Pieter Bruegel, The Tower of Babel

Pieter Bruegel, The Tower of Babel

The Truth About Accents

Setting the Record on Accents Straight

First, I must acknowledge that in science, the term 'truth' is used cautiously, if at all. Science rests on hypotheses, theories, and, more rarely, laws, but even those are not “the truth.” In physics, the relativistic laws, for example, break down at a singularity. Social scientists are not even fortunate enough to have laws that break down, we have theories and hypothesis. We do test them empirically, in case of experimental psychology (in which I am trained and to which I subscribe) in the form of scientific experiments. Because we rely on statistics and probability, can show and demonstrate but not prove consistently. Thus, any evidence from research on accents is either in support or against a theory. Although we have no access to the truth, we have access to scientific evidence, which is much more reliable than anecdotes, opinions, or unsubstantiated claims. However, even scientific evidence is inconsistent, mainly because the topic of examination, accents, is complex and affected by a variety of factors. Nevertheless, here I hope to set the record a bit straighter, and summarize what in truth has been scientifically demonstrated when it comes to accents.

Definitions of Accent

Merriam-Webster dictionary defines accent as “a distinctive manner of expression: as a: an individual’s distinctive or characteristic inflection, tone, or choice of words —usually used in plural b: a way of speaking typical of a particular group of people and especially of the natives or residents of a region”. Dictionary.com provides a similar definition, an accent is “a mode of pronunciation, as pitch or tone, emphasis pattern, or intonation, characteristic of or peculiar to the speech of a particular person, group, or locality: French accent; Southern accent” and “such a mode of pronunciation recognized as being of foreign origin: He still speaks with an accent” and “the characteristic mode of pronunciation of a person or group, esp one that betrays social or geographical origin”.

Unfortunately, these definitions are rather general. An accent is the way one speaks, one’s pronunciation, which is determined by and which betrays the speaker’s social or geographic background. It does not answer the questions of what exactly accent is. Although the concept of “having an accent” or “speaking with an accent” is widely used, it is not a precise phrase. Even linguists disagree as to the exact meaning of speaking with an accent. They usually use aspects of phonology (how sounds are organized and used in language) to explain accents; accents include features of intonation , pitch, stress, tempo, as well as phonological structures, called phonemes , which differ from language to language.

A Note on Non-Native/Foreign Accent Distinction

The distinction between a non-native and foreign accent is quite subtle. “Non-native” refers to an accent of a non-native speaker of the language, whose native (first) language is different from the language in which the speaker has a non-native accent (referred to as “second” language). A “foreign” accent is a broader term: it includes the definition of a non-native accent, but it can also refer to the accent of a speaker whose native language is the same but from a different country. For example, a Japanese person in the United States speaks with a non-native and foreign accent. However, an Australian in the United States speaks the same native language (English) but with a foreign (Australian) accent. For the sake of precision, I use the term non-native accent when referring to non-native speakers, because foreign accent could potentially refer to other native speakers.

The Theory Behind Why Accent Is Important

The lack of an exact formal definition of accent does not interfere with people’s ability to readily determine whether someone has an accent or not. One does not need to know the secrets of linguistic analysis on intonation and phonemes to decide whether someone has an accent. Indeed, native speakers can detect a non-native accent in their language in less than 30 milliseconds and in speech played backwards! Babies as young as five months can perceive a foreign accent in the language that is going to become their native language.

At this point, evolutionary theory proposes the most convincing explanation for this unique sensitivity to traces of non-native pronunciation. The theory suggests that before our ancestors were able to travel long distances, their only contact was with neighboring tribes, who physically looked very much alike. The only way to tell the difference between members of one’s tribe and outsiders, who might be dangerous, was the way they spoke. As a result, we evolved sensitivity for detecting unfamiliar patterns in the way others speak. Speech marked a potential friend or foe, differentiated someone who might help from someone who might harm, and dictated as to whether an individual would be approached or avoided. It was only later on, when people started traveling further distances, that they met others who not only spoke but also looked differently. Taking advantage of the system already developed in the brain that equaled “different” with “potentially dangerous,” race was added as yet another trait to which humans became sensitive.

Why Accent Matters

In itself, this sensitivity is not problematic: we are simply able to detect when people speak our native language whether they are native speakers or not; if they are native speakers, we can further determine if they speak a regional variety of the same language. However, problematic are the consequences of this sensitivity and the processes it initiates. The association between “accent” and “dangerous,” whether conscious or not, may lead a native speaker to be more cautious and suspicious of the non-native speaker, which in turn, may cause stereotyping, prejudice, and discrimination, and ultimately, affect the non-native speaker’s psychological and physical well-being, educational and employment opportunities and advancement, income, and overall quality of life. In other words, the way one speaks matters.

