‘I am Latina’: Exploring how, and the extent to which, Sofia Vergara constructs her identity in different contexts.

Published: 2019/12/11 Number of words: 3841


‘The film and TV industries in the United States have played a crucial role in perpetuating distorted images and stereotypes of Hispanics.’ (Acosta-Belen, 1988: 100). Acosta-Belen blames the industries for the negative perpetration of Hispanic stereotypes, but in the past few years, Sofia Vergara, the highest paid TV actress in Hollywood, has received extensive personal criticism for her role in ‘Modern Family’ for encouraging degrading Latino stereotypes. The Colombian actress plays Gloria Pritchard, the Colombian wife of an older wealthy businessman, and has been criticised for over-sexualised behaviour and playing up to ideas that Latinos are aggressive, outspoken and unintelligent.

However, as Vergara herself said in response to her criticism, ‘Whoever said all Latin women are the same? It’s impossible to caricature millions of women from 20 countries. There are many kinds of Latinas, just like there are many types and styles of women all over the world.’ (Latin Times, 2013) Just because Vergara plays up to stereotypes in her role as Gloria does not mean to say that she truly believes in the stereotypes. It must be recognised that Vergara is creating an on-set persona – she is playing the role of a character. However, it is not just actors who create personas. In day-to-day life, all of us create our own identity. This is achieved largely through language use, which is unique to each individual.

This essay aims to explore the persona Sofia Vergara creates in interviews, how she does this through linguistic features, and the extent to which this identity is conveyed in different contexts.



Vergara creates an identity as a successful Latin woman; however, there are many qualities of this identity that are intrinsically linked, and the way in which they manifest themselves depends on the way in which Vergara constructs this persona. There are many different aspects of identity that are involved in creating a persona. Bucholtz and Hall (2005) argue that ‘identity is the product rather than the source of linguistics and other semiotic practices and therefore is a social and cultural rather than primarily internal psychological phenomenon’ (585). In their exploration of identity creation, they outline five main approaches to identity construction: Emergence, Positionality, Indexicality, Relationality, and Partialness. The most relevant approaches in the way Vergara constructs her identity are the Positionality and Indexicality principles. The Positionality principle argues that identity is created through exploring many different facets of identity simultaneously, for example gender, age and ethnicity, and how these combine to create an identity (591). With regards to Positionality, Vergara explores her ethnicity and femininity in the creation of her persona. The Indexicality principle is based on the idea that people use certain, marked linguistic features to construct an identity (593). This is relevant as Vergara constructs her ethnic identity through emphasis on her accent. Bucholtz and Hall argue that identity construction does not follow any one of these principles, but often a combination of all of them. The features of Vergara’s identity that are important to explore are those which contribute to the Latin Female – or Latina – stereotype. There are many aspects of this Latina stereotype but the one Vergara emphasises most is the female physique. The idea that Latinas are curvy, voluptuous and have sex appeal is something that is inherent in Western culture. For Vergara, this focus on her body often becomes a key feature of her identity, and thus her performance of a feminine identity is prominent in her language use. This essay will analyse her use of gendered language through comparison with the work of Deborah Tannen (1994) and Palomares (2004) which explore gendered language features. In particular, data will be analysed with consideration given to interruptions, politeness and gender salience. Indeed, this idea of gender salience is also important to consider in deconstructing Vergara’s identity. Gender salience is the importance of gendered identity in a given situation, compared with other identities (Palomares, 2004). However, exploring many different identities, using salience as a feature of analyses is incredibly important. Vergara’s identity is a combination of gender, ethnicity and cultural identity but the salience of each of these aspects is inconsistent in different contexts.

Resigi and Wodak (2001) claim that ‘From a social functional point of view, “race” is a social construction’ (2). Vergara constructs a Latin ethnic identity through her language, but what is important to consider is that her language use will be largely influenced by the fact that she is not a native English speaker; her first language is Spanish. In the data being analysed, we see limited instances of code-switching, which must be given attention, but are not a key feature of her language. However, Vergara does have a prominent Spanish accent. In order to analyse whether this emphasis on accent is part of her ethnic identity construction, quantitative analysis into features of her accent means it is important to ascertain whether her accent is more present or not in different situations. This will be done by looking at devoicing of final consonants and final consonant alteration, a feature of Spanish accents outlined by Michael Swan (2001). He notes that ‘Spanish allows only five (or six) word-final consonants: /θ/, /s/, /n/, /r/ and /l/; speakers may omit word-final consonants other than these, or alter them (for example, by turning /m/ to /n/ or /ŋ/).’ This consonant alteration is prominent in Vergara’s accent so instances of this will be recorded and analyse; however, the data is transcribed with this denoted in only a small section of each interview, as in order to transcribe the entire data phonetically to reflect Vergara’s accent would detract from other aspects of her language use such as, for example, lexical choice.


