Listeners’ preferences regarding the regional accent on formal and informal communication context

Abstracts

Purpose

to analyze the preferences of listeners as the regional accent and accent softened in the context of formal and informal communication.

Methods

three TV news presenter recorded vehicle-phrases in situations of regional and soft accent. The recordings were presented to 105 judges, who heard pairs of words and asked which of the two pronunciations preferred to talk about television news presenters (formal context), for native speakers of the local community (informal context) and to the speech itself (context informal).

Results

the presence of listeners preferred the understated accent in a formal context (presentation of television news in all language variants studied (p <0.0001) and, on the other hand, preferred the presence of regional accents (p <0.0001) in informal context. however, speaks for itself, within the informal context, there was a general preference for regional accent or understated, statistical significance only for palatalization of / S / in medial coda (p <0.0001) and not palatalization of dental (p <0.0001), both characteristics of regional accents, and the non-occurrence of monophthongization (p <0.0001) and matching vowel (p <0.0001), characterized as understated accent.

Conclusion

listeners prefer to speaks with an accent softened in a formal communication but prefer the regional accent within an informal context, especially in less educated speakers.

Speech; Language and Hearing Sciences; Speech; Speech Perception; Television

Objetivo

analisar as preferências dos ouvintes quanto ao sotaque regional e sotaque suavizado em contexto formal e informal de comunicação.

Métodos

três telejornalistas gravaram frases-veículo nas situações de sotaque regional e suavizado. As gravações foram apresentadas a 105 juízes, que escutaram os pares de palavras e responderam qual das duas pronúncias preferiam para a fala de apresentadores de telejornal (contexto formal), para falantes nativos da comunidade local (contexto informal) e para a própria fala (contexto informal).

Resultados

os ouvintes preferiram a presença de sotaque suavizado em contexto formal (apresentação de telejornal em todas as variantes linguísticas estudadas (p<0,0001) e, por outro lado, preferiram a presença de sotaque regional (p<0,0001) em contexto informal. Porém, para a própria fala, dentro do contexto informal, não houve uma preferência geral pelo sotaque regional ou suavizado, havendo significância estatística apenas para palatalização do /S/ em coda medial (p<0,0001) e não palatalização das dentais (p<0,0001), ambas características do sotaque regional, e a não ocorrência de monotongação (p<0,0001) e harmonização vocálica (p<0,0001), caracterizados como sotaque suavizado.

Conclusão

os ouvintes preferem à fala com sotaque suavizado em um contexto formal de comunicação, mas preferem o sotaque regional dentro de um contexto informal, principalmente em falantes menos escolarizados.

Fonoaudiologia; Fala; Percepção da Fala; Televisão

INTRODUCTION

The accent is a recurring theme in speech therapy to improve television journalists’ oral communication, since it has been considered a communicative noise for so long in the transmission of news based on the model of Information Theory¹. It is believed that the softening accent is related to the rise of these professionals, there being recovery of specific language variants and rejection of other pronunciation forms.

The practice of speech accent smoothing in the context of television journalism is still very empirical, based on the identification of accent marks present in a given subject, both in terms of pronunciation as prosody, followed by the manipulation of such parameters, seeking a less “marked” style of locution. On the other hand, little is known about the viewer’s perception, and how the presence of different regional characteristics in the speech of reporters and announcers is evaluated.

There are studies showing that listeners even without prior training, can identify the accent of the speaker from short passages of speech, and may even indicate the region of origin and other social categories (occupation, educational and economic levels), although they have more ability to recognize variants used in their area and surrounding regions ^2-12.

This capability reflects the fact that listeners would have mental representations of the linguistic variants and the different categories associated with them. The choices are not random, but based on cognitive categories for dialectal variation, being able to say especially if the accent belongs or not to a speaker of its region ^3,4,8,9,13.