Accent Stereotyping and Prejudice

Heaven is where the police are British, the cooks are French, the mechanics German, the lovers Italian, and it is all organized by the Swiss. Hell is where the chefs are British, the mechanics French, the lover's Swiss, the police German, and it is all organized by the Italians.

Stereotypes refer to specific characteristics, traits, and roles that a group and its members are believed to possess. Stereotypes can be both positive and negative, although negative are more common. The joke above illustrates both. Stereotypes also influence people process and retrieve information. For example, when we meet a man from France, we may make assumptions that he is a great cook, but not a good mechanic, simply because he is from France. He indeed might be a good cook and a bad mechanic, but we do not know that, he could be a terrible cook and a great mechanic, good or bad at both, or simply average.

Stereotypes sometimes result in prejudice, which means having negative attitudes toward a group and its members, which may be based on stereotypic beliefs about the group. Throughout history, various groups have experienced prejudice, and unfortunately, prejudice is still very much part of our lives. Non-native speakers often have to deal with both negative stereotypes and prejudice resulting from speaking with an accent. Decades of research have shown that accents are associated with a range of negative stereotypes and attitudes. On average, native speakers find non-native speakers less intelligent, less competent, less educated, having poor English skills, and unpleasant to listen to. Stereotypical portrayals of accents are also common in the media (see Accents in the Media). People with stronger accents are judged even harsher and native speakers who have trouble understanding accented speech may experience negative feelings toward non-native speakers for refusing to learn to speak the language “properly.” An accent also marks the speaker as an immigrant and immigrants as a group have frequently experienced prejudice from native speakers.

Accent Discrimination

Discrimination refers to specific behaviors or actions directed at a group or its individual members based solely on the group membership. In accent discrimination, non-native speakers’ way of speaking is used as a basis for arbitrary evaluations and judgments. In addition, unlike other forms of discrimination, it is not condemned in the general society. Lippi-Green, an author a book on accent discrimination, writes,

“Accent serves as the first point of gate keeping because we are forbidden, by law and social custom, and perhaps by a prevailing sense of what is morally and ethically right, from using race, ethnicity, homeland or economics more directly. We have no such compunctions about language, however. Thus, accent becomes a litmus test for exclusion, and excuse to turn away, to recognize the other.”(1)

Speakers of accented English often experience discrimination in housing and employment, landlords are less likely to call back speakers who have foreign accents and non-native speakers are more likely to be assigned by employers to lower status positions than are those with standard accents. Although in the United States, Title VII of the Civil Rights Act of 1964 prohibits discrimination based on national origin, implying accents, it does not mention them specifically. It also comes with a caveat: Employers can insist that a person’s accent impairs communication skills that are necessary to the effective business operation and be off the hook. The courts often rely on the employer’s claims or use judges’ subjective opinions when deciding whether the (potential) employee’s accent would interfere with communication or performance, without any objective proof that accent was or might be a hindrance. Perhaps unsurprisingly in a nationwide representative survey of workplaces, 10 percent of employers admitted having discriminated on the basis of a person’s foreign appearance or accent.

Accent discrimination is also widespread among educational institutions, which are at the forefront of preserving the myth of standard language and pronunciation. Non-native speaking graduate students, lecturers, and professors, across college campuses in the US have been blamed for low English proficiency and being unintelligible. There are certainly instances of instructors not having sufficient spoken language proficiency, which interferes with their ability to communicate. However, there has been a growing concern among undergraduate students, parents, administrators, and educators that undergraduates suffer academically because they are taught by non-native speakers of English. On average, however, students taught by non-native English speaker do not underperform when compared to those taught by native speakers of English. Studies have shown the perception of the accent, not the accent by itself, often results in negative evaluations of non-native teachers. In one clever study (Rubin, 1992), students listened to a taped lecture recorded by a native English speaker with a standard accent and were shown a picture of the lecturer. However, half of the students were shown a picture of a Caucasian instructor and the other half saw a picture of an Asian instructor. Students who saw the Asian picture believed that they had heard an accented lecturer and performed worse on a task measuring lecture comprehension. Thus, negative evaluations may reflect the prejudices that one holds rather than real issues with comprehensibility.

If it causes problems – Let’s remove it! – Accent Elimination

The obvious question to ask is that if speaking with an accent leads to being stereotyped, experiencing prejudice and discrimination, then why won’t non-native speakers work really hard at eliminating their accents? The anecdotal stories of the rare instances in which adults have obtained native-like accents abound. The media show actors imitating various accents (with varying degrees of success; see Accents in the Media), which implies that non-native speakers can acquire a native-like accent should they choose to work hard.

