An examination of language attitudes towards motion picture announcers’ voices
Joshua Liez
jliez@sas.upenn.edu
Ling 057
Language and Popular Culture
December 6, 2001
Language
is, of course, the primary means of human communication, in that it transmits
some sort of content from the speaker to the listener. That is, the words a person says have
meaning, so when someone says “bring the ball,” for example, the listener knows
to retrieve a spherical object. But
language is not just a carrier of content, it also is content, as Fishman
points out (Ryan et al. 1982: 2). In
other words, what a person says is not the only way in which language transmits
content; how someone says something is equally content-laden. Depending on the particular listener, a
speaker’s accent, speech patterns, vocabulary, intonation, etc. can all serve
as markers for evaluating that person’s appearance, personality, social status,
and character, among other things.
Listeners, in short, possess language attitudes, which they use to
evaluate the speakers whom they are hearing.
It is these language attitudes that I want to examine in this study.
Ryan et al. define language attitudes as “any affective, cognitive or behavioural index of evaluative reactions toward different language varieties or their speakers” (Ryan et al. 1982: 7), and linguists have long been interested in studying these attitudes. But it was not until the early 1960s before systematic examinations of language attitudes were undertaken. The most effective method of testing such attitudes, investigators quickly realized, was what is known as indirect measurement of language attitudes, where subjects evaluate speakers, and the language attitudes are inferred from these evaluations. The pioneering study of this sort was a 1960 study of French and English accents in Montreal by Lambert, Hodgson, Gardner, and Fillenbaum (Edwards 1982: 22). The investigators introduced a new method in this study, known as matched-guise, which has proved to be one of the most useful means of assessing language attitudes. In a matched-guise study, the same speaker says the same neutral content in at least two different language varieties, and listeners are then asked to evaluate the speakers. In this way, what is being tested is solely the language variety, and not the content or the speech patterns, etc. Matched-guise studies have comprised the majority of the work done on language attitudes, with most of these studies comparing the differing attitudes towards some standard speech and a nonstandard speech, such as a regional dialect, for example. The generally-accepted results from most of these studies have suggested that the standard speakers are generally rated by listeners as being more competent (more intelligent, hard-working, and confident), while nonstandard speakers are rated as having more integrity (more honest and trustworthy) and personality/social attractiveness (more friendly and warm) (Giles 1972: 169-70; Ryan et al. 1982: 8; Edwards 1982: 23). Interestingly, the social status or class of the listener does not seem to change these results significantly. So, even nonstandard speakers tend to rate other nonstandard speakers (even of the same language variety) as having less status, prestige, and competence, and standard speakers as being of a higher status (Edwards 1982: 26). What this suggests, as Edwards and others have noted, is that many of the language attitudes present in a society are the result of social connotations (Edwards 1982: 21; Romaine 1980: 213). There is nothing intrinsically better, of a higher status, or suggesting higher competence in standard speech, as numerous investigators have concluded (Edwards 1982: 21). Rather, the members of a society—both standard and nonstandard speakers—learn the same social prejudices and conventions regarding standard and nonstandard speech, and thus the ideas that standard speech evokes higher status is ingrained in us at a young age (Romaine 1980: 213).
Since the vast majority of studies on language attitudes have focused on standard/nonstandard speech, my interest in this study is to move away from that slightly and move into the realm of mass popular culture, where all speech is, in essence, standard. In media, like television, radio, movies, and other forms of mass media, the language is generally standard speech since it is intended to reach as large an audience as possible. There is simply very little nonstandard speech, unless it is needed for a purpose (for example, an editorial on black civil rights might be better suited to be given by a black person). Instead, what arises in mass media is what we might call “exaggerated standard speech”—speakers whose speech patterns are so good that they sound little like anyone else in the society. Thus, it seems that the speech variables in these mass media speakers are idiosyncrasies and exaggerations on standard speech. In this study, I am interested in examining this exaggerated standard speech in mass media, since very little research has been done on it in the past. Specifically, I will examine some of the language attitudes that are evoked in people upon hearing movie announcers—something most of us hear on tv, in theaters, or on the radio nearly every day. My hypothesis is that in such situations where all the speakers are standard speakers, the speakers whose speech is most exaggerated will, in a sense, assume the role of nonstandard, and thus they will be viewed as being of a lower status, yet more warm and friendly, whereas those speakers whose speech is most standard will be viewed as high-status, highly competent, yet lacking in personality and integrity. I also hypothesize that the listener’s attitudes will have some correlation to the genre of the movie the announcer is speaking for. For example, I imagine a sci-fi movie will employ a serious, deep-voiced announcer whom the listeners will see as serious and competent (intelligent, confident, industrious), and not very warm or kind. On the other hand, announcers for family or children’s movies will probably be rated highly on personality and integrity, yet lacking in competence traits. Comedy announcers will probably be rated very highly on personality traits (funny, friendly).