Studies of speech perception and variation made in the last ten years have sought to understand how lay listeners process and interpret the linguistic variation, coming up to two more general conclusions: that little is yet known about how the listener perceives the change and; that people seem to use their perception of dialect to categorize and assign values to speakers ^3,4,13-21. However, the real challenge in this research is to understand the extent to which different values (positive or negative) are assigned to linguistic variants in various communication styles and contexts, since it is believed that the variation in the language carries a social significance, causing different reactions in the listener ^22,23.

Judgments of evaluative character about the pronunciation of speech sounds are common and part of people’s everyday life, always coming to the fore in most varied and unusual environments and situations. They occur because the language use implies variation and hence allows certain choices, which in turn, result from cultural, dialect, social, psychological, political and pragmatic conditioning, influencing the aesthetic design and option²⁴.

In the context of television journalism, it can be inferred that the viewer makes judgments about the pattern of locution of local reporters, who may or may not present regional marks in speech, comparing it with the pattern propagated by newspapers and setting positive or negative judgment criteria, preferring some dialectal features in detriment of others. This judgment is based on the listeners’ expectations about the different speakers in various communication situations, either professional (formal) or colloquial (informal).

Accordingly, the objective of this research is to analyze the preferences of listeners regarding the accent of their region in a state of formal (television newscast) and informal communication.

METHODS

This study was approved by the Ethics Committee in Research of the Center for Health Sciences, Federal University of Paraíba under the opinion N° 17103. All volunteer participants signed an informed consent form. It is an explanatory, analytical and cross-sectional study.

Initially we selected the linguistic variants to be investigated from the studies conducted by the Project Linguistic Variation in Paraíba State (VALPB), which investigated the linguistic reality of the community in João Pessoa, tracing the linguistic profile of their speakers, including the variants: palatalization of the medial /S/ in medial coda succeeded of dental occlusion, monophthongization, vowel harmonization, palatalization of dental occlusion, assimilation of dental occlusion and weakening of /R/ in medial coda.

Later, from journalistic texts from television were selected the words that represent each of these variables. All words selected were inserted into vehicle-phrases such as “Digo_______baixinho”. This procedure aimed at generating speech samples inserted into similar phonetic-phonological contexts.

Three announcers, native journalists from television recorded the vehicle- phrases in situations of regional accent (RA) and the occurrence of softened accent (SA) regarding the linguistic variables studied. Considering that the recordings were made by three announcers, each word may have occurrence of up to three times, as shown in Figure 1.

Figure 1
Linguistic Variables and their occurrence in the evaluation of speech preference

During recording, the prosodic aspect was controlled mainly with respect to the intonation curve and speech rate, since the purpose was only the analysis of segmental variation. Thus, we tried to prevent that when characterizing the speeches with RS or SA, the announcers performed significant differences in the prosodic aspect and, therefore, the judgment of the listeners were guided by these clues.

Therefore, the announcers were instructed on these issues, being conducted a brief training, including vocal exercises and direct instruction on how to record the text and phrases. Thus, a more even intonation curve, with lower increase of F₀ in non-final stressed syllables, and less difference of F₀ between pre-stressed and stressed final, and post-stressed syllables.

To continue the research, speech excerpts were edited in the Sound Forge software, version 10.0. The target words were cut, preserving all phonemes, pairing up in one audio file according to the announcer and linguistic variable studied. For the latter condition, the pairing was performed in a random sequence of RA and SA patterns. In addition, five pairs of equal words were inserted, either in the RA or SA condition, so-called distractors.

Each file was saved as audio track and arranged randomly for subsequent presentation to the listeners. The words were used to assess the preference of speech.

For validating the words used by listeners for judgment, the audio files were presented to four speech therapists with experience in speech evaluation. Initially, they listened to each word pair (RA vs. SA) and checked whether or not identified differences in pronunciation and which of them corresponded to the RA and SA. For subsequent presentation to listeners, only the word pairs in which at least three evaluators’ perceived differences between the two forms of pronunciation were considered.