This is the reasoning behind many accent reduction classes, which have become a profitable business by guaranteeing non-native speakers a better life if they lose the accent. Yet, the effectiveness of such classes has not been systematically assessed. They are often led by untrained and unlicensed individuals, are not subject to any regulations, and lack an industry-wide standard (although there are professionally trained, licensed speech pathologists that focus on non-native accent acquisition). Even if one’s pronunciation does improve, it is often because one did not learn it correctly in the first place, and similar results could be achieved by self-training by listening to tapes while commuting, often costing less than $50, not the hundreds or thousands of dollars the other programs can cost. There is evidence that people can somewhat improve their pronunciation with long-term practice. Unsurprisingly, the most rapid improvement takes place within the first year of arrival to the host country. However, after the initial year, non-native speakers are highly unlikely to ever completely eliminate their accents, or even to “reduce” them more.

The principles behind such programs are also questionable. Accent reduction, eradication, or elimination, implies that an accent is a bad trait that needs to be clinically treated, listed among speech pathologies. If non-native speakers are unable to acquire native-like accents, then, some researchers have suggested, efforts may be more productively directed at modifying native speakers’ attitudes and training them in accent comprehension. Indeed, there are more non-native speakers of English in the world than there are native speakers, around one billion, and this number is likely to increase as globalization progresses. It may become more effective and more efficient to “eradicate” listeners’ negative attitudes than attempt to eliminate the foreign accents of non-native English speakers.

The Science Behind Accent Acquisition

Most researchers agree that for adults, acquiring a native-like accent in a non-native language is near impossible. The age at which the non-native language is learned is the most significant factor in predicting if a person will acquire a native-like accent and if not, to what degree the non-native accent will be noticeable (or strong). Native language is the first language a person learns. By definition, it is impossible for non-native speakers to become native speakers in the language they did not learn first, they would have to be born again. Only individuals who learn two languages simultaneously from birth, at the same rate, are considered native speakers in two languages.

Pronunciation is the most difficult part of a non-native language to master. In fact, most individuals who speak a non-native language fluently speak it with an accent of their native tongue. The critical period theory states that if learning takes place after the critical period (usually considered puberty) for acquiring native-like pronunciation, an individual is unlikely to acquire a native-like accent. This theory, however, is quite controversial among researchers. Although many subscribe to some form of the critical period, they either place it earlier than puberty or consider it more of a critical “window,” which may vary from one individual to another and depend on factors other than age, such as length of residence, similarity of the non-native language to the native language, and the frequency with which both languages are used. Nevertheless, children as young as 6 at the time of moving to another country, often speak with a noticeable non-native accent as adults. There are also rare instances of individuals who are able to pass for native speakers even if they learned their non-native language in early adulthood. However, neurological constrains associated with brain development appear to limit most non-native speakers’ ability to sound native-like.

As a result, even if individuals truly want to obtain a native-like accent in their non-native language and they work hard at eliminating their non-native accent, they may simply be unable to. An encouraging recent development in the second language acquisition literature advocates for giving up a native speaker as the standard to which non-native speakers are compared, especially in regard to pronunciation. Non-native speakers are unique in that they no longer resemble monolingual speakers of their native language and will never resemble monolingual speakers of their non-native language. Insisting that non-native speakers should strive to become native-like in their pronunciation, and when they do not, judging them as having failed, is like “saying that ducks fail to become swans.”2 A non-native speaker should be considered as neither native nor non-native but as, what has been termed, a multicompetent speaker.

Accent, Competence, and Communication

Speaking about accent invariably brings about issues of language competence and communicative effectiveness. Undoubtedly, the more competent one is in a language, or how well one knows it, the fewer problems one will have communicating and interacting with native speakers. However, as stated above, even speakers who are fluent in a non-native language often retain noticeable accents, which can interfere with communication. This is a quite controversial issue, one that lies at the heart of accent discrimination suits, alleged ineffectiveness of non-native teaching assistants and lecturers, and everyday interactions between native and non-native speakers. Does accent interfere with communication? Yes or no? The answer is not that clear.

Yes, because research demonstrates that non-natively accented speech requires more time to process for native speakers. Accents also vary in how objectively comprehensible they are, meaning how much of the speech a listener actually understands (also referred to as intelligibility). However, researchers have not been able to agree which features of accented speech contribute to a slower processing time and interference with objective comprehensibility.

No, because research shows that although in the beginning accented speech may take longer to process, listeners adapt quickly to the accented speech, usually in less than a minute. Moreover, stronger accents are sometimes easier to understand for listeners than weaker accents, contradicting the claim that stronger accents are harder to understand and making it unclear as to what features of an accent make it difficult to understand.