The first matched-guise study that examined
language attitudes was the 1960 study by Lambert, Hodgson, Gardner, and
Fillenbaum, examining the reactions of English and French speakers in Montreal
to English and French guises.
Interestingly, the investigators found that both English- and French-speaking
listeners rated English speakers as being higher status. This matched-guise study was the model on
which the vast majority of subsequent language attitude studies were based.
Another important investigator on the topic of language attitudes is Giles, who has done numerous studies on British standard and nonstandard accents. Of particular interest to Giles was how the social attitudes of listeners affect their language attitudes. In his 1972 paper, Giles builds upon others that he has done to test listeners’ evaluations of personality content from various types of accents. To examine the different language attitudes based on social attitudes, he used two different social attitude classes: 20 “highly ethnocentric” British English speakers (as defined by the British E Scale) and 20 “lowly ethnocentric” speakers, and played for them six taped voices reading a neutral passage. The voices represented three accents, though all were voiced by two speakers speaking in guise: standard accent British English, South Welsh, and Somerset accents. Thus, each listener heard the two different speakers each speak in three accents (one standard, two nonstandard), thinking they represented six different speakers, and was then asked to rate the speakers personality on 18 traits using a seven point scale. Among the traits, for example, were physical traits, such as good looks; competence traits, such as ambition, intelligence, and self-confidence; and personal integrity traits, such as goodheartedness. Giles hypothesized that ethnocentric listeners would rate nonstandard regional speakers less favorably and standard speakers more favorably than non-ethnocentric listeners across all traits.
The results showed that Ethnocentric listeners rated nonstandard speakers less favorably and standard speakers more favorably than the non-ethnocentric speakers on four traits (all competence traits)—intelligence, determination, ambition, and industriousness. Yet the non-ethnocentric listeners rated all speakers more favorably than did the ethnocentric speakers on social attractiveness and integrity traits, such as sociability, humorousness, popularity, and kindheartedness. Thus Giles concludes that competence is the dominant factor in evaluation of standard and nonstandard accents (favoring standard speakers), and integrity and social attractiveness are important factors favoring nonstandard speakers (though not as important as competence in standard speakers). The results also confirmed that social context does indeed play a major role in language attitudes. Giles suggests that less ethnocentric people are more prepared to perceive favorable qualities in all speakers, whereas highly ethnocentric people tend to rate standard speakers more favorably.
This
is an important paper for my study because it is an example of a matched-guise
study testing language attitudes, and it provides a framework of questions,
methods, etc. on which I can base my study.
Further, the idea that language attitudes are socially-constructed,
which this study supports, is a major part of my study.
Romaine’s (1980) study is a matched-guise experiment examining the reactions of subjects to different Scottish accents. She presented recorded voices from six speakers, each speaking in a reading and a conversation situation (the two guises). The subjects were then asked to evaluate the voices. As in the Giles study, the experimenters found that the reactions of the subjects depended on the guise of the speaker (i.e., whether reading or conversing). Most of the speakers were evaluated less favorably (less well-educated, lower social status, sloppier) when conversing, as opposed to when reading. Furthermore, all the speakers were thought to be more nonstandard, regional speakers when conversing, as opposed to reading—probably due to the fact that conversation speech is generally less neat and grammatically correct than reading speech. When this is compounded with the result that the more Scottish-sounding regional speakers scored higher on personality ratings, it supports the contentions of numerous investigators that regional speakers are generally perceived to be of a lower status, yet are thought to be more honest, warm, and genuine.
One of the values of this paper for the present study is that it introduces certain concepts that are quite useful. For example, Romaine talks about Labov’s belief that perception of language is, in fact, perception of social experience and socially accepted statements about language, which is based upon the idea that perception is mediated through stereotypes of the groups that are believed to speak in that way. So, like Giles’ study, this study supports the idea that language attitudes are social constructs. Romaine’s study is also very valuable as a roadmap for the technical aspect of a matched-guise experiment. She introduces the idea of a semantic differential, which is a method for having subjects evaluate speakers based on a scale between two polar opposites. For example, in order to test a listener’s views on speaker’s competence, Romaine asked the listener to rate the speaker on several characteristics, like intelligence and industriousness, using a seven-point scale. I will use a similar semantic differential in my study, using a five-point scale.
Ryan, Giles, and Sebastian (1982) and Edwards (1982) provide an excellent history and background of the examination of language attitudes over the past 40 years. Ryan, Giles, and Sebastian begin with an explanation of how language variation conveys content to listeners, which is the basis of language attitudes, and they go on to a technical review of the ways linguists measure language attitudes, and a summary of some of the data that have suggested that language varieties associated with high social class, such as standard speech, generally receive high competency marks, on things like intelligence, expertise, and ambition. Edwards gives a summary of the studies that have been done that suggest that language attitudes are based on social connotations, rather than on inherent characteristics of the speech. He follows this with a summary of the major studies that suggest that standard speakers are almost always seen by listeners as being of higher social status and having higher competence characteristics, while nonstandard speakers are seen as having more integrity and better personality characteristics.