A Protocol for Assessment of Speech Preference was prepared to evaluate which of linguistic variants (RA vs. SA) was preferred to the speech of a television newscast announcer (formal situation), a speaker of the local community (informal situation) and the listener’s own speech (informal situation).

For the group of judge, 105 listeners from Joao Pessoa participated in the study, they are students of the Speech Therapy Course, Federal University of Paraíba between the 1^st and 6^th period, aged between 18 and 38 years, 24 male and 81 female, who had no hearing complaints that prevented listening to the audio recorded material. The participation of the listeners was restricted to listening to speech excerpts and completing the protocol for preference speech evaluation.

The word pairs were presented to a group of listeners, using notebook and speakers at intensity referred as sufficient and comfortable by the listeners, being repeated twice. It was requested that after listening to each pair, the judges fill out the protocol for evaluation of speech preference, identifying the preferred pronunciation for television newscast announcer for a speaker from the local community and for the listener himself.

For the data analysis of the preference between RA and SA for the three situations created, tests were performed for proportions, checking whether there were differences between the listeners’ responses.

The significance level was 5% for all analyses. The software used was R, which is free and the most used by the statistical community.

RESULTS

In this study, listeners preferred a speech without regional accent features for television newscast presenter (formal situation) and with occurrence of regional characteristics for speakers of their local community (informal situation), both in general (p <0.0001), as for each linguistic variable studied (p <0.0001) (Table 1). On the other hand, they did not show a general preference for regional accent or understated in the speech itself (informal situation), preferring only the non-occurrence of regional characteristics for the variables monophthongization (p <0.0001) and vowel harmonization (p <0.0001) and occurrence of regional accent for variables palatalization of /S/ in medial coda (p <0.0001) and dental palatalization (p <0.0001) (Table 1).

Table 1
Preference of listeners regarding the presence of regional or softened accent in the announcer’s speech of television newscast, local community and in the speech itself to the different linguistic variables

	PREFERENCE FOR ANNOUNCER					PREFERENCE FOR LOCAL COMMUNITY					PREFERENCE FOR LISTENER
VARIABLE	RA		SA		p value	RA		SA		p value	RA		SA		p value
	n	%	n	%		n	%	n	%		n	%	n	%
General perception	947	12.51	6625	87.49	0.0000*	6427	78.34	1777	21.66	0.0000*	4016	48.94	4190	51.06	0.0562
Palat /S/ in MC	99	23.52	322	76.48	0.0000*	344	68.80	156	31.20	0.0000*	321	64.20	179	35.80	0.0000*
Monophthongization	76	4.48	1621	95.52	0.0000*	1333	76.21	416	23.79	0.0000*	696	39.77	1054	60.23	0.0000*
Vowel harmonization	108	4.77	2155	95.23	0.0000*	1831	77.95	518	22.05	0.0000*	894	38.06	1455	61.94	0.0000*
Dental palat.	438	28.66	1090	71.34	0.0000*	1537	85.15	268	14.85	0.0000*	1227	67.94	579	32.06	0.0000*
Dental assimilation	28	4.99	533	95.01	0.0000*	480	82.47	102	17.53	0.0000*	222	38.14	360	61.86	0.8643
WEAKEN /R/ in MC	30	16.67	150	83.33	0.0000*	142	69.27	63	30.73	0.0000*	106	51.71	99	48.29	0.6752

• Caption: RA = regional accent, SA = softened accent; Palat = palatalization, MC = medial coda; WEAKEN = weakening; n = number of linguistic variants preferred by listeners
• * significant values p <0.05 - Test for proportions

In the protocol Preference for Speech besides the possibility of marking a preference between regional accent or understated, there was the “indifferent” option if the listener judged there being no preference between the speech manners. However, the class “indifferent” was removed from the statistical test among the ratios due to their low frequency, not impairing the final result. Thus, the withdrawal of the class “indifferent” justifies the difference in the sum of “n” of SA and RA regarding the announcer’s speech of the local community and the listener himself.