Goddard and Patterson (2000) note that ‘Language usage and how it is viewed are powerfully affected by context, including the setting and purpose of the interaction’ (98). This essay explores identity construction of a public figure in different interview contexts, and thus the data chosen is from publicly available media sources. It has been transcribed to reflect the interactional and linguistic features of Vergara’s speech. The first interview is an on-set interview in which Vergara is asked a question which she answers and elaborates on, bringing in information that she saw relevant to the question as opposed to simply answering the question. The limitations of an interview of this nature lie in ascertaining how spontaneous Vergara’s speech is, as it could be done in multiple takes with rehearsals and re-recordings. Despite this, the interview is still useful due to the content. She is reflecting on her role as an actress, thus it is likely that different aspects of her identity will be more prominent in this type of identity compared with the other interview selected for analyses. The second interview is a recording of an interview on ‘The Jimmy Kimmel Live’ show, which is filmed in front of an audience and features conversation between Kimmel and Vergara. This interview is more revealing of Vergara’s linguistic features as the speech is spontaneous, regardless of whether she knew what the content of the interview would be beforehand or not. This data has been chosen in contrast to the first interview as it is a different situation in that there are more people present, and as it is live, Vergara is likely to be more conscious of how she wants to present herself. This data does unfortunately only give us insight into the persona Vergara constructs in interview situations, but this is important as these are media resources that are available to the public, so rather than a representation of her whole identity, which could not be assessed without a complete ethnography, it allows for exploration of the identity she performs to a wider audience and how she wants to be perceived in the public eye.

Data Analysis


As outlined in the methodology, this data analysis will focus on accent features, code switching and gendered language styles to determine the persona Sofia Vergara constructs. Perhaps most prominent in her speech is her Spanish accent, a key feature of which is the devoicing or alteration of final consonants. In Data Set 1, in Vergara’s first turn (turn 2), she uses ‘/greɪ/’ instead of ‘/greɪt/ (great). This loss of the final consonant is expected, as Michael Swan (2001) notes, there are only five or six final consonants that are voiced in Spanish, and /t/ is not one of them. However, later in Vergara’s turn she does voice the final consonant /t/ in ‘accent’. In Data Set 2, Vergara’s second turn features devoicing of the final consonant /d/ in ‘did’ and ‘hard’ and the final consonant /t/ in ‘hit’. This devoicing or alteration of consonants is a prominent feature throughout Vergara’s speech, although not consistent across the two data sets. In Data Set 1, out of 15 possible instances of consonant alteration or devoicing, that is words ending with a final consonant not found in Spanish, Vergara only alters 7 of these words, giving a consonant alteration rate of 46.7%, whereas in Data Set 2, out of a possible 24 alterations, Vergara conforms to 17 of these, giving a consonant alteration rate of 70.8%. Given that this is a very noticeable feature of Vergara’s Spanish accent, it is fair to draw the conclusion that her accent is more prominent in Data Set 2. It is particularly important to consider the context of the interviews here. Data Set 1 is the on-set interview with no audience or real interaction with the interviewer, whilst Data Set 2 is in front of a live audience with a conversation-style interview. As Vergara is not physically presenting herself to a large group of people, the persona she creates differs. In front of more people, her accent is more prominent, which is what we would expect if she were perpetuating a strong sense of ethnic identity. By intensifying her accent, which marks her as a Spanish speaker, she is increasing a key aspect of her Latin identity. What is particularly curious about these two data sets is that that Data Set 1 is taken from an interview in December 2011, whilst the interview from Data Set 2 was in January 2014. We would expect that the more time Vergara spends in the United States, the less prominent her accent would be, as she would be likely to pick up features of a North American accent. Instead, what has happened is that Vergara’s accent seems to have gotten more pronounced with time, suggesting she is reluctant to adopt North American accent features. Acosta-Belen (1988) argues that ‘Hispanics have a pattern of resistance to assimilation and a strong sense of ethnic consciousness’ (97) which may account for this phenomenon. However, Vergara’s strong accent, despite long residency in the US, could also be due to the fact that she has found success and popularity by perpetuating her Latin identity both in ‘Modern Family’ and reality, and it is therefore an identity she is comfortable with and proud of creating for a public audience. What is also troubling, and to an extent controversial, about the creation of a Latin identity is that people who consider themselves Latin hold a dual identity: Latino or Latina and their nationality. Acosta-Belen argues that ‘Spanish language has been used by census demographers to lump them all together as ‘Hispanics’, creating a false image of homogeneity and unity, and mystifying cultural, racial, and class differences and disparities among individual groups.’ (84.) Indeed, Vergara refers to herself as ‘Latin’ in turn 2 of Data Set 2 but also as ‘Colombian’ in turn 4. This is problematic, as it means that Hispanics in the United States are forced to pick between a national identity and cultural identity. The reason Vergara often refers to herself as Latin as opposed to Colombian could be because ‘The shorthand label is turning into a symbol of cultural affirmation and identity in an alienating society that traditionally has been hostile and prejudicial to cultural and racial differences.’ (Acosta-Belen, 1998:84.) Whilst Vergara may not experience racial hostility, she is still affirming herself as a member of cultural community as well as addressing her national identity.