Wober’s (1990) look at language and television is one of the few examinations that has been done on language attitudes towards mass media, in this case television. Much of his discussion focuses on the sorts of conversations that tend to be seen on television, and who is having these conversations, and thus is not relevant to this study, but he does touch on regional accents on television. Specifically, he points out that, according to poles, the British public vastly prefers Queen’s English, the standard speech, to other regional accents in tv news anchors, because it somehow has more dignity (Wober 1990: 571). Interestingly, the same audience overwhelmingly welcomed regional accents in drama or comedies on tv, and about 50% would approve of regional accents in news reporters or weathermen (Wober 1990: 572). Another interesting note is that when the audience was asked which accents were associated with a sense of humor, over 60% pointed to the accents of the working-class areas in Liverpool and London, while only 30% suggested Queen’s English. This is in accord with what other researchers have said about regional accents conveying a sense of personality (warmth, humor, etc.), while standard speech was seen as lacking in this respect.
Bradac’s (1990) highly-detailed review of language attitudes in the realm of impression formation begins with a very in-depth review of the history of linguistic work done on language attitudes and moves on to examine specific language attitudes. For example, he points out that there is a recognized attitude difference between male and female voices, with male voices seen as being powerful, loud, and dynamic (more competent), whereas female voices are generally seen as powerless, gentle, polite, emotional (Bradac 1990: 395). He goes on to examine the effect of things like vocabulary, and speech rate on language attitudes.
For my study, I performed a semantic differential study of five different motion picture announcers and a control voice. Each listener was given a sound byte of each of the six voices and was then asked to answer the thirteen questions on the semantic differential questionnaire for each voice (see Figure 1). A copy of the questionnaire is included in the Data section, and a description of the six voices used follows. There were 10 subjects (listeners) who rated the voices, all college students (5 male, 5 female).
The first voice played was the announcer for the movie Ice Age (20th Century Fox, 2001), a computer-generated graphics family movie. The announcer was a male and spoke with an American English accent. His voice is very breathy and high-pitched, and consonants at the ends of words sound very throaty. That is, they vibrate in his throat (almost as if he is swallowing them), and they tend to be exaggerated with an extra syllable, so “else” is pronounced “elshe”, and “them” is pronounced “them-uh.”
The second voice played was the announcer for the movie E.T.: The 20th Anniversary Edition (Universal, 2001; original, 1982), the remake of the 1982 fantasy/sci-fi movie. The announcer was a male and spoke with an American English accent. His voice is breathy and wispy, and sounds much like the Ice Age announcer’s voice, but it is much more low-pitched. Consonants at the ends of words are throaty—they vibrate in the throat as if swallowed, and tend to be exaggerated with an extra (subtle) syllable, so “neighborhood” is pronounced “neighborhood-ih”, “Unfolding” is pronounced “Unfolding-ih,” and “beginning” is pronounced “beginning-ih.” The sound “g” in “again” and “forget” is swallowed and vibrates in bottom of throat. Final “t”s seem to disappear when followed by consonant: “next spring” pronounced “nexspring.” Yet they are pronounced when followed by a vowel, as in: “experience the excitement, as E.T….”, and pronounced when they occur at the end of a sentence (in fact, they are accentuated).
The third voice played was the announcer of the film Monsters, Inc. (Disney and Pixar, 2001), a
computer-animated children’s movie. The announcer was male, and spoke with an American English accent. His voice was very deep and throaty, almost chesty, very similar to that of James Earl Jones. Final “t”s are not very pronounced when followed by consonants, and “r”s are stretched in the middle of words, so “world” is pronounced “worrr-uld.” The sound “c” at the beginning of words are very pronounced, so “closets” is pronounced “k-klosets.” Finally, the tone of the voice becomes significantly higher when the ad becomes humorous.
The forth voice played was that of the announcer of the film Amelie (Miramax, 2001), a French light comedy. The announcer was male and spoke with an American English accent. He pronounces “d”s at the beginning of words very shrilly, almost “t”-like, so “Do you know” is pronounced “t-do you know.” Also, “s”s in the middle of the word are very shrill and sharp: “this-s-s,” “is-s-s,” “dis-s-scovery,” “ques-s-st,” etc. Final “n”s at the ends of words are dragged out and breathy, with an extra subtle syllable added on, so “have in common” at the end of a sentence is pronounced “have in common-ih.” Long “i”s are very low and guttural and are very drawn out.
The fifth voice played was that of the announcer of the film The Lord of the Rings: The Fellowship of the Ring (New Line, 2001), a fantasy film. The announcer was a male, and spoke with a standard British English accent. Final “d”s are swallowed, so “world” is pronounced “worl-ud.” “S”s are pronounced very shrilly, so “centuries” is pronounced “centuries-s-s,” and “person” is pronounced “pers-s-son.” Yet “s”s are not always shrill, as in “tells.” The speech rate is very slow, and the voice is quite breathy. There is very little pitch variation, and the voice is quite low.