DISCUSSION

The fact that listeners express preferences for certain features of pronunciation, either with regional or softened accent, indicates that they have expectations of certain variants used by the speaker¹⁵, and respond to a speech stimulus based on a reference that is stored in their memories for a particular speaker in a specific style, comparing it with the expected pattern for this style.

This expectation is built with the listener exposure over the years to the pattern of television speech, which contributed to the formation of a stereotype for the speech of these professionals^22,25,26. The fact of preferring the speech without regional accent for the television announcer indicates that the softened accent is one of the markers of that particular style of speech.

On the other hand, this is a historical process, in the sense that the smoothing (and sometimes, neutralization) of accent was highly valued for the speeches of television announcers, which spread a pattern of TV narration free or mitigated regarding the regional marks. Consequently, the local TV newscast has adopted these same “rules” for their reporters and announcers.

The use of a regional softened speech became part of the construction style of the oral communication of reporters and announcers, carrying a meaning capable of categorizing the group and style, being recognized by listeners as such.

One of the important conclusions is that this is a feedback system, considering that the choice of a particular variant delineates and disseminates a specific speech style and on the other hand, the listener creates expectations for this speech style, stimulating maintaining these characteristics over time.

In summary, the listener may have made the choice for the non-occurrence of regional speech characteristics to the announcer, because he believes that these variants are stigmatized for a more formal speaking style, such as the presentation of a TV newscast, or, simply because they have expectations for this speech style, which may involve the non-occurrence of these regional characteristics.

Listeners preferred speech of their community with the characteristics of regional accent in general and for all linguistic variables (Table 1), indicating that they can correctly identify the characteristics of their group in regional terms (geographic).

However, comparing the preference for speech to television news announcer and the speakers from the local community, it is observed that the choice for announcers was opposed to the characteristics of regional speech for all variables.

Regarding the self-assessment, the judges chose speech with softened accent only for monophthongization and vowel harmonization variants, and regional characteristics for the palatalization of /s/ in coda and non-palatalization of dental, absence of a general preference for regional or softened accent (Table 1).

The comparison between the choice of listeners regarding the preferred linguistic variants for speech announcers, the local community and the speech itself, allows us to infer that, indeed, the variation may have a direct relationship with the speaking style expected for the group and the accommodation against this expectation.

When the listener evaluates its speech community in regional terms, perhaps he puts into action mechanisms related to the linguistic stereotype²⁵ for the local community, as well as the information recorded in the memory. Thus, when judging the local community as a group, to a situation of informal speech (as directed during collection), he may consider that regional variants have prestige in the local community in an informal context, and that this same variant is stigmatized in a more formal situation.

On the other hand, considering that the group of judges was made up of college students, when they judged the speech itself, they may have taken as reference to a distinct group, their community of practice while students of the third grade, with higher level of education, in which some regional variations can be considered stigmatized.

In general, speakers with more years of education tend to use standardized forms and those with most prestigious. They tend to focus on changes that implement socially acceptable ways, disadvantaging those opposed to the standard form. Thus, the data analysis may indicate that the monophthongization and vowel harmonization processes are less socially accepted, at least in the local community and the palatalization of /S/ in medial coda, along with the non-occurrence of dental palatalization are more acceptable and less stigmatized forms.

Regarding the monophthongization, one study showed that the variable “years of education” was the one that most influenced the execution of diphthongs [aj] and [ej]. The diphthong [ow] occurred regardless of social variables. Thus, the choice of listeners regarding the non-occurrence of this process in the speech itself may be related to the fact that listeners are college students, with a greater number of years of education²⁷.

Regarding the choice of non-occurrence of this process for speech tv announcers, another study that examined the monophthongization of [ow] realized that despite the steady reduction in spontaneous speech, this diphthong can keep in more formal situations, especially when words of lesser frequency in the language are used²⁸.

The identification of palatalization of / S / in medial coda and non-occurrence of dental palatalization as regional marksd in the speech itself, can also lead to questioning that these variants emerge as the speech characteristics most evident in the local community.