Another linguistic feature which marks Vergara as a native Spanish speaker is her use of code-switching, that is, use of Spanish instead of English. We do not see this at all in Data Set 1, but there is one instance in Data Set 2 in Vergara’s first turn (turn 2) in which she talks over Kimmel in order to address the audience. She says ‘hola’ meaning ‘hello’. Whilst code-switching is not a primary feature of her language use, this instance is worth note as it is her first interaction with Kimmel and the audience, and she chooses to speak in Spanish. This suggests that her intentions are to confirm herself as Spanish speaking from the outset of the interview despite the fact that the rest of the interaction is conducted in English. This places further emphasis on her Latin identity. She distances herself from Kimmel and the audience by switching to a code that is not used or understood, particularly as she is in a monoglossic situation, yet it is fair to say that almost all of the people watching the interview would understand ‘hola’ as a greeting. Acosta-Belen (1988) attributes this understanding and acceptance of Spanish code switching to ‘the latinization of Anglo culture as the Hispanic presence becomes more perceptible.’ Thus in using ‘hola’ as her initial address to the audience, Vergara could be furthering this latinization of Anglo culture in the US by representing herself, a successful Hollywood actress, as Hispanic.
This is also worthy of note as it is an interruption, if we consider Deborah Tannen’s (1994) argument that ‘two speakers must act: One must begin speaking, and another must stop.’ (59.) She reports that ‘Research on gender and language has consistently found… females to be cooperative and more likely to avoid conflict.’ (40.) Avoiding interruptions is one such way of avoiding conflict, yet Vergara chooses to interrupt in the very opening of the interaction in Data Set 2. This opens the speech act to scrutiny with relation to gendered language behaviour, as Vergara does not conform to expected female language use here. However, Tannen (1994) emphasises that a lack of interruptions may be normal in Western female speech, but this is not true of all cultures. She notes particularly that a variety of ethnic females, including South Americans, are often viewed as aggressive as they use language styles typical of the Western male. However, that is not to say that Vergara is attempting to reflect male Western speech, but rather just adopting her normal style.

In Data Set 1, Vergara places emphasis on the words ‘Latin’ and ‘accent’ in turn 2. As she is talking about the way she developed her on-screen character in relation to her own identity, it is interesting that she emphasised ‘Latin’ and ‘accent’ – two features of her own identity which she clearly feels are important in the construction of a fictional character. By placing emphasis on aspects of her identity which relate to the fact that she herself is Spanish-speaking, at this point in the interview, her Latin identity becomes the focus. If we consider Palomares’ theory of gender salience, in the same line of argument it could be said that this is an instance of ethnic salience.

Returning to Palomares’ theory, considering the belief that ‘Gender schematic (sex-typed) individuals view themselves as prototypically gendered, prefer using gender-appropriate behaviours and actively avoid gender-inappropriate behaviours’ (563) is incredibly important in assessing her word choice and language use. In Data Set 1, Vergara says ‘I got to dress up sexy.’ The connotations of the word ‘sexy’ in this context are of an attractive woman, provocative clothing and open sexuality. These are all ideas of femininity that Vergara is portraying through this lexical choice in describing her own actions.

In Data Set 2, attention is brought to actions that emphasise this idea of femininity further. Kimmel mentions that Vergara posted pictures of herself in a bikini online. Vergara responds saying ‘at this age I don’t take for granted when I look good in a bikini.’ (Data Set 2, turn 16.) Her use of ‘at this age’ implies that it is not a social norm for older women to look attractive in a bikini and reference to this suggests she is proud of her body. This turn also sees a linguistic feature attributed to feminine language. She goes on to say ‘I want the world to know I’m still…’ but tails off. She does not explicitly state that she is still attractive, thus this is an instance of indirectness. Deborah Tannen (1994) says that indirectness is often seen as a way of women failing to make demands; however, ‘the ability to get one’s demands met without expressing them directly can be a sign of power rather than of the lack of it’ (32). Here it is clear that Vergara is aware that she is attractive and does not feel the need to state it explicitly. The ‘demand’ is that we acknowledge her as attractive, and the fact that she does not need to say it emphasises that she holds the power in this situation. However, the very fact that she has performed this act of indirectness with reference to her physical appearance makes it clear that she considers this to be important, and an area in which she holds power. This contributes to her feminine identity by placing emphasis on her body, a typical feature of the Latina stereotype outlined by Mendible (2007).