The final voice played, the
control, was a newscaster on NPR, and spoke in standard American English with no particular exaggerations.
Physical description questions (this label not shown to subjects)
1) Is the speaker tall or short?
Short 1 2 3 4 5 tall
2) Is the speaker good-looking?
not good-looking 1 2 3 4 5 good-looking
3) How old is the speaker?
definitely young 1 2 3 4 5 definitely old
Competence questions (this label not shown to subjects)
4) Is the speaker intelligent?
not intelligent 1 2 3 4 5 intelligent
5) Is the speaker self-confident?
not confident 1 2 3 4 5 confident
6) Is the speaker a hard worker?
not a hard worker 1 2 3 4 5 hard worker
Personal integrity questions (this label not shown to subjects)
7) Is the speaker honest?
not honest 1 2 3 4 5 honest
8) Is the speaker sincere?
not sincere 1 2 3 4 5 sincere
9) Is the speaker trustworthy?
not trustworthy 1 2 3 4 5 trustworthy
Personality/social attractiveness questions
(this label not shown to subjects)
10) Is the speaker serious or funny?
Serious 1 2 3 4 5 funny
11) Is the speaker kind?
not kind 1 2 3 4 5 kind
12) Is the speaker friendly?
not friendly 1 2 3 4 5 friendly
13) Describe, to the best of
your ability, the speaker’s voice and manner of speaking (ex., nasal, deep, swallows
“d”s, etc.).
Table 1. Response Data,
Voice 1 (Ice Age)
|
Question |
Female Average |
Male Average |
Total Average |
|
1 |
3.2 |
3.4 |
3.3 |
|
2 |
2.6 |
3.0 |
2.8 |
|
3 |
3.0 |
3.2 |
3.1 |
|
4 |
3.4 |
3.0 |
3.2 |
|
5 |
4.2 |
4.6 |
4.4 |
|
6 |
3.4 |
2.8 |
3.1 |
|
7 |
3.2 |
2.0 |
2.6 |
|
8 |
3.2 |
2.0 |
2.6 |
|
9 |
3.2 |
2.0 |
2.6 |
|
10 |
3.7 |
4.6 |
4.15 |
|
11 |
3.4 |
2.8 |
3.1 |
|
12 |
3.8 |
3.6 |
3.7 |
Table 2. Response Data,
Voice 2 (E.T.: The 20th
Anniversary Edition)
|
Question |
Female Average |
Male Average |
Total Average |
|
1 |
4.0 |
3.4 |
3.7 |
|
2 |
3.0 |
3.0 |
3.0 |
|
3 |
3.7 |
3.8 |
3.75 |
|
4 |
3.4 |
3.4 |
3.4 |
|
5 |
4.0 |
3.6 |
3.8 |
|
6 |
3.4 |
2.8 |
3.1 |
|
7 |
3.0 |
2.8 |
2.9 |
|
8 |
3.0 |
2.8 |
2.9 |
|
9 |
3.0 |
3.2 |
3.1 |
|
10 |
2.2 |
2.6 |
2.4 |
|
11 |
3.2 |
3.2 |
3.2 |
|
12 |
3.4 |
3.0 |
3.2 |
Table 3. Response Data,
Voice 3 (Monsters, Inc.)
|
Question |
Female Average |
Male Average |
Total Average |
|
1 |
3.6 |
3.2 |
3.4 |
|
2 |
2.2 |
2.0 |
2.1 |
|
3 |
4.9 |
5.0 |
4.95 |
|
4 |
4.3 |
4.4 |
4.35 |
|
5 |
4.6 |
4.2 |
4.4 |
|
6 |
3.8 |
3.4 |
3.6 |
|
7 |
4.1 |
3.6 |
3.85 |
|
8 |
4.2 |
4.0 |
4.1 |
|
9 |
4.2 |
3.6 |
3.9 |
|
10 |
2.6 |
2.4 |
2.5 |
|
11 |
4.0 |
3.8 |
3.9 |
|
12 |
4.2 |
3.6 |
3.9 |
Table 4. Response Data,
Voice 4 (Amelie)
|
Question |
Female Average |
Male Average |
Total Average |
|
1 |
4.4 |
2.8 |
3.6 |
|
2 |
3.8 |
3.0 |
3.4 |
|
3 |
3.0 |
2.6 |
2.8 |
|
4 |
4.0 |
3.2 |
3.6 |
|
5 |
4.0 |
4.0 |
4.0 |
|
6 |
3.2 |
3.2 |
3.2 |
|
7 |
3.2 |
3.2 |
3.2 |
|
8 |
3.4 |
3.2 |
3.3 |
|
9 |
3.2 |
2.8 |
3.0 |
|
10 |
2.8 |
2.6 |
2.7 |
|
11 |
3.2 |
3.0 |
3.1 |
|
12 |
3.2 |
3.2 |
3.2 |
Table 5. Response Data,
Voice 5 (The Lord of the Rings: The
Fellowship of the Ring)
|
Question |
Female Average |
Male Average |
Total Average |
|
1 |
3.8 |
3.4 |
3.6 |
|
2 |
2.2 |
2.8 |
2.5 |
|
3 |
5.0 |
5.0 |
5.0 |
|
4 |
4.4 |
5.0 |
4.7 |
|
5 |
4.8 |
4.6 |
4.7 |
|
6 |
4.6 |
4.0 |
4.3 |
|
7 |
4.4 |
4.6 |
4.5 |
|
8 |
3.8 |
4.6 |
4.2 |
|
9 |
3.4 |
4.6 |
4.0 |
|
10 |
1.2 |
1.0 |
1.1 |
|
11 |
3.2 |
3.8 |
3.5 |
|
12 |
3.2 |
2.8 |
3.0 |
Table 6. Response Data, Voice
6 (Control-- Newscaster on NPR)
|
Question |
Female Average |
Male Average |
Total Average |
|
1 |
3.8 |
2.6 |
3.2 |
|
2 |
2.9 |
2.4 |
2.65 |
|
3 |
2.6 |
2.8 |
2.7 |
|
4 |
4.7 |
2.8 |
3.75 |
|
5 |
5.0 |
3.4 |
4.2 |
|
6 |
4.3 |
3.4 |
3.85 |
|
7 |
4.0 |
2.8 |
3.4 |
|
8 |
3.9 |
2.4 |
3.15 |
|
9 |
4.0 |
2.4 |
3.2 |
|
10 |
1.2 |
1.8 |
1.5 |
|
11 |
2.8 |
2.2 |
2.5 |
|
12 |
3.3 |
2.2 |
2.75 |
My Study—
Interpretations
The
first voice was that of a standard speaker, and the movie was a family movie (Ice Age). The results suggest that the subjects rated him as having mostly
average competence, especially when it comes to intelligence and