A fact confirming it is that these processes have fewer occurrences in the rest of the country, constituting probably in a border marks of the dialect in João Pessoa. In João Pessoa (local community), the speakers only palatalize the [s] in medial coda, when succeeded by dental occlusion ([t] and [d]), unlike for example the Recife, a nearby town, where palatalization of [s] occurs in medial coda, when succeeded by any occlusion, as well as in final coda.

Regarding the non-occurrence of dental palatalization process before the vowel [i], it is also considered that there is a lower number of communities where this phenomenon is found in Brazil. In the very northeast, the speakers of capitals such as Fortaleza, São Luís, Teresina, Salvador and Sergipe perform this process in the speech.

On the other hand, monophthongization and vowel harmonization are processes likely to occur in most regions of the country, seeming not to be associated, separately, to speakers of specific regions.

Thus, perhaps the palatalization of [s] in medial coda only when succeeded by dental occlusion and the execution of [t] and [d] as affricates ([tʃ] and [dʒ]) are the most contrastive features of the dialect of João Pessoa compared to the speech of other places.

Therefore, it is understood that the occurrence of monophthongization and vowel harmonization relate to the stigmatization of these variables in the regional dialect, besides being less valued in more formal contexts of communication, while the non-occurrence of palatalization of /S/ in the medial coda and dental palatalization are more linked to a stylistic issue, being more valued only for the more formal speaking style. The non-occurrence of the weakening of /R/ in medial coda seems to have a relationship with the style, and the occurrence of dental assimilation seems to be related to the stigmatization of this variant.

The type of methodology used in this study, analyzing the association among surveys of speech preference to a style, comparing them with the judgment of the listeners for their own speech and speakers for their region allows to differentiate the linguistic variants related to a style of which are stigmatized.

Conducting research using the judgment of listeners brings evidence on the parameters that can be worked on individuals in the context of professional communication to achieve certain effects of meaning in specific styles. In terms of dialectal variation (accent), there are few studies in the field of speech therapy, especially in the context of communicative competence and the development of individuals who use communication professionally.

Data from this study subsidize the speech therapy with voice professionals, once it provides parameters from the public receptivity regarding the regional marks in speech and, from that, it becomes possible for the audiologist to elaborate strategies to improve the communication sensitive to style needs of professionals and linguistic preferences noted by listeners/judges.

The work on softened accent in television journalism is focused on the professional development of these individuals, in the sense that, arguably, the viewer expects certain speech characteristics for this style. Thus, the work of the speech therapist, accented with this audience, when taking into account the social aspects and language variation, is oriented to the development of style and market integration, as well as the speaker, who has expectations about this speech.

CONCLUSIONS

Listeners prefer the non-occurrence of regional characteristics of the announcer’s speech and, on the other hand, the occurrence of these marks in the speech of their community. However, they point out that in their own speech there is a general preference for regional or softened accent, even in a situation of informal communication.

