn and Petrick write, “low-status accents are more likely to be interpreted as difficult to understand and indicative of incompetence, whereas high-status accents are more likely to be interpreted as easily understood and suggestive of competence.” For example, Hispanic, African, Asian, or Eastern European accents are often singled out for negative discrimination, while British accents are often heard as connoting high levels of intelligence.
In her seminal paper, “Voices of America,” Mari Matsuda explains that “certain dialects and accents are associated with wealth and power. Others are low-status, with negative associations. In a society with a speech hierarchy of this kind, it is quite common that speakers of the low-status speech variety, by necessity, are able to understand speakers of the high-status variety. Speakers of the high-status variety, on the other hand, frequently report that they cannot understand speakers below them on the speech-status scale.”
2. Linguistic Discrimination – What are the Problems?
While open discrimination based on race, ethnicity, or country of origin may no longer be acceptable, linguistic profiling seems to be fair game. Every instance of speech is met with subtle judgments about our origins, wealth, education, social status or other characteristics. The effects of this kind of profiling are not confined to bad treatment at the ears of others. Profiling can be internalized, leading to fears and self-consciousness and low self-esteem. This can have negative consequences on a person’s day-to-day activities, job performance, and educational progress.
For many people of colour and immigrants, the stigma of speaking differently has meant that their voice has been met with hostility. Some people invest time and money on speech therapy to reduce their accents and the sting of bias, while others seek recourse in the law.
Linguistic discrimination takes place in many contexts, from employment, to housing, to the criminal justice system. The form it takes depends on the situation. For example, Smalls recounts the case of Ferrill v. The Parker Group Inc. (1999), in which a telemarketing firm hired to make get-out-the-vote calls for political candidates assigned callers different scripts and areas, based on their race. Ferrill, an African American employee sued after she was dismissed following the election. The firm argued that voters would respond better to callers that sounded like them and applied this policy uniformly.
Smalls cites a 1994 study in which woman seeking to rent an apartment “called, and they told her that the apartment was rented. And she called [a friend] on the phone and said ‘I’d like for you to call them … because you sound like a white person.’ And the friend called and the apartment was still unrented.”
“Most troubling,” writes Smalls, is the appearance of linguistic profiling in criminal law, where “witnesses are regularly allowed to testify to the race of the speaker based solely on the linguistic inferences gathered from the speaker’s speech, without additional identifying information. The majority of the case law includes cursory references to testimony offered by a witness that a perpetrator ‘sounded black’ or ‘sounded white.’”
Image: Linguistics Professor John Baugh’s personal experience with linguistic profiling when making telephone calls and looking for housing. A short clip of the experiment can be seen here.
3. Legal Paradox – Are There (No) Solutions?
Attempts to address linguistic profiling through the law have led to seemingly contradictory results. Linguistic profiling has been accepted as legal in some instances and illegal in others. For instance, in Clifford v. Commonwealth (1999), the Supreme Court of Kentucky affirmed the conviction of Charles Clifford, an African American man whose conviction was based on linguistic profiling. In Clifford, the police organized a sting operation with an informant. The informant set up a meeting between an undercover officer and Clifford at the informant’s apartment. Another officer, Smith, monitored the sting through a wire. The informant emerged from a bedroom with $75 worth of cocaine. The informant later testified that this cocaine was his and that he had sold it to the officer, not Clifford. Finding that the recording of the operation was inaudible, the court allowed Smith to testify that he heard the voice “of another male, the voice of a female, and then later a fourth voice which ‘sounded as if it was of a male black.’” Under cross examination, Smith’s testimony showed the shakiness of this evidence:
Q. Okay. Well, how does a black man sound?
A. Uh, some male blacks have a, a different sound of, of their voice. Just as if I have a different sound of my voice as Detective Birkenhauer does. I sound different than you.
Q. Okay, can you demonstrate that for the jury?
A. I don’t think that would be a fair and accurate description of the, you know, of the way the man sounds.
Q. So not all male blacks sound alike?
A. That’s correct, yes.
Q. Okay. In fact, some of them sound like whites, don’t they?
A. Yes.
Q. Do all whites sound alike?
A. No sir.
Q. Okay. Do some white people sound like blacks when they’re talking?
A. Possibly, yes.
Smith was sentenced to twenty years in prison as a persistent felony offender. On appeal, the court concluded there was sufficient evidence to reach this conclusion. The dissenting judge would have overturned the conviction because
Race, that is, skin color, must be perceived by sight. To say that a person is capable of ascertaining another’s race solely by hearing his voice is tantamount to saying that one can “hear a color” or “smell a sound” or “taste a noise.” One can no more determine that a person’s skin is pale, cinnamon, or ebony simply by hearing his voice, than one can perceive that an individual will have a British accent, a Portuguese accent, a New York accent or an Appalachian accent simply by gazing at his countenance and the color of his skin. Thus, it was entirely improper to permit Officer Smith to testify that the fourth voice on the videotape “sounded black.” This type of testimony would be improper in any context, but it is all the more so because the defendant was the lone black man sitting at the defense table.
Other courts and tribunals have reached different conclusions. For example, In Gajecki v. Surrey School District No. 36 (1989), the British Columbia Human Rights Tribunal found in favour of a teacher who alleged discrimination based on his accent. Gajecki, a Canadian school teacher originally from Poland, was denied employment despite his satisfactory teaching record. The Tribunal found that discrimination based on his accent fell under the protected grounds of ancestry and place of origin.
Murray Munro argues that, in the Canadian human rights context, claimants face several hurdles in successfully challenging linguistic discrimination. Firstly, human rights acts and codes do not list accent as a protected ground. Claimants must show, like Gajecki did, that discrimination based on accent is really discrimination based on a listed ground. Secondly, tribunals may find that some discrimination based on accent is acceptable, for example, if language proficiency is a bona fide occupational requirement. Perhaps most significantly, Munro notes, because most discrimination goes unreported and most complaints are resolved before they reach the tribunal, “no one knows how often language discrimination occurs in Canada or whether this phenomenon is increasing or decreasing in frequency.”
Conclusion:
Like so many legal problems, linguistic discrimination cannot simply be solved by legislators and judges. Rather, those in the legal profession, as well as all members of our society, need to critically interrogate the ways we hear. If law is to address discrimination based on sound, we need to understand, in Stoever’s words, “how listening operates as an organ of racial discernment, categorization, and resistance in the shadow of vision’s alleged cultural dominance. While vision remains a powerfully defining element of race, scholars have yet to account for how other senses experience racialization and enact race feeling, both alone and in concert with sight.” The solution will not simply be to move towards an aural equivalent of racial colour-blindness but to “jam the sonic color line’s aural signals, enabling more equitable listening practices to emerge.”
All links accessed August 13, 2019.
Further Readings:
Are Accents One of the Last Acceptable Areas for Discrimination?
This is What Housing Discrimination in the U.S. Looks Like
When an Accent Becomes an Issue; Immigrants Turn to Speech Classes to Reduce Sting of Bias – New York Times
You must be logged in to post a comment.