industriousness, though he was rated as being very confident (4.4/5.0). In terms of personal integrity, the speaker
received below-average marks (2.6/5.0 for honesty, sincerity, and
trustworthiness). His personality was
rated significantly higher, with average marks for honesty, but above-average
for humor and friendliness. There was
very little variation between male and female subjects.
These results seem to be somewhat in accordance with much of the literature that has been published on standard and nonstandard speech. The speaker is a standard speaker, though he speaks with clearly exaggerated features that become vocal variables (described in Methods section). Since he is a standard speaker and speaks with no obvious accents, we would expect him to receive the average to above average competence rating he received, based on past studies that have shown standard speakers as receiving higher competence ratings than nonstandard speakers (Giles 1972: 169-70; Ryan et al. 1982: 8; Edwards 1982: 23). Further in accordance with this is that his integrity was viewed as being somewhat below average. Much of the literature on language attitudes suggests that while standard speakers receive high competence marks, they tend to receive lower integrity and personality marks than nonstandard speakers (ibid.). This is, however, what makes the personality data somewhat confusing. According to the literature, as we have seen, standard speakers are generally seen as lacking in personality, such as humor and friendliness, but this speaker was rated quite highly for personality.
I think the way to reconcile this is, as I
suggested in my hypothesis, by considering the movie genre. This announcer was advertising a family
movie, so it comes as no surprise that the voice was rated as being humorous
and friendly. After all, the production
companies want to appeal to a wide audience, especially children who will come
with their parents, so it is vitally important that the announcer representing
the movie be somewhat jovial in order to appeal to the children. So, then, while this speaker is a standard
speaker, and conforms to many of the language attitudes suggested in the
literature, we must keep in mind the movie genre he was advertising.
The
results for voice 2 (E.T.: The 20th
Anniversary Edition), for one, show that he was seen as taller,
better-looking, and older than voice 1.
As Table 2 shows, the speaker’s competence was rated as very similar to
that of voice 1, significantly above average for intelligence (3.4/5.0) and
confidence (3.8/5.0), though only average for industriousness. The integrity and personality results were
only average, though the speaker was seen as slightly less funny than average.
Again, this speaker is a
standard speaker, so the higher-than-average competence is not surprising. While we might expect a standard speaker to
receive lower than average integrity and personality ratings, I think the fact
that this speaker’s ratings were average is testimony to the correlation between
movie genre and language attitudes. The
movie E.T. is a fantasy/sci-fi, but
it is also a family movie, and one of the most beloved family movies of all
time, to top it off. It should, then,
not surprise us that the producers would want a speaker whose personality
appeals to the audience, as this one does to some extent. But, further, this speaker’s speech is
highly exaggerated, as the Description of Voices suggests, so, while he is a
standard speaker, perhaps we are seeing some of the nonstandard speech language
attitudes here. As suggested earlier,
nonstandard speakers are usually seen as having better personalities and being
more trustworthy and honest than standard speakers. The speaker’s voice is far more breathy and wispy than most of
the speakers we hear on television and radio, so perhaps the exaggerated voice
features serve to make the speaker somewhat nonstandard to the listener, in
which case we would expect the average integrity and personality ratings.