In Returning to Tannen’s (1994) idea that women aim to avoid conflict, we could say that avoiding conflict would be female gender-appropriate behaviour. In Data Set 2, we see an instance of potential conflict. Vergara hugs Jimmy Kimmel before the interview and seemingly knocks him across the chest. He opens the interview by saying ‘You have a very strong right cross.’ Vergara immediately identifies this as a source of potential conflict and apologises, saying ‘I’m so sorry, did I hit you hard?’ Her use of the intensifier ‘so’ is intended to increase the validity of her direct apology for potentially injuring Kimmel, which would not be considered feminine behaviour, and thus she has to right it through language in the form of an apology.



Sofia Vergara creates her identity through her language use. What is clear from the data analysis above is that Vergara portrays herself as a Latin woman, and takes into consideration the connotations and stereotypes that come with this identity construction. She flaunts her physique and openly talks about her body, which contributes to a feminine identity that takes pride in aesthetic appearance. As well as this, her accent has become more prominent over time, as she has gained recognition and success as a Colombian actress, thus she may be playing up noticeable features in order to create and sustain this identity as a Spanish-speaking Latin woman.

As a popular figure, Vergara’s perpetuation of a Latina stereotype is controversial. Her focus on her femininity could be seen as detrimental to an extent. Myra Mendible (2007) argues that ‘several forces converge in producing acculturated, gendered bodies and these forces have very real consequences for Latinas in the United States.’ (1) Whilst Vergara has found success in flaunting her physical appearance, perpetuating this stereotype that objectifies the bodies of Latin women, as a Latin women herself, this suggests that Vergara believes this stereotype is acceptable. However, as Mendible argues (2007), a stereotyped Latin female body ‘functions within a social and cultural taxonomy that registers but an echo of the clamor, complexity, and variety of women who embody Latina identities.’ (1) Vergara’s perpetuation of stereotypes has made her popular as an actress and a public figure, but it may have consequences for other Hispanic women seeking success in the United States if they cannot live up to the stereotypes that already exist and are being maintained by successful Latin women.



This project has analysed the language use of Sofia Vergara in two contexts, aiming to ascertain which aspects of her identity are most important to her, portrayed through her linguistic features. By focussing on particular features of her Spanish accent, quantitative analysis has shown her accent to be more pronounced in an on-set interview without a live audience or interaction, where she is more aware of what she is saying as opposed to focussing on the portrayal of an identity to a group of people. By looking at gender salience and features of gendered language, this essay has analysed instances of gendered behaviour such as politeness, interruptions and how this contributes to the feminine aspect of Vergara’s identity. It is impossible to pin-point Vergara’s identity without reference to existing stereotypes which could be seen as problematic; however, at times she does conform to these existing stereotypes, thus furthering them and maintaining them in a society where they can be detrimental.



Goddard, A. and Patterson, L. (2000). Language and Gender. London: Routledge

Reisigi, M. and Wodak, R. (2001). Discourse and Discrimination: Rhetorics of Racism and Antisemitism. London: Routledge.

Swan, M. (2001). Learner English: A Teacher’s Guide to Interference and Other Problems, Volume 1. Cambridge: Cambridge University Press.

Tannen, D. (1994). Gender and Discourse. Oxford: Oxford University Press

Edited Works

Acosta-Belen, E. and Sjostrom, B.R. (eds.) (1988). The Hispanic Experience in the United States: Contemporary Issues and Perspectives. Westport: Praeger Publishers.

Electronic Resources
Bucholtz, M. and Hall, K. (2005). ‘Identity and Interaction: a sociocultural linguistic approach’, Discourse Studies 7: 585-614.
http://www.sagepub.com/upm-data/40469_13a.pdf [Accessed 28th November 2014]

Mendible, Myra (2007). ‘From Bananas to Buttocks: The Latina Body in Popular Film and Culture’.
http://site.ebrary.com/lib/uon/reader.action?docID=10245645 [Accessed 17th November 2014]

Palomares, N. A. (2004). ‘Gender Schematicity, Gender Identity Salience, and Gender-Linked Language Use’, Human Communication Research 30: 556–588.
http://onlinelibrary.wiley.com/store/10.1111/j.1468-2958.2004.tb00745.x/asset/j.1468-2958.2004.tb00745.x.pdf?v=1&t=i2tfpxq8&s=066a4c626068419bbd3d0a8833c83e60f289e72f [Accessed 22nd November 2014]

Valdez, M.G. (2013) ‘Is Sofia Vergara ‘Modern Family’ Character a Stereotype?’. Latin Times.
http://www.latintimes.com/sofia-vergara-modern-family-character-stereotype-actress-defends-latina-gloria-critics-131623 [Accessed 23rd November 2014]



Cite this page

Choose cite format:
Online Chat Messenger Email
+44 800 520 0055