REFERÊNCIAS

• ¹
  Shannon CF, Weaver W. The mathematical theory of communication. Urbana: The University of Illinois Press; 1949.
• ²
  Pisoni DB. Long-term memory in speech perception: some new findings on talker variability, speaking rate, and perceptual learning. Speech Commun. 1993; 13:109-25.
• ³
  Clopper CG, Pisoni DB. Effects of talker variability on perceptual learning of dialects. Lang Speech. 2004;47(3):207-39.
• ⁴
  Clopper CG, Pisoni DB. Some acoustic cues for the perceptual categorization of American English regional dialects. J Phon. 2004;32:111-40.
• ⁵
  Kretzschmar WA. Language variation and complex systems. Am Speech. 2010;85(3):263-86.
• ⁶
  Clopper CG, Bradlow AR. Free classification of American English dialects by native and non-native listeners. J Phon. 2009 37:436-51.
• ⁷
  Evans BG, Iverson P. Vowel normalization for accent: An investigation of best exemplar locations in northerm and southerm British English sentences. J Acoust Soc Am. 2004;115:352-61.
• ⁸
  Tamasi SL. Cognitive patterns of linguistic perceptions [Doctoral dissertation]. Athens: University of Georgia; 2003.
• ⁹
  Williams A, Garret P, Coupland N. Dialect recognition. In: Preston DR. Handbook of perceptual dialectology. Philadelphia: John Benjamins, 1999. p. 369-83.
• ¹⁰
  Remez RE, Fellowes JM, Rubin PE. Talker identification based on phonetic information. J Exp Psychol Hum Percept Perform. 1997;23:651-66.
• ¹¹
  Preston DR. Perceptual dialectology: nonlinguistics views of areal linguistics. Providence: Foris; 1989.
• ¹²
  Preston DR. Five visions of America. Lang Soc. 1986;15:221-40.
• ¹³
  Campbell-Kibler K. Accent, (ing), and the social logic of listeners perceptions. Am Speech. 2007;82(1):32-64.
• ¹⁴
  Eckert P. Variation, convention, and social meaning. Annual Meeting of the Linguistic Society of America; 2005 Jan 6-9; Oakland, CA. Washington: John Benjamins; 2005.
• ¹⁵
  Bell A. Language style as audience design. Lang Soc. 1984;13:145-204.
• ¹⁶
  Eckert P, Rickford J. Style and Sociolinguistic Variation. Cambridge: Standford University, Cambridge Press; 2002.
• ¹⁷
  Foulkes P, Docherty G. The social life of phonetics and phonology. J Phon. 2006;34:409-38.
• ¹⁸
  Oliva MAA, Serrano MJ. Towards a comprehensive view of variation in language: the absolute variable. Lang Commun. In press, 2011.
• ¹⁹
  Campbell-Kibler K. The nature of sociolinguistic perception. Lang Var Change. 2009;21:135-56.
• ²⁰
  Clopper CG, Bradlow AR. Perception of dialect variation in noise: intelligibility and classification. Lang Speech. 2008;51(3):175-98.
• ²¹
  Fiske ST, Neuberg SL. A continuum of impression formation, from category-based to individuating processes: Influences of information and motivation on attention and interpretation. Adv Exp Soc Psychol. 1990;23:1-74.
• ²²
  Edwards J. Refining our understanding of language attitudes. J Lang Soc Psychol. 1999;18(1):101-10.
• ²³
  Garret P. Attitudes to language. Cambridge: Cambridge University Press; 2010.
• ²⁴
  Ferrer RC, Sankoff D. Identity as the primary determinant of language choice in Valencia. J Socioling. 2003;7(1):50-64.
• ²⁵
  Clopper CG, Pisoni DB. Perception of dialect variation: some implications for current ressearch and theory in speech perception. Indiana: Indiana University; 2002. Research on Spoken Language Processing; p. 271-89.
• ²⁶
  Niedzielski N. The effect of social information on the perception of sociolinguistic variables. J Lang Soc Psychol. 1999;18(1):62-85.
• ²⁷
  Battisti E, Hermans B. A palatalização das oclusivas alveolares: propriedades fixas e variáveis. Alfa. 2008;52(2):279-88.
• ²⁸
  Oliveira AM. Inserção e apagamento de [ w ] em posição de coda: uma análise pela geometria dos traços [dissertação]. Rio de Janeiro: Faculdade de Letras da Universidade Federal do Rio de Janeiro; 2006.

Mailing address: Leonardo Wanderley Lopes Departamento de Fonoaudiologia - Centro de Ciências da Saúde Cidade Universitária - Campus I - Bairro Castelo Branco João Pessoa – PB – Brasil CEP: 58051-900 E-mail: [e-mail]

Conflict of interest: non existent

Publication Dates

• Publication in this collection
  may-jun 2014

History

• Received
  25 Sept 2012
• Accepted
  27 Mar 2013