Voice
3 was, I think, vastly different from the first two, and the data would seem to
support this. He was thought to be
definitely old, and the subjects remarks suggested things like “throaty,”
“deep,” and “sounds old.” The speaker’s
speech seemed far less standard than that of the first two speakers, which
might help explain why integrity, and personality were above average. But this speaker received one of the highest
ratings of any of the speakers for all three categories. He was thought to be highly intelligent and
confident, though only above average in industriousness, and very honest and
trustworthy, as well as warm. The only
characteristic that was below average was humor.
These results are somewhat difficult to interpret, since I do not think—and the subjects did not seem to think—that he was a particularly standard speaker. So, if we consider this only, we might conclude that his competence should have been rated far lower than in was, when in reality it was one of the highest competency ratings of any. This apparent aberration could be due, at least in large part, to the advanced age that was associated with the voice. Age is generally associated with things like wisdom, since older people have experienced more than younger people, so the high mark for wisdom comes as no surprise. Along these same lines, perhaps the hardworking characteristic that was associated with the speaker can also be attributed to age, to some degree, as older people have, of coursed, worked for many years, whereas a younger voice may be seen as lazy. The high rate for confidence is more difficult to interpret, as confidence is not usually associated with old age. The likely variable that evoked this sense in the subjects was the tone and pitch of the voice, being deep and low-pitched. Male voices are usually seen as being powerful and loud (Bradac 1990: 395), and the fact that this voice is very deep and “masculine,” as one subject noted, might suggest a good deal of confidence. The fact that the speaker was seen as having much integrity and a warm personality is probably a direct result of the somewhat nonstandard nature of his voice.
The fourth voice received scores for all categories that were largely slightly above average. The speaker’s intelligence, confidence, and trustworthiness, as well as his personal integrity, were about the same as those for the first two speakers, and, indeed, his voice was somewhat similar to those voices. The real surprise in the data for this speaker, though, was his personality/social attractiveness (questions 10-12), which was only in the average range. The movie he was advertising was Amelie, which is a light comedy, and I hypothesized that comedy announcers would elicit high personality marks, especially humor and friendliness, since these characteristics are essential to any comedy. But the data suggest otherwise. The speaker was not at all seen as funny—in fact, he was slightly below average in humor, and his friendliness was only slightly above average. Perhaps the average personality marks can be attributed, to some degree, to the fact that he is a standard speaker, and thus we might expect low personality marks. And, in fact, one of the subjects even commented that the speaker was “smug,” and another that he was not as friendly, both of which are usually associated with standard speakers.
The data on voice 5, like those for voice 3, were very difficult to interpret. Again, as in voice 3, we see that the speaker was seen as old—in fact, every subject recorded 5/5 for age. And, like voice 3, the speaker was rated as being highly-competent and highly sincere. But, unlike voice 3, this speaker received much lower personality marks. In particular, he was seen as being very serious (1.1, where 1.0 was most serious) and of average kindness and friendliness. Like speaker 3, this speaker would probably not be seen as a standard speaker by American listeners, primarily because he speaks with a British accent. Thus, we might attribute his exceptionally high competence to two things. The first is the old age the subjects attributed to him. As I noted for voice 3, age is frequently associated with intelligence and industriousness, so this could account for the high ratings for those two characteristics. The high confidence rating could be due to the deep, masculine voice which many subjects noted, since masculinity is frequently seen as powerful, and thus more confident.
But I imagine this confidence can also be correlated with the movie genre, which is, I think, the second factor giving him such a high competency rating. The Lord of the Rings is a fantasy film, and I hypothesized that the announcers for these sorts of films would tend to be seen as very serious and competent. My reasoning for thinking this was that fantasy films are, of course, fictitious, so no audience has any actual experience or knowledge of the worlds or characters depicted in these movies. The announcer, on the other hand, is the first voice we hear representing the film, and is thus a guide, in many ways, into that fantasy world. It is not surprising that the production companies would want a wise, confident person to “guide” the audience to the movie. The high competence and seriousness, then, do not surprise me, as they can perhaps be attributed to the fact that this is a fantasy film. Nor do the low personality ratings surprise me. As a fantasy film, the producers want to give the film an aura of mystery, not friendliness and warmth, so the average ratings in those categories are to be expected.
What I do find interesting, though, is the high integrity rating—the highest of any of the speakers. Some of this integrity might be attributable to the fact that most of the subjects thought the speaker was old, which might, as in the case of voice 3, suggest integrity. But more likely is that the factor that most greatly influenced the subjects to rate the speaker as very integritive is the fact that he is British. A British speaker is, of course, nonstandard for Americans, thus we might expect such a speaker to be seen as particularly honest and trustworthy, since numerous studies have suggested that nonstandard speakers tend to elicit these traits (Giles 1972: 169-70; Ryan et al. 1982: 8; Edwards 1982: 23).