It is important to remember that understanding of accented speech is not fully dependent on the characteristics of the accent itself. Inability to identify specific features of accents that makes then less understandable suggests that other factors contribute to comprehensibility. One such critical factor is the listeners’ beliefs about their ability to understand accents, regardless of the “objective” comprehensibility of a given accent. If the listener expects accented individuals to be difficult to understand, they will be harder to understand. The perception that because the speaker has an accent it must be hard to understand him or her leads to self-fulfilling prophecy, even if in reality the accented speech is easily comprehensible. Recall the study in the example above, in which researchers showed that students found a lecture harder to understand because they thought the lecturer had an accent, even though the speech was that of a native English speaker with the standard American accent.

The degree to which characteristics of an accent in comparison with the listener’s perceptions and biases hinder comprehension is unclear. Clearly, however, communication involves mutual responsibility, which is responsible for a successful exchange. Speakers try to make themselves understandable and listeners attempt to understand them. Indeed, only listeners can assess whether they understood an utterance, heavier burden falls on them, because they need to let the speaker now if they did not understand something. However, native speakers repeatedly, even if unintentionally, shift the burden onto the non-native speakers and make them responsible not only for producing clear speech but also in assuring that listeners understand it. In extreme cases, they may pretend not to understand or avoid speaking to non-native speakers. However, native speakers, on the other end of extreme, may attempt to go out of their way to “help” a fluent non-native speaker by using so-called foreigner talk by speaking much louder, slower, and using simplified grammar and wording, and peculiar gestures. Their intentions might be noble, but may result in the non-native speaker feel condescended and patronized.

What about Languages Other Than English And Regional Varieties?

On this website, I cover mostly literature relevant to non-native accents, especially in the US and other English-speaking countries. There are several reasons for this limitation. First and the most important is that literature on this topic is limited for non-English speaking countries. However, when it is available (and is in English), findings often mirror those obtained in English-speaking countries. Given the variety of political systems, unique immigration policies and history, and historical and contemporary intergroup relations, without more extensive empirical research, extrapolations to other countries should be made cautiously. Second, English-speaking countries combined have a significantly larger contemporary population of immigrants than countries speaking any other language. Third, English is de facto the international language of communication, and thus spoken most often between those who speak it natively and those who do not, in business, academic, and leisure settings.(3)

Regional varieties of English play different importance in the various English-speaking countries and some are more stigmatized than others. However, such accents interfere less with perceived and actual comprehension. Regionally accented speakers are native speakers of the same language as people with standard accents and they are less likely than non-native speakers to suffer from the negative consequences related to communicative effectiveness. Similarly, individuals with other forms of English accents (e.g., British, Australian, and Canadian) are connected by being native speakers of the same language. Reactions to other native accents are different from responses to non-native accents. Indeed, in the US, the accents of native speakers of English from other countries are often seen as more prestigious, pleasant, and friendly.

But Aren’t Some Accents Better/Nicer/Sexier?

Native speakers often claim that specific accents “just sound nicer, or better, or sexier, or more attractive, etc.” than other accents. In linguistics, all forms and varieties of language are equal: no language, dialect, or accent is better or worse than some other. Most social scientists argue that some accents are found more aesthetically pleasing than others because of the social connotations ascribed to them. Nothing intrinsic to accents makes them more or less pleasing to listen to.

Now What?

The question I am most often asked when summarizing my research and scientific evidence on accents is: Now what? This seems to be a grim view of reality. Non-native speakers start believing that they cannot do anything about their accents, native speakers feel guilty of their inability to understand accented speech. Are we better off? Yes, because in the knowledge rests power. This is the point where I invite you to explore The Linguistic Tolerance Project.

Note: The Truth About Accents is based on my notes for my master’s thesis project, which eventually resulted in an article I published in Personality and Social Psychology Review (for the citation, see Biography). The publisher owns the copyright to distribute the article, but if you are interested in reading it for your personal/educational/research use, feel free to email me.

Sources: The Truth About Accents is based on a summary and synthesis of the findings, arguments, and suggestions from all sources listed in Accent Bibliography.

________________
1 Lippi-Green, 1997, p. 64.
2 Cook, 1999, p. 187.
3 The research on communication in English between non-native speakers has become increasingly important and its own area. Although some of the issues that emerge in interactions between non-native speakers are similar, many are not, and require their own in depth analysis and discussion.

back to The Truth About Accents

Phonemes
1&1 Green Logo Valid CSS! Valid XHTML 1.0 Transitional