Voice six was the control voice, an NPR newscaster, and thus the speaker spoke in relatively normal standard American English, with few noticeable vocal variables. The data from this voice support very strongly the conclusions that previous studies have come to about standard and nonstandard speakers. The speaker’s competence was seen as being very high, while his integrity was average, and his personality was quite low, all of which are generally attributed to standard speakers. In particular, the speaker was seen as very intelligent, confident, and hard-working, while he was seen as very serious (1.5; 1.0 max seriousness) and below average in kindness and friendliness.
The results of this study suggest that my hypothesis was mostly correct. I expected standard and nonstandard speech to have some effect on the language attitudes of the subjects, and, as we have seen, they did. Particularly interesting, I think, is the fact that even though all of these speakers speak very good English (they are, after all, paid to speak), their vocal features are often exaggerated, and those whose language was most exaggerated seemed to assume the role of nonstandard, especially in relation to the more standard speakers among them. The best example of this was voices 3 and 5, the first of which was a deep, cavernous voice, similar to that of James Earl Jones, and the second was a British speaker. In the case of voice 3, the voice was so exaggerated that it sounded quite nonstandard, which helps account for the speaker’s high personality and integrity ratings. In the case of voice 5, the speaker was British, and was thus a nonstandard speaker to an American, even though he may have been a standard British speaker. This “nonstandardness” helped elevate the speaker’s integrity for the subjects listening to his voice. Voices 1, 2, and 4, on the other hand, were exaggerated, but were far more standard than voices 3 and 5, and the fact that all three were perceived as being more competent than integritive or personable is proof of this.
The second aspect of my hypothesis was that listener’s attitudes would also correlate to the genre of the movie. We must remember that movies are products that are meant to make money, so certainly filmmakers would want announcers who will help sell their movie. It is not surprising, then, that the announcers for the movies directed at children or families were perceived as having the most pleasant personalities, or that the announcer for the fantasy film was perceived as serious. The announcer needs to have the same general personality as the movie, so these announcers are probably especially chosen, or modify their voices, depending on the type of movie being advertised.
Having said that, my hypothesis is not perfect, as evinced, for example, by the fact that the comedy announcer was not seen as humorous, something I thought would have been essential to any comedy announcer. And I thought that announcers for children’s/family movies would have received high personality and integrity marks but lower competence marks. But, as we saw, this was not the case. This suggests quite plainly that the factors that affect our language attitudes when it comes to pop culture such as movie announcers are varied and complicated. We cannot simply say that standard speakers will elicit one response and that nonstandard speakers will elicit another, or that comedy announcers will evoke one attitude, while sci-fi announcers will evoke another. But while the factors that influence us are many and not completely understood, as this study shows, clearly many of the attitudes that we do feel towards movie announcers are due, at least in part, to the standardness of the speaker and the genre of the movie.
Bradac, James J. “Language attitudes and impression
formation.” Handbook of Language and
Social Psychology. Ed. Howard Giles and W. Peter Robinson. Chichester, England: John
Wiley & Sons. 387-412.
Edwards, John R. “Language attitudes and their implications among English speakers: Social and
Applied Contexts.” Attitudes towards Language Variation. Ed. Ellen Bouchard Ryan and Howard Giles. London: Edward Arnold, 1982. 20-33.
Giles, Howard. “Evaluation of personality content from accented speech as a function of listeners’
social attitudes.” Perceptual and Motor Skills 34 (1972) : 168-70.
Romaine, Suzanne. “Stylistic variation and evaluative reactions to speech: Problems in the
investigation of linguistic
attitudes in Scotland.” Language and
Speech 23 (1980) : 213-31.
Ryan, Ellen Bouchard, Howard Giles, and Richard J. Sebastian. “An integrative perspective for the study
of attitudes toward language variation.” Attitudes towards Language Variation. Ed. Ellen Bouchard Ryan and Howard Giles. London: Edward Arnold, 1982. 1-19.
Wober, J. Mallory. “Language and Television.” Handbook of Language and Social Psychology. Ed.
Howard Giles and W. Peter Robinson. Chichester, England: John Wiley & Sons, 1990. 561-82.
Voice Study
Please
listen to each recorded voice, and then answer the following questions:
1) Is the speaker tall or short?
Voice 1: Short 1
2 3 4 5 tall
Voice 2: Short 1
2 3 4 5 tall
Voice 3: Short 1
2 3 4 5 tall
Voice 4: Short 1
2 3 4 5 tall
Voice 5: Short 1
2 3 4 5 tall
Voice 6: Short 1
2 3 4 5 tall
2)
Is the speaker good-looking?
Voice 1: not
good-looking 1 2 3 4 5 good-looking
Voice 2: not
good-looking 1 2 3 4 5 good-looking
Voice 3: not
good-looking 1 2 3 4 5 good-looking
Voice 4: not
good-looking 1 2 3 4 5 good-looking
Voice 5: not
good-looking 1 2 3 4 5 good-looking
Voice 6: not
good-looking 1 2 3 4 5 good-looking
3)
How old is the speaker?
Voice 1: definitely
young 1 2 3 4 5 definitely
old
Voice 2: definitely
young 1 2 3 4 5 definitely
old
Voice 3: definitely
young 1 2 3 4 5 definitely
old
Voice 4: definitely
young 1 2 3 4 5 definitely
old
Voice 5: definitely
young 1 2 3 4 5 definitely
old
Voice 6: definitely
young 1 2 3 4 5 definitely
old
4)
Is the speaker intelligent?
Voice 1: not
intelligent 1 2 3 4 5 intelligent
Voice 2: not
intelligent 1 2 3 4 5 intelligent
Voice 3: not
intelligent 1 2 3 4 5 intelligent
Voice 4: not
intelligent 1 2 3 4 5 intelligent
Voice 5: not
intelligent 1 2 3 4 5 intelligent
Voice 6: not
intelligent 1 2 3 4 5 intelligent
5)
Is the speaker self-confident?
Voice 1: not
confident 1 2 3 4 5 confident
Voice 2: not
confident 1 2 3 4 5 confident
Voice 3: not
confident 1 2 3 4 5 confident
Voice 4: not
confident 1 2 3 4 5 confident
Voice 5: not
confident 1 2 3 4 5 confident
Voice 6: not
confident 1 2 3 4 5 confident
6)
Is the speaker a hard worker?
Voice 1: not
a hard worker 1 2 3 4 5 hard worker
Voice 2: not
a hard worker 1 2 3 4 5 hard worker
Voice 3: not
a hard worker 1 2 3 4 5 hard worker
Voice 4: not
a hard worker 1 2 3 4 5 hard worker
Voice 5: not
a hard worker 1 2 3 4 5 hard worker
Voice 6: not
a hard worker 1 2 3 4 5 hard worker
7)
Is the speaker honest?
Voice 1: not
honest 1 2 3 4 5 honest
Voice 2: not
honest 1 2 3 4 5 honest
Voice 3: not
honest 1 2 3 4 5 honest
Voice 4: not
honest 1 2 3 4 5 honest
Voice 5: not
honest 1 2 3 4 5 honest
Voice 6: not
honest 1 2 3 4 5 honest
8)
Is the speaker sincere?
Voice 1: not
sincere 1 2 3 4 5 sincere
Voice 2: not
sincere 1 2 3 4 5 sincere
Voice 3: not
sincere 1 2 3 4 5 sincere
Voice 4: not
sincere 1 2 3 4 5 sincere
Voice 5: not
sincere 1 2 3 4 5 sincere
Voice 6: not
sincere 1 2 3 4 5 sincere
9)
Is the speaker trustworthy?
Voice 1: not
trustworthy 1 2 3 4 5 trustworthy
Voice 2: not
trustworthy 1 2 3 4 5 trustworthy
Voice 3: not
trustworthy 1 2 3 4 5 trustworthy
Voice 4: not
trustworthy 1 2 3 4 5 trustworthy
Voice 5: not
trustworthy 1 2 3 4 5 trustworthy
Voice 6: not
trustworthy 1 2 3 4 5 trustworthy
10)
Is the speaker serious or funny?
Voice 1: Serious
1 2 3 4 5 funny
Voice 2: Serious
1 2 3 4 5 funny
Voice 3: Serious
1 2 3 4 5 funny
Voice 4: Serious
1 2 3 4 5 funny
Voice 5: Serious
1 2 3 4 5 funny
Voice 6: Serious
1 2 3 4 5 funny
11)
Is the speaker kind?
Voice 1: not
kind 1 2 3 4 5 kind
Voice 2: not
kind 1 2 3 4 5 kind
Voice 3: not
kind 1 2 3 4 5 kind
Voice 4: not
kind 1 2 3 4 5 kind
Voice 5: not
kind 1 2 3 4 5 kind
Voice 6: not
kind 1 2 3 4 5 kind
12) Is the speaker
friendly?
Voice 1: not friendly 1 2 3 4 5 friendly
Voice 2: not friendly 1 2 3 4 5 friendly
Voice 3: not friendly 1 2 3 4 5 friendly
Voice 4: not friendly 1 2 3 4 5 friendly
Voice 5: not friendly 1 2 3 4 5 friendly
Voice 6: not friendly 1 2 3 4 5 friendly
13) Describe, to the best of your ability, the speaker’s voice and manner of speaking (ex., nasal, deep, swallows “d”s, etc.).
Voice 1:_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________
Voice 2:_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________
Voice 3: _______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________
Voice 4: _______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________
Voice 5: _______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________
Voice 6: _______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________